Skip to main content

Full text of "Physiology of the fungi"

See other formats


Marine  Biological  Laboratory  Library 

Woods  Hole,  Mas3achusetl5 


Physiology  of  the  Fungi 


McGRAW-HILL  PUBLICATIONS  IN 
THE  BOTANICAL  SCIENCES 

Edmund  W.  Sinnott,  Consulting  Editor 


ARNOLD     An  Introduction  to  Paleobotany 

CURTIS  AND  CLARK     An  Introduction  to  Plant  Physiology 

EAMES     Morphology  of  the  Angiosperms 

EAMES     Morphology  of  Vascular  Plants:  Lower  Groups 

EAMES  AND  MACDANiELS     An  Introduction  to  Plant  Anatomy 

HAUPT     An  Introduction  to  Botany 

HAUPT     Laboratory  Manual  of  Elementary  Botany 

HAUPT     Plant  Morphology 

HILL     Economic  Botany 

HILL,  OVERHOLTS,   POPP,  AND  GROVE       Botauy 

JOHANSEN     Plant  Microtechnique 

KRAMER     Plant  and  Soil  Water  Relationships 

KRAMER  AND  KOZLOWSKi     Physiology  of  Trecs 

LILLY  AND  BARNETT       PhvsiologV  of  the  Fungi 

MAHESHWARi     An  Introduction  to  the  Embryology  of  the  Angiosperms 

MILLER     Plant  Physiology 

POOL     Flowers  and  Flowering  Plants 

SHARP     Fundamentals  of  Cytology 

SINNOTT     Plant  Morphogenesis 

SINNOTT,  DUNN,  AND  DOBZHANSKY     Principles  of  Genetics 

SINNOTT  AND  WILSON     Botany:  Principles  and  Problems 

SMITH     Cryptogamic  Botany 

Vol.     I.     Algae  and  Fungi 

Vol.   II.      Bryophytes  and  Pteridophytes 
SMITH     The  Fresh-water  Algae  of  the  United  States 
SWINGLE     Textbook  of  Systematic  Botany 

WEAVER  AND  CLEMENTS       Plant  Ecology 

There  are  also  the  related  series  of  McGraw-Hill  Publications  in  the  Zoological 
Sciences,  of  which  E.  J.  Boell  is  Consulting  Editor,  and  in  the  Agricultural 
Sciences,  of  which  R.  A.  Brink  is  Consulting  Editor. 


Frontispiece.     Pilobolus  showing  phototropism.     See  page  339. 


L    -VI 


Physiology 
of  the  Fungi 


VIRGIL  GREENE  LILLY 

Professor  of  Physiology,  Department  of  Plant  Pathology 

and  Bacteriology,  West  Virginia  University; 

Physiologist,  West  Virginia  Agricultural 

Experiment  Station 


HORACE  L.  BARNETT 

Professor  of  Mycology,  Department  of  Plant  Pathology 

and  Bacteriology,  West  Virginia  University; 

Mycologist,  West  Virginia  Agricultural 

Experiment  Station 


1951 

McGRAW-HILL   BOOK   COMPANY 
New  York         Toronto        London 


PHYSIOLOGY    OF   THE    FUNGr 

Copyright,  1951,  by  the  McGraw-Hill  Book  Company,  Inc.  Prmtedinthe 
United  States  of  America.  All  rights  reserved.  This  book,  or  parts  thereof, 
may  not  be  reproduced  in  any  form  without  permission  of  the  pubUshers. 


7  89  10  11  12-MAMM-l  09  8 
37865 


This  book  is  dedicated  to  the  mem- 
ory of  Leon  H.  Leonian  and  to  Ernst  A. 
Bessey.  The  guidance  and  inspiration 
of  these  men  in  directing  our  interests 
to  the  study  of  fungi  is  gratefully 
acknowledged. 


PREFACE 

Living  fungi  are  being  studied  more  intensively  than  ever  before. 
This  may  be  attributed  in  part  to  increased  interest  in  the  potentiahties 
of  the  fungi  in  industry  as  well  as  to  the  greater  recognition  of  fungi  as 
important  disease-producing  agents  of  plants  and  animals  and  as  destroy- 
ers of  fabrics  and  other  cellulosic  materials  of  commercial  importance. 
This  has  increased  the  interest  in  the  cultivation  of  the  fungi  and  has 
shown  the  need  for  an  adequate  textbook  covering  the  broad  aspects  of 
physiology  of  the  fungi,  their  growth  requirements,  and  activities.  It 
was  the  intent  of  the  authors  to  prepare  a  textbook  which  would  fulfill 
the  needs  of  students  desirous  of  some  training  in  this  field. 

This  book  is  primarily  a  text  for  the  advanced  student  and  assumes 
some  basic  knowledge  of  the  morphology  of  fungi  and  of  organic  chemistry. 
It  had  its  origin  in  the  lectures  and  laboratory  exercises  used  for  three 
years  by  the  authors  in  a  course  in  physiology  of  the  fungi  offered  to 
graduate  students  at  West  Virginia  University.  The  authors  have 
contributed  equally  of  their  time  and  efforts  in  the  preparation  of  this 
text. 

For  those  who  are  interested  or  are  actively  engaged  in  physiological 
research  on  fungi,  this  textbook  may  serve  as  a  reference  book  and  as  an 
entry  into  the  literature.  The  large  ever-growing  accumulation  of  liter- 
ature  has  also  made  it  desirable  to  bring  together  a  summary  and  dis- 
cussion of  the  information  in  this  field.  However,  no  attempt  has  been 
made  at  complete  documentation  of  the  subjects  discussed.  Certain 
particularly  important  references  are  marked  with  a  star  and  are  recom- 
mended as  required  reading  for  students. 

For  the  most  part,  the  scientific  names  of  the  fungi  are  those  which 
were  used  by  the  investigators  whose  work  has  been  cited.  No  attempt 
has  been  made  to  reduce  these  names  to  synonomy.  Because  of  the 
close  relation  between  fungus  physiology  and  plant  pathology,  plant 
pathogenic  fungi  have  been  used  as  examples  whenever  possible. 

Several  suggested  laboratory  exercises  with  suggested  test  fungi  are 
included  at  the  end  of  the  text,  so  that  other  teachers  might  profit  by  the 
authors'  experience  in  designing  and  conducting  laboratory  work  in 
fungus  physiology. 

All  tables,  graphs,  and  photographs  not  credited  to  other  sources  are 
original. 

It  is  a  pleasure  to  acknowledge  our  indebtedness  to  the  many  individuals 

ix 


X  PREFACE 

who  have  aided  us  with  their  suggestions.  Among  our  colleagues  at 
West  Virginia  University  who  have  read  portions  of  the  manuscript  are: 
J.  G.  Leach,  C.  R.  Orton,  M.  E.  Gallegly,  R.  P.  True,  W.  T.  Jackson, 
B.  G.  Anderson,  R.  B.  Dustman,  and  J.  H.  Hare.  In  addition,  the 
following  have  read  one  or  more  chapters:  B.  W.  Henry,  J.  B.  Routien, 
W.  D.  Gray,  and  J.  B.  Conn.  We  are  also  indebted  to  Mrs.  H.  L. 
Barnett,  who  typed  most  of  the  manuscript. 

We  wish  to  thank  the  many  individuals,  societies,  and  publishers  who 
have  granted  us  permission  to  use  data  and  illustrations.  Particular 
acknowledgment  is  made  in  connection  with  the  material  cited. 

Virgil  Greene  Lilly 
Horace  L.  Barnett 

MORGANTOWN,  W.  Va, 

May,  1951 


CONTENTS 

Preface ix 

1.  Introduction 1 

Fungus  Physiology  in  Relation  to  Other  Sciences — Aims — Scope — Historical 
Development. 

2.  Culture  Media 8 

ICinds  of  Media — Natural  Versus  Synthetic  Media — Choice  and  Preparation 

of  Media — Ways  of  Expressing  Concentration — Comparison  of  Media — - 
Summary. 

3.  Growth 24 

Phases  of  Growth — Rate  of  Growth — Ways  of  Measuring  Growth — Methods 

of  Presenting  Results — Factors  Affecting  Growth — Effect  of  External 
Factors  on  Morphology — Summary. 

4.  Enzymes  and  Enzyme  Action 45 

Classification  of  Enzymes — Chemical  Nature  of  Enzymes — Factors  Affecting 
Enzyme  Activity — Mechanism  of  Enzyme  Action — Adaptive  Enzymes- 
Energy  and  Energy  Utilization  by  Fungi — Summary. 

5.  Essential  Metallic  Elements 65 

Biological  Essential  Elements — The  Essential  Macro  Elements — Essential 
Micro  Elements — Periodicity  of  Biologically  Essential  Elements — Summary. 

6.  The  Essential  Xonmetallic  Elements  Other  Than  Carbon 87 

Hydrogen — Oxygen — Sulfur — Phosphorus — Nitrogen — Other  Nonmetallic 
Elements — Summary. 

7.  Carbon  Sources  and  Carbon  Utilization 116 

Monosaccharides  and  Related  Compounds — Organic  Acids — Glycosides — 
Oligosaccharides — Polysaccharides — Heterotrophic  Utilization  of  Carbon 
Dioxide — Utilization  of  Carbon — Summary. 

8.  Hydrogen-ion  Concentration 149 

Ionization  of  Compounds — The  Meaning  of  pH — Buffers  and  Buffer  Capacity 

— Methods  of  Determining  pH  Values — Effects  on  Fungi — Summary. 

9.  Vitamins  and  Growth  Factors 171 

Part  I.     General  Considerations — Synthesis  of  Vitamins  by  Fungi — Vitamin 

xi 


xii  CONTENTS 

Deficiencies  in  Fungi — Inhibitory  Effects  of  Vitamins — Vitamers — Unidenti- 
fied Growth  Factors,  Part  II.  Specific  Vitamins — Thiamine  and  Its 
Moieties — Biotin — Inositol — Nicotinic  Acid — Pantothenic  Acid — Pyridoxine — 
p-Aminobenzoic  Acid — Riboflavin — Summary. 

10.  Fungi  AS  Test  Organisms 208 

General  Procedures — Vitamin  Assays — Amino-acid  Assays — Assays  for 
Essential  Elements — Sugars — Tests  for  Certain  Metabolic  Products — Testing 
Fabric  Protectants — Summary. 

11.  Metabolic  Antagonists 226 

Antivitamins — Amino-acid  Antagonists — Development  of  Fastness — Sum- 
mary. 

12.  The  Action  of  Fungicides 245 

Copper — Mercury — Sulfur — Organic  Fungicides — Evaluating  Fungicides — 
Summary. 

13.  Metabolic  Products 266 

Decomposition  of  Organic  Materials — Fungi  as  Food — Cultivation  of  Fungi 
for  Food — Fat  Production — Production  of  Vitamins — Enzyme  Production — 
Alcoholic  Fermentation — Organic  Acids — Esters — Antibiotics  and  Drugs — 
Toxins — Pigments — Summary. 

14.  Factors  Influencing  Sporulation  op  Fungi 304 

Environmental  Factors — Other  Physical  Factors — Nutritional  Factors — 
Other  Factors — Summary. 

15.  Spore  Discharge  and  Dissemination 338 

Methods  of  Spore  Discharge — Influence  of  External  Conditions — Spore 
Dissemination — Summary. 

16.  Spore  Germination 355 

Physical  Factors — Nutrients  and  Stimulants — Longevity  of  Spores — Sum- 
mary. 

17.  The  Physiology  of  Parasitism  and  Resistance 372 

Penetration — Parasitism — Resistance — Summ  ary . 

18.  Physiological   Variation    and    Inheritance    of   Physiological  Char- 
acters  400 

Physiological  Variation — Inheritance  of  Physiological  Characters — Summary. 

Suggested  Laboratory  Exercises 419 

Index 441 


CHAPTER  1 
INTRODUCTION 

The  primary  role  of  the  fungi  in  nature  has  been  fittingly  described  in 
the  prophetic  statement  of  B.  O.  Dodge  (1939) : 

.  .  .  the  fungi  are  not  degenerate  organisms  which  are  on  their  way  out  in  a 
scheme  of  evolution,  and  so  of  little  economic  importance  and  scientific  interest. 
The  fungi,  on  the  contrary,  are  progressive,  ever  changing  and  evolving  rapidly 
in  their  own  way  so  that  they  are  capable  of  becoming  readily  adapted  to  every 
condition  of  life.  We  may  rest  assured  that  as  green  plants  and  animals  disappear 
one  by  one  from  the  face  of  the  globe,  some  of  the  fungi  will  always  be  present 
to  dispose  of  the  last  remains. 

The  most  important  role  of  the  fungi  in  the  economy  of  nature  is  to 
act  as  scavengers  in  disposing  of  dead  and  fallen  vegetation.  In  this  way 
the  biologically  essential  elements  are  released  for  reuse,  and  the  balance 
of  nature  is  maintained.  However,  these  are  not  the  only  functions  of  the 
fungi  which  are  of  interest  and  importance  to  man.  Since  the  beginning 
of  agriculture  fungi  have  been  used  to  prepare  bread  and  other  foods, 
as  well  as  fermented  beverages.  Some  fungi  cause  diseases  of  plants 
and  animals.  Knowledge  of  their  role  as  the  causal  agents  of  plant  dis- 
eases long  antedated  the  recognition  of  bacterial  diseases.  While  yeasts 
have  long  been  used  to  produce  alcohol,  the  vast  potentialities  of  other 
species  for  the  industrial  production  of  organic  acids  and  antibiotics  have 
been  recognized  more  recently.  An  understanding  of  life  processes  of 
the  fungi  is  essential  whether  one  wishes  to  control  the  fungi  which  cause 
disease,  to  employ  them  in  industry,  or  to  use  them  in  the  laboratory  to 
unlock  the  secrets  of  nature. 

The  domain  of  physiology  is  the  study  of  functions  or  life  processes. 
Fungus  physiology  is  the  study  of  living  fungi,  their  functions  and  ac- 
tivities, how  they  affect  their  environment  and  how  the  environment 
affects  them.  Like  other  branches  of  science,  fungus  physiology  has 
four  phases  of  development:  (1)  the  discovery  and  verification  of  facts, 
which  are  the  foundation  of  any  science,  (2)  the  organization  of  these 
facts  into  a  systematic  and  coherent  body  of  knowledge,  (3)  the  dissemina- 
tion of  newly  discovered  facts,  and  (4)  use  of  the  newly  discovered  facts 
and  others  already  known  to  formulate  principles.  Facts  are  the  basis  of 
science,  but  facts  alone  are  sterile  unless  they  are  seen  in  relation  to 

1 


2  PHYSIOLOGY  OF  THE  FUNGI 

previous  knowledge.  Organization  and  interpretation  of  facts  are  equally 
as  important  as  the  experimentation  which  reveals  them. 

The  fungi  as  a  group  are  highly  responsive  to  their  environment  and 
are  thus  excellent  test  organisms  for  inquiring  into  the  secrets  of  nature. 
Nature  always  answers  correctly  the  questions  we  ask,  and,  in  this  sense, 
no  experiment  is  a  failure,  although  we  may  fail  to  ask  the  question  we 
intended,  or  we  may  misunderstand  the  answer  given.  Infinite  care  is 
required  to  frame  a  question  so  that  a  definite  answer  may  be  obtained. 
By  observing  fungi  in  nature  we  are  limited  to  questions  asked  by  nature. 
Commonly,  the  environmental  and  nutritional  factors  are  so  complex  that 
the  influence  of  a  single  variable  cannot  be  evaluated.  By  controlling  the 
conditions  under  which  a  fungus  is  placed  in  the  laboratory  it  is  possible 
to  ask  questions  of  great  precision.  Indeed,  the  number  and  scope  of 
the  questions  which  we  may  ask  fungi  are  limited  only  by  the  present-day 
techniques  and  the  curiosity  of  the  investigator. 

Since  most  of  our  knowledge  of  the  physiology  of  the  fungi  has  been 
gained  from  laboratory  investigations,  the  experimental  approach  will 
be  emphasized  in  the  discussions  which  follow.  However,  this  choice  is 
not  meant  to  minimize  the  importance  of  and  need  for  critical  observa- 
tions in  nature.  By  emphasizing  the  results  of  careful  laboratory  re- 
search, we  are  better  able  in  the  following  chapters  to  present  the  facts 
necessary  for  an  understanding  of  the  vital  principles  of  fungus  physi- 
ology, and  also  to  show  that  these  principles,  theories,  and  hypotheses 
are  founded  upon  experimental  evidence. 

FUNGUS  PHYSIOLOGY  IN  RELATION  TO  OTHER  SCIENCES 

Physiology  is  that  branch  of  science  which  deals  with  the  life  processes  or 
the  activities  of  organisms.  The  activities  of  the  whole  organism  or  of 
any  of  its  parts  may  be  hmited  by  its  form  or  structure.  Both  the 
activity  and  the  form  of  an  individual  are  determined  to  a  great  extent 
by  its  genetic  constitution  and  are  modified  by  the  environment  to  which 
the  organism  is  exposed.  Physiology,  therefore,  is  not  an  independent 
subject.  An  understanding  of  physiological  principles  is  based,  in  part, 
upon  facts  and  theories  from  many  other  fields  of  science,  such  as  chem- 
istry, physics,  anatomy,  cytology,  bacteriology,  and  genetics. 

Many  of  the  physiological  principles  which  have  been  established  for 
one  group  of  organisms  apply  equally  well  to  other  groups.  The  vita- 
mins essential  to  the  normal  growth  of  the  fungi  are  the  same  as  those 
required  by  man,  animals,  and  the  higher  plants.  The  general  functions 
of  these  vitamins  appear  to  be  the  same  in  all  organisms.  The  differ- 
ence in  the  vitamin  requirements  seems  to  lie  in  the  different  abilities  of 
these  groups  of  organisms  (or  individuals  within  the  group)  to  synthesize 
these  necessary  compounds.     As  Schopfer  (1943)  has  pointed  out,  the 


INTRODUCTION  3 

vitamin  problem  is  common  to  many  branches  of  science.  Many  other 
problems  investigated  in  fungus  physiology  are  likewise  common  to  other 
related  fields  of  study. 

In  a  similar  way,  a  better  understanding  of  certain  related  fields  is 
gained  by  knowledge  of  fungus  physiology.  The  plant  pathologist  com- 
monly finds  it  necessary  to  study  the  living  parasitic  fungus  apart  from 
its  host  and  must  know  something  of  the  cultural  methods  and  the  spe- 
cific nutritional  requirements  of  the  fungus  at  hand.  The  mycologist 
and  plant  pathologist  are  faced  with  numerous  unsolved  problems  which 
must  be  investigated  by  physiological  methods.  One  of  the  most  chal- 
lenging problems  is  the  cultivation  of  certain  fungi  now  classed  as  obligate 
parasites  on  synthetic  media  of  known  composition.  Until  this  is  accom- 
plished, the  nutritional  requirements  of  these  fungi  cannot  be  fully  deter- 
mined. Such  knowledge  would  without  doubt  lead  to  a  better  under- 
standing of  parasitism  and  resistance. 

The  taxonomic  mycologist  uses  morphological  characters  almost  exclu- 
sively in  his  identification  and  classification  studies,  while  the  bacteriol- 
ogist, being  unable  to  use  distinct  morphological  features  to  any  great 
extent,  emphasizes  the  physiological  characters  in  classifying  bacteria. 
Much  more  information  is  needed  before  it  can  be  determined  whether 
any  physiological  characters  are  sufficiently  valuable  and  uniform  to  be 
used  to  supplement  morphological  characters  in  taxonomy  of  fungi.  It 
seems  logical  that  such  physiological  differences  between  groups  of  fungi 
do  exist,  and  that  the  main  problem  lies  in  the  discovery  and  recognition 
of  these  characters  and  their  application  to  taxonomy.  On  the  other 
hand,  caution  must  be  observed,  for  nutritional  and  environmental  con- 
ditions are  known  to  affect,  to  a  certain  extent,  some  morphological 
characters  used  in  classification. 

The  geneticist  and  the  biochemist  may  find  the  fungi  interesting  and 
suitable  subjects  for  the  study  of  their  respective  problems,  while  the 
bacteriologist  finds  many  points  of  similarity  between  the  physiology  of 
the  bacteria  and  that  of  the  fungi.  Industry  has  used  many  species  of 
fungi  to  its  ovnx  advantage  for  many  decades.  Yeasts  were  used  long 
before  the  physiology  of  the  fungi  became  an  organized  study,  but  the 
search  for  superior  strains  of  yeast  continues.  The  widespread  use  of 
antibiotics  has  brought  under  laboratory  study  many  species  of  fungi 
which  would  otherwise  have  been  ignored.  This  has  created  many  new 
problems  of  nutrition,  especially  with  regard  to  large-scale  cultivation  of 
these  fungi. 

Thus,  knowledge  of  the  life  activities  of  the  fungi  is  important  and  useful 
in  many  related  fields  of  science,  just  as  some  knowledge  of  these  related 
fields  is  essential  to  an  understanding  of  the  fimgi.  The  study  of  fungup 
physiology  is  justified  as  a  separate  field  in  which  the  basic  or  fundamental 


4  PHYSIOLOGY  OF  THE  FUNGI 

principles  are  the  aim,  or  as  a  study  closely  integrated  with  the  fields  of 
science  concerned  with  more  practical  problems.  Often,  the  most 
\aluablc  results  are  obtained  when  research  is  not  restricted  by  the 
boundaries  of  practical  application. 

AIMS 

This  book  is  a  discussion  of  living  fungi,  of  their  life  processes  and  the 
factors  which  influence  them.  It  is  written  primarily  for  the  student  who 
is  acquainted  with  the  structure  of  fungi  and  who  is  beginning  the  study 
of  their  activities.  From  the  discussions  which  follow  the  student  should 
gain  a  knowledge  and  understanding  of  the  basic  principles  of  fungus 
physiology.  To  this  end  a  considerable  amount  of  factual  material  con- 
cerning the  behavior  of  specific  fungi  under  specific  conditions  is  cited. 
The  secondary  aim  of  this  book  is  to  present  a  limited  number  of  selected 
references  which  may  be  of  use  to  the  student  or  investigator  who  wishes 
more  detailed  information.  Where  possible,  review  articles  have  been 
included.  Complete  documentation  is  impossible  because  of  the  tre- 
mendous volume  of  literature.  However,  becoming  familiar  with  the 
literature  is  an  essential  part  of  a  student's  education. 

SCOPE 

As  a  text  this  book  must  cover  many  phases  of  the  subject.  One  of  the 
first  problems  to  be  considered  is  the  choice  and  preparation  of  suitable 
media  for  growth  and  sporulation  of  the  fungi  under  study.  Since  there 
is  no  universal  medium  suitable  for  all  fungi,  a  wise  choice  of  media  for 
the  purpose  at  hand  is  of  fundamental  importance  in  any  investigation. 

Before  a  fungus  can  be  studied  in  any  great  detail  in  the  laboratory,  it 
is  necessary  to  determine  the  conditions  which  affect  growth.  Growth  is 
a  complex  phenomenon,  and  some  discussion  of  the  phenomenon  itself 
and  the  ways  of  measuring  growth  is  necessary  for  the  understanding  of 
these  conditions.  Nutritional  factors,  such  as  source  of  nitrogen,  source 
of  carbon,  the  presence  of  essential  elements  and  vitamins,  and  the  pH  of 
the  substrate,  afTect  growth  in  interrelated  ways.  Each  of  these  factors 
and  its  importance  in  growth  and  other  activities  of  fungi  are  discussed  at 
some  length. 

The  life  processes  of  the  fungi  involve  numerous  chemical  transforma- 
tions. Living  organisms  make  and  use  special  organic  catalysts,  enzymes, 
which  control  these  reactions.  The  actions  of  the  enzymes  in  the  living 
organism  are  coordinated  and  interrelated.  A  knowledge  of  the  princi- 
ples of  enzyme  action  is  essential  to  the  study  of  fungus  physiology. 

The  fungi  are  able  to  make  a  far  greater  contribution  to  the  production 
of  food  and  many  other  valuable  products  than  they  do  at  present.     Both 


INTRODUCTION  5 

the  useful  metabolic  products,  such  as  alcohols,  organic  acids,  and  anti- 
biotics, and  the  harmful  products  (toxins)  are  discussed  at  some  length. 

Certain  fungi  cause  diseases  of  plants  and  animals.  The  action  of 
fimgicides  used  to  control  these  pathogens  will  be  discussed  from  a  theo- 
retical viewpoint,  since  there  is  an  enormous  amount  of  literature  dealing 
with  the  practical  application  of  fungicides.  Too  little  attention  has 
been  devoted  to  the  mechanism  of  fungicidal  action. 

The  production  of  spores,  which  is  of  fundamental  importance  to  the 
fungus  in  the  perpetuation  of  the  species,  affords  many  interesting  prob- 
lems in  fungus  physiology.  Environmental  and  nutritional  factors  play 
important  roles  in  determining  whether  a  fungus  will  spoiiilate  under  a 
given  set  of  conditions.     These  factors  are  discussed  in  some  detail. 

The  latter  portion  of  the  text  emphasizes  the  activities  of  fungi  in 
nature.  These  topics  include  the  discharge,  dissemination,  and  germina- 
tion of  spores  and  the  physiological  aspects  of  parasitism,  variation,  and 
inheritance.  The  physiology  of  parasitism  and  resistance  is  of  special 
interest  to  plant  pathologists  and  medical  mycologists.  Most  of  the 
examples  are  taken  from  the  field  of  plant  pathology.  Perhaps  the  dis- 
cussion of  these  problems  will  stimulate  the  interest  and  curiosity  of  the 
student.  A  better  understanding  of  parasitism  will  surely  lead  to  a  wiser 
choice  of  control  methods  for  certain  fungi. 

No  study  of  fungus  physiology  is  complete  without  experimental  work 
in  the  laboratory.  The  judgment  necessary  to  evaluate  one's  own  work 
is  founded  upon  experience.  Suggested  laboratory  exercises  and  demon- 
strations, with  brief  instructions,  are  given  at  the  end  of  the  text.  These 
are  selected  to  illustrate  important  principles,  many  of  which  can  be 
illustrated  clearly  only  by  direct  observation  of  the  varied  reactions  of 
fungi  to  their  environment. 

HISTORICAL  DEVELOPMENT 

The  development  of  fungus  physiology  is  far  from  complete.  While 
some  of  the  main  outlines  are  clearly  visible,  much  remains  to  be  done. 
Although  space  and  time  do  not  permit  a  complete  review  of  the  history 
of  this  science,  it  is  important  to  realize  that  its  development  was  the 
work  of  many  minds  and  hands.  The  influence  of  the  early  investigators 
continues,  not  only  in  their  published  work  but  also  in  the  students  they 
trained. 

Some  of  the  outstanding  leaders  in  the  development  of  fungus  physi- 
ology are  Avorthy  of  special  mention.  Their  names  and  their  contribu- 
tions are  encountered  frequently  bj'"  all  students  of  this  subject.  Brief 
mention  of  some  of  these  men  and  their  fields  of  interest  and  investigation 
is  made  below. 


6  PHYSIOLOGY  OF  THE  FUNGI 

Louis  Pasteur  (1822-1895),  France.  Pasteur  was  a  chemist  who,  as  a 
result  of  his  interest  in  microorganisms  which  cause  disease  and  fermenta- 
tions, became  a  biologist.  No  other  scientist  has  opened  up  so  many 
fields  of  fruitful  study.  Early  in  his  career  he  discovered  that  fungi  are 
able  to  discriminate  between  the  optical  isomers  of  tartaric  acid.  His 
student  Raulin  devised  the  first  synthetic  medium  for  the  cultivation  of 
fungi  and  published  the  first  thorough  study  of  the  nutritional  require- 
ments of  a  fungus.  Pasteur  discovered  that  some  organisms  are  inhibited 
by  free  oxygen  and  that  some  fungi  change  both  their  morphology  and 
physiology  when  cultivated  anaerobically.  Pasteur's  complete  works 
have  been  collected  and  edited  (1933-1939)  by  his  grandson,  Professor 
Pasteur  Vallery-Radot.  Dubos  (1950)  has  published  an  evaluation  of 
Pasteur's  work. 

Heinrich  Anton  de  Bary  (1831-1888),  Germany.  His  principal  contribu- 
tions to  mycology  dealt  with  life  histories  and  parasitism  of  fungi.  His 
interests  were  primarily  with  biological  adaptations  and  were  more  physi- 
ological than  taxonomic.  De  Bary's  influence  as  a  teacher  attracted 
many  students  who  later  were  responsible  for  much  of  the  development  of 
plant  pathology  and  mycology.  Among  his  writings  was  "  Morphologic 
und  Physiologic  der  Pilze"  (first  edition  1866,  second  edition,  English 
translation,  1887),  which  may  be  considered  as  the  first  book  containing 
discussions  of  the  physiology  of  the  fungi. 

Oscar  Brefeld  (1839-1925),  Germany.  We  owe  a  great  debt  to  this 
patient  investigator,  who  developed  methods  of  ensuring  sterile  media 
and  apparatus  for  pure  culture  work.  He  was  equally  insistent  with 
regard  to  the  purity  of  his  cultures.  His  chief  interest  in  mycology  was 
the  study  of  life  histories  and  development  of  fungi.  This  meant  to  him 
observation  of  a  fungus  from  ''Spore  zu  Spore."  He  was  the  first  to  use 
the  single-spore  technique.  Besides  his  occasional  papers,  he  published 
his  monumental  work  (1872-1912)  in  15  parts.  This  beautifully  illus- 
trated work  is  still  of  great  value. 

Georg  Klehs  (1857-1918),  Germany.  His  important  contributions  to 
the  study  of  fungus  physiology  concerned  problems  related  to  sporulation. 
In  1900  he  summarized  his  conclusions  in  four  statements  or  laws  (Chap. 
14).  No  better  generalizations  on  this  subject  have  appeared  in  the  50 
years  which  have  elapsed  since  they  were  published.  For  an  evaluation 
of  the  significance  of  Klebs'  work,  see  Kauffman  (1929). 

A.  H.  Reginald  Buller  (1874-1944),  England  and  Canada.  Many  of  his 
studies  involved  the  activities  of  fungi  in  relation  to  structure.  His  chief 
interests  lay  in  production  of  fruit  bodies  and  spores,  in  spore  discharge 
and  dissemination,  and  in  the  effects  of  the  environment  on  these  activi- 
ties. His  keen  observations  are  recorded  in  detail  in  seven  volumes, 
"Researches  on  Fungi."     These  volumes  are  written  in  an  interesting, 


INTRODiCTION  7 

readable  style  and  should  be  frequently  consulted  by  all  students  of 
mycology. 

Leon  H.  Leonian  (1888-1945),  United  States.  Trained  as  a  mycologist 
under  Kauffman,  he  was  always  interested  in  discovering  the  potentiali- 
ties of  living  fungi.  His  principal  contributions  were  made  in  the  studj'- 
of  fungus  nutrition  with  emphasis  on  the  factors  which  are  required  by 
fungi  for  growth  and  reproduction.  For  a  bibliography  of  his  papers  see 
Orton  (1946). 

The  number  of  living  investigators  who  have  made  and  are  continuing 
to  make  important  contributions  to  fungus  physiology  is  far  too  great  to 
list  here,  and  for  this  reason  they  have  been  omitted.  An  idea  of  the 
scope  of  their  interests  and  activities  may  be  gained  from  the  references  in 
the  following  chapters. 

REFERENCES 

Bkefeld,  O.:  Botanische  Untersuchungen  iiber  Schimmelpilze,  Hefte  1-4,  1872- 
1881.  Title  changed  to  Botanische  Untersuchungen  iiber  Hefenpilze  Fortset- 
zung  der  Schimmelpilze  for  Heft  5,  1883;  thereafter  Untersuchungen  aus  dem 
Gesamtgebiet  der  Mykologie,  Hefte  6-15,  1884-1912.  Hefte  1-8,  Arthur  Felix, 
Leipzig.     Hefte  9-15,  Heinrich  Schoningh,  Muenster. 

BuLLER,  A.  H.  R.:  Researches  on  Fungi,  Longmans,  Roberts  and  Green,  London. 

Vol.  I,  1909;  Vol.  II,  1922;  Vol.  Ill,  1924;  Vol.  IV,  1931;  Vol.  V,  1933;  Vol.  VI, 

1934;  Vol.  VII,  The  University  of  Toronto  Press,  Toronto,  1950. 

*De  Bary,  a.:  Comparative  Morphology  and  Biology  of  the  Fungi,  Mycetozoa  and 

Bacteria  (trans.  H.  E.  F.  Garnsey),  Oxford  University  Press,  New  York,  1887. 

Dodge,  B.  O.  :  Some  problems  in  the  genetics  of  the  fungi.  Science  90 :  379-385,  1939. 
*DuBos,  R.  J.:  Louis  Pasteur,  Free  Lance  of  Science,  Little,  Bro^Ti  &  Company, 
Boston,  1950. 

Kauffman,  C.  H.:  Klebs'  theory  of  the  control  of  developmental  processes  in 
organisms,  and  its  application  to  fungi,  Proc.  Intern.  Congr.  Plant  Sci.  2 :  1603- 
1611,  1929. 

Klebs,  G.:  Zur  Physiologic  der  Fortpflanzung  einiger  Pilze.  III.  Allgemeine 
Betrachtungen.     Jahrb.  wiss.  Botan.  35 :  80-203,  1900. 

Orton,  C.  R.:  Leon  Hatchig  Leonian.     Phytopathology  36:  241-244,  1946. 

Pasteur,  L.:  Oeuvres  de  Pasteur,  reunies  par  Pastevur  Vallery-Radot,  7  vols., 
Masson  et  Cie,  Paris,  1933-1939. 

ScHOPFER,  W.  H.:  Plants  and  Vitamins,  Chronica  Botanica  Co.,  Waltham,  1943. 


CHAPTER  2 
CULTURE  MEDIA 

Before  discussing  the  nutrition  of  the  fungi  in  detail,  it  will  be  helpful  to 
consider  the  basic  problems  involved.  For  many  purposes  a  knowledge 
of  the  nutrition  of  the  fungi  is  necessary  for  culturing  them  in  the  labora- 
tory or  in  industry.  Like  all  living  organisms  the  fungi  must  obtain  from 
their  environment  the  materials  needed  for  the  synthesis  of  protoplasm 
and  other  cellular  constituents.  Directly  or  indirectly,  the  fungi  as  well 
as  animals  and  most  bacteria  are  dependent  upon  green  plants  for  "food" 
and  energy. 

Not  all  natui'al  substrates  are  equally  suitable  for  all  fungi.  In  nature, 
the  saprophytes  are  more  widely  distributed  than  the  parasites,  which  are 
usually  restricted  to  the  range  of  their  hosts.  Many  of  the  substances 
upon  which  the  fungi  grow  in  nature  are  chemically  complex,  and  some, 
such  as  cellulose,  starch,  and  proteins  are  insoluble  or  are  only  colloidally 
soluble.  Before  such  compounds  can  be  utilized,  they  must  be  changed 
into  low-molecular-weight  compounds  which  are  soluble  in  water.  This 
"digestion"  is  accomplished  by  means  of  enzymes  which  are  excreted  by 
the  fungi.  This  is  analogous  to  digestion  in  animals,  which  is  also  an 
enzymatic  process.  The  complete  utilization  of  a  natural  substrate  is 
frequently  due  to  the  combined  action  of  a  succession  of  microorganisms. 
More  than  one  organism  may  act  at  the  same  time,  and  often  this  simul- 
taneous action  is  more  effective  than  that  of  a  single  organism. 

One  may  ask.  Do  the  fungi  simply  incorporate  within  their  own  proto- 
plasm the  suitable  elements  and  compounds  found  in  the  medium,  or  do 
they  transform  the  compounds  of  the  medium  before  building  their  own 
structures?  Apparently  the  fungi  do  both.  The  essential  elements  such 
as  potassium  and  magnesium  are  taken  up  as  ions,  although  these  ele- 
ments may  be  in  the  state  of  chemical  combination  in  the  substrate  and 
also  in  the  fungus  cells.  Certain  organic  compounds,  such  as  the  vita- 
mins, are  undoubtedly  absorbed  as  such  from  the  medium  by  vitamin- 
deficient  fungi;  otherwise,  these  fungi  would  derive  no  benefit  from  them. 
The  same  statement  is  true  for  other  necessary  compounds  which  the 
various  fungi  are  unable  to  synthesize. 

By  far  the  greater  part  of  the  compounds  utilized  by  the  fungi  are 
modified  or  changed  either  before  or  after  they  are  taken  into  the  cells. 
Outside  the  fungus  cells,  these  changes  are  largely  in  the  direction  of 
simplifying  the  molecular  structure  of  compounds  used.  Within  the 
fungus  cells  some  of  the  metabolite  molecules  are  oxidized  to  carbon 

8 


CULTURE  MEDIA  9 

dioxide  and  water  or  to  intermediate  products.  By  this  process  the 
fungus  obtains  the  chemical  energy  which  it  requires  for  the  processes  of 
synthesis. 

KINDS  OF  MEDIA 

No  one  knows  when  man  began  to  cultivate  fungi,  but  certainly  it  was 
many  thousands  of  years  ago.  This  cultivation  was  no  doubt  uninten- 
tional at  first  and  was  later  developed  into  an  art,  in  connection  with  the 
preparation  of  foodstuffs  and  beverages.  The  use  of  leaven  (yeast) 
extends  back  to  the  beginning  of  agriculture.  The  yeast  culture  was 
preserved  in  a  piece  of  dough  which  in  turn  was  added  to  the  next  batch, 
much  as  buckwheat  batter  is  prepared  today.  In  the  Orient,  species  of 
Mucor  and  Aspergillus  have  been  used  from  the  dawn  of  civilization  in 
preparing  food  from  rice  and  soybeans.  Brewers  used  yeast  many  cen- 
turies before  it  was  learned  that  yeast  is  a  living  organism.  On  the  other 
hand,  the  science  of  growing  fungi  in  pure  culture  is  fairly  recent. 

Natural  media.  It  was  quite  natural  that,  when  mycologists  and 
others  began  to  cultivate  fungi  in  the  laboratory,  they  should  turn  to 
natural  materials  as  media.  A  natural  medium  is  one  which  is  composed 
entirely  of  complex  natural  materials  of  unknown  composition.  Among 
the  natural  substances  so  used  are  the  following:  plant  parts,  malt,  yeast, 
peptone,  manure,  bread,  wort,  fruit,  and  vegetables.  Many  of  these 
substances  are  used  in  the  form  of  extracts,  infusions,  or  decoctions.  The 
very  diversity  of  these  natural  media  is  strong  testimony  to  the  fact  that 
different  species  have  different  nutritional  requirements.  Brefeld  (1881) 
was  among  the  first  to  grow  fungi  in  pure  culture,  and  many  of  his  tech- 
niques are  in  use  today.  Since  his  interest  in  cultivating  the  fungi  was 
largely  for  the  purpose  of  observing  their  development,  it  was  necessary 
for  him  to  select  suitable  media.  He  found  two  natural  media  to  be  of 
great  utility:  a  decoction  prepared  from  dried  plums  or  raisins  and  a 
manure  extract.  This  latter  medium  he  considered  "als  Universal- 
nahrlosung  flir  Pilzculturen."  This  medium  is  still  used  in  some  labora- 
tories. Natural  media  have  many  advantages.  They  are  cheap  and 
easy  to  prepare.  In  many  instances  it  is  necessary  only  to  add  water  to 
the  base  material  and  autoclave.  More  important  yet  is  the  fact  that 
many  fungi  grow  well  upon  a  wide  variety  of  natural  media. 

Certain  of  the  more  fastidious  fungi  have  never  been  cultivated  in  the 
laboratory.  These  obligate  parasites  live  only  upon  or  ^^^thin  the  living 
tissues  of  their  hosts.  Puccinia  graminis  tritici  lives  only  on  wheat,  some 
species  of  grasses,  and  some  species  of  barberry.  These  host  plants  when 
killed  will  no  longer  support  growth  of  this  fungus.  However,  many 
species,  which  in  the  past  were  considered  to  be  obligate  parasites,  have 
since  been  cultured  on  nonliving  media. 


10  PHYSIOLOGY  OF  THE  FVSGI 

Semisynthetic  media.  A  semisynthetic  medium  is  one  Avhich  is  com- 
posed in  part  of  natural  materials  of  unknown  composition.  Such  media 
are  made  by  adding  compounds  of  known  composition  to  one  or  more 
natural  materials.  The  widely  used  potato-glucose  (dextrose)  medium  is 
an  example  of  this  type.  The  addition  of  agar  to  an  otherwise  synthetic 
medium  introduces  a  natural  material  of  unknown  composition.  Media 
Avhich  contain  agar  cannot  be  classed  strictly  as  synthetic  media.  Semi- 
synthetic media  may  be  used  for  many  types  of  physiological  investigations. 

The  composition  of  a  given  natural  or  semisynthetic  medium  is  not 
constant.  Potato-glucose  medium  may  vary  greatly  in  composition 
depending  upon  whether  or  not  the  potatoes  were  peeled  and  upon  the 
variety  and  age  of  the  potatoes  used.  Neuberger  and  Sanger  (1942) 
found  a  twofold  difference  in  the  amide  nitrogen  (asparagine  and  gluta- 
mine)  among  varieties.  In  this  laboratory  we  have  found  that  the 
amount  of  potato  pulp  Avhich  is  allowed  to  enter  this  medium  exerts  a 
marked  influence  upon  growth  and  reproduction  of  certain  fungi.  These 
differences,  which  may  seem  minor,  are  great  enough  to  make  comparisons 
between  work  done  in  different  laboratories  difficult. 

Synthetic  media.  As  the  term  is  used  in  this  book,  a  synthetic  medium 
is  one  of  known  composition  and  concentration.  It  does  not  mean  that 
every  compound  used  is  a  product  of  the  chemist's  art.  Some  of  the 
constituents,  such  as  the  sugars,  may  be  of  natural  origin.  The  important 
condition  is  that  the  compounds  used  be  pure,  and  this  is  difficult  to 
attain  in  practice.  "Chemically  pure"  compounds  are  usually  far  from 
being  pure,  as  a  glance  at  the  labels  will  show.  The  ideal  of  using  pure 
compounds  is  seldom  realized,  but  the  closer  it  is  approached,  the  more  we 
shall  learn  about  the  nutrition  of  the  fungi. 

Natural  media  and  most  semisynthetic  media  are  of  limited  usefulness 
in  studying  nutrition  of  the  fungi.  The  chief  value  of  synthetic  media  is 
for  nutritional  studies.  However,  growth  and  reproduction  are  fre- 
quently poorer  on  a  synthetic  medium  than  on  one  containing  some 
natural  material.  For  example,  Aspergillus  niger  grew  well  in  a  synthetic 
medium  composed  of  sucrose,  ammonium  nitrate,  magnesium  sulfate,  and 
dipotassium  hydrogen  phosphate  (Steinberg,  1939).  In  addition  to 
these  major  constituents,  iron,  zinc,  copper,  manganese,  molybdenum, 
and  gallium  salts  were  present.  Extraordinary  care  was  taken  in  prepar- 
ing the  medium.  The  concentration  of  every  constituent  was  so  balanced 
that  a  decrease  in  concentration  of  any  constituent  resulted  in  diminished 
growth.  Growth  and  sporulation  were  excellent  upon  this  medium. 
When  20  mg.  per  liter  of  either  peptone  or  yeast  extract  was  added,  the 
rate  of  growth  was  greatly  increased  and  the  time  required  for  sporulation 
was  decreased.  The  small  amount  of  yeast  extract  or  peptone  used 
could  have  added  only  an  insignificant  amount  of  material  from  which 


CULTURE  MEDIA  ii 

the  fungus  could  synthesize  protoplasm  or  derive  energy.  Steinberg's 
synthetic  medium  was  adequate  but  not  optimum  for  most  rapid  growth 
and  sporulation.  We  may  suppose  that  the  yeast  extract  and  peptone 
contained  compounds  the  synthesis  of  which  constituted  a  limiting  effect 
upon  the  rate  of  growth  and  sporulation. 

Synthetic  media  may  be  simple  or  complex  but  must  contain  the  essen- 
tial elements  in  utilizable  form.  Brefeld  (1881)  gave  the  following  direc- 
tions for  preparing  a  synthetic  medium:  Add  cigar  ashes  dissolved  in 
nitric  or  citric  acid  to  a  solution  containing  a  soluble  carbohydrate,  such 
as  glucose,  and  an  ammonium  salt.  The  amount  of  ashes  was  not  speci- 
fied.    The  first  vsynthetic  medium  was  devised  by  Raulin  (1869). 

Table  1.     Composition  of  the  First  Synthetic  Medium  for  CuLTrvATiNG  Fungi 

(Raulin,  Ann.  sci.  nat.,  Ser.  V,  11,  1869.) 

Ammonium  nitrate 4.0    g. 

Ammonium  phosphate 0.6    g 

Magnesium  carbonate 0.4    g 

Potassium  carbonate 0.6    g 

Ammonium  sulfate 0 .  25  g 

Zinc  sulfate 0 .  07  g, 

Iron  sulfate 0 .  07  g 

Potassium  silicate 0 .  07  g, 

Sucrose 70        g, 

Tartaric  acid 4        g 

Water 1,500        ml. 

However,  not  enough  information  is  given  in  Table  1  for  the  duplication 
of  this  medium.  Which  ammonium  phosphate,  (NH4)H2P04  or  (NH4)2- 
HPO4,  was  used  by  Raulin  in  the  original  work?  Which  zinc  sulfate, 
ZnS04-7H20  or  ZnS04-H20,  was  used?  Was  the  iron  sulfate  FeS04, 
FeS04-7H20,  or  re2(S04)  3?  Did  he  use  D-tartaric,  L-tartaric,  DL-tartaric, 
or  weso-tartaric  acid?  These  questions  are  asked  for  the  purpose  of 
emphasizing  the  need  for  exactness  in  reporting  the  composition  of  media 
used  in  experimental  work.  These  uncertainties  creep  into  the  literature 
through  ignorance  or  carelessness,  or  both.  Nor  are  these  ambiguities  to 
be  found  only  in  the  older  literature,  for  they  are  present  in  papers  pub- 
lished only  yesterday.  Either  the  specific  name  or  the  formula,  or  both, 
should  be  stated.  If  it  is  stated  that  dipotassium  phosphate,  K2HPO4, 
was  used,  the  reader  is  certain  of  the  identity  of  the  compound.  Potas- 
sium phosphate  may  designate  at  least  five  distinct  chemical  compounds. 

NATURAL  VERSUS  SYNTHETIC  MEDIA 

In  addition  to  the  fact  already  noted  that  the  composition  of  natural 
media  is  unknown,  natural  and  synthetic  media  differ  in  two  further 
respects.     Natural  media  are  more  complex;  i.e.,   they  contain  more 


12  PHYSIOLOGY  OF  THE  FUNGI 

chemical  compound.s   than  synthetic  media.     They  also  contain  com 
pounds  ordinarily  not  present  in  synthetic  media. 

Specific  metabolites.  Only  certain  chemical  compounds  are  utilized 
by  fungi,  but  not  all  fungi  are  able  to  utilize  the  same  compounds.  Any 
compound  utilized  by  a  fungus  is  called  a  metabolite.  Some  fungi  are 
unable  to  synthesize  certain  essential  metabolites  and  are  said  to  be 
"deficient"  for  the  specific  metabolites  they  are  unable  to  synthesize. 
In  order  to  cultivate  such  deficient  fungi,  these  metabolites  must  be  pres- 
ent in  the  medium.  Natural  media  usually  contain  these  metabolites. 
If  a  fungus  grows  upon  a  natural  medium  and  fails  to  grow  upon  a  variety 
of  simple  synthetic  media,  it  may  be  suspected  that  specific  metabolites 
are  involved  in  its  nutrition. 

The  following  example  will  illustrate  the  role  of  specific  metabolites  in 
fungus  nutrition.  Fellows  (1936)  investigated  the  ability  of  Ophiobolus 
grayninis  to  utilize  different  nitrogen  compounds  for  growth.  A  sucrose- 
mineral  salts  solution  was  used  as  the  basal  medium  to  which  various 
nitrogen  sources  were  added.  Only  complex  nitrogen  sources  such  as  egg 
albumen,  peptone,  casein,  and  nucleic  acid  allowed  growth.  Under  the 
conditions  used  no  growth  resulted  when  ammonium  compounds,  nitrates, 
nitrites,  and  amino  acids  were  tested.  0.  graminis  in  the  presence  of  egg 
albumen  utilized  glucose,  maltose,  lactose,  fructose,  xylose,  starch,  and 
dextrin,  in  addition  to  sucrose.  From  these  experimental  results  it  was 
concluded  that  0.  graminis  requires  a  complex  source  of  nitrogen  for 
growth. 

Later,  White  (1941)  found  that  this  fungus  requires  two  specific 
metabolites,  thiamine  and  biotin.  When  these  vitamins  were  added  to 
synthetic  media  containing  simple  nitrogen  sources  (sodium  nitrate, 
ammonium  nitrate,  asparagine,  or  glycine),  good  growth  was  obtained. 
Thus,  it  appears  obvious  that  0.  graminis  does  not  require  a  complex 
nitrogen  source,  but  that  it  is  unable  to  synthesize  two  specific  chemical 
compounds.  These  papers  illustrate  the  fact  that  fungus  physiology  is  a 
young  and  developing  science.  Much  of  the  early  work  needs  reevalua- 
tion  in  the  light  of  recent  discoveries.  A  student  should  strive  to  develop 
a  critical  attitude  toward  the  work  of  others,  but  he  should  be  no  less 
critical  with  regard  to  his  own  work.  The  evaluation  of  experimental 
results  depends  upon  the  conditions  under  which  the  work  was  done,  and 
among  these  conditions  the  medium  used  is  of  first  importance. 

Complexity  of  media.  It  is  a  common  experience  to  find  that  a  trace  of 
some  crude  natural  product  stimulates  the  rate  of  growth  and  sporulation 
of  a  fungus.  This  stimulation  frequently  occurs  with  fungi  which  grow 
well  on  synthetic  media  and  which  are  not  deficient  for  vitamins  or  amino 
acids.  It  appears  that  the  complexity  of  natural  media  offers  a  clue  to 
understanding  this  stimulatory  effect.     If  a  fungus  is  grown  upon  a  simple 


CULTURE  MEDIA  13 

synthetic  medium  which  has  only  one  source  of  carbon  and  one  source  of 
nitrogen,  it  must  synthesize  many  complex  chemical  compounds  from 
constituents  present  in  the  medium.  It  may  be  suspected  that  these 
biochemical  syntheses  are  slowed  up  under  these  conditions.  When  a 
mixture  of  many  carbon  and  nitrogen  sources  is  present,  the  fungus  may 
function  more  efficiently,  because  the  biochemical  syntheses  are  easier 
since  some  of  the  intermediates  are  furnished.  These  speculations 
receive  some  support  from  evidence  to  be  presented  in  Chaps.  6  and  7_ 

CHOICE  AND  PREPARATION  OF  MEDIA 

Considerable  care  is  needed  in  the  selection  of  a  suitable  medium.  A 
medium  may  be  excellent  for  growth  and  unsuitable  for  reproduction  or 
the  production  of  an  antibiotic.  The  method  of  preparation  may  influ- 
ence the  composition  of  a  medium  in  unsuspected  ways. 

Choice  of  media.  In  selecting  a  medium  the  purpose  for  which  it  is  to 
be  used  should  be  kept  in  view.  For  many  purposes  a  natural  medium  is 
the  one  of  choice.  This  is  especially  true  for  routine  maintenance  of 
cultures,  for  isolations,  and  for  preliminary  investigations.  The  composi- 
tion of  natural  media  may  be  varied  by  choosing  different  substrates. 
Frequentl}^,  a  combination  of  natural  products  may  be  used  to  advantage, 
e.g.,  malt  and  yeast  extracts.  In  addition,  these  natural  substrates  may 
be  fortified  with  one  or  more  pure  chemical  compounds.  The  constitu- 
ents of  natural  media  are  fixed  by  the  substances  used,  but  the  amounts 
used  may  be  changed  at  will. 

More  judgment  enters  into  the  selection  of  synthetic  media.  The 
essentials  of  a  synthetic  medium  may  be  stated  as  follows :  sources  of  car- 
bon and  nitrogen  in  utilizable  forms;  phosphate  and  sulfate  ions;  the 
metallic  ions  potassium,  magnesium,  iron,  zinc,  manganese,  and  others 
which  are  usually  present  as  impurities  in  the  chemicals  used.  These  are 
the  essential  elements  and  will  be  considered  at  length  in  later  chapters. 
Most  fungi  utilize  glucose,  so  this  sugar  is  frequently  used  as  the  carbon 
source.  More  fungi  utilize  nitrogen  in  organic  combinations  than  in 
inorganic  compounds.  The  question  of  specificity  enters  into  the  choice 
of  the  carbon  and  nitrogen  sources,  and  this  can  be  determined  only  by 
experiment.  In  order  to  cultivate  deficient  fungi  on  synthetic  media,  the 
specific  metabolites  for  which  the  fungi  are  deficient  must  be  added. 
Since  synthetic  media  are  used  to  study  nutrition,  the  development  of  a 
suitable  synthetic  medium  for  a  specific  fungus  may  require  considerable 
investigation.  In  our  laboratory  we  commonly  first  use  a  glucose-casein 
hydrolysate  medium  containing  the  essential  inorganic  elements.  This 
medium  has  been  very  useful  in  vitamin  studies.  Its  composition  is 
given  in  Chap.  10. 


14  PHYSIOLOGY  OF  THE  FUNGI 

Solid  versus  liquid  media.  Both  solid  and  liquid  media  are  used  in 
cultivating  fungi.  Media  solidified  with  agar,  or  semisolid  substrates 
such  as  corn  meal,  offer  many  advantages  in  that  the  culture  vessels  can 
be  freely  handled  without  disturbing  the  fungus.  This  feature  is  particu- 
larly valuable  when  one  wishes  to  follow  the  development  of  a  fungus. 
Microscopic  examination  is  facilitated,  and  contaminants  are  more  easily 
detected.  Single-spore  isolations  can  be  made  more  easily  from  solid 
media.  Agar  media  are  used  to  maintain  stock  cultures  and  are  recom- 
mended for  many  preliminary  experiments. 

Frau  Hesse  (Kitchens  and  Leikind,  1939)  introduced  the  use  of  agar 
into  microbiological  procedures  in  1881.  Agar,  which  is  obtained  from 
various  marine  red  algae,  is  a  complex  polysaccharide  sulfate  ester 
(Pigman  and  Goepp,  1948).  It  forms  colloidal  solutions  at  elevated 
temperatures  and  sets  to  a  gel  at  temperatures  around  45°C.  On  acid 
hydrolysis  both  D-galactose  and  its  enantiomorph,  L-galactose,  as  well 
as  sulfuric  acid,  are  formed.  Agar  must  exist  in  the  form  of  a  salt 
(Ca,  Mg,  Na,  K,  etc.)  to  form  a  gel.  Agar  introduces  physiologically 
active  elements  into  media.  It  may  contain  significant  amounts  of 
zinc  (Leonian  and  Lilly,  1940)  and  other  micro  essential  elements. 
Mulder  (1940)  found  that  magnesium  could  be  efficiently  removed  from 
agar  by  repeated  soakings  in  10  per  cent  sodium  chloride  solution,  followed 
by  washing  with  distilled  w^ater  until  the  filtrate  was  free  from  chloride 
ion.  Agar  also  contains  growth  factors  such  as  thiamine  (Day,  1942) 
(see  Fig.  1).  Many  fungi  make  some  growth  on  water  agar,  which  indi- 
cates that  agar  or  the  ''impurities"  contained  in  it  are  utilized  by  fungi. 
Robbins  (1939)  found  that  leaching  agar  with  5  per  cent  aqueous  pyridine 
removed  many  of  the  physiologically  active  compounds. 

Liquid  media  should  be  used  for  precise  investigations  where  it  is 
desired  to  control  as  many  variables  as  possible.  The  composition  of  the 
medium  may  be  controlled  and  the  amounts  used  measured  accurately. 
Cultures  may  be  aerated  by  shaking  or  by  blowing  sterile  air  through  the 
media.  Weighing  the  mycelium  is  facilitated.  When  it  is  desired  to 
study  the  metabolic  by-products  of  fungus  metabolism  (except  gaseous 
products),  it  is  almost  necessary  to  use  liquid  media.  Isolation  of 
by-products  is  less  complicated  when  liquid  media  are  used.  Studies  of 
various  metabolite  deficiencies  and  many  microbiological  assays  (Chap. 
10)  almost  always  require  the  use  of  liquid  media.  The  choice  between 
the  use  of  solid  or  liquid  media  should  be  made  on  the  basis  of  the  known 
advantages  and  disadvantages  of  both  and  with  regard  to  the  purpose  of 
the  problem  under  investigation. 

Designating  media.  It  is  common  to  find  references  to  a  medium  by 
the  name  of  the  investigator  who  first  used  it.  These  names  have  served 
as  convenient  abbreviations  and  commemorate  the  pioneers  in  the  art  of 


CULTURE  MEDIA 


15 


cultivating  fungi.  Some  of  these  names  are  Blakeslee,  Uschinsky,  Coons, 
Czapek,  Leonian,  Sabouraud,  Richard,  Thaxter,  Shear,  Raulin.  From  a 
historical  standpoint  this  practice  has  much  to  recommend  it.  However, 
this  usage  has  many  disadvantages.  These  distinguished  names  give  no 
clue  to  the  composition  of  these  media.  The  original  formulas  have  in 
many  instances  been  changed.     Some  of  these  modifications  have  received 


A  B 

Fig.  1.  Growth  of  Phycomyces  blakesleeanus  on  vitamin-free  liquid  medium  solidified 
with  two  different  brands  of  agar.  Growth  in  A  indicates  relatively  high  content  of 
thiamine  of  this  agar.  The  trace  of  growth  in  B  shows  that  this  agar  is  relatively 
free  of  thiamine. 


hyphenated  names:  e.g.,  Czapek-Dox.  Frequently  the  originator  of  a 
medium  modified  it  from  time  to  time.  This  introduces  a  further  uncer- 
tainty as  to  its  composition.  In  our  opinion  the  use  of  personal  names  to 
designate  media  should  be  abandoned.  It  is  much  more  helpful  to 
designate  media  by  descriptive  titles  than  by  names  which  tell  nothing  of 
the  composition.  The  carbon  and  nitrogen  sources  are  important  con- 
stituents of  every  medium.  Thus,  sucrose-nitrate  medium,  glucose- 
asparagine  medium,  or  malt  extract-yeast  extract  medium  are  preferred 
to  Czapek's  medium,  Schopfer's  medium,  or  Leonian's  medium.  These 
descriptive  terms  afford  valuable  information  that  personal  names  do  not. 
Even  when  the  reader  is  familiar  with  the  composition  of  a  named  medium, 


16  PHYSIOLOGY  OF  THE  FUNGI 

there  is  a  tendency  to  fail  to  associate  experimental  results  with  the 
composition. 

Effect  of  autoclaving.  Media  are  commonly  and  effectively  sterilized 
by  autoclaving.  It  should  be  noted,  however,  that  such  high  tempera- 
tures may  cause  destruction  or  alteration  of  some  constituents  in  the 
media.  These  changes  are  not  serious  for  many  uses;  at  least  media  pre- 
pared in  this  way  are  satisfactory.  Sugars  are  among  the  substances 
most  easily  altered  by  autoclaving.  The  extent  of  decomposition  depends 
upon  the  specific  sugar  used,  the  other  constituents  of  the  medium,  and 
the  time  of  autoclaving.  It  is  desirable  to  adopt  a  uniform  schedule  for 
autoclaving  media.  An  increase  in  the  amount  of  caramelization  occurs 
as  the  time  of  heating  is  increased.  Maillard  (1912)  showed  that  a  brown 
color  results  when  reducing  sugars  (glucose,  fi-uctose,  etc.)  are  autoclaved 
with  amino  acids.  Hill  and  Patton  (1947)  have  shown  that  growth  of 
Streptococcus  faecalis  is  reduced  when  tryptophane  is  autoclaved  with 
sugars.  Margolin  (1942)  found  that  no  one  method  of  sterilization 
resulted  in  best  growth  for  all  of  the  14  species  tested.  Phythophthora 
erythroseptica  made  three  times  the  amount  of  growth  on  glucose  sterilized 
by  filtration  as  when  the  entire  medium  was  autoclaved.  Syncephalastrum 
racemosum,  however,  made  more  growth  on  autoclaved  than  on  sterile- 
filtered  glucose  (Table  2).  The  organisms  most  sensitive  to  heated 
glucose  appear  to  be  various  species  of  Cytophaga,  which  failed  to  grow 
on  glucose  which  had  been  heated  to  50°C.  (Stanier,  1942).  These 
organisms  utilized  glucose  which  had  been  sterilized  by  filtration.  Phos- 
phates, a  universal  constituent  of  media,  are  active  in  converting  glucose 
into  ketoses  and  other  products  (Englis  and  Hanahan,  1945)  during 
autoclaving. 

Complex  sugars  and  polysaccharides  undergo  some  hydrolysis  during 
autoclaving.  The  amount  of  hydrolysis  is  dependent  upon  the  carbo- 
hydrate, the  time  and  temperature  of  autoclaving,  and  the  pH  of  the 
medium.  Sucrose,  when  autoclaved  in  acidic  media,  may  undergo  suffi- 
cient hydrolysis  to  support  some  growth  of  species  unable  to  utilize 
sucrose.  This  possibility  must  be  guarded  against  in  experiments  on  the 
availability  of  complex  sugars. 

Other  substances  used  in  media  may  be  destroyed  during  autoclaving. 
To  minimize  or  avoid  such  effects,  heat-sensitive  substances  may  be  auto- 
claved separately,  or  they  may  be  sterilized  using  special  bacteriological 
filters.  The  Berkefeld  and  Chamberland  filters  are  less  used  than 
formerly,  while  at  present  Seitz  and  fritted-glass  filters  are  widely  used. 
Fritted-glass  filters  are  best  for  most  purposes,  inasmuch  as  the  asbestos 
pad  used  in  the  Seitz  filter  may  adsorb  active  compounds.  All  methods 
of  sterilization  w^hich  depend  upon  filtration  are  slow  and  can  be  used  only 
with  liquid  media.     Various  volatile  chemical  sterilization  agents  such  as 


CULTURE  MEDIA 


17 


alcohol  and  acetone  have  been  used.  Hansen  and  Snyder  (1947)  have 
recommended  the  use  of  propylene  oxide  for  the  sterilization  of  plant 
parts  used  for  culture  media.  Frequently  a  seemingly  insignificant 
change  in  the  method  of  preparing  a  medium  may  result  in  significant 
changes  in  the  composition  of  the  medium,  which  in  turn  may  be  reflected 
in  the  behavior  of  the  organisms  grown  upon  it.  Even  the  volume  of 
medium  in  culture  vessels  affects  the  amount  of  decomposition  during 
autoclaving.  Cotton  plugs  may  introduce  hnt  into  the  medium.  Less 
refined  grades  of  cotton  release  a  volatile  substance  which  affects  the 

Table  2.     The  Effect  of  Different  Methods  of  Sterilizing  Glucose  upon 

THE  Growth  of  Sex  Fungi,  at  25°C. 

Growth  reported  as  milligrams  of  dry  mycelium. 

The  entire  medium,  containing  a  mixture  of  amino  acids,  was  autoclaved  in  the 
control  experiment.  In  the  other  experiments  the  glucose  was  sterilized  by  either 
Seitz  filtration  or  treatment  with  acetone  and  added  aseptically  to  the  remainder  of 
the  sterile  medium.      (Margolin,  thesis.  West  Virginia  University,  1942.) 


Species 

Days  of 
incuba- 
tion 

Control, 

entire 

medium 

autoclaved 

Glucose 

sterilized  by 

filtration 

Glucose 
sterilized  by 

treating 
with  acetone 

Phycomyces  blakesleeanus 

Rhizopus  suinus 

7 

6 

5 

12 

15 

15 

130 
122 
103 

79 

84 

142 

140 

123 

68 

241 

88 
147 

132 
115 

80 
192 

85 
103 

Syncephalastrum  racemosxun 

Phytophthora  erythroseptica 

Diplodia  macrospora 

Phytophthora  cadonim 

germination  of  some  spores  {Phycomyces  blakesleeanus,  Robbins  and 
Schmitt,  1945).  Paper  or  aluminium  caps  may  be  used  to  replace  cotton 
plugs.  Residual  soap  films  on  improperly  rinsed  glassware  may  cause 
trouble  in  some  cases. 

Preparation  of  media.  Directions  for  the  preparation  of  specific  media 
are  given  at  the  end  of  the  text  in  the  section  Suggested  Laboratory 
Exercises.  Additional  details  concerning  various  media  are  to  be  found 
in  Riker  and  Riker  (1936)  and  Rawlins  (1933). 

WAYS  OF  EXPRESSING  CONCENTRATION 

Concentrations  are  frequently  expressed  in  the  literature  as  percentages. 
Unless  the  basis  for  calculating  these  values  is  given,  percentage  is  an 
ambiguous  way  of  reporting  concentration.  Buchanan  and  Fulmer 
(1928)  have  pointed  out  that  there  are  six  ways  of  calculating  the  percent- 
age composition  of  a  solution.  A  10  per  cent  sulfuric  acid  solution  may 
represent  six  different  concentrations.     For  any  precise  work  it  is  best  to 


18  PHYSIOLOGY  OF  THE  FUNGI 

avoid  the  use  of  percentages,  but  for  routine  work,  where  the  composition 
of  media  is  of  less  importance,  the  use  of  percentages  may  be  allowed. 
Before  the  same  medium  can  be  prepared  repeatedly,  it  is  necessary  to 
know  what  constituents  are  used  and  the  amount  of  each. 

Two  general  methods  are  used  for  reporting  the  composition  of  media. 
Either  the  weights  of  the  constituents  and  the  volume  of  water  used  are 
given,  or  the  weights  of  the  constituents  are  given  and  the  medium  made 
up  to  a  definite  volume.  The  first  method  is  in  common  use;  its  sim- 
plicity conceals  its  disadvantages.  The  volume  of  a  medium  prepared  by 
this  method  is  never  the  same  as  the  volume  of  water  used.  It  is  neces- 
sary to  measure  the  volume  of  the  medium  after  preparation  in  order  to 
calculate  the  amount  of  any  constituent  in  an  alicjuot. 

The  method  of  choice  in  accurate  work  is  to  weigh  the  constituents  and 
make  the  medium  up  to  a  given  volume.  The  amovmt  of  any  constituent 
in  any  volume  of  medium  may  then  be  calculated.  If  a  liter  of  medium 
contains  25  g.  of  sucrose,  and  25-ml  lots  are  dispensed,  each  lot  contains 
^/iooo  X  25,  or  0.625  g.  of  sucrose. 

Direct  units.  The  units  of  volume  most  used  are  the  liter  (1.)  and  the 
milliliter  (ml.).  A  cubic  centimeter  (cc.)  is  nearly,  but  not  exactly, 
equivalent  to  a  milliliter.  Its  use  should  be  discouraged.  The  formulas 
for  media  are  usually  given  on  the  basis  of  a  liter.  This  practice  is  to  be 
encouraged,  as  the  liter  is  a  convenient  volume  in  preparing  media.  The 
weights  of  solid  constituents  should  be  reported  as  grams  (g.)  or  decimal 
divisions  thereof.  The  most  commonly  used  decimal  fractions  of  the 
gram  are  the  milligram  (mg.),  the  microgram  (jug),  and  the  millimicrogram 
(m/ig),  each  of  which  is  one-thousandth  of  the  preceding  weight.  Since 
it  is  easy  to  make  mistakes  in  reading  small  decimals,  it  is  recommended 
that  no  decimals  smaller  than  0.1  be  used.  The  use  of  12  mg.  is  preferable 
to  0.012  g.,  although  both  mean  exactly  the  same.  It  is  easier  to  read 
5  jug  than  0.000005  g.  One  milligram  of  a  substance  in  a  Uter  of  solution 
equals  one  part  per  million  (p. p.m.).  Each  milliliter  of  such  a  solution 
wiU  contain  1  fxg  of  the  substance.  Similarly,  a  microgram  of  a  substance 
in  a  liter  of  solution  is  present  as  one  part  per  billion.  The  microgram 
has  also  been  called  the  gamma  (7),  but  this  usage  should  be  abandoned 
inasmuch  as  gamma  is  not  a  regular  prefix  used  in  the  metric  system. 
The  necessity  of  using  such  small  units  of  weight  arises  from  the  physio- 
logical activity  of  certain  compounds  and  elements.  For  example,  a 
concentration  of  1  mg.  of  biotin  in  a  liter  of  medium  is  a  relatively  enor- 
mous concentration. 

Derived  units.  Derived  units  must  be  used  in  comparing  the  effect  of 
compounds  which  have  different  molecular  weights.  Among  these 
derived  units  the  mole  is  the  most  useful.  A  mole  is  the  molecular  weight 
of  a  chemical  compoimd  expressed  in  grams.     A  mole  of  glucose  is  180  g., 


\^   K^  ±^  J.     \^  X\/X-J        1.WX  JUt  X-^X  iX. 


while  a  mole  of  sucrose  is  342  g.  A  liter  of  solution  containing  one  mole 
of  a  compound  is  said  to  be  one  molar  (il/).  Equimolar  solutions  contain 
the  same  number  of  molecules.  In  problems  in  physiology,  such  as 
osmotic  pressure,  which  have  to  do  with  the  numbers  of  molecules  it  is 
necessary  to  use  this  way  of  expressing  concentration.  If  it  is  desired  to 
compare  the  effect  of  the  osmotic  pressure  due  to  glucose  and  sucrose,  the 
concentration  must  be  expressed  in  terms  of  molar  strengths,  for  the 
osmotic  pressure  is  a  function  of  the  number  of  molecules  of  solute  in  a 
solution.  If  it  is  the  purpose  to  compare  the  effect  of  glucose  and  sucrose 
on  the  amount  of  growth  of  a  fungus,  this  method  of  expressing  concen- 
trations should  not  be  used.  Media  of  equal  molarity  with  respect  to 
sucrose  and  glucose  do  not  contain  the  same  amount  of  carbon.  The 
first  contains  twice  as  much  carbon  as  the  second.  Just  as  a  milligram  is 
one-thousandth  of  a  gram,  a  millimole  is  one-thousandth  of  a  mole.  The 
meaning  of  micromole  and  millimicromole  should  be  obvious. 

If  the  weight  of  a  compound  is  given  in  grams,  this  datum  may  be  con- 
verted into  moles.  If  a  medium  contains  50  g.  of  glucose  per  liter,  the 
glucose  concentration  may  be  expressed  as  50/180  or  5/18ilf.  Con- 
versely, if  the  concentration  of  sucrose  in  a  medium  is  stated  to  be  0.15ilf, 
the  weight  of  sucrose  is  0.15  X  342  or  51.3  g.  per  liter.  These  conversions 
imply  that  the  molecular  weight  is  known  or  can  be  calculated.  In  pre- 
paratory work  compounds  are  weighed  on  a  balance  as  grams,  not  as 
moles,  and  unless  the  interpretation  of  the  results  demands  conversion  to 
moles,  it  is  better  to  record  the  weights  than  to  convert  these  data  to 
derived  units.  The  mole  and  molar  solutions  are  particularly  useful  in 
dealing  with  non-ionizing  compounds. 

Another  derived  unit,  the  equivalent,  is  frequently  used  to  express  the 
concentration  of  ionized  compounds.  An  equivalent  is  the  atomic  weight 
of  an  ion  expressed  in  grams  divided  by  the  valence  of  the  ion.  If  an  ion 
is  composed  of  more  than  one  atom,  the  ion  weight  is  computed  by  adding 
together  the  atomic  weights.  It  is  important  to  remember  that,  if  an 
element  has  more  than  one  valence,  the  equivalent  weight  depends  upon 
the  valence.  An  equivalent  of  ferrous  (Fe++)  ion  is  55.8/2  or  27.9  g., 
while  an  equivalent  of  ferric  (Fe+++)  ion  is  55.8/3  or  18.6  g.  A  normal 
solution  (A^)  is  one  which  contains  one  equivalent  in  a  liter  of  solution. 
In  dealing  with  small  amounts  it  is  convenient  to  use  milliequivalents  or 
microequivalents. 

In  preparing  a  series  of  media  for  the  purpose  of  comparing  the  growth 
of  a  fungus  on  different  nitrogen  sources,  the  nitrogen  content  of  the  media 
should  be  equal.  If  urea,  CO(NH2)2,  and  aspartic  acid,  HOOC — CH2 — 
CH(NH2) — COOH,  are  used,  it  is  obvious  that  different  weights  of  these 
nitrogen  sources  must  be  used  if  the  media  are  to  contain  equal  amounts  of 
nitrogen.     Whenever  media  are  modified  by  replacing  one  compound  by 


1 


20 


PHYSIOLOGY  OF  THE  FUNGI 


another,  it  should  be  done  in  such  a  way  that  the  same  amount  of  the 
essential  element  is  present  in  all  the  media.  If  this  is  not  done,  the  basis 
upon  which  the  replacement  was  made  should  be  stated.  If  25  g.  of 
glucose,  C6H12OC  is  replaced  by  25  g.  of  sucrose,  C12H22O11,  it  should  be 
realized  that  the  carbon  contents  of  the  two  media  are  different.  It  is 
frequently  difficult  or  impossible  to  find  out  from  some  papers  in  the 
literature  how  substitutions  in  the  media  were  made. 

Table  3.     A  Compakison  of  Two  Synthetic  Media  upon  the  Basis  of  Amounts 

OF  Essential  Elements  and  Compounds  Present  in  One  Liter 

Both  media  were  made  with  double-distilled  water. 


Glucose-asparagine  * 

Sucrose-ammonium  nitrate  f 

Element  or 
compound 

Unit  of 
meas. 

Source 

Unit  of 
meas. 

Source 

c                  

G. 

4.0 

0.427 

0.049 

0.065 

0.287 

0.228 

Mg. 

0.2 
0.2 
0.1 

^g 

100 

5 

D-Glucose,  10  g. 
L-Asparagine,  2  g. 
MgS04-7H.,0,  0.5  g. 
MgS04-7H.20,  0.5  g. 
KH2PO4,  1.0  g. 
KH,P04,  1.0  g. 

As  sulfate 
As  sulfate 
As  sulfate 

G. 

21.4 
0.720 
0.025 
0.032 
0.125 
0.062 

Mg. 

0.3 

0.3 

0.075 

0.075 

0.02 

0.02 

Sucrose,  50  g. 

N 

NH4NO3,  2.06  g. 

Mg 

MgS04-7HoO,  0.25  g. 

S 

MgS04-7H,0,  0.25  g. 

K           

K2HPO4,  0.35  g. 

P 

K2HPO4,  0.35  g. 

Fe 

As  chloride 

Zn 

As  chloride 

Mn 

Cu 

As  chloride 
As  chloride 

Mo 

As  chloride 

Ga   

As  chloride 

Thiamine 

hydrochloride 

Biotin 

*  Medium  5,  Suggested  Laboratory  Exercises, 
t  Steinberg,  1941. 

Finally,  it  should  be  noted  that  the  common  practice  of  using  one  com- 
pound as  the  source  of  two  essential  elements  does  not  permit  perfect 
freedom  in  adjusting  the  composition  of  a  medium.  If  magnesium  sulfate 
heptahydrate  is  used  to  supply  both  magnesium  and  sulfur,  it  is  obvious 
that  the  ratio  Mg/S  is  fixed.  If  it  is  desired  to  vary  the  amounts  of 
magnesium  and  sulfur  independently,  it  is  necessary  to  use  different  com- 
pounds of  magnesium  and  sulfur;  e.g.,  magnesium  chloride  and  sodium 
sulfate.     This  practice  introduces  other  elements  into  the  medium. 


CULTURE  MEDIA  21 


COMPARISON  OF  MEDIA 


Media  differ  only  in  constituents  and  amounts  used.  It  is  desirable  to 
be  able  to  compare  media  in  some  uniform  way.  To  do  this,  it  is  neces- 
sary to  know  not  only  the  amounts  of  the  elements  present,  but  also  the 
compounds  in  which  these  elements  occur.  A  comparison  of  two  syn- 
thetic media  is  given  in  Table  3. 

From  Table  3  it  will  be  noted  that  these  media  contain  the  same  essen- 
tial elements.  Copper,  molybdenum,  and  gallium  do  not  appear  in  the 
composition  of  the  glucose-asparagine  medium,  but  it  should  not  be  con- 
cluded that  these  elements  were  not  present,  since  only  c.p.  chemicals 
were  used  to  prepare  this  medium.  Stout  and  Arnon  (1939)  note  that  a 
distinction  must  be  made  between  ordinary  chemical  purity  and  biological 
purity.  This  will  be  considered  in  detail  in  Chap.  5.  The  two  features 
which  make  these  media  quite  distinct  are  the  different  sources  of  carbon 
and  nitrogen  used  and  the  addition  of  two  vitamins  to  the  glucose- 
asparagine  medium.  The  latter  medium  is  suitable  for  the  growth  of 
more  species  of  fungi  than  is  the  sucrose-ammonium  nitrate  medium. 

SUMMARY 

Fungi  secure  food  and  energy  from  the  substrates  upon  which  they  live 
in  nature.  In  order  to  culture  fungi  in  the  laboratory,  it  is  necessary  to 
furnish  in  the  medium  those  essential  elements  and  compounds  they 
require  for  the  synthesis  of  their  cell  constituents  and  for  the  operation  of 
their  life  processes.  The  synthetic  abilities  of  fungi  differ.  Some  fungi 
are  unable  to  s5Tithesize  certain  key  compoimds  that  they  require  and 
must  obtain  them  from  the  medium  upon  which  they  grow.  All  the  fungi 
require  much  the  same  essential  elements  but  differ  widely  in  their  ability 
to  utilize  compounds  in  which  these  elements  occur.  There  is  no  uni- 
versal natural  substrate  or  artificial  medium  upon  which  all  fungi  will 

grow. 

On  the  basis  of  composition  there  are  three  general  types  of  media: 
natural  media,  which  are  composed  entirely  of  natural  products;  semi- 
synthetic media,  which  are  composed  in  part  of  natural  substances;  and 
synthetic  media,  which  are  of  kno^\^^  composition.  Natural  media  are 
most  useful  for  routine  work,  while  synthetic  media  and,  to  a  limited 
extent,  semisynthetic  media  are  used  to  investigate  the  nutritional 
requirements  of  the  fungi.  Media  differ  only  with  respect  to  constituents 
and  concentrations. 

The  compounds  and  the  amounts  used  in  preparing  a  medium  must  be 
specified  exactly.  Media  should  be  designated  by  naming  the  carbon 
and  nitrogen  sources  used,  e.g.,  glucose-asparagine  medium.     The  use 


22  PHYSIOLOGY  OF  THE  FUNGI 

of  proper  names  to  designate  the  composition  of  a  medium  should  be 
avoided. 

The  selection  of  a  suitable  medium  depends  upon  the  fungus  under 
study  and  the  purpose  of  the  experiment.  Not  all  media  are  equally 
suitable  for  all  fungi,  nor  is  one  medium  suitable  for  a  complete  physio- 
logical study  of  one  fungus. 

REFERENCES 

Brefeld,   O.:  Botanische   Untersuchungen  liber  Schimmelpilze,  Heft,  IV  Verlag 

Arthur  Felix,  Leipzig,  1881. 
Buchanan,  R.  E.,  and  E.  I.  Fulmer:  Physiology  and  Biochemistry  of  Bacteria, 

Vol.  I,  The  Williams  &  Wilkins  Company,  Baltimore,  1928. 
Day,  D.:  Thiamin  content  of  agar.  Bull.  Torrey  Botan.  Club  69:  11-20,  1942. 
Englis,  D.  T.,  and  D.  Hanahan:  Changes  in  autoclaved  glucose,  Jour.  Am.  Chevi. 

Soc.  67 :  51-54,  1945. 
Fellows,  H.:  Nitrogen  utilization  by  Ophiobolus  graminis,  Jour.  Agr.  Research  53: 

765-769,  1936. 
*Hansen,  H.  N.,  and  W.  C.  Snyder:  Gaseous  sterilization  of  biological  materials  for 

use  as  culture  media,  Phytopathology  37:  369-371,  1947. 
Hill,  E.  G.,  and  A.  R.  Patton:  The  Maillard  reaction  in  microbiological  assay. 

Science  105 :  481-482,  1947. 
HiTCHENS,  A.  P.,  and  M.  C.  Leikind:  The  introduction  of  agar-agar  into  bacteri- 
ology. Jour.  Bad.  37:  485-493,  1939. 
Leonian,  L.  H.,  and  V.  G.  Lilly:  Studies  on  the  nutrition  of  fungi.     IV.  Factors 

influencing  the  growth  of  some  thiamin-requiring  fungi.  Am.  Jour.  Botany  27: 

18-26,  1940. 
Maillard,  L.  C:  Action  des  acides  amines  sur  les  sucres;  formation  des  melanoldines 

par  voie  methodique,  Conipt.  rend.  acad.  sci.  154:  66-68,  1912. 
Margolin,  A.  S.:  The  effect  of  various  carbohydrates  upon  the  growth  of  some 

fungi,  thesis.  West  Virginia  University,  1942. 
Mulder,  E.  G.  :  On  the  use  of  micro-organisms  in  measuring  a  deficiency  of  copper, 

magnesium  and   molybdenum  in  soils,   Antonie  van  Leeuwenhoek  6 :   99-109, 

1939-1940. 
Neuberger,  a.,  and  F.  Sanger:  The  nitrogen  of  the  potato,  Biochem.  Jour.  36: 

662-671,  1942. 
Pigman,  W.  W.,  and  R.  M.  Goepp,  Jr.:  Chemistry  of  the  Carbohydrates,  Academic 

Press,  Inc.,  New  York,  1948. 
Raulin,  J.:  Etudes  chimiques  sur  la  v^g6tation,  Ann.  sci.  nat.,  Ser.  V,  11:  93-229, 

1869. 
Rawlins,  T.  E.:  Phytopathological  and  Botanical  Research  Methods,  John  Wiley 

&  Sons,  Inc.,  New  York,  1933. 
RiKER,  A.  J.,  and  R.  S.  Riker:  Introduction  to  Research  on  Plant  Diseases,  John  S. 

Swift  Co.,  St.  Louis,  1936. 
*RoBBiNS,  W.  J.:  Growth  substances  in  agar.  Am.  Jour.  Botany  26:  772-778,  1939. 
Robbins,  W.  J.,  and  M.  B.  Schmitt:  Effect  of  cotton  on  the  germination  of  Phyco- 

myces  spores,  Bull.  Torrey  Botan.  Club  72 :  76-85,  1945. 
Stanier,  R.  Y.  :  The  Cytophaga  group:  a  contribution  to  the  biology  of  myxobacteria, 

Bact.  Revs.  6 :  143-196,  1942. 
Steinberg,  R.  A.:  Relation  of  carbon  nutrition  to  trace-element  and  accessory 

requirements  of  Aspergillus  niger,  Jour.  Agr.  Research  59:  749-763,  1939. 


CULTURE  MEDIA  23 

Steinberg,  R.  A.:  Sulfur  and  trace  element  nutrition  of  Aspergillus  niger,  Jour. 

Agr.  Research  63:  109-127,  1941. 
Stout,  P.  R.,  and  D.  I.  Arnon:  Experimental  methods  for  the  study  of  the  role  of 

copper,  manganese,  and  zinc  in  the  nutrition  of  higher  plants,  Am.  Jour.  Botany 

26:  144-149,  1939. 
*■  White,  N.  H.:  Physiological  studies  of  the  fungus  Ophiobolus  graminis  Sacc,  Jour. 

Council  Sci.  Ind.  Research  14:  137-146,  1941. 


CHAPTER  3 
GROWTH 

Growth  may  be  considered  either  as  an  increase  in  cell  number  or  as  an 
increase  in  mass.  Usually  both  these  processes  are  concurrent  in  the 
phenomenon  called  growth.  To  a  limited  degree,  fungus  cells  may 
divide  and  form  new  cells  without  an  increase  in  mass.  A  spore  may 
germinate  in  distilled  water  and  give  rise  to  a  germ  tube,  but  in  the 
absence  of  nutrients  this  process  soon  stops.  A  few  cell  divisions  exhaust 
the  reserve  material  originally  present  in  the  spore,  and  growth  soon 
ceases  unless  these  new  cells  obtain  nutrients  from  the  external  environ- 
ment. Under  certain  conditions  fungus  cells  may  increase  their  store  of 
reserve  materials,  and  thus  their  mass,  without  an  increase  in  cell  number, 
but  this  process  is  also  limited.  Growth,  excluding  the  limited  meanings 
given  above,  involves  an  increase  in  both  the  number  and  the  mass  of  cells. 

This  definition  of  growth  neither  ''explains"  the  processes  involved  nor 
indicates  their  complexity.  Rahn  (1932)  has  expressed  doubt  that  we 
will  ever  fully  understand  the  process  of  growing.  A  yeast  cell  which 
buds  and  produces  a  daughter  cell  illustrates  one  of  the  striking  features 
of  growth :  growth  involves  duplication.  From  a  dozen  or  so  simple  chemi- 
cal substances  present  in  the  medium  the  parent  cell  synthesizes  at  least 
a  portion  of  the  protoplasm  of  the  daughter  cell.  The  daughter  cell  has 
the  same  genetic  constitution  as  the  parent  cell,  and  thus  a  duplication  of 
genes  is  a  feature  of  cell  multiplication.  The  compounds  which  comprise 
protoplasm,  enzymes,  genes,  and  other  substances  are  extraordinarily 
complex.  Our  meager  knowledge  concerning  the  chemical  architecture 
of  these  substances  only  confirms  this  view.  In  the  synthesis  of  such 
compounds  we  may  assume  that  the  chemical  reactions  which  produce 
them  are  perfectly  timed  and  coordinated,  for  no  series  of  uncorrelated 
reactions  could  produce  such  compounds. 

The  growth  processes  of  the  filamentous  fungi  are  still  more  complex 
than  those  of  yeast,  because  of  greater  differentiation  in  structure.  In 
those  species  of  fungi  which  produce  aerial  mycelium  these  parts  are 
nourished  through  the  mycelium  in  contact  with  the  medium.  This 
involves  translocation  of  nutrients  over  considerable  distances.  This  is 
especially  true  of  sporangiophores  and  aerial  fruit  bodies.  The  develop- 
ment of  fruiting  structures  and  spores  is  growth,  in  that  the  formation  of 
new  cells  is  involved.     The  formation  of  fruit  bodies  in  many  species 

24 


GROWTH  25 

takes  place  at  the  expense  of  reserve  materials  and  protoplasm  formed  by 
and  stored  in  the  vegetative  mycelium. 

PHASES  OF  GROWTH 

Growth  in  the  fungi,  as  in  other  organisms,  follows  a  definite  pattern. 
The  way  this  development  takes  place  depends  upon  the  species  and  the 
environmental  and  nutritional  conditions.  In  the  present  discussion,  it 
will  be  assumed  that  the  external  conditions  are  favorable  and  that  growth 
takes  place  in  a  limited  volume  of  medium. 

Unicellular  organisms.  The  bacteriologists  have  long  been  interested 
in  the  mathematical  analysis  of  the  phenomenon  of  growth.  The  student 
is  referred  to  Buchanan  and  Fulmer  (1928)  and  to  Rahn  (1932,  1939)  for 
further  information  on  this  subject.  Among  the  fungi,  the  yeasts  have 
somewhat  the  same  type  of  development  as  the  bacteria.  Since  bacteria 
multiply  by  fission  and  the  yeasts  (except  Schizosaccharomyces)  by 
budding,  we  cannot  expect  the  growth  pattern  of  yeasts  to  fit  exactly  the 
same  formulas  which  have  been  developed  for  bacteria.  But,  in  a  general 
way,  yeasts  follow  closely  the  phases  of  growth  shown  by  bacteria. 
These  phases  of  growth  are  as  follows:  (1)  Stationary  phase.  When  cells 
are  inoculated  into  a  medium,  there  is  a  period  of  time  following  inocu- 
lation when  there  appears  to  be  no  change  in  number.  The  stationary 
phase  may  be  long  or  short  depending  upon  the  age  and  vigor  of  the 
inoculum,  the  medium,  and  other  factors.  (2)  Phase  of  accelerated  growth. 
Not  until  cell  division  is  established  and  new  protoplasm  is  being  formed 
from  the  constituents  of  the  medium  may  growth  be  considered  as  begun. 
This  phase  is  characterized  by  an  increase  in  the  rate  of  cell  division,  i.e., 
the  generation  time  is  decreasing.  (3)  Exponential  or  logarithmic  phase. 
This  phase  is  clearly  defined  for  bacteria  and  approached  by  yeasts.  It 
is  characterized  by  a  constant  generation  time.  If  the  logarithms  of  the 
cell  numbers  are  plotted  against  time,  the  curve  is  a  straight  line.  (4) 
Phase  of  declining  acceleration.  As  the  nutrients  become  exhausted,  or 
as  toxic  by-products  accumulate,  the  average  generation  time  increases. 
A  combination  of  these  and  other  factors  results  in  a  lessened  rate  of 
growth.  If  fresh  medium  were  continuously  supplied  and  toxic  by-prod- 
ucts removed,  it  is  possible  that  this  phase  would  never  be  attained.  (5) 
Maximum  stationary  phase.  This  marks  the  attainment  of  maximum 
weight,  or  numbers  of  living  cells.  It  is  quite  likely  that  the  death  of  old 
cells  is  balanced  by  new  growth.  The  duration  of  this  phase  is  dependent 
upon  the  organism  and  upon  the  composition  of  the  medium  at  this  time. 
(6)  Phase  of  decline  or  autolysis.  Sooner  or  later,  following  attainment  of 
maximum  development,  autolysis  sets  in.  As  the  cells  die,  the  cellular 
enzymes  begin  to  digest  the  various  cell  constituents.  Only  the  more 
resistant  portions  of  the  cell  remain.     Microscopic  examination  at  this 


26  PHYSIOLOGY  OF  THE  FUNGI 

time  reveals  that  many  cells  are  devoid  of  protoplasm.  It  is  quite  possi- 
ble that  some  of  the  materials  released  by  autolysis  are  used  by  the 
remaining  living  cells. 

Filamentous  fungi.  With  exception  of  the  third  phase  of  growth  dis- 
cussed above,  the  filamentous  fungi  follow  the  same  order  of  development 
as  the  yeasts.  The  most  obvious  difference  between  the  filamentous 
fungi  and  unicellular  organisms  is  the  failure  to  attain  an  exponential  rate 
of  growth.  Usually,  the  exponential  phase  is  replaced  by  a  more  or  less 
linear  phase  of  growth.  Emerson  (1950)  found  a  straight-line  relation 
between  the  cube  root  of  the  weight  of  mycelium  produced  by  Neurospora 
crassa  grown  in  nonagitated  liquid  medium  and  the  time  of  incubation. 
This  relation  held  for  three  surface-volume  ratios.  A  comparison  of  the 
linear,  logarithmic,  and  cube-root  growth  curves  indicates  that  this  fungus 
has  a  cube-root  phase  of  growth  during  the  interval  when  the  linear  graph 
is  concave  upward.  Growth  in  the  filamentous  fungi  is  limited  to  the 
tips  of  the  hyphae.  The  influence  of  neighboring  cells  which  compete  for 
nutrients  is  a  much  more  important  factor  in  the  growth  of  filamentous 
fungi  than  in  submerged  unicellular  organisms.  In  unagitaged  cultures  a 
portion  of  the  mycelium  is  usually  aerial  at  some  stage  of  growth.  The 
aerial  mycelium  derives  its  nutrients  from  the  submerged  cells,  which 
involves  the  transport  of  these  substances  over  some  distance. 

RATE  OF  GROWTH 

To  study  growth,  it  is  necessary  to  consider  both  the  rate  and  amount  of 
production  of  cells  formed  during  incubation.  The  average  rate  of 
growth  is  obtained  by  measuring  the  amount  of  growth  at  two  intervals  of 
incubation  and  dividing  the  difference  by  the  time  interval.  If  the 
weight  of  a  fungus  colony  increased  from  50  to  98  mg.  between  the  fourth 
and  sixth  days  of  incubation,  the  average  rate  of  growth  is  24  mg.  per  day, 
or  1  mg.  per  hr.  In  experimental  work,  measurements  of  growth  should 
be  made  sufficiently  often  during  the  period  of  incubation  so  that  a 
smooth  graph  (growth  curve)  can  be  plotted  from  the  data.  The  inter- 
vals between  measurements  of  growth  may  be  as  short  as  1  day  for  a 
rapidly  growing  fungus  and  as  long  as  a  week  for  species  which  grow 
slowly.  The  rate  of  growth  at  any  time  may  be  determined  by  finding 
the  slope  (tangent)  of  the  curve.  The  growth  rates  of  fungi  differ,  as  is 
illustrated  in  Fig.  2. 

Since  growth  is  a  process  which  takes  place  in  time,  it  can  be  studied 
only  by  making  many  growth  measurements  during  the  period  of  incuba- 
tion. Such  a  study  is  not  complete  until  the  phase  of  autolysis  is  attained. 
Much  of  the  information  in  the  literature  is  incomplete  because  growth 
was  measured  only  at  one  time.  Many  of  the  potentialities  of  the  fungi 
can  be  discovered  only  by  prolonged  observation. 


GROWTH 


27 


WAYS  OF  MEASURING  GROWTH 

The  discussion  of  phases  of  growth  presupposes  methods  of  measuring 
growth.  In  choosing  a  method  of  measuring  growth,  or  any  other  physio- 
logical process,  the  accuracy  and  type  of  information  desired  must  be  kept 
in  mind.  For  some  purposes  the  simplest  methods  are  satisfactory;  for 
others  the  most  accurate  methods  should  be  chosen. 

Visual  inspection.  The  simplest  way  to  measure  growth  is  by  inspec- 
tion and  comparison.  The  value  of  this  method  lies  in  the  speed  with 
which  growth  measurements  are  made.     Elaborate  equipment  is  not 


400 


300 


•  200 


T5 

'o 

S 


too 


16 


18 


20 


4  6  8  10  12  14 

Doys  of  incubation 

Fig.  2.     Growth  of  four  fungi  under  the  same  conditions,  in  25  ml.  of  liquid  glucose- 
casein  hydrolysate  medium  at  25°C. 


needed,  as  test  tubes  and  Petri  dishes  are  satisfactory  culture  vessels. 
This  method  has  the  further  advantage  that  the  same  cultures  may  be 
kept  under  observation.  It  is  frequently  the  method  of  choice  for  pre- 
liminary experiments,  for  the  very  appearance  of  the  mycelium  is  a  clue 
to  the  amount  of  growth.  Growth  under  varying  conditions  may  be 
compared  if  some  condition  is  used  as  a  standard  for  comparison  (see 
Suggested  Laboratory  Exercises).  It  is  obvious  that  a  great  deal  of  sub- 
jective judgment  enters  into  this  method  of  estimating  growth,  but  it  is 
veiy  useful  where  fine  distinctions  are  not  required. 

Linear  growth.  A  second  widely  used  method  of  measuring  growth 
consists  in  growing  fungi  in  Petri  dishes  and  measuring  either  the  diameter 
or  the  area  of  the  colony.     This  is  a  useful  method  in  some  instances  but 


28 


PHYSIOLOGY  OF  THE  FUNGI 


almost  useless  in  others.  At  least  these  measurements  can  be  made  in  an 
objective  way.  In  this  method,  the  diameter,  radius,  or  area  of  a  colony 
is  used  to  express  the  amount  of  growth,  while  the  daily  increase  repre- 
sents the  rate  of  growth.  It  is  obvious  that  this  method  neglects  the 
thickness  of  the  colony.  Worley  (1939)  has  proposed  to  take  the  thick- 
ness of  the  mycelium  into  account  when  growth  is  measured  by  this 
method.  Such  measurements  are  difficult  and  neglect  the  mycelium 
buried  in  the  agar.  The  rate  of  linear  growth  of  some  fungi  has  little 
relation  to  the  composition  of  the  medium.  The  rapid  extension  of 
mycelium  on  water-agar  medium  may  serve  as  a  familiar  example. 

It  has  been  frequently  assumed  that  fungi  grow  at  a  constant  rate  when 
maintained  under  constant  environmental  conditions.  This  assumption 
is  not  necessarily  true,  for  the  growth  of  Aspergillus  rugulosiis  and  many 
other  fungi  is  self-limited  under  cultural  conditions.  Two  factors  may 
contribute  to  cause  nonuniform  rates  of  growth:  (1)  the  change  in  con- 
centration of  nutrients  due  to  diffusion  and  utilization;  (2)  the  excretion 
of  inhibitory  metabolic  products  into  the  medium. 

The  same  fungus  may  have  a  constant  rate  of  growth  at  one  tempera- 
ture and  not  at  another.  The  rate  of  growth  is  frequently  not  constant 
when  fungi  are  cultured  at  temperatures  higher  than  optimum.  Fawcett 
(1921)  found  the  rate  of  growth  of  Phytiacystis  citrophthora,  Phytophthora 
terrestris,  Phoviopsis  citri,  and  Diplodia  natalensis  to  decrease  with  time 
when  these  fungi  were  cultivated  above  the  optimum  temperature.  Some 
of  Fawcett's  data  which  illustrate  this  phenomenon  are  given  in  Table  4. 

Table  4.     The  Effect  of  Temperature  upon  the  Rate  of  Growth 

OF  Three  Fungi 
The  daily  increase  in  the  average  radius  of  the  colonies  is  given  in  milhmeters, 
(From  the  data  of  Fawcett,  Univ.  Calif.  {Berkeley)  Pubs.  Agr.  Sci.  4,  1921.) 


Phytiacystis 

Phytophthora 

Phomopsis 

Days  of 
incubation 

citrophthora 

terrestris 

citri 

23.5°C. 

31.0°C. 

30.0°C. 

35.5°C. 

27.5°C. 

32.0°C. 

1 

5.4 

6.3 

5.5 

4.8 

4.6 

0.9 

2 

10.0 

5.5 

13.8 

4.2 

8.0 

0.3 

3 

10.2 

3.5 

13.3 

2.6 

8.0 

0.2 

X 

10.5 

1.5 

13.2 

2.5 

8.5 

0 

5 

10.5 

0.5 

10.9 

0 

8.5 

0 

If  the  rate  of  growth  under  a  given  condition  does  not  change  with 
time,  this  method  is  useful  and  simple.  It  permits  observation  of  the 
same  culture  for  the  duration  of  the  experiment.  Ryan  et  al.  (1943)  have 
proposed  the  use  of  an  ingenious  growth  tube  in  which  linear  growth  can 


GROWTH  29 

be  measured  with  ease  and  accuracy.     This  growth  tube  is  illustrated  in 
Fig.  3. 

These  authors  (Ryan  et  al,  1943)  found  the  rate  of  linear  growth  of 
Ncurospora  sitophila  in  such  a  growth  tube  to  be  constant  for  200  hr. 
The  growth-tube  method  has  been  used  to  study  the  effect  of  temperature, 
pH,  vitamin  content,  and  other  variables  upon  Neurospora.  These 
special  tubes  have  another  advantage  over  Petri  dishes  in  that  cultures 
are  well  protected  from  contamination.  The  same  culture  may  be 
exposed  to  a  variety  of  environmental  conditions  such  as  hght  and  tem- 
perature. These  tubes  have  the  disadvantage  that  it  is  more  difficult  to 
remove  mycelium  or  fruit  bodies  for  examination.  In  addition,  aeration 
may  be  poor  and  become  a  limiting  factor  for  some  fungi. 


Fig.  3.     Growth  tube  patterned  after  those  described  by  Ryan,  Beadle,  and  Tatum 
{Am.  Jour.  Botany  30:  784-799,  1943)  for  measuring  linear  growth. 

Dry  weight.  By  weighing  the  mycelium  and  spores  produced,  an 
accurate  and  objective  measure  of  growth  is  obtained.  For  precise  work 
it  is  the  method  of  choice.  Where  any  significant  weight  of  spores  is  pro- 
duced, either  Gooch  or  Alundum  crucibles  may  be  used  to  collect  both 
mycelium  and  spores.  For  most  purposes  the  mycelium  may  be  filtered 
from  the  culture  medium  by  use  of  a  finely  woven  cloth  and  then  trans- 
ferred to  weighing  bottles  or  small  aluminum  cups.  The  excess  medium 
should  be  removed  by  washing  and  pressing  the  mycelium,  which  is  then 
dried  to  constant  weight  at  80  to  100°C.  After  the  mycelium  is  dry,  it  is 
weighed  on  an  analytical  balance.  It  is  usually  sufficient  to  record  the 
weight  to  the  nearest  milligram. 

Some  fungi  make  better  growth  and  sporulate  more  readily  on  agar 
than  in  liquid  medium.  It  is  desirable  to  have  an  objective  measure  of 
growth  of  agar  cultures.  Fries  (1943)  and  Day  and  Hervey  (1946)  have 
obtained  the  dry  weight  of  cultures  grown  on  agar.  This  technique 
should  be  more  widely  used.  The  mycelium  is  freed  from  agar  by  briefly 
autoclaving  the  cultures,  filtering  off  the  mycelial  mats,  and  washing  with 


:36 


PHYSIOLOGY  OF  THE  FUNGI 


hot  water.  Frequently  the  mat  can  be  removed  from  the  melted  agar 
with  a  pair  of  forceps  instead  of  by  filtering.  Autoclaving  removes  some 
soluble  constituents  from  the  mycelium,  but  if  a  uniform  procedure  is 
adopted,  the  results  are  comparable. 

Measuring  yeast  growth.  The  growth  of  yeasts  may  be  measured  by 
four  methods.  (1)  Yeast  cells  may  be  counted  in  an  aliquot  of  the 
medium  by  the  use  of  a  hemocytometer  or  other  counting  chamber.  The 
method  is  tedious.  (2)  The  volume  of  yeast  cells  in  a  given  volume  of 
medium  may  be  measured  in  special  graduated  centrifuge  tubes.      Yeast 


80r 


3.1 


6.2 


0.8  1 .6 

iig.  fhiomine  per  culture 

Fig.  4.  Direct  comparison  between  diameters  and  dry  weights  of  the  same  10-day- 
old  cultures  of  Ceratostomella  fimbriata  in  the  presence  of  varying  amounts  of  thiamine. 
Cultures  were  grown  in  Petri  dishes  on  25  ml.  of  glucose-casein  hydro lysate  agar 
at  25°C. 

cells  are  large  and  easily  separated  from  the  medium  by  centrifuging. 
This  method  is  less  tedious  than  counting.  (3)  Turbidity  may  be  used  to 
measure  the  amount  of  yeast  growth.  Accurate  determinations  by  this 
method  require  the  use  of  a  photoelectric  photometer.  This  method  is 
rapid  and  sufficiently  accurate  for  many  purposes.  Lindegren  and  Raut 
(1947)  have  cultivated  yeasts  in  colorimeter  tubes  and  have  followed  the 
rate  and  amount  of  growth  for  as  long  as  desired.  (4)  Yeast  cells  may  be 
filtered  under  vacuum,  washed,  dried,  and  weighed.  Selas  porcelain 
crucibles  with  fritted  bottoms  are  suitable.  This  method  is  accurate  but 
somewhat  time-consuming. 

Comparison  of  methods.     It  should  be  clearly  recognized  that   one 
method  of  measuring  growth  may  not  agree  with  another.     This  is  illus- 


GROWTH  31 

trated  by  Fig.  4,  where  two  methods  of  measuring  the  amount  of  growth 
of  Ceratostomella  fimbriata  were  used.  This  figure  demonstrates  that  the 
diameter  of  a  colony  may  be  a  very  poor  measure  of  the  amount  of  growth. 
Fries  (1943)  grew  Ophiostoma  {Ceratostomella)  ulmi  on  agar  medium  and 
measured  the  radii  of  the  colonies  and  also  w^eighed  the  mycelium  after 
removing  the  agar.  After  5  days  the  average  radius  of  cultures  without 
pyridoxine  was  16.3  mm.,  while  the  average  radius  of  cultures  receiving 
pyridoxine  was  12.3  mm;  the  weights  of  mycelium  produced  under  these 
two  conditions  were  5.2  and  18.1  mg.,  respectively.  It  is  clear  from  these 
examples  that  different  methods  of  measuring  growth  do  not  always  give 
comparable  results.  Before  valid  conclusions  can  be  reached,  it  is  neces- 
sary to  use  valid  methods  of  measuring  the  quantities  involved. 

METHODS  OF  PRESENTING  RESULTS 

The  data  obtained  in  a  well-planned  and  carefully  executed  experiment 
have  value  in  themselves,  but  more  frequently  data  are  a  means  to  an  end. 
Experimental  data  form  the  basis  upon  which  conclusions  are  reached  and 
serve  as  a  guide  to  further  investigation.  A  conclusion  is  sound  only  if 
the  data  are  sound.  To  be  of  greatest  value,  data  must  be  presented  in  an 
understandable  manner.  Extensive  data  may  be  presented  either  as 
tables  or  graphs;  each  method  has  certain  advantages. 

Tables.  The  utility  and  conciseness  of  tables  make  them  desirable  for 
many  purposes.  Tables  are  especially  suitable  in  comparing  the  amount 
of  growth  (or  any  other  function  under  study)  of  a  number  of  fungi  under 
standard  conditions  or  under  a  number  of  conditions.  They  give  the 
reader  the  same  basic  and  fundamental  information  available  to  the 
original  investigator.  The  utility  of  such  information  can  be  appreciated 
only  when  one  attempts  to  assess  the  reports  in  the  literature. 

Derived  data,  such  as  ratios  or  percentages,  may  be  needed  for  the  pur- 
poses of  interpretation  and  study,  and  as  such  they  are  entirely  proper. 
However,  the  original  data  from  which  the  derived  data  were  calculated 
should  always  be  published.  The  original  data  frequently  have  values 
which  are  not  perceived  or  considered  by  the  original  investigator. 
Derived  data  as  such  afford  no  clue  as  to  the  original  magnitudes. 
Without  the  original  data  no  comparison  can  be  made  with  other  experi- 
ments, whether  in  the  same  or  other  laboratories.  The  usefulness  of 
many  publications  is  severely  limited  because  the  author  presented  only 
ratios  or  percentages  instead  of  the  original  data.  If  a  datum  represents 
an  average  value,  the  number  of  determinations  upon  which  it  is  based 
should  be  stated.  It  is  desirable  to  indicate  the  range  of  variation  among 
replicates,  or  the  standard  deviation  should  be  given  if  the  number  of 
observations  is  large. 


32  PHYSIOLOGY  OF  THE  FUNGI 

Graphs.  The  significance  of  data  is  frequently  best  appreciated  when 
presented  in  graphical  form.  A  graph  reminds  one  that  growth  is  a  con- 
tinuous function  in  time,  whereas  a  table  may  suggest  a  discontinuous 
process.  Growth  curves  are  especially  suited  to  illustrate  the  rate  and 
amount  of  growth  as  a  function  of  time.  In  Fig.  2  the  growth  curves  of 
four  fungi  illustrate  differences  among  species.  Growth  curves  are 
equally  applicable  to  the  study  of  a  single  species  under  different  condi- 
tions. The  points  representing  the  data  should  be  given,  so  that  the 
reader  may  see  how  closely  the  curve  fits  the  data. 

Three-dimensional  graphs  may  be  used  to  represent  the  relations  among 
three  variables.  Three-dimensional  graphs  take  the  form  of  a  surface. 
Rahn  (1939)  has  given  concise  directions  for  constructing  such  graphs  and 
models.  Schopfer  (1943)  has  used  such  graphs  to  represent  the  growth  of 
Phycomyces  hlakesleeanus  with  respect  to  the  amount  of  thiamine  and 
asparagine  in  the  medium  as  a  function  of  time  of  incubation  (Fig.  33). 

Another  way  of  showing  the  relations  among  the  variables  involves  the 
use  of  a  triangular  graph.  Such  a  presentation  is  effective  if  one  desires, 
for  example,  to  show  the  effect  of  the  concentrations  of  three  constituents 
of  a  medium  upon  growth.  For  examples  of  the  use  of  triangular  graphs 
see  Haenseler  (1921)  and  Pratt  and  Hok  (1946). 

Photographs.  The  presentation  of  experimental  results  is  frequently 
improved  by  the  judicious  use  of  photographs.  Photographs  are 
particularly  useful  in  comparing  the  behavior  of  fungi  under  different 
experimental  conditions.  The  behavior  of  different  species  under 
identical  conditions  may  be  effectively  compared  by  the  use  of  photo- 
graphs. Well-labeled  photographs  also  make  excellent  permanent 
records  of  certain  types  of  experimental  results. 

FACTORS  AFFECTING  GROWTH 

All  the  separate  factors  comprising  the  internal  and  external  environ- 
ment may  affect  either  the  rate  or  the  amount  of  growth,  or  both.  Among 
the  internal  factors  are  the  genetic  constitution  and  the  internal  modifica- 
tions due  to  age  and  to  the  previous  external  environment.  While  more 
is  known  about  the  external  factors  which  affect  growth  than  about  the 
internal  factors,  it  should  always  be  remembered  that  the  external  envi- 
ronment acts  by  modifying  the  internal  environment. 

Internal  factors.  One  species  differs  from  another,  and  even  one  isolate 
of  a  species  may  differ  from  another  in  genetic  composition.  Many 
mutations  have  been  produced  in  the  laboratory  by  the  action  of  X  rays, 
ultraviolet  rays,  and  certain  chemicals  (see  Chaps.  10  and  18).  These 
mutants  of  a  single  species  produced  in  the  laboratory  differ  from  the 
parent  type  in  one  or  more  biochemical  or  morphological  characteristics 
and  thus  correspond  to  the  different  isolates  of  a  species  found  in  nature. 


GROWTH  33 

There  is  no  reason  to  suppose  that  mutants  produced  in  the  laboratory- 
differ  fundamentally  from  those  isolated  in  nature. 

The  potentiahties  of  a  fungus  are  limited  by  its  genetic  constitution. 
The  realization  of  these  potentialities  may  be  denied  or  favored  by  the 
external  environment,  and  only  as  the  environment  is  suitable  do  these 
inherent  factors  find  expression.  Diversity,  rather  than  uniformity,  in 
behavior  among  species  and  isolates  is  the  rule. 

Only  a  small  amount  of  inoculum  is  used  in  most  studies.  It  is  impor- 
tant to  learn  if  the  age,  history,  or  kind  of  inoculum  has  any  effect  on  the 
subsequent  development  of  the  fungus.  All  these  factors  may  influence 
the  rate  and  amount  of  growth  and  other  functions  of  the  fungi.  Young 
and  vigorously  growing  inoculum  is  most  suitable,  since  old  cells  as  a 
general  rule  are  slow  to  start  growth.  Apparently  one  of  the  first  func- 
tions a  cell  loses  is  the  power  of  division.  From  this  standpoint  such  cells 
are  "dead,"  although  they  may  be  still  capable  of  performing  many  vital 
functions,  such  as  respiration.  Difficulty  is  frequently  experienced  in 
making  subcultures  from  old  cultures.  Certain  species  are  difficult  to 
maintain  in  culture  unless  they  are  frequently  subcultured.  In  general, 
these  species  do  not  readily  form  resting  cells.  Among  these  are  various 
species  of  Pythium  and  Phytophthora,  test-tube  cultures  of  Choanephora 
cucurhitarum,  and  others. 

In  experimental  work  of  the  highest  precision  neither  the  temperature 
nor  the  medium  upon  which  the  inoculum  is  grown  may  be  neglected. 
Zikes  (1919)  investigated  the  generation  time  of  six  strains  of  yeast  and 
found  that  the  storage  temperature  of  the  inoculum  affected  the  time 
required  for  cell  division.  These  original  cultures  were  grown  at  8°C.  and 
25°C.,  and  subcultures  were  incubated  over  a  range  of  temperatures. 
When  the  inoculum  which  was  grown  and  stored  at  8°C.  was  subcultured 
at  low  temperatures,  the  generation  time  was  less  than  that  of  the  culture 
grown  and  stored  at  25°C.  At  temperatures  above  25°C.  the  generation 
time  of  the  high-temperature  yeast  was  less  than  that  of  the  low-tempera- 
ture yeast.  In  some  way,  yeast  cells  cultured  over  long  periods  of  time 
at  a  certain  temperature  become  adapted  to  this  temperature,  and  when 
such  cells  are  transferred  to  other  temperatures,  the  influence  of  the 
original  temperature  of  incubation  persists  for  a  time.  It  is  evident  that 
some  change  in  the  internal  environment  has  occurred. 

Comparable  studies  on  the  filamentous  fungi  are  rare.  From  Fawcett's 
data  on  the  rate  of  linear  growth  of  four  citrus  pathogens  it  appears  that 
the  same  phenomenon  takes  place  with  some  filamentous  fungi.  Fawcett 
grew  the  inoculum  at  20°C.,  and  on  subculturing  at  7.5°C.  the  linear  rate 
of  growth  increased  with  time,  as  is  shown  in  Table  5. 

Many  fungi  have  latent  abilities  to  synthesize  various  essential  metab- 
olites.    In  the  virtual  absence  of  these  compounds  in  the  medium  and 


34 


PHYSIOLOGY  OF  THE  FUNGI 


after  a  shorter  or  longer  period  of  incubation,  a  fungus  may  begin  to 
synthesize  these  essential  metabolites,  and  growth  then  takes  place  in  a 
normal  way.  This  is  especially  true  of  the  yeasts  with  respect  to  vitamins. 
Many  fungi  lose  their  pathogenicity  Avhen  cultured  for  a  long  time  on 
laboratory  medium.  Host  passage  frequently  restores  pathogenicity. 
The  indiscriminate  use  of  inoculum  from  a  variety  of  substrates  and  of 
different  ages  may  introduce  unexpected  variation  in  experimental  work 
and  should  be  guarded  against. 

Table  5.     Daily  Increase  (in  Millimeters)  in  Diameter  of  Colonies 

OF  Four  Fungi 
Inoculum  grown  at  20°C.;  subcultures  incubated  at  7.5°C.     (From  the  data  of 
Fawcett,  Univ.  Calij.  {Berkeley)  Pubs.  Agr.  Sci.  i,  1921.) 


Species 


Phythiacystis  citrophthora 
Phytophthora  terrestris . .  . 

Phomopsis  citri 

Diplodia  natalensis 


1st 
day 


0.04 
0.02 
0.01 
0.05 


2d 
day 


0.4 
0.14 
0.16 
1.9 


3d 
day 


0.6 
0.21 
0.83 
2.1 


4th 
day 


0.8 
0.7 
0.9 


5th 
day 


1.2 
0.8 
1.0 


External  factors.  Among  the  external  factors  which  influence  the 
growth  of  fungi,  temperature  plays  an  extremely  important  role.  Tem- 
perature affects  almost  every  function  of  the  fungi.  For  each  fungus 
there  is  a  temperature  below  which  it  will  not  grow,  the  minimum  tem- 
perature. Likewise  there  is  a  temperature  above  which  growth  ceases, 
the  maximum  temperature.  These  two  temperatures  indicate  the  tem- 
perature range  of  an  organism.  A  few  fungi  are  capable  of  growing  below 
0°C.,  but  for  most  species  the  minimum  temperature  is  0  to  5°C.  The 
maximum  temperature  varies  from  27°C.  ior  Phacidium  infestans  (Pehrson, 
1948)  and  Sclerotinia  cameUiae  (Barnett  and  Lilly,  1948)  to  45  or  50°C.  for 
Aspergillus  fumigatus  (Thom  and  Raper,  1945).     The  maximum  tempera- 


Table  6.     Cardinal  Temperatures  for  Various 

Fungi 

Species 

Minimum, 

Optimum, 
°C. 

Maximum, 
°C. 

Citation 

Neurospora  sitophila 

Ceratostomella  pilifera 

C.  ips 

4 

5 

5 

-3 

2 

12.0 
0.5 

36 
25-30 

30 

15 

18-21 

31.5 

25-30 

44 

35 

40 

27 

26 

36.1 

40 

Ryan  et  al,  1943 
Lindgren,  1942 
Lindgren,  1942 

Phacidium  infestans 

Phytophthora  infestans 

P.  terrestris 

Pehrson,  1948 
Crosier,  1933 
Fawcett,  1921 

Various  yeasts 

Zikes,  1919 

GROWTH  35 

ture  is  sometimes  an  important  factor  limiting  the  attack  of  plant 
pathogens. 

The  cardinal  temperatures  of  a  few  fungi  are  given  in  Table  6.  A  more 
extensive  compilation  is  given  by  Wolf  and  Wolf  (1947).  The  character- 
istic effect  of  different  temperatures  on  the  rate  of  growth  of  two  fungi  is 
shown  in  Fig.  5.  Further  examples  may  be  found  in  the  work  of  Lindgren 
(1942). 

Most  reports  on  the  effect  of  light  on  the  fungi  have  been  concerned 
with  reproduction  rather  than  vegetative  growth.  However,  Elfving 
(1890)  found  strong  diffuse  daylight  to  depress  the  growth  of  Penicillium 
glaucum  and  a  species  of  Briarea.  The  amount  of  inhibition  was  least 
when  the  culture  medium  contained  complex  nutrients  such  as  peptone. 
Greater  inhibition  resulted  when  the  media  contained  glucose,  mannitol, 
and  malic  acid.  Scattered  observations  indicate  that  the  depressing 
effect  of  strong  light  may  be  rather  common.  In  the  old  literature  some 
mention  is  made  of  the  favorable  effect  of  light  on  red  yeasts.  The 
sporangiophores  of  Phycomyces  hlakesleeanus  attain  a  greater  length  in 
darkness  than  in  intense  light.  The  role  of  light  in  the  sporulation  of 
some  fungi  is  discussed  in  Chap.  14. 

Conclusive  evidence  that  light  affects  the  amount  of  growth  of  Karlingia 
(Rhizophijlctis)  rosea,  one  of  the  lower  Chytridiales,  was  presented  by 
Haskins  and  Weston  (1950).  This  fungus  when  grown  in  liquid  glucose- 
nitrate  medium  produced  twice  the  amount  of  dry  weight  of  cells  when 
cultured  in  light  than  when  the  cultures  were  kept  in  total  darkness. 
With  the  exception  of  the  factor  of  illumination,  the  experimental  con- 
ditions were  the  same.  Approximately  twice  as  much  glucose  was 
utilized  by  cultures  exposed  to  light  as  those  kept  in  darkness.  On  the 
other  hand,  when  K.  rosea  was  grown  in  a  liquid  cellobiose-nitrate 
medium,  more  growth  resulted  in  total  darkness  than  in  light.  The 
explanation  for  this  behavior  of  K.  rosea  is  not  known. 

The  moisture  requirements  of  fungi  differ.  Most  species  in  nature  live 
on  substrates  which  are  not  saturated  with  water.  The  low  moisture 
content  of  a  substrate  is  often  a  factor  which  limits  the  growth  of  fungi. 
Particularly  is  this  true  of  the  species  which  live  on  wood  or  in  soil.  As  a 
general  rule,  wood  which  contains  less  than  20  per  cent  moisture  is  immune 
to  fungus  decay.  A  difference  of  a  few  per  cent  in  the  moisture  content 
may  determine  whether  a  species  will  be  able  to  grow  or  not.  Lindgren 
(1942)  has  reported  that  Ceratostomella  pilifera,  a  wood-staining  fungus, 
does  not  grow  in  pine  wood  having  a  moisture  content  of  23  per  cent  but 
develops  in  wood  containing  24.5  per  cent  moisture.  The  maximum  rate 
of  penetration  was  attained  on  wood  having  a  moisture  content  of  29  per 
cent  or  more.  Jute  sacking  is  subject  to  fungus  attack  only  if  the  mois- 
ture content  exceeds  17  per  cent. 


36 


PHYSIOLOGY  OF  THE  FUNGI 


In  physiological  studies  dealing  with  high  concentrations  of  nutrients, 
it  is  important  to  distinguish  between  osmosis  and  osmotic  pressure. 
Osmosis  is  the  transfer  of  water  through  a  membrane  permeable  to  water 


160 


120 


^80 

E 


40 


^ 

i"^"^    ^^^ 

\ 

/ 

\ 

\ 

\ 

1 

\ 

\ 

/ 

/ 

- 

10 


20 


Degrees  centigrade 


30 


40 


10  20  30 

Temperature  in  degrees  centigrade 

Fig.  5.  A,  the  effect  of  temperature  on  the  dry  weight  of  mycelium  produced  by 
Glomerella  cingulata  after  5  days  in  25  ml.  of  liquid  glucose-asparagine  medium. 
(Drawn  from  the  data  of  I.  G.  Bennett,  1951.)  B,  the  effect  of  temperature  on  the 
rate  of  linear  growth  of  Neurospora  crassa.  (Courtesy  of  Ryan,  Beadle,  and  Tatum, 
Am.  Jour.  Botany  30 :  785,  1943.) 

but  not  to  the  solute  molecules.  In  simple  systems  water  passes  from  a 
dilute  to  a  more  concentrated  solution.  Osmotic  pressure  is  the  force 
necessary  to  restrain  the  movement  of  water  from  a  dilute  to  a  concen- 
trated solution  through  a  semipermeable  membrane.  The  osmotic  pres- 
sure which  a  solution  is  capable  of  developing  is  a  function  of  the  number 


GROWTH 


37 


of  ions  and  molecules  of  solute  contained  in  a  unit  \'olume  of  solution.  A 
mole  of  a  non-ionized  compound  in  1,000  g.  of  water  at  0°C.  has  an 
osmotic  pressure  of  22.4  atm.  if  separated  from  pure  water  by  a  semiper- 
meable membrane.  For  a  fuller  discussion  of  osmosis  and  osmotic  pres- 
sure the  student  is  referred  to  Gortner  (1949),  Seifriz  (1936),  and  Meyer 
and  Anderson  (1948). 

If  concentration  were  the  sole  factor  which  determines  whether  growth 
is  possible,  all  solutions  having  the  same  osmotic  pressure  would  be 
equally  inhibitory.     Table  7  indicates  that  this  is  not  true. 

Table  7.     Highest  Osmotic  Pressures  (Atmospheres)   op  Solutions  of  Four 
Compounds  in  Which  Various  Fungi  Grew 
(Hawkins,  Jour.  Agr.  Research  7,  1916.) 


Species 

Glucose* 

Sucrose 

Potassium 
nitrate 

Calcium 
nitrate 

Plenodomus  destruens 

Diplodia  tubericola 

Rhizopus  nigricans 

Botrytis  cinerea ....        

58.3 
63.2 
63.2 
63.2 
63.2 

47.4 
42.1 
42.1 
47.4 
47.4 

54.5 
58.8 
27.5 
54.5 
54.5 

33.6 
33.6 

15.9 
27.7 

Ceratoslomella  fimhriata 

19.5 

*  Limiting  concentrations  not  used. 

These  data  and  others  show  that  the  limiting  osmotic  pressure  depends 
upon  the  fungus  and  the  compounds  used.  It  is  difficult  to  evaluate  the 
effects  of  osmotic  pressure  upon  the  fungi,  for  the  cell  membrane  is  per- 
meable to  other  compounds  in  addition  to  water.  Calculations  of  osmotic 
pressure  are  made  by  assuming  that  an  indifferent  semipermeable  mem- 
brane separates  solutions  of  different  concentrations.  The  effect  of 
osmotic  pressure  upon  the  fungi  cannot  be  considered  as  a  simple  physio- 
chemical  process.  However,  the  ability  of  many  fungi  to  grow  in  solu- 
tions having  high  osmotic  pressures  is  advantageous.  Parasitic  fungi 
characteristically  have  a  higher  osmotic  pressure  than  the  cell  sap  of  the 
plants  they  parasitize  (Thatcher,  1939).  For  further  references  to  the 
effect  of  osmotic  pressure  on  fungi,  see  Kroemer  and  Krumbholz  (1931). 

Another  process  involved  in  the  entrance  of  water  into  fungus  cells  is 
imbibition.  Gortner  (1949)  has  defined  imbibition  as  the  process  whereby 
colloidal  substances  such  as  protoplasm  take  up  water,  and  imbibition 
pressure  as  the  pressure  against  which  a  colloid  will  imbibe  liquid. 
Raciborski  (1905)  grew  a  species  of  Torula  in  saturated  lithium  chloride 
(1,000  atm.)  and  Aspergillus  glaucus  in  a  saturated  sodium  chloride 
solution. 

Aside  from  osmotic  effects,  the  concentration  of  the  medium  has  a  great 
effect  on  the  rate  and  amount  of  growth  of  fungi.     The  concentration  of 


38 


PHYSIOLOGY  OF  THE  FUNGI 


nutrients  which  is  most  favorable  for  growth  may  be  poor  in  other 
respects,  e.g.,  for  reproduction.  The  concentration  may  be  varied  in  two 
ways:  (1)  by  dihiting  the  entire  medium,  whereby  the  ratios  among  the 
constituents  remain  unchanged,  and  (2)  by  varying  the  concentration  of 
one  constituent.  These  methods  are  not  equivalent  and  yield  different 
results. 

When  an  entire  medium  is  diluted,  it  might  be  expected  that  the  de- 
crease in  amount  of  mycelium  produced  would  be  directly  proportional 
to  the  amount  of  dilution.  Such  is  not  always  the  case.  When  Chae- 
tomium  convolutum  was  grown  in  full-strength  medium  and  in  medium 
diluted  to  one-fourth  and  one-sixteenth  full  strength,  the  maximum 
weights  of  mycelium  produced  were  220,  75,  and  22  mg.,  respectively 
(Lilly  and  Barnett,  1949).  C  convolutum  grew  most  efficiently  in  the 
most  dilute  medium.  This  principle  appears  to  be  generally  valid  and  is 
also  illustrated  by  Ceratostomella  fimbriata  (Table  57) . 


Table  8. 


The  Effect  of  Different  Volumes  of  Medium  upon  the  Rate  and 
Maximum  Amount  of  Growth  of  Sordaria  fimicola 
Dry  weight  of  mycelium  in  milligrams. 


Days  of 

Ml.  medium  per  250-ml.  Erlenmeyer  flask 

incubation 

6.25 

12.5 

25.0 

50.0 

3 

47 

80 

63 

22 

4 

75 

99 

129 

99 

5 

71 

113 

166 

160 

6 

65 

100 

156 

238 

9 

57 

107 

168 

269 

When  the  concentration  of  one  constituent  in  the  medium  is  changed, 
over  a  certain  range,  the  amount  of  growth  will  be  proportional  to  the 
concentration.  Above  a  certain  concentration  there  will  be  no  further 
increase  in  the  amount  of  growth.  This  is  due  to  the  limiting  concentra- 
tion of  some  other  constituent  in  the  medium.  This  is  the  principle  upon 
which  fungi  are  used  in  vitamin  and  other  assays  (Chap.  10). 

The  maximum  weight  of  mycelium  which  is  obtained  from  a  given  vol- 
ume of  medium  depends  upon  the  type  and  size  of  the  culture  vessels  used. 
The  rate  of  growth  is  also  affected.  These  results  appear  to  be  due 
mainly  to  differences  in  aeration,  and  perhaps  to  a  lesser  degree  to  diffu- 
sion. The  effect  of  depth  of  medium  on  rate  and  amount  of  growth  in  non- 
agitated  cultures  may  be  demonstrated  by  using  a  constant  volume  of 
medium  in  different-sized  flasks,  or  by  varying  the  volume  of  medium  in 
flasks  of  the  same  size.  Data  illustrating  this  latter  condition  are  pre- 
sented in  Table  8.     The  slow  initial  rate  of  growth  when  the  mycelium  is 


GROWTH  39 

entirely  submerged  is  due  to  lack  of  an  adequate  supply  of  oxygen.  The 
efficiency  of  Sordaria  fimicola  in  converting  the  constituents  of  the 
medium  into  mycelium  decreased  as  the  depth  of  the  medium  increased. 
This  fungus  was  less  than  half  as  efficient  when  grown  in  50  ml.  of  medium 
as  when  grown  in  6.25  ml. 

EFFECT  OF  EXTERNAL  FACTORS  ON  MORPHOLOGY 

While  the  study  of  morphology,  as  such,  is  not  within  the  province  of 
physiology,  there  is  a  close  connection  between  these  two  aspects  of 
mycology.  Form  and  function  are  the  two  ways  in  which  the  poten- 
tialities of  organisms  come  to  expression.  The  morphology  of  a  fungus 
may  be  modified  by  environmental  factors  to  such  a  degree  as  to  be 
unrecognizable.  These  changes  in  morphology  may  be  microscopic  as 
well  as  grossly  visible. 

Pasteur  (1879)  noted  that  species  of  Mucor,  when  grown  submerged  in 
liquid  and  in  the  absence  of  air,  assumed  a  yeast-like  form.  Not  only  did 
they  resemble  yeasts,  but  under  these  conditions  they  fermented  sugar  to 
alcohol.  Under  aerobic  conditions  no  detectable  amounts  of  alcohol  were 
formed.  Reproductions  of  Pasteur's  drawings  have  been  published  by 
Foster  (1949). 

When  yeasts  are  cultured  in  liquid  media  and  allowed  to  age  undis- 
turbed, a  film  or  membrane  frequently  covers  the  surface  of  the  liquid. 
Film  formation  frequently  starts  as  a  ring  of  cells  on  the  wall  of  the  flask 
at  the  air-liquid  interface.  The  morphology  of  the  yeast  cells  in  such 
films  is  unusual  in  that  the  cells  are  joined  together  in  filaments.  The 
supply  of  oxygen  must  play  an  important  role  in  the  formation  of  fila- 
ments. The  temperature  range  within  which  film  formation  occurs  varies 
with  the  species  of  yeast  and  is  usually  considerably  less  than  the  tempera- 
ture range  for  growth.  Most  species  of  yeasts  forms  films  only  between 
6  and  30°C.,  although  Zikes  (1919)  found  Monilia  Candida  and  Mycoderma 
cerevisiae  to  form  films  at  37°C.  The  early  literature  on  this  subject  has 
been  summarized  by  La  Far  (1911). 

Nickerson  and  Van  Rij  (1949)  have  reviewed  the  mechanisms  of  fila- 
ment formation  in  yeast  and  conclude  that  the  processes  of  cell  elongation 
and  cell  division  are  controlled  by  different  enzyme  systems.  Appar- 
ently, the  sulfhydryl  enzymes  which  regulate  the  process  of  cell  division 
may  be  inhibited  without  greatly  interfering  with  cell  elongation.  Among 
the  agents  which  inhibit  cell  division  are  cobalt,  iodoacetate,  and  peni- 
cillin. The  effect  of  penicillin  on  Saccharomyces  cerevisiae  is  shown  in 
Fig.  6.  Camphor  and  other  narcotizing  agents  produce  somewhat  the 
same  changes  in  morphology  of  yeast  cells  (Levan,  1947). 

Many  pathogenic  fungi  which  cause  disease  in  man  are  dimorphic. 
These  fungi  are  usually  yeast-like  in  the  host  but  frequently  form  myce- 


40 


PHYSIOLOGY  OF  THE  FUNGI 


Fig.  6.  Saccharomyces  cerevisine,  camera  lucida  drawings  of  cells  from  agar  cultures. 
A,  culture  treated  with  penicillin;  B,  culture  treated  with  penicillin  plus  cysteine. 
(Courtesy  of  Nickerson  and  Van  Rij,  Biochim.  et  Biophijs.  Acta  3:  461-475,  1949. 
Published  by  permission  of  Elsevier  Book  Company,  Inc.) 


iw 


\       .  \*' 


i/^/\   "" 


A  B 

Fig.  7.  The  effect  of  hydrogen-ion  concentration  on  the  morphology  of  cells  of 
Sordaria  fimicola.  A,  rounded  swollen  cells  produced  in  glucose-casein  hydrolysate 
medium  at  initial  pH  3.6.  B,  normal  mycelium  from  the  same  culture  a  few  days 
after  a  drop  of  NaOH  was  added. 

Hum  in  culture.  Blastomyces  dermatitidis  and  B.  hrasiliensis  exhibit 
thermal  dimorphi.sm  (Nickerson  and  Edwards,  1949).  When  these  fungi 
are  cultured  on  certain  media  at  37°C.,  they  are  yeast-like,  while  at  lower 
temperatures  of  incubation  they  form  mycelium.  This  change  in  mor- 
phology is  accompanied  by  changes  in  the  rate  of  respiration  and  type  of 


GROWTH 


41 


Fig.  8.  The  effect  of  environment  on  the  morphology  of  fruit  bodies  of  Forties 
applanatus.  A,  normal  fruit  body  developed  in  nature;  B,  C,  malformed  fruit  bodies 
of  the  same  (?)  fungus  developed  under  water  in  abandoned  coal  mines.  The  "nodes  " 
in  B  are  believed  to  be  caused  by  different  water  levels. 


42  PHYSIOLOGY  OF  THE  FUNGI 

metabolism.  Chemical  agents  may  favor  or  prevent  similar  morphologi- 
cal changes.  Trichophyton  ruhrum  produces  two  metabolic  products  of 
unknown  constitution  which  inhibit  the  transformation  of  Candida 
albicans  to  the  mycelial  form  (Jillson  and  Nickerson,  1948).  The  addi- 
tion of  excessive  amounts  of  inositol  to  the  culture  medium  causes 
Ophiostoma  (Ceratostomella)  muUiannulatum  to  grow  almost  entirely  in 
the  form  of  conidia  (Fries,  1949).  The  morphology  of  the  vegetative 
mycelium  and  sporangia  of  various  species  of  Phytophthora  was  found  to 
depend  upon  the  medium  used  (Leonian,  1925). 

The  form  of  mycelial  growth  of  many  species,  when  grown  on  agar 
media,  is  an  aid  in  identification.  The  colony  form  may  be  altered 
beyond  recognition  when  cultures  are  grown  in  agitated  liquid  medium. 
In  general,  spherical  colonies  or  balls  form  in  agitated  medium.  Burk- 
holder  and  Sinnott  (1945)  investigated  colony  form  of  a  large  number  of 
species  when  subjected  to  agitation. 

The  acidity  of  the  medium  affects  the  size  and  shape  of  the  vegetative 
cells  of  some  fungi.  In  a  medium  so  acid  as  to  allow  only  very  slow 
growth  the  cells  often  become  swollen  or  nearly  spherical  in  shape,  much 
like  chlamydospores,  but  the  wall  remains  thin  (Fig.  7).  This  may  be 
accompanied  by  excessive  branching. 

Unusual  environmental  conditions  often  affect  the  morphology  of  both 
vegetative  and  reproductive  structures.  The  environment  which  exists 
in  coal  mines  is  unnaturally  uniform  with  respect  to  temperature,  mois- 
ture,'and  absence  of  light.  Basidiomycetes  growing  on  old  mine  timbers 
either  fail  to  fruit  or  produce  odd-shaped  sterile  fruit  bodies  (Fig.  8). 

SUMMARY 

Normal  growth  results  in  an  increase  in  cell  number  and  mass.  Limited 
growth  may  result  from  either  of  these  two  processes  alone.  Growth  is  a 
phenomenon  which  requires  time  for  its  various  manifestations.  Growth 
follows  a  pattern  which  differs  from  species  to  species,  but  the  general 
sequence  of  phases  is  much  the  same  for  all  fungi.  Growth  studies  are 
based  upon  measuring  both  the  amount  and  the  rate  of  growth.  The  rate 
and  amount  of  growth  are  controlled  by  the  internal  and  external  environ- 
ment. The  potentialities  of  a  fungus  are  limited  by  its  genetic  constitu- 
tion, but  the  expression  of  these  potentialities  is  controlled  by  external 
factors  such  as  temperature,  light,  composition,  and  concentration  of  the 
medium.  Even  the  size  and  shape  of  the  culture  vessels  used  affect  the 
rate  and  amount  of  growth. 

The  amount  of  growth  can  be  estimated  by  visual  comparison  or  meas- 
ured by  determining  the  diameter  of  a  colony  or  by  harvesting  the  myce- 
lium and  weighing  it  after  drying  to  constant  weight.  The  amount  of 
yeast  growth  may  be  measured  by  counting  the  numbers  of  cells  produced, 


GROWTH  43 

by  centrifuging  and  measuring  the  volume  of  cells,  by  turbidity,  or  by 
weighing.  The  most  direct  way  of  measuring  growth  of  either  yeast  or 
filamentous  fungi  is  by  weighing  the  crop  produced.  The  various  meth- 
ods of  measuring  growth  are  not  strictly  comparable. 

The  morphology  of  a  fungus  may  be  changed  by  environmental  factors 
so  that  it  becomes  unrecognizable.  The  processes  of  cell  elongation  and 
cell  division  are  controlled  by  different  enzyme  systems.  In  some 
instances  it  has  been  possible  to  inhibit  cell  division  without  interrupting 
cell  elongation.  Frequently  a  change  in  physiology  accompanies  a 
change  in  morphology. 

REFERENCES 

*Barnett,  H.  L.,  and  V.  G.  Lilly:  The  interrelated  effects  of  vitamins,  temperature, 
and  pH  upon  vegetative  growth  of  Sclerotinia  camelliae,  At7i.  Jour.  Botany  35: 
297-302,  1948. 

Bennett,  I.  G.:  Thesis,  West  Virginia  University,  1951. 

Buchanan,  R.  E.,  and  E.  I.  Fulmer:  Physiology  and  Biochemistry  of  Bacteria, 
Vol.  I,  The  Williams  &  Wilkins  Company,  Baltimore,  1928. 

Burkholder,  p.  R.,  and  E.  W.  Sinnott:  Morphogenesis  of  fungus  colonies  in  sub- 
merged shaken  culture.  Am.  Jour.  Botany  32:  424-431,  1945. 

Crosier,  W.:  Studies  in  the  biology  of  Phytophthora  infestans  (Mont.)  de  Bary, 
Cornell  Univ.  Agr.  Expt.  Sta.  Mem.  155,  1933. 

Day,  D.,  and  A.  Hervey:  Phycomyces  in  the  assay  of  thiamine  in  agar.  Plant 
Physiol.  21 :  233-236,  1946. 

Elfving,  F.:  Studien  iiber  die  Einwirkung  des  Lichtes  auf  die  Pilze,  Helsingfors 
Central-Druckerei,  Helsingfors,  1890. 

Emerson,  S.:  The  growth  phase  in  Neurospora  corresponding  to  the  logarithmic 
phase  in  unicellular  organisms,  Jour.  Bad.  60:  221-223,  1950. 
*Fawcett,  H.  S.  :  The  temperature  relations  of  growth  in  certain  parasitic  fungi, 
Univ.  Calif.  (Berkeley)  Pubs.  Agr.  Sci.  4:  183-232,  1921. 

Foster,  J.  W.:  Chemical  Activities  of  Fungi,  Academic  Press,  Inc.,  New  York,  1949. 

Fries,  N.:  Die  Einwirkung  von  Adermin,  Aneurin  und  Biotin  auf  das  Wachstum 
einiger  Ascomyceten.     Symbolae  Botan.  Upsalienses  7(2):  1-73,  1943. 

Fries,  N.:  Ophiostoma  multiannulatum  (Hedge,  and  Davids)  as  a  test  object  for  the 
determination  of  pyridoxin  and  various  nucleotide  constituents.  Arkiv  fiir 
Botanik  1:  271-287,  1949. 

Gortner,  R.  a.:  Outlines  of  Biochemistry,  3d  ed.,  John  Wiley  &  Sons,  Inc.,  New 
York,  1949. 

Haenseler,  C.  M.:  The  effect  of  salt  proportions  and  concentrations  on  the  growth 
of  Aspergillus  niger,  Am.  Jour.  Botany  8:  147-163,  1921. 
*Haskins,  R.  H.,  and  W.  H.  Weston,  Jr.:  Studies  in  the  lower  Chytridiales.     I.  Fac- 
tors affecting  pigmentation,  growth,  and  metabolism  of  a  strain  of  Karlingia 
(Rhizophlyctis)  rosea,  Am.  Jour.  Botany  37 :  739-750,  1950. 

Hawkins,  L.  A.:  Growth  of  parasitic  fungi  in  concentrated  solutions,  Jour.  Agr. 
Research!:  255-260,  1916. 

JiLLSON,  O.  F.,  and  W.  J.  Nickerson:  Mutual  antagonism  between  pathogenic 
fungi.  Inhibition  of  dimorphism  in  Candida  albicans,  Mycologia  40:  369-385, 
1948. 

Kroemer,  K.,  and  G.  Krumbholz:  Untersuchung  iiber  Osmophile  Sprosspilze. 
I.  Mitteilung,  Beitrage  zur  Kenntniss  der  Garungsvorgange  und  der  Garungs- 
erreger  der  Trockenbeerenauslesen,  Arch.  Mikrobiol.  2:  352-410,  1931. 


44  PHYSIOLOGY  OF  THE  FUNGI 

La    Far,    F.:    Tochnical    Mycology.     Vol.    II,    Eumycetic    Fermentation    (trans. 

C.  T.  C.  Salter),  Chas.  Griffin  &  Co.,  Ltd.,  London,  1911. 
Leonian,  L.  H.:  Physiological  studies  on  the  genus  Phytoyhthora,  Am.  Jour.  Botany 

12:444-498,  1925. 
Lev  AN,  A.:  Studies  on  the  camphor  reaction  of  yeast,  Heredilas  33:  457-514,  1947. 
Lilly,  V.  G.,  and  H.  L.  Barnett:  The  influence  of  concentrations  of  nutrients, 

thiamin,  and  biotin  upon  growth  and  formation  of  perithecia  and  ascospores  by 

Chaetomium  convolutum,  Mycologia  41 :  186-196,  1949. 
LiNDEGREN,  C.  C,  and  C.  Raut:  The  effect  of  the  medium  on  apparent  vitamin- 
synthesizing   deficiencies   of  microorganisms.     A  direct   relationship  between 

pantothenate  concentration  and  the  time  required  to  induce  the  production  of 

pantothenate-synthesizing  "mutants"  in  yeasts,  Ann.  Missouri  Botan.  Garden 

34 :  75-90,  1947. 
LiNDGREN,  R.  M. :  Temperature,  moisture,  and  penetration  studies  of  wood-staining 

Ceratostomellae  in  relation  to  their  control,  U.S.  Dept.  Agr.  Tech.  Bull.  807, 

1942. 
Meyer,  B.  S.,  and  D.  B.  Anderson:  Plant  Physiology,  D.  Van  Nostrand  Company, 

Inc.,  New  York,  1948. 
NiCKERSON,   W.  J.,  and  G.  A.  Edwards:  Studies  on  the  physiological  bases  of 

morphogenesis.     I.  The  respiratory  metabolism  of  dimorphic  pathogenic  fungi, 

Jour.  Gen.  Physiol.  33:  41-55,  1949. 
NiCKERSON,  W.  J.,  and  N.  J.  W.  van  Rij:  The  effect  of  sulfhydryl  compounds, 

penicillin,  and  cobalt  on  the  cell  division  mechanism  of  yeasts,  Biochim.  et 

Biophys.  Acta  3:  461-475,  1949. 
*Pasteur,  L.:  Studies  on  Fermentation.     The  Diseases  of  Beer,  Their  Causes,  and 

the  Means  of  Preventing  Them  (trans,  from  Etudes  sur  la  biere  by  F.  Faulkner 

and  D.  C.  Robb.),  Macmillan  &  Co.,  Ltd.,  London,  1879. 
Pehrson,  S.  O.:  Studies  of  the  growth  physiology  of  Phacidium  infestans  Karst., 

Physiologia  Plantarum.  1 :  38-56,  1948. 
Pratt,  R.,  and  K.  A.  Hok:  Influence  of  the  proportions  of  KH2PO4,  MgS04,  and 

NaNOs  in  the  nutrient  solution  on  the  production  of  penicillin  in  submerged 

cultures,  Am.  Jour.  Botany  33:  149-156,  1946. 
Raciborski,  M.:  Ueber  die  obere  Grenze  des  osmotischen  Druckes  der  lebenden 

Zelle,  Bull,  internat.  acad.  sci.  Cracovie,  CI.  sci.  math,  et  nat.  7:  461-471,  1905. 
Rahn,  0.:  Physiology  of  Bacteria,  The  Blakiston  Company,  Philadelphia,  1932. 
Rahn,  O.:  Mathematics  in  Bacteriology,  Burgess  Publishing  Co.,  Minneapolis,  1939. 
Ryan,  F.  J.,  G.  W.  Beadle,  and  E.  L.  Tatum:  The  tube  method  of  measuring  the 

growth  rate  of  Neurospora,  Atn.  Jour.  Botany  30:  784r-799,  1943. 
ScHOPFER,  W.  H.:  Plants  and  Vitamins,  Chronica  Botanica  Co.,  Waltham,  1943. 
Seifriz,  W.:  Protoplasm,  McGraw-Hill  Book  Company,  Inc.,  New  York,  1936. 
*Thatcher,  F.  S.:  Osmotic  and  permeability  relations  in  the  nutrition  of  fungus 

parasites.  Am.  Jour.  Botany  26:  449-458,  1939. 
Thom,  C,  and  K.  B.  Raper:  A  Manual  of  the  Aspergilli,  The  Williams  &  Wilkins 

Company,  Baltimore,  1945. 
Wolf,  F.  A.,  and  F.  T.  Wolf:  The  Fungi,  Vol.  II,  John  Wiley  &  Sons,  Inc.,  New 

York,  1947. 
WoRLEY,  C.  L.:  Interpretation  of  comparative  growths  of  fungal  colonies  on  different 

solid  substrata.  Plant  Physiol.  14 :  589-593,  1939. 
ZiKEs,  H.:  Ueber  den  Einfluss  der  Temperature  auf  verschiedene  Funktionen  der 

Hefe,  Cent.  Bakt.,  Abt.  II,  49:  353-373,  1919. 


CHAPTER  4 
ENZYMES  AND  ENZYME  ACTION 

The  fungi,  in  common  with  other  hving  organisms,  possess  tools  or 
reagents  far  more  specific,  more  deHcate,  and  more  powerful  than  those 
available  in  the  laboratory.  The  most  complex  natural  substances  such 
as  proteins,  polysaccharides,  and  Hpoids  are  degraded  into  simpler  com- 
pounds which  are  soluble  in  water.  Fungi  also  synthesize  similar  com- 
plex compounds  from  relatively  simple  molecules.  These  transformations 
are  carried  out  under  such  mild  conditions  of  temperature  and  pressure 
and  in  such  low  concentrations  of  acid  and  alkali  that  it  is  certain  the 
means  used  are  of  a  peculiar  kind.  For  in  the  absence  of  these  special 
agents  formed  by  the  living  organisms,  these  reactions  do  not  take  place 
or  do  so  at  a  very  slow  rate.  These  organic  catalysts  produced  by  living 
organisms  are  called  enzymes.  The  life  processes  of  organisms  are  con- 
trolled and  directed  by  a  complicated  and  interrelated  series  of  enzymes 
or  enzyme  systems  (Dixon,  1949). 

Some  enzymes  formed  by  fungi  are  excreted  and  normally  perform  their 
functions  outside  the  cells  that  produce  them.  These  are  termed  exo- 
enzymes  (extracellular  enzymes),  such  as  cellulase,  amylase,  and  pectinase. 
Exoenzymes  perform  the  functions  of  digestion;  i.e.,  the  degradation  of 
complex  food  materials  into  low-molecular-weight  compounds  which  are 
able  to  enter  the  cell.  After  entering  the  cell,  these  metabolites  are  acted 
upon  by  the  enzymes  within  the  cell.  These  enzymes  are  called  endo- 
enzymes  (intracellular  enzymes). 

Naturally  enough,  exoenzymes  were  recognized  and  studied  first.  In 
the  early  literature,  these  exoenzymes  were  called  unorganized  ferments 
because  of  their  solubility.  In  contrast  to  these  unorganized  ferments  it 
was  recognized  that  other  ferments  (enzymes)  occurred  in  an  insoluble 
organized  form.  These  were  called  organized  ferments.  Pasteur  (1875) 
still  spoke  of  yeast  as  "ferment  alcoohque  ordinaire  du  vin."  Thus,  the 
name  organized  ferment  took  on  a  dual  meaning,  that  of  a  living  organism 
and  the  various  chemical  reactions  caused  by  these  organisms.  In  1878 
Kiihne  suggested  that  the  word  enzyme  be  used  to  replace  the  terms 
organized  and  unorganized  ferments.  Enzyme  is  derived  from  the  Greek 
phrase,  en  zyme,  which  means  in  yeast  or  leaven.  For  excellent  sum- 
maries of  the  historical  development  of  the  relation  between  fermentation 

45 


46  PHYSIOLOGY  OF  THE  FUNGI 

and  the  action  of  microorganisms,  see  Stephenson  (1939)  and  Harden 
(1932). 

It  was  not  until  late  in  the  nineteenth  century  that  Buchner  (1897) 
succeeded  in  releasing  certain  enzymes  from  yeast  cells  and  demonstrating 
that  the  endoenzyme(s)  in  yeast  causing  fermentation  was  also  active 
entirely  apart  from  the  living  yeast  cells.  While  yeast  juice  prepared 
according  to  the  method  of  Buchner  contained  a  variety  of  enzymes,  it 
contained  fermentative  enzymes  never  before  obtained  apart  from  the 
living  cell.  These  enzymes  cleaved  sugar  into  alcohol  and  carbon  dioxide. 
This  was  truly  a  monumental  step  in  the  science  of  enzymes,  for  it 
afforded  a  way  of  studying  "life"  processes  apart  from  the  terrible  com- 
plexity of  the  living  organism.  The  study  of  isolated  enzyme  systems 
has  led  to  important  advances  in  our  knowledge  and  understanding  of 
life  processes;  yet  the  student  should  be  reminded  that  life  is  more  com- 
plex than  its  parts.     Leibowitz  and  Hestrin  (1945)  say: 

...  it  has  become  clear  that  the  risk  involved  in  translating  results  from  lifeless 
to  living  systems  is  a  two-way  one:  not  only  may  mechanisms  which  operate 
in  vivo  be  absent  in  vitro;  mechanisms  may  be  present  in  vitro  and  yet  not  neces- 
sarily function  in  vivo.  In  fermentative  physiology,  as  in  biology  generally, 
selective  and  restrictive  activity  by  the  living  organism  must  always  be  taken  into 
account. 

The  rate  of  many  chemical  reactions  is  changed  by  the  presence  of 
traces  of  substances  which  do  not  appear  to  enter  into  permanent  chemi- 
cal combination  with  the  reactants  and  which  appear  unchanged  when 
the  reaction  has  come  to  equilibrium.  Substances  which  alter  the  rates 
of  chemical  reactions  are  called  catalysts,  and  the  process  catalysis. 
Enzymes  are  catalysts  of  a  very  special  kind,  and  many  of  them  catalyze 
but  a  single  reaction.  For  example,  lactose  reacts  with  water  to  form 
glucose  and  galactose.  Unless  a  catalyst  is  present,  this  reaction  occurs 
at  a  very  slow  rate.  Even  at  100°C.  a  long  time  is  required  for  an  appreci- 
able amount  of  lactose  to  react  with  water.  If,  however,  some  acid  is 
added  to  the  lactose  solution,  the  rate  of  the  reaction  is  greatly  increased, 
varying  in  degree  with  the  amount  and  kind  of  acid  used.  This  same 
reaction  is  catalyzed  w^hen  the  enzyme,  lactase,  produced  by  some  yeasts 
and  certain  other  fungi,  is  added  to  a  solution  of  lactose.  In  general, 
enzymes  are  specific  catalysts.  There  is  no  stoichometric  relation 
between  the  amount  of  catalyst  (acid  or  enzyme)  and  the  amount  of  sub- 
strate decomposed.  Within  limits,  the  amount  of  substrate  decomposed 
per  unit  of  time  is  dependent  upon  the  amount  of  catalyst  present. 

For  a  given  set  of  conditions  there  is  a  position  of  equilibrium  where  the 
rate  of  reaction  of  the  reactants  is  equal  and  opposite  to  the  rate  of  com- 
bination of  the  products.     The  position  of  equilibrium  is  not  changed  by 


ENZYMES  47 

the  presence  of  a  catalyst.  The  same  catalyst  will  effect  synthesis  as  well 
as  decomposition;  the  position  of  equilibrium  as  well  as  the  relative  con- 
centrations of  reactants  and  products  determines  which  reaction  pre- 
dominates. It  is  possible  to  choose  conditions,  in  some  instances,  so  the 
equilibrium  conditions  favor  synthesis.  Bourquelot  (1915)  demonstrated 
a-methylglucoside  was  readily  formed  from  methyl  alcohol  and  glucose  in 
the  presence  of  yeast  juice. 

CLASSIFICATION  OF  ENZYMES 

It  is  more  important  to  classify  enzymes  upon  the  basis  of  function 
rather  than  the  site  of  action  (endo-  and  exoenzymes).  Many  enzymes 
catalyze  reactions  in  which  water  is  either  a  product  (synthesis)  or  a 
reactant  (degradation).  These  enzymes  are  called  hydrolases.  These 
reactions  usually  involve  only  moderate  energy  changes.  Another  class 
of  enzymes,  usually  intracellular,  catalyze  oxidation  and  reduction  reac- 
tions and  reactions  involving  the  scission  (or  formation)  of  carbon-to-car- 
bon linkages.  These  enzymes  are  known  as  desmolyzing  enzymes  and 
include  oxidases,  dehydrogenases,  and  desmolases.  Energy  changes 
involved  in  these  reactions  are  usually  large.  For  more  detailed  classifi- 
cations of  enzymes  see  Gortner  (1949)  and  Sumner  and  Somers  (1947). 

Since  an  enzyme  acts  upon  a  restricted  number  of  compounds,  it  is 
convenient  to  name  enzymes  with  reference  to  the  substrate  acted  upon. 
In  general,  enzymes  are  named  either  by  adding  the  suffix  -ase  to  the 
name  of  the  substrate  or  by  replacing  the  final  syllable  of  the  name  of  the 
substrate  by  this  suffix.  The  following  examples  give  the  substrate  fol- 
lowed by  the  name  of  the  enzyme:  maltose,  maltase;  lactose,  lactase; 
cellulose,  cellulase;  starch  (amylum),  amylase;  protein,  proteinase;  pectin, 
pectinase.  The  suffix  -ase  is  also  used  to  designate  classes  of  enzymes. 
Thus,  esterases  are  members  of  that  group  of  enzymes  which  catalyze  the 
hydrolysis  and  synthesis  of  esters;  oxidases  are  enzymes  which  activate 
oxygen,  and  dehydrogenases  are  enzymes  which  activate  the  hydrogen  of 
various  metabolites.  An  enzyme  may  have  several  names.  The  enzyme 
which  catalyzes  the  hydrolysis  of  sucrose  is  known  also  as  saccharase  and 
invertase.     Amylase  is  also  called  diastase. 

Hydrolases.  The  hydrolases  catalyze  a  wide  variety  of  reactions  in 
which  water  is  either  a  reactant  or  a  product.  Hydrolysis  is  generally 
thought  of  as  a  process  whereby  complex  molecules  react  with  water  to 
form  simpler  substances.  Many  hydrolases  are  exoenzymes  which  func- 
tion by  preparing  the  substrate  for  assimilation.  Among  these  the  follow- 
ing should  be  noted :  cellulase,  amylase,  pectinase,  various  disaccharidases, 
proteinases,  and  peptidases.  Others  are  endoenzymes  (the  same  enzymes 
in  some  instances),  which  catalyze  the  same  or  similar  reactions  within  the 
cells.     It  would  be  expected  that  the  process  of  synthesis  within  the  cells 


48 


PHYSIOLOGY  OF  THE  FUXGI 


would  be  of  much  more  common  occurrence  than  outside  the  cells.  In  the 
medium  the  process  of  degradation  may  be  expected  to  go  more  or  less  to 
completion,  since  the  soluble  products  of  the  reaction  are  assimilated  by 
the  organism  and  hence  equilibrium  is  not  reached.  Within  the  cell, 
however,  the  reverse  may  be  true.  Here,  the  products  of  hydrolysis  may 
accumulate,  a  situation  which  would  tend  to  favor  the  reverse  reaction,  or 
synthesis.  Therefore,  synthesis  within  the  cell  would  be  expected  to 
occur  when  a  plentiful  supply  of  simple  metabolite  molecules  continue  to 
reach  the  cell.  When  few,  if  any,  metabolite  molecules  are  entering  the 
cell,  the  hydrolysis  of  reserve  materials  would  take  place.  These  prod- 
ucts of  hydrolysis  within  the  cell  are  then  used  in  other  metabolic  processes 
until  the  store  of  reserve  material  is  exhausted.  Some  of  these  functions 
are  illustrated  in  scheme  I. 

Scheme  I.     General  Scheme  of  Starch  Utilization 
Outside  the  cell 


Starch 


->-  Maltose 


*-  Glucose 


amylase  ^^■^^  ^-^^  maltase 

Fungus  cells 


many  enzymes 


Carbon  dioxide, 
alcohol  and  other 
products  of  anaerobic 
respiration 


Within  the  cell 


Glucose  -<- 


many  enzymes 


Carbon  dioxide, 
water  and  other 
products  of  aerobic 
respiration 


Glycogen,  or 
other  storage 
products 

Esterases.  These  enzymes  catalyze  the  hydrolysis  of  esters,  an  acid 
and  an  alcohol  being  formed.  The  most  important  natural  esters  are  the 
fats,  which  are  the  glycerol  esters  of  the  long-chain  fatty  acids.  Enzymes 
which  catalyze  the  hydrolysis  of  fats  are  called  lipases.     Both  exo-  and 


ENZYMES  49 

endolipases  are  known.  Many  fungi  store  fat  as  reserve  material,  and 
presumably  the  first  step  in  utilization  is  hydrolysis. 

Phosphatases  are  classified  as  esterases  because  of  the  fact  that  they 
catalyze  the  hydrolysis  of  esters  of  phosphoric  acid.  Phosphorus  is  an 
essential  element  which  enters  into  many  metabolic  processes  and  is  a 
constituent  of  many  physiologically  important  compounds.  Many 
coenzymes  are  esters  of  pyrophosphoric  acid  (thiamine  pyrophosphate, 
and  diphosphopyridine  nucleotide,  DPN),  while  triphosphoric  acid  is  a 
constituent  of  triphosphopyridine  nucleotide,  TPN.  The  synthetic 
capacity  of  the  phosphatases  has  been  rarely  demonstrated.  Other 
enzymes,  phosphorylases,  are  apparently  the  catalytic  agents  active  in 
forming  many  phosphate  esters.  In  many  instances  the  substrates  from 
which  these  esters  are  formed  are  different  from  the  products  of  phos- 
phatase hydrolysis. 

Carbohydrases.  The  enzymes  which  catalyze  the  hydrolysis  of  com- 
plex carbohydrates,  or  polysaccharides,  are  called  carbohydrases.  These 
enzymes  appear  to  be  highly  specific ;  thus  each  of  the  common  disaccha- 
rides  requires  a  different  enzyme  for  hydrolysis.  Sucrase  is  found  in 
many  fungi,  including  the  common  strains  of  Saccharomyces  cerevisiae, 
although  it  is  apparently  absent  in  Schizosaccharomyces  octosporus.  The 
enzyme  which  hydrolyzes  maltose  to  glucose  is  called  maltase.  Maltase 
is  very  widely  distributed  among  the  fungi.  The  enzyme  which  catalyzes 
the  hydrolysis  of  lactose  to  glucose  and  galactose  is  called  lactase.  While 
this  enzyme  is  less  widely  distributed  among  the  fungi  than  sucrase  and 
maltase,  it  is  produced  by  many  species. 

While  it  is  doubtless  correct  to  assume  that  the  more  complex  and  in- 
soluble carbohydrates  must  be  hiydrolyzed  before  utilization,  this  assump- 
tion may,  in  some  instances,  be  false  with  regard  to  the  disaccharides.  It 
is  possible  that  some  fungi  may  employ  a  phosphorylative  degradation  of 
the  disaccharides  rather  than  hydrolysis.  For  a  critical  review  of  carbo- 
hydrate utilization  without  preliminary  hydrolysis,  see  Hestrin  (1948). 

In  addition  to  the  water-soluble  polysaccharides  there  is  a  wide  variety 
of  water-insoluble  high-molecular-weight  carbohydrates  which  are  utilized 
by  many  fungi  as  carbon  sources.  Only  two  of  these  complex  polysaccha- 
rides will  be  considered  here.  The  empirical  formula  for  cellulose  is 
(C6Hio05)„.  On  complete  hydrolysis  by  acids,  glucose  is  the  only  prod- 
uct. Less  complete  hydrolysis  produces  a  disaccharide  known  as  cello- 
biose.  The  majority  of  fungi,  according  to  Norman  and  Fuller  (1942), 
are  able  to  attack  cellulose.  The  early  work  is  reviewed  by  Thaysen  and 
Bunker  (1927).  With  respect  to  the  fungi  which  attack  cellulose,  a  great 
deal  of  variation  in  cellulolytic  ability  is  found  (see  White  et  al.,  1948). 
The  enzyme  which  catalyzes  the  hydrolysis  of  cellulose  is  called  cellulase. 

While  starch  has  the  same  empirical  formula  as  cellulose,  it  is  more 


50  PHYSIOLOGY  OF  THE  FUNGI 

easily  hydrolyzed.  Glucose  is  likewise  the  end  product  of  hydrolysis. 
The  enzyme  (or  enzymes)  which  catalyzes  the  hydrolysis  of  starch  is 
called  amylase.  In  general,  the  end  product  of  enzymatic  hydrolysis  of 
starch  is  maltose  and  glucose.  The  various  intermediate  degradation 
products  are  called  dextrins. 

Starch  appears  to  be  composed  of  two  main  types  of  compounds: 
amylose  (20  to  25  per  cent)  and  amylopectin.  Amylose  appears  to  con- 
sist of  long,  unbranched  molecules  containing  some  300  glucose  residues, 
whereas  amylopectin  has  a  branched  structure.  There  are  two  types  of 
amylase:  jS-amylase,  which  hydrolyzes  off  two  glucose  residues  at  a  time 
to  form  maltose,  and  oi-amylase,  which  attacks  the  1,4-glucosidic  linkages 
in  such  a  way  as  to  produce  starch  fragments  (dextrins)  as  the  primary 
products.  The  dextrins  are  further  hydrolyzed  to  form  maltose  and  some 
glucose.  The  primary  function  of  a-amylase  is  thus  liquefaction;  that  of 
the  /3-amylase  is  saccharification.  The  Aspergillus  amylases  are  of  the 
alpha  type.  The  student  is  referred  to  the  excellent  reviews  of  Hopkins 
(1946)  and  Myrback  (1948)  for  critical  summaries  of  amylase  activity. 
Amylase  is  widely  distributed  among  the  fungi  but  is  not  universal. 

Pectinase.  The  pectins  are  colloidal  carbohydrate-like  compounds 
found  in  fruits  and  in  the  middle  lamellae  of  plants.  Many  fungi  produce 
pectinase,  which  catalyzes  the  hydrolysis  of  pectin.  When  the  pectin  is 
hydrolyzed,  the  cells  fall  apart.  Harter  and  Weimer  (1921)  tested  the 
ability  of  nine  species  of  Rhizoyus  to  produce  pectinase  in  culture  but  were 
unable  to  correlate  the  pathenogenicity  of  these  species  with  the  amount 
of  pectinase  secreted.  In  fact,  some  of  the  pathogenic  species  {R. 
nigricans  and  R.  autocarpi)  secreted  less  pectinase  than  did  two  non- 
pathogenic species  {R.  chinensis  and  R.  microsporus) . 

Pectins  were  formerly  believed  to  yield  a  considerable  variety  of  hydro- 
lytic  products,  including  acetic  acid,  galactose,  and  arabinose  in  addition 
to  methyl  alcohol  and  D-galacturonic  acid.  More  recent  work  indicates 
that  pectins  are  methylated  polymers  of  D-galacturonic  acid  (Schneider 
and  Bock,  1937).  The  chemistry  and  physiology  of  the  pectins  have  been 
reviewed  by  Bonner  (1936). 

Proteinases  and  peptidases.  These  enzymes,  also  called  proteolytic 
enzymes,  catalyze  the  hydrolysis  (and  synthesis)  of  proteins  and  peptides. 
These  enzymes  have  been  separated  into  two  groups  upon  the  basis  of 
ability  to  attack  native  protein.  Those  enzymes  which  act  upon  intact 
proteins  are  called  proteinases,  while  those  which  attack  peptides  are 
called  peptidases.  It  seems  that  the  fundamental  difference  between 
these  two  classes  of  enzymes  lies  in  the  point  of  attack.  The  proteinases 
attack  the  protein  molecule  in  such  a  way  as  to  produce  various  peptides 
as  well  as  amino  acids,  while  the  peptidases  act  only  on  the  ends  of  the 
peptide  chains.     This  is  analogous  to  the  action  of  the  two  amylases. 


ENZYMES  51 

The  proteolytic  enzymes  are  a  very  complex  group  of  hydrolases.  In 
view  of  the  complexity  of  protein  structure  this  is  not  unexpected.  The 
question  of  specificity  of  the  proteolytic  enzymes  has  been  considered  by 
Bergmann  (1942),  who  emphasizes  that  the  specificity  of  a  given  enzyme 
for  a  certain  substrate  may  be  modified  by  the  presence  of  a  second  sub- 
strate. Johnson  and  Berger  (1942)  have  reviewed  the  enzymatic  proper- 
ties of  the  peptidases,  including  those  produced  by  the  fungi. 

Oxidases,  hydrogenases,  and  desmolases.  One  of  the  central  prob- 
lems in  metabolic  processes  is  how  and  by  what  means  oxidation  of 
metabolites  to  carbon  dioxide  and  water  is  brought  about.  Some 
organisms  (bacteria)  are  inhibited  or  killed  by  free  oxygen  (anaerobes). 
Others  may  live  either  in  the  presence  or  absence  of  free  oxygen  (faculta- 
tive anaerobes),  while  others  require  free  oxygen  (aerobes)  to  carry  on 
their  metabohc  processes  and  to  maintain  life.  Thus,  one  organism  may 
degrade  a  substrate  only  partially,  and  these  intermediate  oxidation 
products  become  substrates  for  other  organisms.  In  the  end  complete 
oxidation  takes  place.  In  other  instances  an  organism  may  first  carry 
out  a  partial  degradation  and  complete  it  later.  Thus,  yeast  produces 
alcohol  by  fermentation.  In  the  presence  of  oxygen,  alcohol  is  utilized 
for  the  synthesis  of  cellular  constituents  and  as  a  source  of  energy.  Many 
fungi  possess  two  ways  of  obtaining  energy  by  the  degradation  of  metabo- 
lites: an  anaerobic  (fermentative)  and  an  aerobic  (oxidative)  pathway. 
Both  may  function  in  the  same  organism  at  the  same  time,  although 
external  conditions  may  favor  one  process  at  the  expense  of  the  other,  or  a 
substance  may  inhibit  one  without  affecting  the  other. 

Biological  oxidations  are  carried  out  in  two  ways:  by  the  removal  of 
hydrogen  from,  or  by  the  addition  of  oxygen  to,  substrates.  The  name 
of  Wieland  is  associated  with  the  process  of  dehydrogenation,  and  that  of 
Warburg  with  the  second  process. 

The  theory  of  Wieland  stressed  the  importance  of  the  enzyme  systems 
which  activated  hydrogen  or  removed  hydrogen  from  substrate  molecules, 
while  Warburg's  theory  focused  attention  upon  the  enzyme  systems 
which  activated  oxygen  and  which  carried  oxygen  to  the  substrates. 
These  two  theories  might  seem  irreconcilable,  but  today  they  are  con- 
sidered as  mutually  complementary.  Both  types  of  enzymatic  oxidation 
are  known  for  the  same  organism.  For  further  discussion  of  this  problem 
the  student  is  referred  to  Elvehjem  and  Wilson  (1944)  and  Meyerhof  et  al. 
(1942) .  For  a  classification  of  the  respiratory  enzymes  see  Gortner  (1949) 
and  Sumner  and  Somers  (1947).  For  the  electronic  mechanism  involved 
in  biological  oxidation-reduction  see  Michaelis  (1946). 

Some  representative  dehydrogenases  and  oxidases  are  aerobic  dehj^dro- 
genases  (xanthine  oxidase,  and  uricase);  anaerobic  dehydrogenases, 
(succinic  dehydrogenase,  glucose  dehydrogenase,  triose  phosphate  dehy- 


52  PHYSIOLOGY  OF  THE  FUNGI 

drogenase) ;  oxidases  (cytochrome  oxidase,  tyrosinase,  polyphenol  oxi- 
dase). Succinic  acid  dehydrogenase  oxidizes  succinic  acid  to  fumaric 
acid  by  the  removal  of  two  hydrogens ;  but  this  reaction  takes  place  only 
in  the  presence  of  another  system  (cytochromes)  which  "carries"  the 
hydrogen  to  an  oxidizing  enzyme,  which  converts  the  hydrogen  to  water 
and  regenerates  the  cytochrome  system  so  that  it  can  transport  more 
hydrogen.  In  the  cell,  succinic  acid  dehydrogenase  is  said  to  be  cyto- 
chrome-linked.  In  the  laboratory,  hydrogen  carriers  other  than  cyto- 
chrome may  be  used.  Various  other  dehydrogenases  are  linked  to  the 
cytochrome  system. 

Another  oxidase,  tyrosinase,  is  found  in  many  fungi.  It  is  well  estab- 
lished that  copper  is  an  essential  constituent  of  this  enzyme  system 
(Kubowitz,  1937)  and  may  be  removed  by  dialyzing  the  enzyme  against 
cyanide  solutions.  The  activity  which  is  lost  by  this  treatment  is  restored 
by  cupric  ion,  Cu++,  but  other  divalent  metals  do  not  replace  copper. 
Various  reagents  which  react  with  copper,  such  as  cyanide,  diethyl 
dithiocarbamate,  salicylaldoxine,  and  carbon  monoxide,  inhibit  the  action 
of  tyrosinase.  Among  the  fungi  which  produce  tyrosinase  are  the  follow- 
ing species  (Nelson  and  Dawson,  1944) :  Boletus  luridis,  Russula  foetens, 
R.  niger,  Lactarius  piperatus,  and  PsalUota  campestris.  It  is  probable  that 
the  darkening  and  coloration  of  the  fruit  bodies  of  these  fungi  depend  upon 
the  activity  of  tyrosinase. 

Pyruvic  acid,  CHs — CO — COOH,  is  a  key  compound  in  carbohydrate 
utilization,  and  perhaps  in  other  metabolic  processes  as  well.  The 
enzyme,  carboxylase,  catalyzes  the  decomposition  of  pyruvic  acid  in 
the  following  way: 

carboxylase 

CHs— CO— COOH >  CO.  +    CHs— CHO 

Pyruvic  acid  Carbon  dioxide       Acetaldehyde 

The  carbon  dioxide  formed  escapes,  while  the  acetaldehyde  formed  may 
be  either  oxidized  to  acetic  acid  or  reduced  to  ethyl  alcohol.  The  enzyme 
which  catalyzes  the  decarboxylation  of  pyruvic  acid  to  carbon  dioxide  and 
acetaldehyde  is  abundant  in  yeast  and  other  fungi.  This  enzyme  con- 
sists of  three  moieties,  a  specific  protein,  a  magnesium  ion,  and  thiamine 
pyrophosphate. 

CHEMICAL  NATURE  OF  ENZYMES 

In  the  past  there  has  been  a  great  deal  of  controversy  over  the  chemical 
nature  of  enzymes.  Sumner  (1926)  was  the  first  to  isolate  an  enzyme 
(urease)  in  pure  crystalline  condition.  Since  then  a  dozen  or  more 
enzymes  have  been  prepared  in  pure  crystalline  form.  All  the  enzymes 
which  have  been  isolated  in  pure  crystalline  condition  have  proved  to  be 
proteins. 


ENZYMES  53 

Some  enzymes  are  specific  proteins  requiring  neither  coenzymes  nor 
metals  for  activity.  These  enzymes  must  contain  as  an  integral  part  of 
their  structure  the  specific  groups  whereby  they  react  with  the  substrate. 
Other  enzymes  consist  of  two  moieties,  a  specific  protein  and  a  specific 
nonprotein  compound  which  can  be  detached  from  the  protein.  In  the 
process  of  purifying  an  enzyme  by  dialysis  the  activity  may  be  lost  and 
later  restored  by  adding  to  the  dialyzed  material  some  boiled  juice  from 
the  tissue  under  investigation.  These  specific  nonprotein  compounds  are 
known  as  coenzymes.  Neither  the  specific  protein  nor  the  coenzyme  alone 
functions  as  the  enzyme;  both  are  required  for  activity.  The  specific 
protein  is  called  the  apoenzyme,  while  the  combination  of  apoenzyme  and 
coenzyme  is  called  the  holoenzyme.  Still  other  holoenzymes  consist  of  an 
apoenzyme,  a  coenzyme,  and  a  metallic  ion.  Coenzymes,  being  non- 
protein in  nature,  have  proved  to  be  more  easily  isolated  and  studied  than 
the  specific  protein  moieties  of  enzymes.  Coenzymes  are  a  varied  group 
of  compounds,  some  relatively  simple  in  structure  and  others  more  com- 
plex. The  vitamins  are  known  to  enter  into  the  structure  of  some 
coenzymes,  and  it  is  supposed  it  is  through  such  coenzyme  molecules  that 
the  vitamins  exert  their  specific  effects.  The  same  coenzyme  may  com- 
bine with  many  specific  proteins  to  form  different  enzymes. 

FACTORS  AFFECTING  ENZYME  ACTIVITY 

Some  of  the  factors  influencing  enzyme  activity  affect  the  intact  organ- 
ism as  well  as  isolated  enzyme  systems.  While  the  situation  within  the 
intact  organism  is  more  complex,  a  knowledge  of  the  behavior  of  isolated 
systems  will  be  useful  in  interpreting  the  behavior  of  living  fungi.  The 
factors  which  will  be  discussed  are  temperature,  hydrogen-ion  concentra- 
tion (pH),  chemical  reagents  (activators  and  inhibitors),  and  radiation. 

Temperature.  The  rate  of  many  reactions  is  approximately  doubled 
for  each  10°C.  increase  in  temperature.  The  rate  of  reactions  catalyzed 
by  enzymes  also  increases  with  temperature.  This  increase  is  not  main- 
tained indefinitely,  for  enzymes  are  destroyed  by  temperatures  of  less 
than  100°C.  Although  there  are  some  reports  in  the  literature  of  the  rate 
of  enzymatic  reactions  being  increased  as  much  as  fivefold  by  a  10°C. 
increase  in  temperature,  for  most  enzymatic  reactions  the  increase  in  rate 
is  less  than  twofold.  This  increase  between  two  temperatures  10°C. 
apart  is  called  the  temperature  coefficient,  or  Qio.  Since  the  increase  in 
rate  is  not  exactly  constant,  it  is  desirable  to  specify  the  temperatures 
involved;  e.g.,  Q20-30. 

A  reaction  with  a  Qio  of  2  proceeds  sixteen  times  faster  at  40°C.  than 
at  0°C.  Or,  the  transformation  of  a  given  amount  of  substrate  which 
requires  16  hr.  at  0°C.  will  occur  within  1  hr.  at  40°C.  Figure  9  shows  the 
theoretical  effect  of  temperature  upon  the  amount  of  substrate  trans- 


54 


PHYSIOLOGY  OF  THE  FUNGI 


formed  when  Qio  is  2,  3,  and  4.  It  was  assumed  that  one  unit  of  substrate 
was  transformed  per  unit  of  time  at  0°C.  For  a  reaction  with  a  Qio  of  2 
an  increase  in  temperature  from  28  to  30°C.  causes  as  great  an  increase  in 
the  amount  of  substrate  transformed  as  does  the  increase  from  0  to  10°C. 
A  small  increase  in  temperature  in  the  range  25  to  35°C.  has  a  greater 
effect  on  the  rate  of  reaction  than  a  much  greater  increase  in  temperature 
in  the  lower  temperature  range. 

80 


70 
E 

O60 

C 
QJ 

'^SO 

OJ 

E 
_o 

£40 
o 

a> 

I  30 

"in 

3 

o  20 
trt 

c 
3 

10 


0 


1 

1 

0,0=4  - 

7 

/ 

J 

1  0,0=3-- 

J 

/ 

/ 

/ 

/ 

/ 

/ 

0,0^2-^ 

^ 

_^ 

y^ 

^ 

10  20  30 

Temperature  in  degrees  centjgrode 


40 


I'  ;g.  9.     The  theoretical  effect  of  temperature  on  the  rate  of  enzymatic  reactions  for 
different  assumed  values  of  Qio. 

Enzymes  are  inactivated  by  heat.  The  inactivation  may  be  reversible 
or  irreversible  depending  upon  the  enzyme  involved,  the  duration  of  heat- 
ing, and  other  factors.  The  temperature  at  which  the  increased  rate  of 
reaction  is  balanced  by  destruction  of  an  isolated  enzyme  is  the  so-called 
optimum  temperature  (Bayliss,  1925). 

The  life  processes  of  a  fungus  are  mediated  by  a  large  number  of  en- 
zymes, which  differ  in  their  sensitivity  to  heat.  Fungi  cease  to  grow  or 
reproduce  at  temperatures  lower  than  that  required  to  kill  them.  It  may 
be  assumed  that  the  enzymes  most  sensitive  to  heat  are  gradually  inac- 
tivated as  the  temperature  increases.  This  situation  in  the  li^'ing  fungus 
is  different  from  that  of  an  isolated  system  in  that  the  enzyme  is  in  its 
natural  surroundings  and  the  fungus  is  able  to  synthesize  or  repair  the 
vital  enzymes  in  question.  At  some  temperature  we  may  suppose  that 
the  rate  of  synthesis  or  repair  of  the  enzyme  system  is  exceeded  by  the 
rate  of  inactivation.  When  this  temperature  is  reached,  or  exceeded,  the 
activity  of  these  enzyme  systems  decrease.     This  decreased  activity  is 


ENZYMES  55 

reflected  in  a  lowered  rate  of  growth  or  may  be  seen  in  other  behavior  of 
the  fungus.  With  further  increases  in  temperature,  the  enzyme  systems 
become  less  and  less  operative.  So  long  as  the  temperature  does  not 
exceed  the  point  which  produces  irreversible  inactivation,  lowering  the 
temperature  will  enable  the  fungus  to  resume  growth  or  other  activity. 
The  temperature  of  inactivation  is  not  fixed  unless  the  length  of  exposure 
is  also  considered. 

The  effect  of  temperature  upon  growth  is  shown  in  Figs.  5  and  39. 
The  portions  of  the  curves  in  the  optimum  temperature  range  represent  a 
balance  between  inactivation  and  increased  rate  of  reaction.  Above 
optimum  temperature,  the  rate  of  growth  falls  off  abruptly.  In  a  general 
way  the  rate  of  growth  parallels  that  expected  of  enzymatic  processes. 

Hydrogen-ion  concentration.  Long  ago  it  was  recognized  that  strong 
acids  and  alkalies  were  destructive  to  enzymes.  A  second  effect  was  also 
recognized:  some  enzj^mes  exhibited  maximum  activity  only  in  the  pres- 
ence of  weakly  acidic  or  alkaline  solutions  (see  Chap.  8  for  a  discussion 
of  pH).  The  effect  of  pH  on  the  activity  of  urease  is  shown  in  Fig.  25. 
It  should  be  noted  that  the  pH  optimum  is  dependent  upon  the  concen- 
tration of  urea. 

Haldane  (1930)  compiled  the  pH  optima  of  105  enzymes  and  found 
that  the  range  extended  from  pH  2  to  10.  However,  all  but  nine  of  these 
enzymes  had  pH  optima  between  4  and  8.  Most  fungi  grow  between 
these  limits.  The  effect  of  the  pH  of  the  medium  upon  the  pH  of  the  cell 
contents  is  unknown  in  most  instances.  Biinning  (1936)  has  reported 
that  the  internal  pH  of  the  cells  of  Aspergillus  niger  is  influenced  by  the 
pH  of  the  medium.  The  activities  of  the  exoenzymes  are  affected  by  the 
pH  of  the  medium. 

Chemical  reagents.  Some  enzymes  are  inactive  or  nearly  so  until  they 
have  been  treated  with  certain  reagents.  A  group  of  the  plant  proteinases 
which  includes  papain  and  bromelin  are  activated  by  hydrogen  sulfide  and 
hydrogen  cyanide  (inhibitors  for  many  enzymes),  glutathione,  and  other 
thiol  compounds.  These  various  activators  do  not  act  by  removing 
heavy  metals  (inactivators  for  many  enzymes)  but  by  reducing  the  disul- 
fide linkage,  — S — S — ,  to  thiol  (sulf hydril) ,  — SH.  Neutral  salts 
activate  some  enzymes  (emulsin,  pancreatic  amylase).  The  mode  of 
activation  by  neutral  salts  is  unknown.  Many  of  the  metallic  ions 
(Mg++,  Ca++,  Fe++,  Cu++,  Mn++)  are  required  for  enzyme  activity,  but 
it  seems  better  to  consider  them  as  essential  parts  of  some  enzymes  rather 
than  activators. 

Inhibitors  are  substances  which  reduce  or  destroy  enzyme  activity. 
Inhibition  may  be  reversible  or  irreversible.  A  few  enzyme  inhibitors  are 
cyanides,  monoiodoacetate,  fluoride,  and  the  hea\^  metals  (lead,  copper, 
mercury,  silver,  etc.).     An  inhibitor  is  active  against  certain  enzymes  and 


50 


PHYSIOLOGY  OF  THE  FUNGI 


not  others.  There  appears  to  be  a  close  relation  between  the  chemical 
constitution  of  the  prosthetic  group  of  the  enzyme  and  the  inhibitors 
which  inactivate  it.  We  may  postulate  that  inactivation  results  from  a 
chemical  reaction  between  the  inhibitor  and  the  prosthetic  group  of  an 
enzyme. 

One  characteristic  of  an  oxidase  is  inhibition  by  cyanide  and  hydrogen 
sulfide.     This  points  to  some  common  moiety  in  these  enzymes  which  is 


/Vo  cyanide 


0.95 X 10'^ M  NaCN 


o — '-' — o- 
2.5xlO'^M  NaCN 


U P — o — o-  — o — o, 


l2.4xlO'^MNaCN 


50  75  100 

Oxygen  tension  (mm  Hg) 

Fig.  10.  The  effect  of  cyanide  on  yeast  respiration.  (Courtesy  of  Winzler,  Jour. 
Cellular  Comp.  Physiol.  21:  238,  1943.  Published  by  permission  of  Wistar  Institute 
of  Anatomy  and  Biology.) 

able  to  react  with  cyanide.  The  oxidases  are  metalloproteins,  and  in 
view  of  the  property  of  cyanides  of  reacting  with  metals  to  form  complexes, 
it  would  appear  likely  that  cyanide  reacts  with  the  metal  to  form  inactive 
little-ionized  compounds.  The  typical  properties  of  ferrous  and  ferric 
ions  are  masked  by  cyanide.  Tyrosinase,  a  copper-containing  enzyme,  is 
inactivated  by  cyanide.  Hydrogen  sulfide  acts  on  many  of  the  same 
enzymes  which  are  inhibited  by  cyanide ;  the  action  may  be  assumed  to  be 
due  to  the  formation  of  insoluble  metal  compounds  rather  than  the  forma- 
tion of  non-ionized  complexes. 

Winzler  (1943)  studied  the  effect  of  different  concentrations  of  cyanide 
upon  the  respiration  of  yeast  maintained  under  different  oxygen  tensions. 
The  effect  of  cyanide  on  yeast  respiration  is  shown  in  Fig.  10.     It  may  be 


ENZYMES  57 

noted  that  tlie  percentages  of  inhibition  of  respiration  (oxygen  uptake) 
depend  upon  two  conditions,  the  amount  of  oxygen  available  and  the 
concentration  of  cyanide  present.  We  may  assume  that  the  cyanide 
inhibited  one  or  more  respiratory  enzymes  and  that,  as  the  concentration 
of  cyanide  increased,  more  and  more  of  these  enzymes  were  inactivated. 
When  the  oxygen  tension  was  reduced,  these  effects  were  increased. 

While  it  is  kno^^^l  that  salts  of  the  heavy  metals  may  denature  proteins, 
and  this  explanation  has  been  advanced  to  account  for  enzyme  inactiva- 
tion  by  them,  recent  opinion  inclines  to  the  \'iew  that  the  heavy  metals 
inactivate  enzymes  either  by  combining  with  — SH  groups,  or,  under 
alkaline  conditions,  by  oxidizing  thiol  sulfur  to  disulfide.  Mercuric  ions, 
especially,  may  combine  with  specific  metabolites  which  contain  — SH 
groups  (glutathione,  thioamino  acids),  as  found  by  Fildes  (1940).  Cer- 
tain metals  may  inactivate  enzymes  by  replacing  the  normal  metal,  ren- 
dering the  enzyme  inoperative.  It  is  noteworthy  that  many  enzymes 
which  are  inactivated  by  heavy  metals  may  be  either  "protected"  or 
restored  to  activity  by  the  addition  of  thiol  compounds.  We  may  assume 
for  the  purpose  of  illustration  that,  when  a  heavy  metal  combines  with  an 
enzyme,  an  inactive  complex  or  compound  is  formed  as  shown  in  scheme 
II.  Two  factors  would  influence  the  effectiveness  of  thiol  compounds  in 
preventing  or  reversing  enzyme  inactivation,  the  relative  affinity  of  the 
enzyme  — SH  groups  and  the  thiol  compound  for  mercury,  and  the  rela- 
tive concentration  of  enzyme  and  thiol  compound. 

Scheme  II.     A  Scheme  Illustrating  a  Possible  Mechanism  of  Inactivation 

OF  A  Sulfhydril  Enzyme  by  Mercuric  Ion  and  Reactr'ation  of  the  Inactive 

Enzyme-Mercury  Complex  by  the  Addition  of  a  Thiol  Compound 

Inactivation 

Enzyme — S 

2(Enzyme— SH)  +  Hg++^  Hg  +  2H+ 

Enzyme — S 
Active     enzyme  Inactive  enzyme  complex 

Reactivation 
Enzyme— S  RS 

\  \ 

Hg  +  2RSH^  2  (Enzyme— SH)  +  Hg 

Enzyme— S  RS 

Active  enzyme 

Radiation.  Many  reports  are  to  be  found  in  the  literature  that  radia- 
tion affects  enzymes  adversely  (see  the  review  of  Schomer,  1936).  Radia- 
tion may  affect  not  only  the  enzymes  of  an  organism  but  also  the  sub- 
strates. Ionizing  short-wave  radiations  may  cause  the  formation  of 
hydrogen  peroxide  from  water.     Barron  et  al.  (1947)  were  able,  by  adding 


58 


PHYSIOLOGY  OF  THE  FUNGI 


glutathione,  to  reactivate  phosphoglyceral  dehydrogenase  which  had  been 
inactivated  by  X  rays. 

Whether  radiation  is  absorbed  or  not  depends  upon  the  chemical  con- 
stitution of  the  absorbing  molecule  and  the  wave  length  of  the  radiation. 
The  energy  thus  obtained  may  disrupt  the  molecule  or  may  merely 
increase  its  ability  to  react.  These  generalizations  are  not  very  helpful  in 
either  predicting  the  effect  of  light  upon  living  fungi  or  interpreting  the 
observed  effects  of  light  on  growth  and  reproduction.  It  is  probable  that 
light  acts  on  various  enzyme  systems.  Light  is  known  to  affect  one 
specific  enzyme  system  (cytochrome-cytochrome  oxidase).  Warburg 
(1926)  showed  that  the  respiration  of  baker's  yeast  was  inhibited  to  the 
extent  of  70  per  cent  in  the  dark  when  exposed  to  carbon  monoxide  con- 
taining 5  per  cent  oxygen,  while  respiration  was  inhibited  only  14  per  cent 
in  light.  The  same  effect  of  light  on  carbon  monoxide  inhil)ition  of 
respiration  has  been  demonstrated  with  larvae  of  Tenehris  molitor  and  the 

heart  of  embryo  trout. 

Ultraviolet  radiation  and  X  rays  have  a 
lethal  effect  on  fungi.  A  small  percentagr^ 
of  the  spores  which  survive  exposure 
to  ultraviolet  radiation  may  produce 
mutants.  It  has  been  noted  recently 
(Kelner,  1949)  that  the  lethal  effect  of 
ultraviolet  radiation  upon  spores  of  Strep- 
tomyces  griseus  is  overcome  to  a  considera- 
ble extent  by  exposing  irradiated  spores 
to  visible  light.     Whether  this  is  due  to 


o      o 

o         "" 

o 

r^    O 

^ Q 

o       o 

/ 

^        ^ 

\p 

o  o/ 

V€ 

°  ( 

Enzyme 

V 

° 

molecule 

o 

°  V 

Jo 

o    \ 

/r^          O 

o    o 

o   o 

o  o 

0 

o 

Fig.  11.  Diagrammatic  illustra- 
tion of  the  mechanism  of  enzy- 
matic hydrolysis.  The  substrate 
molecules  are  represented  by 
small  circles,  the  products  of 
hydrolysis  by  semicircles.  (Cour- 
tesy of  Van  Slyke,  Advances  in 
Enzymol.  2  :  38,  1942.  Published 
by  permission  of  Interscience 
Publishers,  Inc.) 


reactivation  of  certain  enzyme  systems  is 
not  known. 

MECHANISM   OF  ENZYME  ACTION 


The  most  generally  accepted  theory  of 
enzyme  action  postulates  that  the  enzyme 
and  substrate  unite  to  form  a  molecular 
compound  or  complex  (enzyme-substrate 
complex).  In  favorable  instances  the  ex- 
istence of  such  enzyme-substrate  complexes  has  been  demonstrated  (Stern, 
1936) .  During  this  temporary  union  the  substrate  molecule  is  "  strained" 
or  activated  so  that  it  undergoes  reaction.  The  products  of  the  reaction 
have  less  affinity  for  the  enzyme  surface  than  the  substrate  molecules  and 
hence  diffuse  away,  and  other  substrate  molecules  unite  temporarly  with 
the  enzyme  and  the  process  continues.  If  the  product  molecules  are 
present  in  excess,  they  may  compete  more  successfully  for  the  enzyme 
surface  than  the  substrate  does.     During  synthesis,  when  the  reactants 


ENZYMES 


59 


(products)  are  present  in  solution  in  greater  than  equilibrium  concentra- 
tions, the  reactants  combine  with  the  enzyme,  unite,  and  diffuse  away. 
Figure  11  gives  a  diagram  which  is  helpful  in  visualizing  these  processes. 

ADAPTIVE  ENZYMES 

Some  fungi  produce  certain  enzymes  only  in  response  to  particular 
environmental  conditions.  Such  enzymes  are  called  adaptive  enzymes. 
\\^ether  they  are  produced  under  all  cultural  conditions,  but  in  such  small 
amounts  as  to  be  undetectable,  or  whether  they  are  produced  de  novo  is 
questionable.  How^ever,  this  phenomenon  is  of  great  importance.  Two 
types  of  behavior  may  be  noted  when  a  fungus  is  placed  upon  unsuitable 
medium  for  the  first  time.  Either  the  fungus  may  die,  owing  to  lack  of 
ability  to  synthesize  the  enzymes  to  cope  with  the  new  environment;  or 


240 


160 


co^ 


80 


Strain  CI 


-d^ 


^-0- 


o  -Aerobic  adaptation 
•  -Anaerobic  adaptation 

■4-1-*        I        *      ■ 


_1_ 


./ 


8 


16 


32 


40 


24 
Hours 
Fig.  12.     Rate  of  adaptation  of  a  strain  of  Saccharomyces  carlsbergensis  to  galactose 
under  aerobic  and  anaerobic  conditions.      (Courtesy  of  Spiegelman,  Jour.  Cellular 
Comp.  Physiol.   25:    128,    1945.     Published  by  permission  of  Wistar   Institute  of 
Anatomy  and  Biology.) 

after  a  time  it  may  synthesize  the  necessary  enzymes,  and  the  fungus  is 
then  able  to  grow  and  function  under  the  new  surroundings.  Wlrether  or 
not  the  fungus  is  able  to  synthesize  "new"  enzymes  depends  upon  its 
genetic  constitution.  The  biochemical  and  physiological  responses  of  an 
organism  may  change  when  it  is  placed  on  a  different  kind  of  medium. 
These  changes  ordinarily  are  called  forth  by  deficiencies  in  the  medium. 
The  substrate  upon  which  the  inoculum  grew-  may  be  very  important  in 
governing  the  various  responses  of  the  organism. 

Spiegelman  (1945)  has  shown  that  the  adaptation  of  yeasts  to  galactose 
is  affected  by  aerobic  and  anaerobic  conditions.  Adaptation  is  more 
rapid  in  air  than  in  nitrogen,  and  some  strains  of  yeast  are  unable  to  adapt 
to  galactose  in  the  absence  of  oxygen.  Figure  12  shows  that  only  some 
30  min.  is  required  for  Saccharomyces  carlshergensis  to  begin  to  utilize 
galactose  under  aerobic  conditions,  while  about  20  hr.  are  required  under 
anaerobic  conditions. 

The  effect  of  composition  of  the  medium  on  the  readaptation  of  panto- 


()0  PHYSIOLOGY  OF  THE  FUNGI 

thenate-dependent  strains  of  yeast  to  the  synthesis  of  pantothenate  has 
been  studied  in  some  detail  (Lindegren  and  Rant,  1947;  Lindegren,  1949). 
Changes  to  pantothenate  independence  occurred  by  an  adaptation,  which 
was  transmitted  vegetatively,  and  by  a  gene  mutation.  The  adaptation 
occurred  only  in  the  media  of  low  pantothenate  content,  while  the  muta- 
tions were  apparently  not  affected  by  the  concentration  of  pantothenate. 
Leonian  and  Lilly  (1943)  studied  the  induced  ability  of  eight  strains  of 
Saccharoniyces  cerevisiae  to  synthesize  various  vitamins  for  which  they 
were  normally  deficient.  This  was  accomplished  by  long  "training"  in 
media  "free"  from  various  vitamins.  The  ability  of  various  yeast  strains 
to  synthesize  a  given  vitamin  varied.  These  yeasts  which  had  been 
trained  "reverted"  to  their  deficient  status  when  cultured  for  6  months  on 
media  containing  vitamins  and  yeast  extract. 

ENERGY  AND  ENERGY  UTILIZATION  BY  FUNGI 

Fungi  need  energy,  as  well  as  certain  elements  and  chemical  compounds, 
for  life,  growth,  and  reproduction.  Since  the  life  processes  of  the  fungi 
are  controlled  by  interlocking  systems  of  enzymes,  the  utilization  of 
energy  is  also  an  enzymatic  process.  The  chemical  reactions  which 
accompany  or  underlie  life  processes  may  be  divided  into  those  which 
yield  energy  (exergonic)  and  those  which  require  energy  (endergonic) 
(Coryell,  1940).  The  oxidation  reactions  whereby  such  substrate  mole- 
cules as  glucose  are  converted  into  carbon  dioxide  and  alcohol  or  carbon 
dioxide  and  water  yield  energy,  while  the  reactions  involved  in  the  syn- 
thesis of  protoplasm  and  reserve  materials  require  energy.  Let  us  con- 
sider an  analogy  first.  When  water  falls  from  a  higher  to  a  lower  level, 
there  is  a  decrease  in  energy  content,  and  this  decrease  in  energy  content  is 
the  same  whether  the  water  has  passed  through  a  turbine  or  not.  The 
water  that  passes  over  a  spillway  does  no  useful  work,  while  the  water  that 
turns  a  turbine  makes  part  of  the  energy  available  (as  mechanical  or 
electrical  power)  for  doing  useful  work.  The  energy  given  up  by  the 
falling  water  is  the  same  in  both  cases,  but  only  where  the  proper  mecha- 
nism is  available  is  any  useful  work  obtained. 

A  similar  situation  occurs  when  a  fungus  oxidizes  glucose  to  water  and 
carbon  dioxide.  If  energy-requiring  synthetic  reactions  are  coupled  with 
the  degradation  reactions,  a  portion  of  the  available  energy  becomes  useful 
to  the  fungus.  The  remainder  of  the  energy  liberated  appears  as  heat, 
which  is  unavailable  to  the  fungus  for  lack  of  suitable  mechanisms  to 
utilize  it. 

Winzler  and  Baumberger  (1938)  have  investigated  the  liberation  of 
energy  by  yeast  cells  during  metabolism.  Washed  yeast  cells  were  sus- 
pended in  a  phosphate  buffer  containing  glucose  but  no  nitrogen.  The 
reaction  vessel  was  placed  in  an  adiabatic  calorimeter,  and  the  heat 


ENZYMES 


61 


evolved  and  the  amount  of  oxygen  absorbed  and  of  carbon  dioxide  evolved 
were  measured.  In  the  absence  of  a  nitrogen  source  the  synthesis  of 
protoplasm  was  avoided.  The  rate  at  which  heat  was  evolved  was  con- 
stant until  all  the  glucose  was  consumed  (exogenous  respiration),  after 
which  the  rate  of  heat  formation  decreased  (endogenous  respiration)  (see 
Fig.  13). 


20 


50 


60 


30  40 

Time  in  minutes 
Fig.  13.  Heat  produced  from  glucose  oxidation  by  yeast  in  the  absence  of  a  nitrogen 
source;  10,  20,  and  30  mg.  of  glucose  were  added  at  zero  time  in  curves  I,  II,  and  III, 
respectively.  In  all  cases,  only  26  per  cent  of  the  expected  amount  of  heat  was 
evolved  before  the  endogenous  respiration  rate  was  resumed.  (Courtesy  of  Winzler 
and  Baumberger,  Jour.  Cellular  Comp.  Physiol.  12:  199,  1938.  Published  by  per- 
mission of  Wistar  Institute  of  Anatomy  and  Biology.) 

In  this  experiment  the  theoretical  amount  of  heat  could  be  calculated 
for  the  amounts  of  glucose  used.  Only  26  per  cent  of  the  theoretical  heat 
was  produced  before  endogenous  respiration  set  in.  The  volume  of  oxy- 
gen used  was  equal  to  the  volume  of  carbon  dioxide  evolved,  i.e.j  the  R.Q. 
was  1.  These  data  may  be  interpreted  as  follows:  For  every  molecule  of 
glucose  oxidized  to  carbon  dioxide  and  water,  three  molecules  were  syn- 
thesized into  a  carbohydrate,  presumably  glycogen.  When  sodium 
acetate  was  the  substrate,  about  59  per  cent  of  the  theoretical  heat  was 
evolved,  but  in  the  presence  of  dinitrophenol  the  theoretical  amount  of 
heat  was  evolved.  This  inhibitor,  therefore,  blocked  the  assimilative 
mechanism  but  not  the  oxidative  processes. 

Within  recent  years  it  has  been  discovered  that  certain  phosphate  esters 
may  play  a  very  important  role  in  energy  transfer.  The  student  is 
referred  to  the  review  of  Lipmann  (1941)  for  further  information  on  this 
subject. 

The  utilization  of  energy  derived  from  degradation  reactions  depends 
upon  such  energy-yielding  reactions  being  coupled  with  energy-requiring 
reactions.     Degradation  reactions  which  are  not  so  coupled  (blocked) 


62  PHYSIOLOGY  OF  THE  FUNGI 

waste  energy  in  the  form  of  heat  which  is  not  iitihzed  by  the  fungi.  The 
efficiency  of  utiHzation  depends  upon  the  substrate  utihzcd  and  upon  the 
nature  of  the  coupled  reactions.  In  any  case  only  a  part  of  the  energy 
available  in  the  sulDstrate  does  useful  chemical  work  for  the  fungus  utiliz- 
ing it.  The  application  of  these  ideas  with  any  rigor  requires  a  sound 
knowledge  of  thermodynamics. 

SUMMARY 

The  chemical  reactions  which  underlie  the  life  processes  of  fungi  and 
other  organisms  are  initiated  by  organic  catalysts,  or  enzymes.  Enzymes 
catalyze  synthetic  as  well  as  degradation  reactions  and  are  mediators  of 
energy  transfer  as  well. 

Enzymes  are  specific  proteins  which  in  some  instances  require  certain 
metallic  ions  or  organic  coenzymes,  or  both,  before  they  are  active.  In 
general,  an  enzyme  controls  but  a  single  type  of  reaction.  In  living 
organisms  these  enzyme-controlled  reactions  are  correlated  and  integrated 
to  a  high  degree. 

Among  the  external  factors  which  modify  the  action  of  enzymes  the 
following  are  especially  important:  temperature,  hydrogen-ion  concentra- 
tion, concentration  of  substrate  and  products,  and  inhibitors.  The 
effects  of  these  factors  on  isolated  enzymes  and  intact  organisms  are  much 
the  same. 

While  the  role  of  enzymes  in  maintaining  life  processes  in  fungi  and 
other  organisms  is  well  established,  the  application  of  this  information  to 
living  fungi  must  be  made  with  due  caution  and  the  realization  that  a 
living  organism  is  more  complex  than  its  parts. 

REFERENCES 

Barron,  E.  S.  G.,  S.  Dickman,  and  T.  P.  Singer:  On  the  inhibition  ol  enzymes  by 

ionizing  radiations,  Fed.  Proc.  6 :  236,  1947. 
*Bayliss,  W.  M.:  The  Nature  of  Enzyme  Action,  2d  ed.,  Longmans,  Roberts  and 

Green,  London,  1925. 
Bergmann,  M.:  A  classification  of  proteolytic  enzymes.  Advances  in  Enzymol.  2: 

49-68,  1942. 
Bonner,  J.:  The  chemistry  and  physiology  of  the  pectins,  Botan.  Rev.  2:  475-497, 

1936. 
Bourqxtelot,  E.:  La  Synthese  biochimique  des  d-glucosides  d'alcools  monovalents. 

II.  AlcooW-glucosides  a,  Ann.  chim.,  Ser.  IX,  3:  287-337,  1915. 
Buchner,  E.:  Alcoholische  Garung  ohne  Hefezellen,  Ber.  d.  deut.  chem.  Ges.  30: 

117-124,  1897. 
BtJNNiNG,  E.:  Ueber  die  Farbstoff-  und  Nitrataufnahme  bei  Aspergillus  niger,  Flora 

131:87-112,  1936. 
Coryell,  C.  D.:  The  proposed  terms  "exergonic"  and  "endergonic"  for  thermo- 
dynamics. Science  92  :  380,  1940. 
DrxoN,  M.:  Multi-enzyme  Systems,  Cambridge  University  Press,  New  York,  1949. 
Elvehjem,   C.  a.,  and  P.   W.   Wilson   (Editors):  Respiratory  Enzymes,   Burgess 

Publishing  Co.,  Minneapolis,  1944. 


ENZYMES  63 

IiLDES,   P.:  The  mechanism  of  the  anti-bacterial  action  of  mercury,  Brit.  Jour. 

Exptl.  Path.  21:  G7-73,  19-40. 
GoRTNER,  R.  A.:  Outlines  of  Biochemistry,  3d  ed.,  John  Wiley  &  Sons,  Inc.,  New 

York,  1949. 
Haldane,  J.  B.  S.:  Enzymes,  Longmans,  Roberts  and  Green,  London,  1930. 
Harden,   A.:   Alcoholic   Fermentation,   4th  ed.,  Longmans,    Roberts  and   Green, 

London, 1932. 
Harter,  L.  L.,  and  J.  L.  Weimer:  A  comparison  of  the  pectinase  produced  by 

different  species  of  Rhizopus,  Jour.  Agr.  Research  22:  371-377,  1921. 
Hestrin,  S.:  The  fermentation  of  disaccharides.     I.  Reducing  disaccharides  and 

trehalose,  Wallerstein  Labs.  Communs.  11 :  193-206,  1948. 
Hopkins,  R.  H.:  The  action  of  the  amylases.  Advances  in  Enzymol.  6:  389-414,  1946. 
Johnson,  ]\L  J.,  and  J.  Berger:  The  enzymatic  properties  of  peptidases.  Advances 

in  Enzymol.  2 :  69-92,  1942. 
Kelner,  a.:  Effect  of  visible  light  on  the  recovery  of  Streptomyces  griseus  conidia 

from  ultraviolet  irradiation  injury,  Proc.  Natl.  Acad.  Sci.  U.S.  35:  73-79,  1949. 
KuBowiTZ,    r.:    Ueber    die    chemische    Zusammensetzung    der    Kartoffeloxydase, 

Biocnem.  Zeit.  292 :  221-229,  1937. 
Leibowitz,  J.,  and  S.   Hestrin:  Alcoholic  fermentation  of  the  oligosaccharides, 

Advances  in  Enzymol.  5:  87-127,  1945. 
Leonian,  L.  H.,  and  V.  G.  Lilly:  Induced  autotrophism  in  yeast.  Jour.  Bad.  45: 

329-339,  1943. 
Lindegren,  C.  C.:  The  Yeast  Cell,  Its  Genetics  and  Cytology,  Educational  Pub- 
lishers, St.  Louis,  1949. 
Lindegren,  C.  C,  and  C.  Ralt:  A  direct  relationship  between  pantothenate  con- 
centration and  the  time  required  to  induce  the  production  of  pantothenate- 

synthesizing  ''mutants"  in  yeasts.     Ann.  Missouri  Botan.  Garden  34:  85-93, 

1947. 
Lipmann,   F.:    Metabolic   generation  and  utilization  of  phosphate  bond  energy, 

Advances  in  Enzymol.  1:  99-162,  1941. 
*Meyerhof,  O.,  et  al.:  Symposium  on  Respiratory  Enzymes,  University  of  Wisconsin 

Press,  Madison,  1942. 
MicHAELis,  L. :  Fundamentals  of  oxidation  and  reduction  in  Currents  in  Biochemical 

Research  (edited  by  D.  E.  Green),  Interscience  Publishers,  Inc.,  New  York, 

1946. 
Myrback,  K.:  The  structure  of  starch,  Wallerstein  Labs.  Communs.  11:  209-218, 

1948. 
Nelson,  J.  M.,  and  C.  R.  Dawson:  Tyrosinase,  Advances  in  Enzymol.  4:  99-152, 

1944. 
Norman,  A.  G.,  and  W.  H.  Fuller:  Cellulose  decomposition  by  microorganisms, 

Advances  in  Enzymol.  2 :  239-264,  1942. 
Pasteur,  L.:  Etudes  sur  la  vin,  Librairie  F.  Savy,  Paris,  1875. 
Schneider,  G.  G.,  and  H.  Bock:  Ueber  die  Konstitution  der  Pektinstoffe,  Ber.  d. 

deut.  chem.  Ges.  70:  1617-1630,  1937. 
Schomer,  H.  a.  :  The  effects  of  radiation  on  enzymes  in  Biological  Effects  of  Radi- 
ation (edited  by  B.  M.  Duggar),  McGraw-Hill  Book  Company,  Inc.,  New  York, 

1936. 
*Spiegelman,  S.  :  The  effect  of  anaerobiosis  on  adaptation  to  galactose  fermentation 

by  yeast  cells,  Jour.  Cellular  Comp.  Physiol.  25:  121-131,  1945. 
Stephenson,  M.:  Bacterial  Metabolism,  2d  ed.,  Longmans,  Roberts  and  Green, 

London, 1939. 
Stern,  K.  G.:  On  the  mechanism  of  enzyme  action.     A  study  of  the  decomposition 


64  PHYSIOLOGY  OF  THE  FUNGI 

of  nioiioctliyl  hydrogon  peroxide  by  catalase  and  of  an  intermediate  enzyme- 
substrate  eompound,  Jour.  Biol.  Chem.  114:  473-494,  1936. 

Sumner,  J.  B.:  The  isolation  and  crystallization  of  the  enzyme  urease,  Jour.  Biol. 
Chem.  69:  435-441,  1926. 
*SuMNER,  J.  B.,  and  G.  F.  Somers:  Chemistry  and  Methods  of  Enzymes,  Academic 
Press,  Inc.,  New  York,  1947. 

Thaysen,  a.  C,  and  H.  J.  Bunker:  The  Microbiology  of  Cellulose,  Hemicelluloses, 
Pectins  and  Gums,  Oxford  University  Press,  New  York,  1927. 
*Van  Slyke,  D.  D.:  The  kinetics  of  hydrolytic  enzymes  and  their  bearing  on  methods 
for  measuring  enzyme  activity.  Advances  in  Enzymol.  2 :  33-47,  1942. 

Warburg,  O.  :  Ueber  die  Wirkung  des  Kohlenoxyds  auf  den  Stoffwechsel  der  Hefe, 
Biochetn.  Zeit.  177:  471-486,  1926. 

White,  W.  L.,  R.  T.  Darby,  G.  M.  Stechert,  and  K.  Sanderson:  Assay  of  cellulo- 
lytic  activity  of  molds  isolated  from  fabrics  and  related  items  exposed  in  the 
tropics,  Mycologia  40:  34-84,  1948. 

WiNZLER,  R.  J. :  A  comparative  study  of  the  effects  of  cyanide,  azide,  and  carbon 
monoxide  on  the  respiration  of  bakers  yeast.  Jour.   Cellular  Conip.  Physiol. 
21 :  229-252,  1943. 
*WiNZLER,  R.  J.,  and  J.  P.  Baumberger:  The  degradation  of  energy  in  the  metabo- 
lism of  yeast  cells.  Jour.  Cellular  Comp.  Physiol.  12:  183-211.  1938. 


CHAPTER  5 
ESSENTIAL  METALLIC  ELEMENTS 

The  fungi  need  about  17  elements  to  supply  their  nutritional  require- 
ments. These  elements  are  utilized  in  the  form  of  specific  compounds,  as 
ions,  and  as  free  elements.  Some  of  the  essential  elements  are  required  by 
all  fungi.  Other  elements  are  required  only  by  certain  species.  In  a 
general  way,  the  elements  required  by  the  fungi  are  the  same  ones 
required  by  bacteria,  green  plants,  and  animals.  There  are,  however, 
striking  differences  in  the  essential-element  requirements  of  different 
groups  of  organisms  (Table  15).  Differences  in  ability  to  utilize  specific 
compounds  containing  these  essential  elements  are  common  in  the  fungi 
and  bacteria. 

BIOLOGICALLY  ESSENTIAL  ELEMENTS 

Before  seeking  to  determine  which  elements  are  essential,  it  is  necessary 
to  define  what  is  meant  by  the  term  hiologically  essential  element.  An 
essential  element  is  indispensable  in  that  no  other  element  may  entirely 
replace  it.  Without  these  essential  elements  life  is  impossible.  An  ele- 
ment needed  in  extremely  small  amounts  may  be  just  as  essential  as  car- 
bon, which  comprises  almost  half  the  weight  of  a  fungus. 

There  are  some  92  chemical  elements  (if  we  exclude  the  recently 
isolated  trans uranic  elements),  most  of  which  are  known  to  exist,  or  may 
exist,  as  a  mixture  of  isotopes.  So  far  as  is  known,  all  the  isotopes  of  an 
element  (with  the  possible  exception  of  the  isotopes  of  hydrogen)  have  the 
same  chemical  and  biological  properties.  Even  radioactive  isotopes, 
before  they  decay,  exhibit  the  same  biological  properties  as  the  stable 
isotopes.  The  biological  effects  of  radiation  in  inducing  mutations  are 
considered  briefly  in  Chapter  18.  In  spite  of  the  limited  number  of  ele- 
ments, the  question  of  essentiality  is  not  settled  completely  for  all. 

The  problem  of  determining  which  elements  are  essential  for  the  fungi 
has  been  approached  from  the  standpoint  of  ultimate  analysis  of  mycelium 
and  spores.  If  certain  elements,  such  as  carbon,  potassium,  and  mag- 
nesium, are  always  found  in  all  samples  analyzed,  irrespective  of  the  sub- 
strates upon  which  these  fungi  grew,  it  may  be  concluded  with  a  high 
degree  of  probability  that  these  elements  are  essential  for  the  fungi. 
Some  of  the  analytical  results  of  ultimate  analyses  of  mycelium  and  spores 
have  been  collected  by  Buchanan  and  Fulmer  (1928)  and  Foster  (1949). 

65 


66  PHYSIOLOGY  OF  THE  FUNGI 

Organic  materials  are  dried  before  analysis.  On  the  average  about  75  per 
cent  of  the  fresh  weight  of  mycelium  is  water,  while  spores  contain  only 
about  40  per  cent  water.  It  is  probable  that  the  water  driven  off  when 
fungus  cells  are  dried  to  constant  weight  is  in  part  free  water  and  in  part 
water  bound  to  various  colloidal  cell  constituents. 

Ultimate  analyses  of  mycelium  and  spores  always  reveal  the  presence  of 
carbon  and  nitrogen.  On  the  average  about  45  per  cent  of  dry  mycelium 
is  carbon.  This  high  content  of  carbon  makes  it  certain  that  carbon  is  an 
essential  element.  The  percentage  of  nitrogen  found  is  quite  variable. 
Phosphorus,  potassium,  magnesium,  calcium,  sodium,  sulfur,  and  iron  are 
found  in  the  ash  that  remains  after  burning  mycelium  and  spores.  More 
refined  methods  of  analysis  reveal  that  fungus  ash  contains  still  other 
elements.  Richards  and  Troutman  (1940)  investigated  the  composition 
of  yeast  ash  by  spectrographic  analysis  and  found  the  following  elements : 
iron,  sodium,  boron,  bismuth,  barium,  magnesium,  manganese,  copper, 
zinc,  tin,  lead,  tellurium,  silver,  chromium,  potassium,  gold,  and  lan- 
thanum. However,  the  mere  presence  of  an  element  in  fungus  cells  does 
not  necessarily  mean  that  it  is  essential. 

Since  many  of  these  elements  in  fungus  ash  occur  in  minute  traces  only, 
it  is  desirable  to  approach  the  problem  of  essentiality  in  another  way. 
This  is  done  by  omitting  from  the  medium  the  element  in  question. 
Raulin  (1869)  was  apparently  the  first  to  use  this  method.  He  found 
that  the  omission  of  phosphorus,  sulfur,  magnesium,  zinc,  or  iron  from  the 
basal  medium  allowed  very  little  growth  of  Aspergillus  niger.  These  ele- 
ments are  thus  shown  to  be  essential  by  the  two  methods  of  investigation. 
In  general,  the  experimental  work  in  which  specific  elements  have  been 
omitted  from  the  medium  is  more  convincing  than  the  method  of  ultimate 
analysis.  This  is  the  preferred  method  of  testing  the  essentiality  of  ele- 
ments required  in  small  amounts. 

Functions  of  the  essential  elements.  Thatcher  (1934)  has  attempted 
to  classify  the  essential  elements  into  groups:  structural,  functional,  and 
those  utilized  in  the  transfer  of  energy.  This  classification  has  some 
validity  and  may  serve  to  fix  attention  upon  the  more  salient  biological 
features  of  an  element.  However,  most,  if  not  all,  of  the  essential  ele- 
ments play  many  roles  in  the  life  processes  of  the  fungi.  In  general,  the 
nonmetallic  elements  may  be  classified  as  structural  elements.  This 
means  that  the  compounds  which  make  up  the  structural  units  such  as 
the  protoplasm  are  largely  composed  of  the  nonmetallic  essential  ele- 
ments: carbon,  nitrogen,  hydrogen,  oxygen,  sulfur,  and  phosphorus.  The 
functional  uses  of  these  elements  by  the  fungi  are  no  less  important.  The 
essential  metallic  elements  may  be  classified  as  functional  elements,  but 
this  does  not  mean  that  these  metallic  elements  have  no  structural 
functions. 


ESSENTIAL  METALLIC  ELEMENTS  67 

The  elements  are  in  the  form  of  chemical  compounds,  some  of  which 
are  relatively  simple,  Avhile  others  are  complex.  With  the  exception  of 
oxygen  the  essential  elements  are  usually  utilized  in  the  form  of  com- 
pounds or  ions.  An  essential  element  may  exist  in  a  chemical  compound 
and  be  unavailable.  The  properties  of  a  chemical  compound  are  deter- 
mined by  all  the  atoms  that  compose  it  and  by  the  way  in  which  atoms 
are  joined  together  in  the  compound.  It  is  convenient  to  consider  the 
essential  elements  one  by  one,  but  this  is  done  only  to  simplify  the 
approach  to  a  complex  subject.  These  separate  factors  must  be  con- 
sidered in  relation  to  the  organism  as  a  whole. 

A  fungus  is  no  more  capable  of  growth  on  an  iron-free  medium  than  on 
a  carbon-  or  nitrogen-free  medium.  Yet,  in  a  balanced  medium  the  ratio 
of  iron  to  carbon  is  in  the  neighborhood  of  1  to  50,000.  The  essential 
metallic  elements  function  in  conjunction  with  enzyme  systems  (Chap.  4). 
This  accounts  for  the  small  amounts  of  these  elements  required.  If  a 
vital  enzyme  system  lacks  an  essential  metal  ion,  it  will  not  function.  It 
appears  that  in  processes  such  as  growth  a  suboptimal  amount  of  an 
essential  metal  will  stop  growth  because  the  apoenzymes  or  coenzymes 
synthesized  will  lack  the  necessary  activating  metal.  The  ratios  as  well 
as  the  amounts  of  the  various  essential  metallic  ions  affect  certain 
metabolic  processes  other  than  growth. 

The  absolute  amounts  of  the  essential  metallic  elements  required  differ 
widely.  Raulin  (1869)  found  that  Aspergillus  niger  required  1  g.  of 
potassium  to  produce  64  g.  of  mj^celium,  while  1  g.  of  magnesium  sufficed 
for  the  synthesis  of  200  g.  of  mycelium.  Recent  work  of  Steinberg  (1946) 
with  .4.  niger  indicates  still  higher  yields  per  gram  of  these  two  elements. 
The  yield  of  mycelium  per  gram  of  iron  and  zinc  was  in  the  neighborhood 
of  55,000  g. 

The  list  of  metallic  elements  knoTvni  to  be  essential  to  fungi  has  in- 
creased over  the  years.  The  list  now  includes  potassium,  magnesium, 
iron,  zinc,  copper,  calcium,  gallium,  manganese,  molybdenum,  vanadium, 
and  scandium.  Others  will  probably  be  added  as  cultural  methods 
become  more  refined  and  more  species  are  studied.  It  is  unfortunate  that 
only  a  few  fungi  have  been  investigated  thoroughly  with  respect  to  mineral 
nutrition.  In  stating  that  the  above  elements  are  essential,  the  reserva- 
tion must  be  made  that  they  are  essential  for  some  fungi  under  certain 
conditions.  WTiile  it  may  be  assumed  that  all  fungi  require  the 
same  essential  elements,  experimental  evidence  is  lacking  for  most 
species. 

For  the  purpose  of  discussion  the  essential  metallic  elements  will  be 
divided  into  two  groups,  macro  and  micro  metallic  elements.  This 
grouping  is  made  solely  for  convenience  and  on  the  basis  of  the  amounts 
ordinarily  employed  in  culturing  fungi  under  laboratory  conditions. 


68  PHYSIOLOGY  OF  THE  FUNGI 

THE  ESSENTIAL  MACRO  ELEMENTS 

Potassium.  This  element  is  essential  for  all  organisms,  so  far  as  is 
known.  There  is  an  immense  amount  of  information  on  the  specific 
effects  of  potassium  on  green  plants  and  animals,  l)ut  such  data  are  not 
common  for  the  fungi.  The  quantitative  relation  between  the  amount  of 
potassium  in  the  medium  and  the  weight  of  mycelium  produced  by 
Aspergillus  niger  was  studied  by  Steinberg  (1946).  This  work  was  done 
with  extraordinary  care  using  a  highly  purified  optimal  medium  (except 
potassium).  The  optimum  amount  of  potassium  w-as  150  mg.  per  liter. 
The  relative  amounts  of  mycelium  formed  increased  as  the  potassium 
content  of  the  medium  decreased.  The  fungus  produced  almost  three 
times  as  much  mycelium  per  milligram  of  potassium  when  15  instead  of 
150  mg.  per  liter  were  used.  Jarvis  and  Johnson  (1950)  have  reported 
that  Penicillium  chrysogenum  Q176  requires  40  mg.  of  potassium  and  8 
mg.  of  magnesium  per  liter  of  medium  for  optimum  growth. 

The  physiological  effects  of  potassium  on  fungi  have  been  studied  but 
little.  The  enzymes  in  yeast  maceration  juice  which  ferment  glucose  are 
activated  by  either  potassium  or  ammonium  ions  (Muntz,  1947).  Mol- 
liard  (1920)  noted  that  a  low  potassium  content  of  the  medium  resulted  in 
increased  synthesis  of  oxalic  acid  by  A.  mger.  The  chemical  composition 
of  A.  niger  mycelium  varies,  depending  upon  the  amount  of  potassium  in 
the  medium  (Rippel  and  Behr,  1934). 

The  problem  of  biological  substitution  arose  early  in  the  study  of  fungus 
nutrition.  Biological  substitution  means  that  one  element  can  replace 
another,  in  whole  or  in  part.  The  possibility  of  biological  substitution 
was  investigated  by  Steinberg  (1946)  using  A.  niger  as  the  test  fungus. 
This  investigation  was  made  to  determine  whether  the  alkali  metals 
(lithium,  sodium,  rubidium,  or  cesium)  could  replace  potassium,  and 
whether  the  alkaline-earth  metals  (calcium,  beryllium,  strontium,  or 
barium)  could  replace  magnesium.  Under  these  conditions  sodium  and 
beryllium  gave  increased  yield  of  mycelium  in  media  containing  sub- 
optimal  amounts  of  potassium  and  magnesium.  These  effects  are 
illustrated  in  Table  9. 

Some  increases  in  weight  of  mycelium  were  noted  under  certain  con- 
ditions with  some  of  the  other  metallic  ions  tested,  but  the  effects  of 
these  elements  were  ascribed  to  ion  antagonism. 

Studies  of  biological  substitution  require  great  care  and  a  detailed  and 
extensive  knowledge  of  the  composition  of  the  media  and  of  the  behavior 
of  the  fungus  under  the  experimental  conditions  used. 

Magnesium.  This  element  is  one  of  the  alkaline-earth  group.  It  is 
essential  for  green  plants  and  animals  as  well  as  for  fungi  and  bacteria. 
Aspergillus  niger  has  been  more  carefully  investigated  with  respect  to  the 


ESSENTIAL  METALLIC  ELEMENTS 


69 


effects  of  magnesium  than  any  other  fungus.  Within  certain  limits  of 
concentration,  the  amount  of  growth  of  A.  niger  is  proportional  to  the 
concentration  of  magnesium  in  the  medium.  This  has  been  demon- 
strated by  Steinberg  (1946),  Lavollay  and  Laborey  (1938),  and  others. 
The  application  of  this  principle  to  the  microbiological  assay  of  magne- 
sium is  discussed  in  Chap.  10.  Penicillium  glaucum,  Botrytis  cinerea,  and 
Alternaria  tenuis  failed  to  grow  in  the  absence  of  magnesium  (Rabinovitz- 
Sereni,  1933).  Excess  magnesium  was  not  harmful  to  these  three  fungi 
until  the  concentration  of  magnesium  sulfate  in  the  medium  reached  about 
40  per  cent.  These  three  species  were  able  to  grow  in  the  presence  of 
traces  of  magnesium  but  sporulated  only  when  the  concentration  of  mag- 
nesium was  increased.  Respiration  also  increased  as  the  magnesium  con- 
tent of  the  medium  increased.  Failure  to  sporulate  unless  sufficient 
magnesium  is  available  is  probably  to  be  expected  with  many  fungi. 

Table  9.     The  Effect  of  50  Milligrams  of  Sodium  on  the  Amount  of  Mycelium 

Produced  by  Aspergillus  niger  in  an  Optimal  Medium  Containing  Twice 

THE  Normal  Amounts  of  INIicro  Elements  when  the  Concentration 

OF  Potassium  Was  Varied 
(Steinberg,  Avi.  Jour.  Botany  33,  1946.) 


Potassium, 
mg.  per  liter 

Control,  mg. 
myceUum 

Sodium  added, 

50  mg.  per  liter, 

mg.  mycelium 

15 

256.3 

401.3 

30 

446.1 

783.1 

45 

641.2 

896.7 

60 

823.4 

1,042.0 

75 

955,2 

1,089.0 

90 

988.0 

1,070.0 

105 

1,065.2 

1,093.1 

120 

1,059.2 

1,095.5 

135 

1,113.9 

1,084.9 

150 

1,145.9 

1,146.5 

Most  of  the  magnesium  in  the  mycelium  of  Aspergillus  niger  can  be 
extracted  by  means  of  dilute  acids  (Ripple  and  Behr,  1930),  which  indi- 
cates that  this  element  does  not  form  stable  organic  compounds.  A  rela- 
tion between  the  optimum  concentrations  of  magnesium  and  phosphorus 
for  A.  niger  was  discovered  by  Laborey  et  at.  (1941).  Some  36  phosphate 
ions  are  required  for  every  ion  of  magnesium.  Many  enzyme  sj^stems  are 
activated  by  magnesium  ion,  and  in  view  of  the  role  of  phosphate  in 
enzymatic  transformations  it  is  not  surprising  that  there  should  be  a  close 
relation  between  magnesium  and  phosphate  concentrations.  Magnesium 
is  involved  in  many  of  the  enzymatic  reactions  involved  in  fermentation 


70  PHYSIOLOGY  OF  THE  FUNGI 

(Sumner  and  Somers,  1947).  It  is  equally  likely  that  magnesium  is 
involved  in  aerobic  oxidation  of  carbohydrate.  Low  concentrations  of 
magnesium  in  the  medium  led  to  increased  synthesis  of  riboflavin  by  A. 
niger  (Lavollay  and  Laborey,  1938). 

One  ion  may  affect  the  physiological  action  of  another.  This  is  called 
ion  antagonism.  In  nature  and  in  the  laboratory  fungi  come  in  contact 
with  compounds  of  both  essential  and  nonessential  elements.  Many  of 
the  nonessential  elements  are  toxic,  although  toxicity  is  not  limited  to  the 
nonessential  elements.  Copper  is  an  essential  element,  but  it  is  toxic  to 
most  fungi  when  the  concentration  exceeds  certain  limits  (Chap.  12). 
The  toxic  effect  of  an  ion  may  be  overcome  by  the  presence  of  one  or  more 
other  ions  in  the  medium.  Gortner  (1949)  has  reviewed  this  subject  from 
the  standpoint  of  colloidal  chemistry  and  suggests  that  the  relative 
concentrations  of  various  metallic  ions  may  regulate  the  process  of 
adsorption. 

As  an  example  of  ion  antagonism  Lohrmann  (1940)  described  the  toxic 
action  of  mercuric  chloride  and  boric  acid  on  Aspergillus  niger,  A.  flavus, 
Mucor  pusillus,  Penicillium  glaucum,  Fusarium  coeruleum,  Cunning- 
hamella  elegans,  Ahsidia  cylindrospora,  and  Rhizopus  nigricans.  The 
inhibition  caused  by  either  of  these  toxic  compounds  was  overcome  in 
part  by  increasing  the  concentration  of  magnesium  sulfate.  Similarly, 
the  toxic  effects  of  high  concentrations  of  magnesium  sulfate  were  over- 
come by  mercuric  chloride.  Either  mercuric  chloride  or  boric  acid  in 
certain  concentrations  "stimulated"  growth  in  the  nutrient  solution  used. 
This  is  not  evidence  that  either  boron  or  mercury  is  an  essential  element, 
but  it  does  show  that  the  nutrient  solution  used  was  unbalanced.  The 
effect  of  sodium  and  calcium  ions  upon  growth  and  respiration  of  A.  niger 
depended  upon  the  ratio  of  these  nonessential  ions  present  in  the  medium. 
A  sodium-calcium  ratio  of  19  to  1  gave  the  highest  rate  of  respiration, 
while  a  ratio  of  4  to  1  was  most  favorable  for  growth  (Gustafson,  1919). 
Aluminum  inhibits  the  production  of  itaconic  acid  by  A.  terreus.  This 
inhibition  is  overcome  by  magnesium  sulfate  (Lockwood  and  Reeves, 
1945).  Nickerson  (1946)  found  the  inhibitory  effects  of  zinc  ion  on  the 
rate  of  respiration  of  Epidermophyton  floccosum  to  be  reversed  by  calcium 
or  magnesium  ions. 

The  phenomenon  of  antagonism  is  not  confined  to  ions.  Organic 
compounds  present  in  media  may  modify  the  activity  of  ions,  and  organic 
compounds  may  antagonize  the  physiological  activity  of  other  organic 
compounds  (Chap.  11).  All  these  possibihties  exist.  Whether  a  given 
ion  or  compound  will  be  physiologically  active  depends  upon  the  other 
constituents  of  the  medium  and  the  metabolic  compounds  excreted  by  the 
fungus  under  study. 


ESSENTIAL  METALLIC  ELEMENTS  71 

ESSENTIAL  MICRO  ELEMENTS 

These  elements  have  been  called  heavy-metal  nutrients,  trace  elements, 
micronutrients,  and  minor  elements.  The  literature  on  this  subject  is 
extensive  and  often  conflicting.  Reviews  of  this  subject  are  given  by 
Perlman  (1949),  Foster  (1939),  and  Steinberg  (1939).  A  collection  of 
10,000  abstracts  on  the  effects  of  the  micro  elements  on  green  plants  and 
animals  has  been  published  by  the  Chilean  Nitrate  Educational  Bureau 
(1948). 

In  spite  of  Raulin's  (1869)  discovery  that  iron  and  zinc  are  essential  for 
Aspergillus  niger,  there  arose  a  school  of  investigators  who  considered  tho 
micro  elements  to  be  stimulatory  rather  than  essential.  This  view  is  no 
longer  held.     There  are  a  number  of  reasons  for  this  misinterpretation: 

(1)  The  failure  to  realize  that  the  "chemically  pure"  compounds  used  in 
preparing  media  are  grossly  contaminated  from  the  biological  standpoint 
and  that  rigorous  purification  of  media  is  essential  in  work  of  this  kind. 

(2)  Distilled  water  is  often  a  source  of  metallic  ions  unless  it  has  been 
redistilled  in  Pyrex,  or  preferably  quartz,  stills.  (3)  Many  kinds  of 
chemical  glassware  are  sufficiently  soluble  to  furnish  the  fungi  all  or  a  part 
of  the  micro  elements  required.  (4)  The  inoculum,  whether  mycelium  or 
spores,  may  introduce  sufficient  micro  elements  to  obscure  the  need  for 
these  elements.  Serial  transfer  using  media  free  from  the  element  in 
question  and  the  use  of  small  inocula  minimizes  this  source  of  error. 

Steinberg  (1936)  has  indicated  that  the  optimum  concentration  of  the 
essential  micro  metallic  elements  for  A.  niger  ranges  from  0.3  mg.  of  iron 
to  0.02  mg.  of  gallium  per  liter  of  medium.  Lest  the  reader  conclude  that 
these  concentrations  are  so  small  as  to  be  meaningless,  it  is  revealing  to 
calculate  the  number  of  atoms  of  iron  in  0.3  mg.  From  the  atomic  weight 
of  iron  and  Avagadro's  number  it  may  be  calculated  that  there  are  about 
3  X  10^^  atoms  in  0.3  mg.  of  iron.  If  the  number  of  cells  produced  by 
A.  niger  under  these  conditions  were  known,  the  number  of  iron  atoms 
available  for  each  cell  could  be  calculated.  In  lieu  of  this  information  we 
may  use  data  from  experiments  on  the  number  of  yeast  cells  produced  in 
a  liter  of  medium.  Under  favorable  conditions  there  are  roughly  500 
billion  yeast  cells  produced  in  a  liter  of  medium  (Stark  et  al.,  1941).  If 
A.  niger  produces  the  same  number  of  cells  per  liter  as  yeast,  there  would 
be  available  6.4  X  10^  atoms  of  iron  per  cell. 

The  prime  essential  in  investigations  dealing  with  the  effects  of  the 
essential  micro  elements  is  a  medium  free  from  the  element  under  study. 
This  ideal  is  difficult  to  attain  in  practice.  Equal  care  is  necessary  in  the 
choice  of  culture  vessels,  for  it  is  wasted  effort  to  remove  an  element  from 
the  medium  rigorously  and  then  contaminate  it  by  using  glassware  which 


72  PHYSIOLOGY  OF  THE  FUNGI 

furnishes  the  metal.  The  culture  A^essels  should  be  of  quartz  for  work  of 
the  most  exacting  kind,  although  Pyrex  or  other  suitable  glassware  may 
be  used.  It  is  desirable  in  any  event  to  use  a  few  quartz  culture  vessels  as 
controls.  In  part,  the  long  controversy  over  the  effect  of  zinc  on  fungi 
was  due  to  the  liberation  of  sufficient  amounts  of  this  element  from  certain 
kinds  of  glassware  used  as  culture  vessels.  Javillier  (1914)  showed  that 
the  addition  of  zinc  to  cultures  of  A .  niger  has  little  effect  when  Jena  glass 
culture  flasks  were  used.  When  quartz  vessels  were  used,  the  crop 
increased  from  291  mg.  in  the  control  without  added  zinc  to  1,624  mg. 
when  zinc  was  added.  Steinberg  (1919)  found  essentially  the  same 
results  except  that  zinc  deficiency  could  be  demonstrated  for  A.  niger 
when  Pyrex  vessels  were  used  (Table  10) . 

Table  10.     The  Average  Weight  of  Five  Cultures  of  Aspergillus  niger  Culti- 
vated ON  the  Same  Basal  Medium  in  Three  Makes  of  Glassware 
(Steinberg,  Am.  Jour.  Botany  6,  1919.) 


Make  of  glassware 


Jena 

Kavalier  Bohemian . 
Pyrex 


Mg.  mycelium 


Zinc,  10  mg.  per  liter 

987 
943 
957 


Purification  of  culture  media.  Progress  in  the  study  of  essential  micro 
elements  depends  upon  methods  of  removing  them  from  media.  As  long 
as  these  elements  occur  in  the  ingredients  of  the  media,  their  need  may  be 
unnoticed  and  unsuspected.  In  1919  Steinberg  devised  a  useful  method 
of  reducing  the  concentration  of  heavy  metals,  especially  iron  and  zinc,  in 
media.  In  essentials,  this  method  consists  in  autoclaving  the  complete 
medium  with  15  g.  per  liter  of  calcium  carbonate.  The  hot  solution  after 
autoclaving  is  filtered  through  paper  or  a  fritted-glass  filter,  or  allowed  to 
cool  and  the  supernatant  liquid  decanted  off.  The  precipitate  must  be 
removed;  otherwise  the  essential  elements  will  be  released  by  the  fungi. 
Calcium  oxide  and  magnesium  carbonate  may  replace  calcium  carbonate 
in  some  applications  (Steinberg,  193oa) .  The  mode  of  purification  appears 
to  be  as  follows:  During  autoclaving,  heavy-metal  carbonates  or  their 
hydroxides  are  formed.  The  excess  calcium  carbonate  serves  to  adsorb 
these  insoluble  compounds  The  composition  of  a  medium  is  somewhat 
changed  by  this  treatment,  part  of  the  phosphate  being  removed  as  cal- 
cium phosphate.  In  practice  this  is  compensated  by  using  an  excess  of 
phosphate.  A  medium  which  is  treated  by  this  process  is  essentially 
neutral  in  reaction,  which  may  lead  to  some  changes  in  the  sugar  during 
autoclaving. 


ESSENTIAL  METALLIC  ELEMENTS  73 

Sugars  are  frequently  highly  contaminated  with  metallic  compounds. 
Steinberg  (1937)  has  reported  a  sample  of  glucose  to  contain  the  following 
elements:  lithium,  sodium,  strontium,  calcium,  rubidium,  potassium, 
manganese,  aluminum,  iron,  rhodium,  nickel,  silver,  copper,  magnesium, 
tin,  boron,  and  silicon.  The  metallic  contamination  of  non-ionic  com- 
pounds such  as  the  sugars  can  be  sharply  reduced  by  a  variety  of  mild 
procedures.  Shu  and  Johnson  (1948)  give  these  details  for  an  aluminum 
hydroxide  coprecipitation  method:  To  140  g.  of  glucose  contained  in  500 
ml.  of  solution,  1.25  g.  of  Al2(S04)3'18H20  were  added.  Dilute  ammo- 
nium hydroxide  was  added  until  the  pH  rose  to  9,  and  the  precipitate  of 
A1(0H)3  Avith  the  adsorbed  impurities  was  filtered  off.  This  treatment 
was  repeated  until  the  desired  degree  of  purification  was  attained.  Non- 
ionic  substances  such  as  glucose  and  urea  may  be  purified  by  treatment 
with  cation-exchange  materials  operating  on  the  hydrogen  cycle.  Perl- 
man  (1945)  used  Zeo-karb  H  (Permutit  Corporation)  for  this  purpose. 
Various  ion-exchange  materials  are  used  to  purify  beet  juice  in  the  manu- 
facture of  sugar.  Mulder  (1939-1940)  found  the  combination  of  ammo- 
nium sulfide  and  Norit  to  be  efficient  in  removing  copper  from  media. 
The  sulfide  ion  forms  insoluble  heavy-metal  sulfides  while  the  activated 
carbon  serves  as  a  "gatherer." 

Complex-forming  reagents  such  as  diphenylthiocarbazone  (dithizone) 
(Stout  and  Arnon,  1939),  and  8-hydroxyquinoline  (Waring  and  Werkman, 
1943)  are  useful  in  removing  heavy-metal  ions  from,  or  testing  the  purity 
of,  salts  used  in  preparing  media.  These  reagents  and  the  metal  com- 
plexes they  form  are  removed  from  solutions  by  extraction  with  chloro- 
form or  other  organic  solvents.  The  chemistry  of  complex  formation 
between  organic  compounds  and  ions  is  treated  by  Yoe  and  Sarver  (1941). 
Others  have  merely  added  such  complex-forming  reagents  to  the  media, 
in  which  the  various  metallic  ions  combine  with  the  reagent  to  form  non- 
available  compounds.  The  specificity  of  the  reagent,  concentrations  of 
reagent  and  the  metallic  ion  or  ions,  the  pH  of  the  medium,  as  well  as  the 
stability  of  the  complex,  enter  into  the  success  of  this  type  of  treatment. 
Hickey  (1945)  found  that  2,2'-bipyridine  inactivated  ferrous  iron  in  media 
treated  with  this  reagent.  It  is  better  to  remove  metallic  impurities  from 
the  media  by  extraction  than  to  depend  upon  complexing  compounds  to 
hold  these  ions  in  non-ionic  combination. 

Certain  compounds  used  in  making  media  such  as  the  amino  acids  and 
hydroxy  acids  form  non-ionized  complexes  with  various  metallic  ions. 
Media  containing  these  types  of  compounds  are  difficult  to  free  from 
metallic  contamination.  In  addition,  some  fungi  excrete  hydroxy  acids, 
such  as  citric  acid,  which  may  modify  the  availability  of  the  essential 
micro  elements. 

Media  may  be  freed  of  essential  micro  elements  by  a  biological  process. 


74  PHYSIOLOGY  OF  THE  FUNGI 

If  a  fungus  is  grown  on  a  medium,  it  will  absorb  and  utilize  the  essential 
elements  present  in  the  medium.  The  success  of  this  procedure  depends 
upon  having  a  low  initial  concentration  of  the  essential  micro  elements, 
which  soon  become  exhausted  so  that  the  culture  liquid  no  longer  supports 
growth.  Removal  of  the  mycelium  will  thus  remove  the  elements  which 
have  been  taken  up.  The  culture  filtrate  may  then  be  used  as  a  medium 
relatively  free  of  micro  elements.  However,  fungi  excrete  various  com- 
pounds which  may  affect  the  results.  MacLeod  and  Snell  (1947)  have 
recently  utilized  this  method  in  studying  the  mineral  nutrition  of  some 
lactic  acid  bacteria. 

Iron.  Raulin's  claim  that  iron  was  essential  for  fungi  was  questioned 
at  first,  but  his  findings  were  soon  confirmed.  So  far  as  is  known,  iron  is 
essential  for  all  fungi.  It  may  be  noted  that,  in  the  absence  of  another 
essential  element  in  the  medium,  iron  alone  may  cause  little  or  no  response. 
If  the  zinc  content  of  a  medium  is  low,  the  addition  of  iron  to  an  iron-free 
medium  will  have  little  effect.  This  situation  is  true  of  any  essential 
nutrient.  Only  one  element  may  be  studied  at  a  time,  but  all  the  other 
essential  nutrients  must  be  present  before  the  effect  of  the  nutrient  under 
investigation  can  be  studied.  Some  results  of  Steinberg  (1919)  with 
Aspergillus  niger  on  media  purified  by  the  calcium  carbonate  method  are 
given  in  Table  11.  Neither  iron  nor  zinc  alone  had  much  effect  on  the 
growth  of  A.  niger,  since  both  of  these  elements  are  essential  for  this 
fungus. 

Table  11.     The  Effect  of  Iron  and  Zinc,  Singly  and  in  Combination,  on  the 

Amount  of  Growth  of  Aspergillus  niger 

(Steinberg,  Am.  Jour.  Botany  6,  1919.) 
Essential  Micro  Element  Mg.  Mycelium 

Added 

Control  (none  added) 18 

Iron 44 

Zinc 40 

Iron  plus  Zinc 731 

Little  interest  has  been  shown  in  recent  years  in  proving  iron  to  be  an 
essential  element  for  a  large  number  of  fungi.  In  view  of  the  almost  uni- 
versal occurrence  of  a  group  of  iron-containing  enzymes  (catalase,  the 
cytochromes,  cytochrome  oxidase,  etc.),  the  essential  role  of  iron  is  taken 
for  granted. 

The  most  obvious  effect  of  suboptimal  iron  concentrations  upon  fungi 
is  decreased  growth.  This  result  is  probably  due  to  the  decreased  and 
limited  amounts  of  iron-containing  enzymes  formed  under  these  condi- 
tions. It  was  shown  by  Yoshimura  (1939-1940)  that  the  amount  of 
catalase  produced  by  Aspergillus  oryzae  increased  as  the  amount  of  iron 
in  the  medium  increased.     Lilly  and  Leonian  (1945)  showed  that  a  rela- 


ESSENTIAL  METALLIC  ELEMENTS 


to 


tion  existed  between  the  amount  of  iron  supplied  in  the  medium  and  the 
ability  of  Rhizohium  trijolii  to  synthesize  certain  vitamins.  In  the  pres- 
ence of  suboptimal  concentrations  of  iron  the  addition  of  certain  vitamins 
replaced  iron  to  a  certain  degree.  A  quantitative  study  of  the  ^ntamins 
synthesized  by  Torulopsis  utilis  has  shown  the  iron  concentration  to  be 
important  (Lewis,  1944).  Increased  amounts  of  thiamine,  riboflavin, 
nicotinic  acid,  and  pyridoxine  were  synthesized  on  media  low  in  iron,  while 
the  amounts  of  biotin,  inositol  and  p-aminobenzoic  acid  were  decreased. 


L)  10  15  20 

Mg.  ferric  sulfate  per  liter 

Fig.  14.  The  effect  of  iron  [Fe2 (804)3]  in  overcoming  the  inhibitory  action  of  copper 
(CuS04-5H20)  on  the  production  of  penicilUn  by  Penicillium  chnjsogenum  X-1612. 
An  amount  of  copper  sufficient  to  inhibit  peniciUin  production  entirely  did  not  affect 
the  amount  of  growth.  The  fungus  was  cultured  submerged  in  a  lactose-starch- 
dextrin-ranmonium  sulfate  medium  for  7  days.  (Curves  drawn  from  data  of  Koffler 
et  al.,  Jour.  Bad.  53 :  120,  1947.  Pubhshed  by  permission  of  The  Williams  &  Wilkins 
Company.) 

There  has  been  a  great  deal  of  interest  in  the  effects  of  iron  and  other 
metallic  ions  on  various  microbiological  processes.  Perlman  et  al.  (1946) 
have  shown  that  the  iron  concentration  is  an  important  factor  in  citric 
acid  fermentation  by  Aspergillus  niger.  The  optimum  iron  concentration 
for  citric  acid  production  varied  over  tenfold  for  different  strains  of 
A.  niger.  The  effect  of  iron  on  penicillin  production  has  been  studied  by 
Kofl^ler  et  al.  (1947),  who  concluded  that  the  effect  of  the  ash  of  corn  steep 
is  due  to  iron  and  phosphate.  Chromium  increased  penicillin  production 
above  that  obtained  with  iron  and  phosphate,  presumably  by  neutralizing 
the  effect  of  other  ions.  Similarly  an  antagonism  was  shown  to  exist 
between  copper  and  iron.  The  antagonistic  effect  of  copper  and  iron  on 
the  production  of  penicillin  by  Penicillium  chrysogenum  X-1612  is  shoA\Ti 
in  Fig.  14. 


76  PHYSIOLOGY  OF  THE  FUNGI 

The  iron  concentration  of  the  medium  has  been  shown  to  affect  the 
amount  of  pigmentation  of  Torulopsis  pulchcrrima  (Roberts,  1946). 

Zinc.  This  element  is  essential  for  Aspergillus  niger  (Raulin,  1869; 
Steinberg,  1919).  Foster  (1939)  lists  Trycliophytoninterdigitale,  Rhizopus 
nigricans,  and  Saccharomyces  cerevisiae  as  recjuiring  zinc,  and  Roberg 
(1928)  found  zinc  to  be  essential  for  A.  fumigatus  and  A.  oryzae.  Blank 
(1941)  reported  the  amount  of  growth  oi  Phymatotrichum  omnivorum  to  be 
increased  by  the  addition  of  zinc  to  a  medium  treated  with  calcium  car- 
bonate, and  Perlman  (1948)  noted  that  the  sclerotia  of  Sclerotium 
delphinii  are  more  highly  pigmented  in  the  presence  of  added  zinc. 

Zinc  ions  activate  (and  inhibit)  various  enzymes  such  as  enolase  and 
dipeptidase.  Zinc  is  contained  in  carbonic  anhydrase,  an  enzyme  which 
catalyzes  the  decomposition  of  carbonic  acid  to  carbon  dioxide  and  water. 
In  addition  to  these  specific  uses  the  zinc  concentration  has  a  decided 
effect  on  a  number  of  physiological  or  biochemical  processes  in  fungi. 
Foster  and  Waksman  (1939)  found  that  the  production  of  fumaric  acid 
from  glucose  by  Rhizopus  nigricans  varied  according  to  the  amount  of  zinc 
added  to  the  medium.  Fumaric  acid  was  produced  most  efficiently  when 
the  concentration  of  zinc  was  low  (1.2  mg.  per  liter).  Higher  concentra- 
tions of  zinc  resulted  in  increased  growth  and  decreased  production  of 
fumaric  acid.  From  these  results  it  appears  that  zinc  plays  a  role  in  the 
utilization  of  glucose,  the  completeness  of  oxidation  and  assimilation 
being  favored  by  relatively  high  concentrations  of  zinc.  A  somewhat 
similar  effect  of  zinc  on  the  production  of  lactic  acid  by  Rhizopus  sp.  has 
been  noted  (Waksman  and  Foster,  1938).  Zinc  was  found  to  cause 
increased  growth  and  a  decrease  in  the  production  of  lactic  acid,  while  the 
effect  of  iron  is  to  increase  the  yield  of  lactic  acid.  For  a  further  dis- 
cussion of  the  mechanism  of  zinc  in  fungus  metabolism,  see  Foster  (1949). 

Copper.  This  element  is  essential  for  animals,  green  plants,  and  fungi. 
From  the  work  of  Steinberg  (1936)  it  appears  that  0.04  mg.  of  added 
copper  per  liter  of  purified  medium  is  sufficient  for  the  maximum  growth 
of  Aspergillus  niger.  Under  these  conditions  omission  of  copper  decreased 
the  yield  only  from  984.8  to  774.3  mg.  It  is  probable  that  purification  of 
the  medium  by  the  calcium  carbonate  treatment  is  not  very  satisfactory 
for  this  element.  The  ^veight  of  metal  needed  to  obtain  maximum  growth 
with  A.  niger  is  much  less  for  copper  than  for  iron  or  zinc.  The  experi- 
mental difficulties  increase  as  the  amount  of  a  micro  element  needed 
becomes  less.  Apparently  it  is  very  difficult  to  prepare  a  copper-free 
medium.  Roberg  (1931)  made  use  of  Bortel's  method  of  adding  a  trace 
of  ammonium  sulfide  to  convert  heavy  metals  to  sulfides  and  adsorbing 
these  impurities  with  charcoal.  This  treatment  is  very  efficient  in  remov- 
ing iron  and  zinc  but  somewhat  less  satisfactory  for  removing  copper. 
The  essential  nature  of  copper  for  A.  flavus  and  Rhizopus  nigricans  was 


ESSENTIAL  METALLIC  ELEMENTS 


77 


shown  by  McHargue  and  Calfee  (1931).  The  full  effect  of  copper  was 
dependent  upon  the  presence  of  other  essential  elements.  The  coloration 
of  conidia  of  A.  niger  has  been  shown  to  depend  upon  the  copper  content 
of  the  medium  (Javillier,  1939). 

Although  copper  is  an  essential  element,  it  is  a  constituent  of  many 
fungicides  (Chap.  12).  The  concentration,  therefore,  is  a  very  important 
consideration  in  studying  the  effect  of  this  element.  The  phenomenon  of 
ion  antagonism  must  also  be  considered,  for  the  effect  of  a  given  amount 
of  copper  is  dependent  upon  the  other  constituents  of  the  medium. 
Marsh  (1945)  investigated  the  antagonistic  effects  of  three  salts  upon 
copper   as   it   affected   germination   of   conidia   of   Sclerotinia  fructicola 

Table  12.     The  Antagonistic  Effect  of  Three  Salts  on  Copper  as  Shown  by 

THE  Germination  of  Conidia  of  Sclerotinia  fructicola 

(Marsh,  Phytopathology  35,  1945.) 


Salt  concen- 
tration 

Percentage  Germination  in 

4  X  10-^/  CuS04,  plus 

0.01%  glucose 

MgS04 

CaCh 

KCl 

0.0 

1.2 

0.8 

2.9 

10-^M 

54.0 

31.0 

3.9 

10-^il/ 

67.0 

62.0 

2.6 

IQ-'M 

78.0 

83.0 

3.9 

10-2M 

— 

59.0 

(Table  12).  It  was  shown  that  the  mechanism  of  the  protective  action  of 
these  salts  was  to  decrease  absorption  of  copper.  There  is  no  reason  to 
assume  that  the  absorption  and  utilization  of  copper  from  nutrient  solu- 
tions would  not  be  affected  similarly.  Thus,  the  amount  of  copper  added 
to  a  nutrient  solution  may  reflect  only  imperfectly  the  amount  absorbed 
and  used  by  a  fungus. 

It  was  noted  in  Chap.  4  that  copper  is  an  essential  constituent  of  certain 
enzymes,  including  tyrosinase,  which  occurs  in  many  fungi.  Nelson  and 
Dawson  (1944)  suggest  that  tyrosinase  functions  in  the  respiration  chain 
as  an  oxygen  shuttle. 

Manganese.  The  classification  of  this  element  as  essential  rests  upon 
the  experimental  findings  that  omission  of  this  element  from  media 
results  in  decreased  yields.  The  multiplication  of  examples  strengthens 
the  validity  of  this  conclusion,  although  most  investigators  have  confined 
their  attention  to  a  relatively  few  species.  The  results  of  Robbins  and 
Hervey  (1944)  with  Pythiomorpha  gonapodyoides  indicate  that  investiga- 
tion of  fungi  other  than  Aspergillus  niger  with  regard  to  micro-element 


78 


I'lIYSlOWGY  OF  THE  FUSGl 


nutrition  may  be  rewarding.  It  was  unnecessary  to  resort  to  elaborate 
methods  of  medium  purification  to  demonstrate  that  manganese  is  essen- 
tial for  P.  gonapodyoides.  This  situation  occurred  only  when  reagent 
magnesium  sulfate  of  a  certain  manufacture  was  used.  Substitution  of 
another  brand  of  magnesium  sulfate  revealed  heavy  (biological)  con- 
tamination by  manganese  (Fig.  15).  The  inoculum  was  found  to  carry 
sufficient  manganese  and  other  micro  elements  to  influence  the  amount  of 
growth  in  the  first  passage.     No  growth  resulted  in  the  third  passage  in 


A  B 

Fig.  15.  Pythiomorpha  gonapodyoides  growing  in  a  basal  solution  \Yith  no  added 
mineral  supplements.  A,  medium  prepared  with  Baker's  Analyzed  magnesium 
sulfate.  B,  medium  prepared  with  Mallinckrodt's  magnesium  sulfate  analytical 
reagent.  Age,  5  days.  Note  the  extensive  white  mj'celium  in  A  and  the  slight 
growth  in  B.  (Courtesy  of  Robbins  and  Hervey,  Bull.  Torrey  Botan.  Club  71:  263, 
1944.) 

the  absence  of  added  manganese.  The  range  of  manganese  concentra- 
tions for  optimum  growth  was  narrow  and  appeared  to  depend  upon  the 
concentration  of  other  micro  elements  present,  particularly  zinc.  Stein- 
berg (1935)  found  manganese  to  be  essential  for  A.  niger.  McHargue  and 
Calfee  (1931,  1931a)  noted  that  growth  of  A.  flainis,  Rhizopus  nigricans, 
and  Saccharomyces  cerevisiae  increased  in  the  presence  of  added 
manganese.  Steinberg  (1945)  showed  that  omission  of  manganese  from 
a  balanced  medium  resulted  in  a  decrease  in  yield  of  A.  niger  from 
1,084.8  to  356.6  mg.  No  spores  formed  when  manganese  was  omitted. 
It  is  interesting  to  note  that,  as  the  numbers  of  spores  used  for  inoculum 


ESSENTIAL  METALLIC  ELEMENTS  79 

decreased,  A.  niger  became  more  sensitive  to  micro-element  deficiencies 
in  the  medium.  The  favorable  effect  of  adding  biotin  to  the  medium 
when  only  a  few  spores  were  used  as  inoculum  suggests  an  intimate  con- 
nection between  micro-element  nutrition  and  the  synthesis  of  this  vita- 
min. Whether  the  decreased  yield  due  to  small  inoculum  was  due  to 
other  deficiencies  or  to  a  decreased  rate  of  growth  is  not  entirely  clear,  as 
all  harvests  were  made  after  4  days. 

Manganese  (Mn++)  has  been  shown  to  be  the  natural  activator  of 
yeast  arginase.  Other  enzymes  are  activated  by  this  element  (Sumner 
and  Somers,  1947).  In  view  of  the  small  amounts  of  manganese  required 
by  fungi,  it  may  be  assumed  that  manganese  functions  as  a  constituent  of 
various  enzymes. 

Molybdenum.  The  study  of  the  role  of  this  element  emphasizes  the 
similarity  in  certain  physiological  processes  throughout  the  plant  king- 
dom. The  most  striking  feature  of  this  essential  element  is  its  role  in 
nitrogen  metabolism.  The  utilization  of  nitrate  nitrogen  by  green  plants 
and  fungi  and  the  fixation  of  atmospheric  nitrogen  by  bacteria  {Azotohac- 
terchroococcum,  Clostridium  pasteimanum)  is  dependent  upon  molybdenum 
(Bortels,  1930,  1936). 

Our  knowledge  of  the  effect  of  molybdenum  on  fungi  is  largely  confined 
to  Aspergillus  niger.  Steinberg  (1936,  1937)  found  that  more  molybde- 
num was  required  by  A.  niger  for  maximum  growth  in  media  containing 
nitrate  nitrogen  than  in  media  with  ammonium  nitrogen.  Steinberg 
expressed  the  opinion  that  molybdenum  is  essential  for  A .  niger  even  when 
ammonium  nitrogen  is  available.  Additional  studies  on  A.  niger  and 
other  organisms  (Mulder,  1948)  indicated  that  an  increased  need  for 
molybdenum  is  associated  with  nitrate  utilization.  It  may  be  assumed 
that  the  enzymatic  reduction  of  nitrate  is  carried  out  by  enzymes  which 
require  molybdenum  as  an  activator. 

In  view  of  the  important  role  of  molj^bdenum  in  the  utilization  of 
nitrate  nitrogen,  care  should  be  used  in  comparing  the  value  of  different 
nitrates.  Unless  sufficient  molybdenum  is  present,  misinterpretations 
may  result.  Steinberg  (1937)  found  the  amount  of  molybdenum  present 
as  an  impurity  in  various  nitrates  to  vary.  One  sample  of  calcium  nitrate 
contained  enough  molybdenum  to  support  maximum  growth  of  A.  niger. 
Perhaps  the  report  of  Young  and  Bennett  (1922)  that  many  fungi  made 
better  growth  on  calcium  than  on  potassium  nitrate  may  be  partially 
explained  on  the  basis  of  the  molybdenum  content  of  these  two  salts. 
This  explanation,  of  course,  must  allow  for  the  effect  of  calcium,  which  is 
now  known  to  be  essential  for  certain  fungi. 

Calcium.  This  element  was  one  of  the  first  to  be  recognized  as  essen- 
tial for  green  plants  and  animals.  In  1922,  Young  and  Bennett  reported 
that  Rhizocionia  solani  made  no  growth  in  the  absence  of  this  element 


80 


PHYSIOLOGY  OF  THE  FUNGI 


This  report  apparently  attracted  little  attention  since  most  investigators 
working  on  this  problem  confined  their  attention  to  Aspergillus  niger. 
The  value  of  using  more  than  one  fungus  to  demonstrate  the  essential 
nature  of  calcium  was  strikingly  shown  by  Steinberg  (1948,  1950).  These 
data  are  given  in  Table  13. 

It  is  evident  from  the  data  in  Table  13  that  the  essential  nature  of  cal- 
cium for  certain  fungi  is  established.  The  concentrations  of  calcium 
required  for  maximum  growth  varied  from  2  to  G  mg.  per  liter  of  medium. 
On  the  other  hand,  neither  A.  niger  nor  Fusariuni  oxysporum  needs  more 

Table   13.     Effect  of  the   Omission   of   Calcium  from  the   Medium  on  the 

Growth  of  Seven  Fungi 
(Steinberg,  Science  107,  1948.) 


Fungus 


Aspergillus  niger 

Rhizoctonia  solani 

Sclerotium  I'olfsii 

Cercos-pora  nicotianae 

Fusarium  oxysporum  var.  nicotianae. 

Pythium  irregulars 

Thielaviopsis  basicola 


Calcium  added, 

Calcium  not  added, 

mg.  mycelium 

percentage  of  yield 

1,250.0 

100.0 

1,215.1 

14.3 

1,082.3 

49.5 

1,380.2 

90.1* 

823.3 

100.0 

459.0 

60.1* 

364.2 

82.0* 

*  Asparagine  of  unknown  purity  was  used  as  a  sovirce  of  nitrogen. 

than  spectroscopic  traces  of  calcium,  if  they  require  this  element  at  all. 
Steinberg  is  of  the  opinion  that  further  advances  in  purity  of  nutrient 
solutions  will  reveal  more  uniformity  in  the  essential  element  require- 
ments of  organisms. 

Lindeberg  (1944)  has  demonstrated  a  synergistic  effect  between  manga- 
nese and  calcium  upon  the  growth  of  various  species  of  Marasmius. 

Table  14.     The  Effect  of  Increasing  Concentrations  of  Calcium  and  Manga- 
nese, Alone  and  in  Combination,  on  the  Growth  of  Marasmius  epiphyllus 
(Lindeberg,  Symbolae  Botan.  Upsalie7isis,  8:  2,  1944.) 
(Dry  weight  mycelium  in  milligrams.) 


Mn,  millimoles 
per  liter 

Ca,  millimoles  per  liter 

0 

0.005 

0.05 

0.5 

0.0 

0.0005 

0.005 

0.05 

0.5 

10.1 

11.1 

10.7 

18.2 
20.3 

19.8 
18.0 
18.3 
20.8 

35.8 

33.0 
38.4 
35.7 
47.6 
48.6 

73.5 

83.8 
78.5 
77.0 
52.6 

ESSENTIAL  METALLIC  ELEMENTS  81 

Within  limits,  the  growth  of  M.  aUiaceus  and  M.  epiphyllus  was  propor- 
tional to  the  concentration  of  either  of  these  elements,  and  the  response 
to  each  element  was  modified  by  the  presence  of  the  other.  The  data  in 
Table  14  illustrate  this  effect. 

In  addition  to  the  essential  micro  elements  discussed  above,  there  is 
some  evidence  which  indicates  the  essentiality  of  other  metallic  elements 
for  the  fungi.     Certain  of  these  elements  are  essential  for  other  organisms. 

Gallium.  Under  certain  conditions  Steinberg  (1938)  was  able  to  show 
that  omission  of  this  element  from  the  medium  led  to  decreased  yield  and 
sporulation  of  Aspergillus  niger.  Extraordinary  care  was  needed  to 
demonstrate  gallium  deficiency.  The  chemicals  used  were  spectroscopi- 
cally  pure  with  the  exception  of  traces  of  iron,  calcium,  and  sodium.  The 
sucrose,  after  6-hr.  extraction  with  alcohol,  contained  only  0.0014  per  cent 
ash.  The  water  used  was  triple-distilled,  the  last  distillation  being  made 
in  a  quartz  still.  Spectroscopically  pure  calcium  oxide  w^as  used  to  purify 
the  sucrose  further.  Under  these  conditions  the  yield  of  A.  niger 
increased  from  814  mg.  to  1,053  mg.  when  gallium  (0.02  mg.  per  liter)  was 
added  to  the  medium.  The  salts  of  76  other  chemical  elements  were 
tested,  and  none  was  found  to  replace  gallium.  In  view  of  the  similar 
chemical  behavior  of  gallium  and  aluminum,  Steinberg  considers  it  possi- 
ble that  the  biologic  activity  sometimes  attributed  to  aluminum  may  in 
reality  be  due  to  gallium. 

Scandium.  In  the  discussion  of  the  role  of  manganese  in  nitrogen 
metabolism  it  w^as  noted  that  the  amount  of  manganese  required  was 
determined  by  the  nitrogen  source  used.  In  a  somewhat  similar  fashion, 
Steinberg  (1939)  found  that  scandium  appeared  to  be  essential  when 
glycerol  was  used  as  a  carbon  source  for  Aspergillus  niger.  Growth  was 
poor  on  this  carbon  source ;  omission  of  copper  or  manganese  increased  the 
yield  somewhat.  Omission  of  scandium  decreased  the  yield  from  269.4  to 
107.4  mg.  Interestingly  enough,  scandium  appeared  to  have  no  effect  on 
growth  when  sucrose  was  used  as  a  source  of  carbon.  Addition  of  lysine 
or  proline  (20  mg.  per  liter)  to  the  glycerol  medium  increased  growth  and 
at  the  same  time  prevented  the  effect  of  scandium.  These  results  suggest 
that  the  need  for  certain  elements  may  be  shoA\Ti  only  under  certain 
nutritional  conditions. 

Vanadium.  Bertrand  (1943)  reported  the  presence  of  this  element  in 
all  fungi  examined.  Amanita  muscaria  contained  from  61  to  156  mg.  of 
vanadium  per  kilogram.  Bertrand  (1941)  considers  vanadium  as  an 
essential  element  for  Aspergillus  niger. 

Cobalt.  Whether  fungi  require  some,  or  all,  of  the  other  metallic  ele- 
ments required  by  other  organisms  is  not  kno^vn.  Cobalt  is  required  by 
animals.  Lack  of  sufficient  amounts  of  this  element  in  the  soil  causes 
severe  cobalt  deficiency  in  animals  which  are  pastured  on  such  soils. 


82 


PHYSIOLOGY  OF  THE  FUNGI 


Recently,  a  cobalt-containing  vitamin  (B12)  was  isolated.  This  vitamin 
is  synthesized  by  Streptomyces  griseus  (Rickes  et  al.,  1948)  and  some 
bacteria.  Whether  S.  griseus  requires  cobalt  as  an  essential  element  for 
growth  or  reproduction  is  not  known.  The  synthesis  of  this  vitamin  is 
necessarily  dependent  upon  a  supply  of  cobalt.  Some  bacteria  are  known 
to  be  deficient  for  vitamin  B12. 


PERIODICITY  OF  BIOLOGICALLY  ESSENTIAL  ELEMENTS 

Steinberg  (1938a),  Frey-Wyssling  (1935),  and  others  have  considered 
the  problem  of  biologically  essential  elements  in  relation  to  the  structure 

Table  15.     A  Portion  of  the  Periodic  Table  of  Elements  Based  on  Atomic 

Number 
The  biologically  essential  elements  are  set  in  italics.     Those  elements  essential  for 
fungi  are  marked  with  an  asterisk. 


Group 

Group 

Group 

Group 

Group 

Group 

Group 

Group 

Group 

0 

1 

2 

3 

4 

5 

6 

7 

8 

H* 

1 

He 

Li 

Ce 

B 

C* 

N* 

0* 

F 

2 

3 

4 

5 

6 

7 

8 

9 

Ne 

Na 

Mg* 

Al 

Si 

P* 

S* 

CI 

10 

11 

12 

13 

14 

15 

16 

17 

A 

7v'* 

Ca* 

Sc* 

Ti 

V* 

Cr 

Mn* 

Fe* 

Co       Ni 

18 

19 

20 

21 

22 

23 

24 

25 

26 

27       28 

Cu* 

Z71* 

Ga* 

Ge 

As 

Se 

Br 

29 

30 

31 

32 

33 

34 

35 

Kr 

Rb 

Sr 

Y 

Zr 

Cb 

Mo* 

Tc 

Ru 

Rh      Pd 

36 

37 

38 

39 

40 

41 

42 

43 

44 

45       46 

Ag 

Cd 

In 

Sn 

Sb 

Te 

I 

47 

48 

49 

50 

51 

52 

53 

and  atomic  number  of  the  elements.  The  biologically  essential  elements 
are  in  italics  in  Table  15.  It  is  noteworthy  that  the  essential  elements 
tend  to  occur  in  groups  with  consecutive  atomic  numbers.  Atomic  num- 
ber is  a  fundamental  property  of  atoms  and  denotes  the  number  of  excess 
positive  charges  on  the  nucleus.  Only  those  elements  which  have  certain 
configurations  are  required  by  organisms.  Why  some  organisms  require 
certain  elements  not  required  by  others  is  not  known. 


ESSENTIAL  METALLIC  ELEMENTS  83 

SUMMARY 

The  role  of  the  essential  metallic  elements  is  primarily  functional  rather 
than  structural.  Presumably  these  ions  usually  function  in  ionizable 
combinations,  but  some  compounds  containing  metals  in  non-ionizable 
compounds  have  been  isolated  from  fungi.  It  may  be  assumed  that 
many  of  these  metallic  ions  activate  enzyme  systems,  while  others  are 
integral  parts  of  enzymes  and  other  essential  organic  compounds.  An 
element  is  essential  because  some  of  its  vital  functions  cannot  be  replaced 
by  any  other  element.  Some  functions  may  be  performed  by  other  closely 
related  elements. 

The  concentration  of  an  essential  element  affects  many  life  processes 
besides  growth,  which  is  the  usual  criterion  of  essentiality.  The  concen- 
trations of  various  essential  ions  influence  the  formation  of  pigments,  the 
synthesis  of  vitamins  and  other  products,  and  the  dissimilation  of  carbo- 
hydrates. While  the  essential  elements  may  be  supposed  to  participate 
uniquely  in  certain  life  processes,  the  concentrations  of  other  ions,  both 
of  essential  and  nonessential  elements,  modify  the  action  of  a  given  ele- 
ment. The  phenomenon  of  ion  antagonism  no  doubt  exists  among  all 
ions,  and  in  evaluating  the  effects  of  any  element  it  is  necessary  to  con- 
sider the  other  constituents  present  in  the  medium.  It  is  probable  that 
the  mechanism  involved  is  one  of  modified  adsorption  rather  than  any 
direct  chemical  reaction  in  the  medium. 

The  widespread  use  of  Aspergillus  niger  as  a  test  fungus  in  micro- 
element studies  has  had  the  advantage  that  the  work  in  many  laboratories 
may  be  compared.  The  careful  and  long-continued  studies  by  Steinberg 
are  especially  valuable.  The  almost  exclusive  use  of  this  fungus  has  also 
had  its  disadvantages.  Comparatively  little  is  known  about  the  need  of 
other  species  for  micro  elements.  Other  fungi  may  require  some  of  these 
elements  in  amounts  which  make  it  comparatively  easy  to  demonstrate 
deficiency.  The  evidence  for  the  essentiality  of  iron,  zinc,  copper, 
manganese,  molybdenum,  and  calcium  is  impressive  in  most  instances, 
but  the  need  for  the  elements  on  the  part  of  all  fungi  under  all  cultural 
conditions  has  not  been  established.  In  a  few  instances  the  evidence  is 
confined  to  a  single  fungus.  The  micro-element  nutrition  of  a  wide  range 
of  species  needs  further  study. 

REFERENCES 

Bertrand,  D.:  Le  Vanadium  comme  facteur  de  croissance  pour  V Aspergillus  niger, 

Bull.  soc.  chim.  biol.  23:  467-471,  1941. 
Bertrand,  D.:  Le  Vanadium  chez  les  champignons  et  plus  sp^cialement  chez  les 

Amanites,  Bull.  soc.  chim.  biol.  25:  194-197,  1943. 
Blank,  L.  M.:  Response  of  Phymatotrichum  omnivorum  to  certain  trace  elements, 

Jour.  Agr.  Research  62:  129-159,  1941. 


84  PHYSIOLOGY  OF  THE  FUNGI 

BoRTELS,   IT.:  Molybdfln  als  Katalysator  bei  der  biologischen   Stickstoffbindung, 

Arch.  Mikrohiol.  1 :  333-342,  1930. 
BoRTELS,  H. :  Weitere  Untersucluingen  fiber  die  Bedeutung  von  IVrol3'bdan  Vanadium, 

Wolfram    uiid   anderen   Erdaschenstoffen   fiir   stickstoffbindende   und   andere 

Mikroorganismen.     Cent.  Bakt.,  Abt.  II  95:  193-218,  1936. 
Buchanan,  R.  E.,  and  E.  I.  Fulmer:  Physiologj^  and  Biochemistry  of  Bacteria, 

Vol.  I,  The  Williams  &  Wilkins  Company,  Baltimore,  1928. 
Chilean  Nitrate  Educational  Bureau:  Bibliography  of  the  Literature  on  the 

Minor  Elements  and  Their  Relation  to  Plant  and  Animal  Nutrition,  New  York, 

1948. 
*Foster,  J.  W.:  The  heavy  metal  nutrition  of  fungi,  Bolan.  Rev.  5 :  207-239,  1939. 
Foster,  J.   W.:  Chemical  Activities  of  Fungi,  Academic  Press,  Inc.,  New  York, 

1949. 
Foster,  J.  W.,  and  S.  A.  Waksman:  The  specific  effect  of  zinc  and  other  heavy 

metals  on  growth  and  fumaric  acid  production  by  Rhizopus,  Jour.  Bad.  37: 

599-617,  1939. 
Frey-Wyssling,  A.:  Die  unentbehrlichen  Elemente  der  Pflanzennjihrung,  Natur- 

wissenschaften  23:  767-769,  1935. 
GoRTNER,  R.  H.:  Outlines  of  Biochemistry,  3d  ed.,  John  Wiley  &  Sons,  Inc.,  New 

York,  1949. 
GusTAFSON,  F.  G.:  Comparative  studies  on  respiration.     IX.  The  effects  of  antago- 
nistic salts  on  the  respiration  of  Aspergillus  niger,  Jour.  Gen.  Physiol.  2:  17-24, 

1919. 
HicKEY,  R.  J. :  The  inactivation  of  iron  by  2,2'-bipyridine  and  its  effects  on  riboflavin 

synthesis  by  Clostridium  acetobutylicum,  Arch.  Biochem.  8:  439-447,  1945. 
Jarvis,  F.  G.,  and  M.  J.  Johnson:  The  mineral  nutrition  oi  Penicillium  chrysogenum 

Q176,  Jour.  Bad.  59:  51-60,  1950. 
Javillier,  M.:  Une  cause  d'erreur  dans  I'etude  de  Taction  biologique  des  elements 

chimiques:  la  presence  de  traces  de  zinc  dans  la  verre,  Compt.  rend.  acad.  sci. 

158:  140-143,  1914. 
Javillier,  M.:  Cuivre  et  Aspergillus  niger.     Rappel  de  quelques  faits  anciens,  Ann. 

fermentations  5:  371-381,  1939. 
*KoFFLER,  II.,  S.  G.  Knight,  and  W.  C.  Frazier:  The  effect  of  certain  mineral  ele- 
ments on  the  production  of  penicillin  in  shake  flasks,  Jour.  Bad.  53:  115-123, 

1947. 
Laborey,  F.,  J.  Lavollay,  and  J.  Neumann:  Coefficient  d'action  du  magnesium 

vis-a-vis  d' Aspergillus  niger:  variation  lineaire  de  ce  coefficient  avec  la  concen- 
tration en  phosphore,  Compt.  rend.  acad.  sci.  212:  624-626,  1941. 
Lavollay,  J.,  and  F.  Laborey:  Sur  les  circonstances  d'apparition  de  pigments  jaunes 

dans  le  liquide  de  culture  d' Aspergillus  niger,  Compt.  rend.  acad.  sci.  206  :  1055- 

1056,  1938. 
Lewis,  J.  C. :  Relationship  of  iron  nutrition  to  the  synthesis  of  vitamins  by  Torulopsis 

utilis,  Arch.  Biochem.  4:  217-228,  1944. 
Lilly,  V.  G.,  and  L.  H.  Leonian:  The  interrelationship  of  iron  and  certain  factors  in 

the  growth  of  Rhizohium  trifolii.  strain  205,  Jour.  Bad.  50 :  383-395,  1945. 
Lindeberg,  G.:  Ueber  die  Physiologic  Ligninabbauender  Boden  Hymenomyzeten, 

Symbolae  Botan.  Upsalienses  8(2):  1-183,  1944. 
LocKwooD,  L.  B.,  and  M.  D.  Reeves:  Some  factors  affecting  the  production  of 

itaconic  acid  by  Aspergillus  terreus,  Arch.  Biochem.  6 :  455-469,  1945. 
Lohrmann,  W.  :  Untersuchungen  liber  die  antagonistische  Wirkung  von  Magnesium 

gegeniiber  Bor  und  Quecksilber  bei  einigen  Pilzen,  Arch.  Mikrohiol.  11 :  329-367, 

1940. 


ESSENTIAL  METALLIC  ELEMENTS  85 

McHargue,  J.  S.,  and  R.  K.  Calfee:  Effect  of  manganese,  copper  and  zinc  on 

growth  and  metabolism  of  Aspergillus  flavus  and  Rhizopus  nigricans,  Botan.  Gaz. 
91:  183-193,  1931. 
McHargue,  J.  S.,  and  R.  K.  Calfee:  Effect  of  manganese,  copper  and  zinc  on  the 

growth  of  yeast,  Pkmt  Physiol.  6:  559-5G6,  1931a. 
MacLeod,  R.  A.,  and  E.  E.  Snell:  Some  mineral  requirements  of  the  lactic  acid 

bacteria.  Jour.  Biol.  Chem.  170:  351-3G5,  1947. 
Marsh,  P.  B. :  Salts  as  antidotes  to  copper  in  its  toxicity  to  the  conidia  of  Sclerotinia 

fructicola,  Phytopathology  35:  54-Gl,  1945. 
Molliard,    M.:   Influence   d'une   dose   reduite   de   potassium   sur  les   characteres 

physiologiques  du  Sterigmatacystis  nigra,  Compt.  rend.  acad.  sci.  170:  949-951, 

1920. 
Mulder,  E.  G.  :  On  the  use  of  microorganisms  in  measuring  a  deficiency  of  copper, 

magnesium  and  molybdenum  in  soils,   Antonie  van  Leeuwenhoek  6:  99-109, 

1939-1940. 
Mulder,  E.  G.:  Importance  of  molybdenum  in  the  nitrogen  metabolism  of  micro- 
organisms and  higher  plants,  Plant  and  Soil  1:  94-119,  1948. 
MuNTZ,  J.  A. :  The  role  of  potassium  and  ammonium  ions  in  alcoholic  fermentation, 

Jour.  Biol.  Chem.  171:  653-665,  1947. 
Nelson,  J.  M.,  and  C.  R.  Dawson:  Tyrosinase,   Advances  in  Enzymol.  4:  99-152, 

1944. 
NiCKERSON,  W.  J. :  Inhibition  of  fungus  respiration :  a  metabolic  bio-assay  method, 

Science  103 :  484-486,  1946. 
Perlman,  D.  :  Some  effects  of  metallic  ions  on  the  metabolism  of  Aerobacter  aerogenes, 

Jour.  Bad.  49:  167-175,  1945. 
Perlman,  D.:  On  the  nutrition  of  Sclerotium  delphinii,  Am.  Jour.  Botany  35:  360- 

363,  1948. 
*Perlman,  D.:  Effects  of  minor  elements  on  the  physiology  of  fungi,  Botan.  Rev.  15: 

195-220,  1949. 
Perlman,  D.,  D.  A.  Kita,  and  W.  Peterson:  Production  of  citric  acid  from  cane 

molasses.  Arch.  Biocheni.  11:  123-129,  1946. 
Rabinovitz-Sereni,  D.:  Influenza  del  magnesio  suUo  sviluppo  di  alcuni  funghi,  Dol. 

R.  Staz.  Veg.  13 :  203-226,  1933. 
Raulin,  J.:  Etudes  chimiques  sur  la  vegetation,  Ann.  sci.  nat.,  Ser.  V,  11:  93-299, 

1869. 
Richards,  O.  W.,  and  M.  C.  Tkoutman:  Spectroscopic  analysis  of  the  mineral  con- 
tent of  yeast  grown  on  synthetic  and  natural  media.  Jour.  Bad.  3D :  739-746, 

1940. 
Rickes,   E.  L.,  N.  G.  Brink,  F.  R.  Koniuszy,  T„  R.  Wood,  and  K.  Folkers: 

Crystalline  vitamin  B12,  Science  107 :  396-397,  1948. 
RippEL,  A.,  and  G.  Behr:  Ueber  die  Verteilung  des  Magnesiums  im  Pilzmycel,  Arch. 

Mikrohiol.  1 :  271-276,  1930. 
RippEL,  A.,  and  G.  Behr:  Ueber  die  Bedeutung  des  Kaliums  im  Stoffwechsel  von 

Aspergillus  niger,  Arch.  Mikrohiol.  5:  561-577,  1934. 
RoBBiNS,  W.  J.,  and  A.  Hervey:  Response  oi  Pythiomorpha  gonapodyoides  to  manga- 
nese. Bull.  Torrey  Botan.  Club  71 :  258-266,  1944. 
Roberg,  M.:  Ueber  die  Wirkung  von  Eisen-,  Zink-,  und  Kupfersalzen  auf  Aspergillen, 

Cent.  Bakt.,  Abt.  II.  74:  333-371.  1928. 
RoBERG,  M.:  Weitere  Untersuchungen  iiber  die  Bedeutung  des  Zinks  fur  Aspergillus 

niger.  Cent.  Bakt.,  Abt.  II,  84:  196-230,  1931. 
Roberts,  C.  :  The  effect  of  iron  and  other  factors  on  the  production  of  pigment  by  the 

yeast  Torulopsis  pulcherrima,  Atn.  Jour.  Botany  33 :  237-244,  1946, 


86  PHYSIOLOGY  OF  THE  FUNGI 

Shu,  p.,  and  M.  J.  Johnson:  The  interdependonce  of  modium  oonstituents  in  citric 

acid  production  by  submerged  fermentation,  Jour.  Bad.  66:  577-585,  1948. 
Stark,  W.  H.,  P.  J.  Kolichov,  and  H.  F.  Willkie:  Some  factors  affecting  yeast 

propagation,  Proc.  Am.  Soc.  Brexving  Chemists  4:  49-56,  1941. 
■^Steinberg,  R.  A.:  A  study  of  some  factors  in  the  chemical  stimulation  of  the  growth 

of  Aspergillus  niger,  Am.  Jour.  Botany  6:  330-372,  1919. 
Steinberg,  R.  A.:  The  nutritional  requirements  of  the  fungus  Aspergillus  niger, 

Bull.  Torrey  Botan  Chib  62:  81-90,  1935. 
Steinberg,  R.  A.:  Nutrient-solution  purification  for  the  removal  of  heavy  metals  in 

deficiency  investigations  with  Aspergillus  niger,  Jour.  Agr.  Research  51 :  413-424, 

1935o. 
Steinberg,  R.  A. :  Relation  of  accessory  growth  substances  to  heavy  metals,  includ- 
ing molybdenum,  in  the  nutrition  of  Aspergillus  niger,  Jour.  Agr.  Research  52: 

439-448,  1936. 
Steinberg,  R.  A. :  Role  of  molybdenum  in  the  utilization  of  ammonium  and  nitrate 

nitrogen  by  Aspergillus  niger,  Jour.  Agr.  Research  55:  891-902,  1937. 
Steinberg,  R.  A.:  The  essentiality  of  gallium  to  growth  and  reproduction  of  Asper- 
gillus niger.  Jour.  Agr.  Research  57:  569-574,  1938. 
Steinberg,  R.  A. :  Correlations  between  biological  essentiality  and  atomic  structure 

of  the  chemical  elements.  Jour.  Agr.  Research  57:  851-858,  1938a. 
♦Steinberg,  R.  A.:  Growth  of  fungi  in  synthetic  nutrient  solutions,  Botan.  Rev.  5: 

327-350,  1939. 
Steinberg,  R.  A.:  Relation  of  carbon  nutrition  to  trace  element  and  accessory 

requirements  of  Aspergillus  niger.  Jour.  Agr.  Research  59:  749-763,  1939a. 
Steinberg,  R.  A.:  A  dibasal  (minimal  salt,  maximum  yield)  solution  for  Aspergillus 

niger;  acidity  and  magnesium  optimum,  Plant  Physiol.  20 :  600-608,  1945. 
Steinberg,  R.  A. :  Specificity  of  potassium  and  magnesium  for  the  growth  of  Asper- 
gillus niger.  Am.  Jour.  Botany  33:  210-214,  1946. 
♦Steinberg,  R.  A.:  Essentiality  of  calcium  in  the  nutrition  of  fungi.  Science  107: 

423,  1948. 
Steinberg,  R.  A. :  Growth  on  synthetic  nutrient  solutions  of  some  fungi  pathogenic 

to  tobacco,  Am.  Jour.  Botany  37:  711-714,  1950. 
Stout,  P.  R.,  and  D.  I.  Arnon:  Experimental  methods  for  the  study  of  the  role  of 

copper,  manganese,  and  zinc  in  the  nutrition  of  higher  plants,  Am.  Jour.  Botany 

26:  144-149,  1939. 
Sumner,  J.  B.,  and  G.  F.  Somers:  Chemistry  and  Methods  of  Enzymes,    Academic 

Press,  Inc.,  New  York,  1947. 
Thatcher,  R.  W.:  A  proposed  classification  of  the  chemical  elements  with  respect 

to  their  function  in  plant  nutrition,  Science  79:  463-466,  1934. 
Waksman,  S.  a.,  and  J.  W.  Foster:  Respiration  and  lactic  acid  production  by  a 

fungus  of  the  genus  Rhizopus,  Jour.  Agr.  Research  57:  873-900,  1938. 
Waring,  W.  S.,  and  C.  H.  Werkman:  Growth  of  bacteria  in  an  iron-free  medium. 

Arch.  Biochem.  1 :  303-310,  1943. 
YoE,  J.  H.,  and  L.  A.  Sarver:  Organic  Analytical  Reagents,  John  Wiley  «fe  Sons, 

Inc.,  New  York,  1941. 
YosHiMURA,  F. :  The  action  of  some  heavy  metals  upon  the  production  of  catalase  in 

Aspergillus,  Japan.  Jour.  Botany  10:  75,  1939-1940. 
Young,  H.  C,  and  C.  W.  Bennett:  Growth  of  some  parasitic  fungi  in  synthetic 

culture  media,  Am.  Jour.  Botany  9:  459-469,  1922. 


CHAPTER  6 

THE  ESSENTIAL  NONMETALLIC  ELEMENTS  OTHER 

THAN  CARBON 

Fungus  mycelium  and  spores  are  composed  mainly  of  compounds  of  the 
nonmetallic  elements.  As  a  rule,  more  than  95  per  cent  of  the  fungus 
consists  of  hydrogen,  oxygen,  carbon,  nitrogen,  sulfur,  and  phosphorus. 
The  nonmetallic  essential  elements  are  both  structural  and  functional. 
The  cell  wall,  which  is  composed  mainly  of  chitin  or  cellulose,  appears  to 
be  the  most  stable  structure  of  the  fungus.  Protoplasm  is  highly  labile, 
and  the  constituent  compounds  of  protoplasm  are  continually  undergoing 
destruction,  repair,  and  synthesis.  The  various  structural  and  functional 
compounds  of  organisms  are  in  a  state  of  continual  flux  (Hevesy,  1947). 
The  turnover  of  essential  elements  in  functional  compounds  is  more  rapid 
than  in  structural  compounds. 

The  terms  utilization,  assimilation,  and  dissimilation  are  frequently 
used  in  physiology.  Utilization  is  a  broad  term  and  implies  that  an 
organism  uses  or  gains  some  benefit  from  a  specific  substance.  Fungi 
utilize  water  as  a  solvent  but  derive  neither  energy  nor  substance  from  it. 
Assimilation  is  the  incorporation  of  substances  or  their  degradation 
products  into  cellular  materials.  Assimilation  implies  synthesis.  Dis- 
similation is  the  degradation,  or  breakdown,  of  complex  compounds  into 
simpler  ones.  This  term  is  particularly  applied  to  those  processes  such  as 
alcoholic  fermentation  where  intermediate  metabolic  products  accumu- 
late in  the  medium.  Frequently  dissimilation  must  precede  assimilation 
and  may  be  considered  as  the  first  phase  of  utilization. 

HYDROGEN 

Hydrogen  enters  into  the  composition  of  nearly  all  organic  compounds 
of  interest  to  physiology  except  carbon  dioxide.  This  is  true  of  the 
organic  nutrients  used  by  fungi  as  well  as  of  the  fungus  protoplasm  and 
other  cellular  compounds.  Elemental  hydrogen  is  not  used  by  fungi. 
All  the  hydrogen  utilized  by  fungi  is  in  chemical  combination.  Certain 
bacteria  (hydrogen  bacteria),  however,  are  able  to  obtain  energy  by 
oxidizing  hydrogen. 

The  importance  of  water  for  all  living  organisms  is  so  great  that  it 
seems  impossible  to  conceive  of  life  without  water.  The  formula  H2O  is 
really  the  formula  of  steam.     In  the  liquid  state  these  simple  molecules 

87 


88  PHYSIOLOGY  OF  THE  FUNGI 

associate  to  form  polymers.  At  room  temperature  water  consists  mostly 
of  (H20)3,  which  is  sometimes  called  trihydroL  For  a  further  discussion 
of  water  see  Barnes  (1937). 

The  chemistry  of  life  processes  is  largely  confined  to  reactions  which 
take  place  in  the  presence  of  water  or  in  solution.  In  addition  to  being  a 
solvent  of  remarkable  powers,  water  is  associated  with  the  colloids  which 
comprise  protoplasm.  Gortner  (1949)  has  distinguished  between  "free" 
and  "bound"  water.  Free  water  is  mobile  within  the  cell  and  serves  as  a 
solvent  and  for  the  purpose  of  translocation  of  the  various  products  of 
metabolism.  Bound  water  is  firmly  adsorbed  by  protoplasm,  and  in  this 
form  water  does  not  freeze.  This  property  of  bound  water  enables  cells 
to  withstand  low  temperatures.  The  ability  of  fungus  spores  to  with- 
stand low  temperatures  may  well  be  due  to  their  having  most  of  their 
water  content  in  the  bound  form. 

Water  ionizes  to  form  hydrogen  (H+)  and  hydroxyl  (0H~)  ions. 
The  effects  of  these  ions  on  biological  processes  are  so  important  that  they 
will  be  discussed  in  detail  in  Chap.  8. 

OXYGEN 

Apparently  none  of  the  fungi  are  obligate  anaerobes.  Many  are 
strictly  aerobic,  and  some  are  facultatively  anaerobic.  An  aerobic 
organism  requires  uncombined  oxygen,  while  a  facultative  anaerobe  may 
use  combined  oxygen  in  addition  to  free  oxygen.  The  amount  of  oxygen 
required  for  optimum  growth  varies  with  the  species.  It  is  common  to 
express  the  amount  of  oxygen  available  in  terms  of  millimeters  of  mercury. 
Approximately  21  per  cent  of  air  is  oxygen.  The  amount  of  oxygen  may 
be  regulated  by  controlling  the  air  pressure  within  the  culture  vessel.  If 
the  barometric  pressure  is  740  mm.  Hg,  the  partial  pressure  due  to  oxygen 
is  ^Hoo  X  740,  or  155.4  mm.  Hg.  If  the  pressure  within  a  culture  vessel 
is  reduced  to  100  mm.  Hg,  the  partial  pressure  of  oxygen  amounts  to  21 
mm.  Hg.  Tamiya  (1942)  has  reported  that  Aspergillus  onjzae  has  a 
maximum  rate  of  respiration  when  the  partial  pressure  of  oxygen  is  500 
to  630  mm.  Hg.  Such  partial  pressures  of  oxygen  are  readily  obtained  by 
using  oxygen-nitrogen  mixtures.  Ternetz  (1900)  reported  the  following 
effects  of  reduced  oxygen  supply  on  Ascophanus  carneus:  at  10  mm.  Hg  the 
mycelium  grew  with  difficulty;  at  20  mm.  Hg  growth  was  good,  but  no 
spores  formed;  at  40  mm.  Hg  some  fructification  occurred;  at  120  to  140 
mm.  Hg  growth  was  somewhat  better  than  at  atmospheric  pressure. 

The  ability  of  certain  soil  fungi  to  exist  under  conditions  of  low  oxygen 
supply  is  important  for  survival.  The  amount  of  oxygen  in  soil  depends 
upon  the  soil  type  and  the  amount  of  water  present.  Soil  saturated  with 
water  contains  but  a  trace  of  free  oxygen.  Hollis  (1948)  found  Fusarium 
oxysporum  to  survive  under  essentially  anaerobic  conditions  for  13  weeks, 


ESSENTIAL  NONMETALLIC  ELEMENTS  89 

while  F.  eumartii  perished  within  3  weeks  when  exposed  to  the  same  condi- 
tions, The  mycehum  of  F.  oxysporuni  grown  under  reduced  oxygen  ten- 
sion was  abnormal  in  its  morphology.  For  further  information  on  the 
effect  of  reduced  oxygen  tension,  see  Fellows  (1928)  and  Scheffer  and 
Livingston  (1937). 

Enormous  amounts  of  sterile  air  must  be  supplied  to  the  10,000-  to 
15,000-gal.  tanks  used  in  the  production  of  penicillin  and  other  antibiotics. 
In  the  laboratory,  aeration  is  provided  by  shaking  machines  of  the  rotat- 
ing or  reciprocal  type.  Aeration  under  these  conditions  is  more  uniform 
than  is  possible  in  stationary  cultures,  W'here  submerged  and  aerial 
hyphae  obtain  different  amounts  of  oxygen.  This  was  sho\Mi  by  Tamiya 
(1942)  who  reported  that  the  enzyme  systems  of  submerged  mycelium  of 
Aspergillus  oryzae  are  more  easily  poisoned  by  cyanide  than  are  those  of 
aerial  mycelium. 

In  a  broad  sense,  respiration  denotes  all  the  enzymatic  processes  which 
occur  in  cells  involving  a  release  of  energy.  There  are  two  general  ways 
in  which  energy  is  released  by  living  cells:  (1)  Cells  obtain  energy  from 
chemical  reactions  in  which  free  oxygen  is  a  reactant.  The  oxidation  of 
metabolite  molecules  by  this  process  is  generally  called  respiration,  or 
more  specifically  aerobic  respiration.  This  process  is  characterized  by  the 
intake  of  free  oxygen  and  the  formation  of  carbon  dioxide.  If  the  com- 
pound being  oxidized  is  composed  of  carbon,  hydrogen,  and  oxygen  only, 
the  products  are  carbon  dioxide,  water,  and  energJ^  (2)  Cells  also  obtain 
energy  from  chemical  reactions  in  which  free  oxygen  is  not  a  reactant. 
This  process  is  called  anaerobic  respiration,  or  fermentation.  Metabolic 
processes  of  this  kind  are  characterized  by  the  production  of  carbon 
dioxide,  the  incomplete  oxidation  of  substrate  molecules,  and  the  release 
of  a  small  amount  of  energy. 

The  reactions  involved  in  the  aerobic  respiration  of  glucose  may  be 
summarized  in  a  single  equation: 

CsHisOe  +  60.-^  6CO2  +  6H2O  +  673,000  cal. 

This  equation  gives  no  indication  of  the  intermediate  stages  in  this  reac- 
tion or  how  the  energy  is  utilized  by  the  organism  performing  the  oxida- 
tion. The  number  and  variety  of  intermediate  reactions  do  not  affect 
the  total  amount  of  energy  released.  The  reactions  involved  in  the 
alcoholic  fermentation  of  glucose  are  summarized  in  the  following  equation : 

CeHisOs-*  2CH3CH2OH  +  2CO2  +  25,000  cal. 

This  equation,  like  the  preceding  one,  gives  no  indication  of  the  inter- 
mediate reactions  involved.  To  obtain  the  same  amount  of  energy,  more 
of  a  compound  must  be  fermented  than  when  it  is  completely  oxidized. 
Not  all  of  the  energy  released  by  either  of  these  processes  is  available  to 
the  organism  (Chap.  4). 


90  PHYSIOLOGY  OF  THE  FUNGI 

A  knowledge  of  the  amounts  of  oxygen  consumed  and  carbon  dioxide 
evolved  by  organisms  is  the  basis  of  a  useful  method  of  study  in  many 
phases  of  physiology.  The  principles  of  such  measurements  are  simple. 
In  aerobic  respiration  both  the  oxygen  and  carbon  dioxide  may  be  meas- 
ured. The  ratio  of  the  moles,  or  volumes,  of  carbon  dioxide  evolved  and 
oxygen  used  is  called  the  respiratory  quotient  (R.Q.)  and  is  written 
CO2/O2.  From  the  respiratory  quotient  the  nature  of  the  substrate 
being  oxidized  may  be  deduced.  A  respiratory  quotient  of  1  is  character- 
istic of  aerobic  oxidation  of  carbohydrate.  The  complete  oxidation  of  a 
fat  may  be  represented  as  follows: 

(C,8H3602)3C3H5  +  81.50-2^  57CO0  +  55H.2O 

The  respiratory  quotient  for  this  fat  is  57/81.5,  or  0.7.  If  fungus  cells  are 
suspended  in  a  buffer  in  the  absence  of  nutrients,  and  the  respiratory 
quotient  determined,  it  is  possible  to  deduce  the  type  of  compound  within 
the  cells  being  used  as  a  source  of  energy.  Oxidation  of  the  stored  com- 
pounds within  the  cell  is  called  endogenous  respiration.  The  oxidation  of 
substrate  molecules  from  the  medium  is  called  exogenous  respiration. 
Since  both  types  of  respiration  may  occur  simultanously  in  the  presence 
of  nutrients,  it  is  necessary,  in  order  to  determine  exogenous  respiration, 
to  subtract  the  value  for  endogenous  respiration  from  that  obtained  in  the 
presence  of  nutrients. 

The  rate  and  amount  of  respiration  are  determined  by  instruments 
known  as  respirometers.  Various  types  of  respirometers  have  been  used 
to  investigate  different  phases  of  fungus  metabolism  and  nutrition.  In 
principle  a  respirometer  is  a  closed  vessel  of  known  volume  in  which 
fungus  cells  are  suspended  in  a  buffer  or  other  solution.  The  carbon 
dioxide  evolved  is  absorbed  in  a  concentrated  solution  of  potassium 
hydroxide.  The  change  in  volume  due  to  the  consumption  of  oxygen  is 
measured  by  the  use  of  suitable  manometers.  At  the  end  of  the  experi- 
ment the  amount  of  carbon  dioxide  evolved  is  measured  after  the  potas- 
sium hydroxide  solution  is  treated  with  a  mineral  acid.  Carbon  dioxide 
alone  may  be  measured  by  passing  a  stream  of  carbon  dioxide-free  air 
through  a  culture  and  absorbing  the  carbon  dioxide  evolved  in  barium 
hydroxide  or  other  suitable  reagent.  The  results  of  such  experiments  are 
reported  on  the  basis  of  the  volumes  of  oxygen  used  and  carbon  dioxide 
evolved  per  milligram  of  dry  weight  per  hour.  These  values  are  reported 
as  Q02  and  Qco2  (see  Umbreit  et  at.,  1945). 

A  modern  respirometer  is  illustrated  in  Fig.  16.  The  various  manipula- 
tive details  will  not  be  discussed.  For  an  adequate  treatment  of  these 
see  Umbreit  etal.  (1945)  and  Dixon  (1943).  These  methods  are  extremely 
useful  in  studying  a  wide  range  of  problems.  Hawker  (1944)  used 
manometric   techniques  in  studying  the   effect   of  excess   thiamine   on 


ESSENTIAL  NONMETALLIC  ELEMENTS 


91 


glucose  utilization  by  Melanospora  destruens  and  Phycomyces  nitens. 
The  papers  of  Siu  and  Mandels  (1950)  and  Mandels  and  Siu  (1950) 
should  be  consulted  for  details  concerning  a  simple  differential  manometer. 
This  manometer  is  designed  to  measure  the  respiration  of  intact  growing 
cultures  of  filamentous  fungi.     Dorrell  (1948)  investigated  the  effect  of 


Fig.  16.  A  constant-temperature  bath  and  shaking  device  for  micro  respiration 
studies.     (Courtesy  of  American  Instrument  Company.) 

dinitrophenol  on  endogenous  and  exogenous  respiration  of  Fusarium 
graminearum  (Gibberella  zeae).  As  usually  carried  out,  respiration  experi- 
ments last  only  a  few  hours.     The  initial  state  of  the  cells  or  mycelium 

Table  16.     The  Effect  of  Age  of  Zygosaccharomyces  acidifaciens  Cells  on  the 

Amount  of  Aerobic  Respiration 
(Nickerson  and  Carroll,  Jour.   Cellular  Cornp.  Physiol.   22,   1943.     Published  by 
permission  of  the  Wistar  Institute  of  Anatomy  and  Biology.) 


Age  of  cells, 
hr. 

Qo.* 

Glucose  substrate 

No  substrate 
(endogenous) 

24 

48 

72 

60 
35 
35.5 

16 
7.3 

7.0 

*  Q02  equals  lil  O2  per  hr.  per  mg.  dry  cells. 


92  PHYSIOLOGY  OF  THE  FUNGI 

used  has  a  great  effect  on  the  results  obtained.  Nickerson  and  Carroll 
(1943)  have  indicated  that  the  culture  history  of  the  cells  used  influences 
the  amount  of  aerobic  respiration.  Some  of  their  data  for  Zygosaccharo- 
myces  acidifaciens  are  shown  in  Table  16. 

SULFUR 

Not  all  compounds  which  contain  an  essential  element  are  equally 
useful.  In  fact,  some  compounds  are  useless  because  the  essential  ele- 
ment is  unavailable.  Among  the  factors  which  may  affect  availability  is 
the  state  of  oxidation  of  the  essential  element.  This  is  particularly  true 
of  sulfur,  phosphorus,  and  nitrogen.  Among  the  organic  compounds, 
structure  is  enormously  important.  The  situation  is  further  complicated 
in  that  not  all  fungi  utilize  the  same  compounds.  Many  examples  of  thi, 
will  be  cited  in  connection  with  nitrogen  and  carbon  nutrition.  Atten- 
tion must  be  given  the  sources  of  the  essential  elements  as  well  as  the  uses 
fungi  make  of  them. 

Sources  of  sulfur.  This  element  is  present  in  many  types  of  com- 
pounds, both  inorganic  and  organic.  The  state  of  oxidation  of  sulfur,  as 
well  as  the  specific  structure  of  organic  sulfur  compounds,  affects  utiliza- 
tion. Sulfate  sulfur,  SO4"",  is  the  most  common  source  of  sulfur  used  in 
media.  Some  fungi,  however,  require  specific  organic  sources  of  sulfur. 
Steinberg  (1936,  1941)  has  made  an  exhaustive  study  of  sulfur  sources  for 
Aspergillus  niger  and  reached  the  general  conclusions  that  inorganic  sulfur 
compounds  containing  oxidized  sulfur  are  utilized,  while  sulfide  and 
disulfide  sulfur  are  not  utiHzed.  Of  the  organic  compounds  containing 
sulfur,  the  alkyl  thioalcohols,  sulfides,  and  disulfides  are  not  used. 
Alkyl  sulfonates  and  sulfinates  are  excellent  sources  of  sulfur.  Steinberg 
is  of  the  opinion  that  oxidized  sulfur  is  reduced  to  suKoxylate  before  it 
enters  the  normal  metabolic  channels.  An  exception  to  the  nonutiliza- 
tion  of  reduced  sulfur  was  noted  for  compounds  which  occur  as  normal 
metabolites,  such  as  cysteine,  cystine,  methionine,  and  homocystine. 
These  are  assumed  to  enter  normal  metabolic  channels  without  pre- 
liminary modification.  An  exception  to  this  statement  was  noted  with 
thiamine  (thiazole  sulfur),  but  the  enormous  (physiologically)  amounts 
used  may  have  upset  the  metabolic  activities  of  the  fungus. 

In  spite  of  the  general  utility  of  sulfate  sulfur  in  fungus  nutrition,  many 
fungi  either  utilize  organic  sulfur  contained  in  natural  metabolites  to  bet- 
ter advantage  or  require  these  compounds  as  a  source  of  sulfur.  Leonian 
and  Lilly  (1938)  reported  that  the  addition  of  cystine  to  a  synthetic 
medium  was  necessary  for  the  grovv^th  of  Saprolegnia  mixta,  Achlya  con- 
spicua,  Isoachlya  monilijera,  and  Aphanomyces  camptostylus.  Since  other 
naturally  occurring  sulfur-containing  amino  acids  were  not  tested,  it 
should  not  be  concluded  that  these  species  are  deficient  for  cystine. 


ESSENTIAL  NONMETALLIC  ELEMENTS  93 

Volkoiisky  (1933,  1934)  observed  that  certain  of  the  aquatic  Phycocomy- 
cetes  failed  to  utiHze  sulfate  sulfur.  These  species  were  Saprolegnia 
parasitica,  Isoachlya  monilifera,  Achlya  prolifera,  A.  polya7idra,  A.  oblon- 
gata, A.  conspicua,  Dichtyuchus  monosporus,  and  Aphanomyces  sp.  A 
total  of  26  isolates  failed  to  utilize  sulfate  sulfur.  This  investigator 
(1933a)  designates  ability  to  utilize  6-valent  sulfur  as  euthiotrophy  and 
inability  to  utilize  sulfate  sulfur  and  ability  to  utilize  reduced  sulfur  as 
parathiotrophy . 

Fries  (1946)  was  able  to  induce  mutation  in  Ophiostoma  (Ceratostomella) 
rmiltianmilatum  by  irradiating  the  ascospores  with  X  rays.  Among 
these  mutants  13  strains  were  unable  to  utilize  sulfate  sulfur.  Only  five 
of  these  strains  regained  this  ability  when  cultivated  on  media  containing 
sulfate.  These  parathiotrophic  strains  of  0.  multiannulaium  utilized 
ammonium  sulfide  as  well  as  cystine  and  cysteine  as  sources  of  sulfur. 
From  the  fact  that  these  mutants  could  utilize  sulfide  sulfur,  it  is  evident 
that  these  strains  were  not  deficient  for  specific  sulfur-containing  amino 
acids.  Bonner  (1946)  has,  however,  found  induced  mutants  of  Peni- 
cillium  to  be  deficient  for  specific  sulfur-containing  amino  acids.  Blasto- 
cladia  pringsheimii  has  been  reported  to  require  methionine  (Cantino, 
1949). 

Fries  (1948)  has  reported  the  occurrence  of  natural  mutants  of  Ophi- 
ostoma multiannulatiim  which  require  reduced  sulfur,  and  also  mutants 
which  are  unable  to  synthesize  methionine.  Of  a  total  of  51,037  single- 
conidium  cultures,  2  required  reduced  sulfur  and  30  required  methionine. 

The  role  of  sulfur.  The  use  fungi  make  of  sulfur  may  be  deduced  from 
the  sulfur-containing  compounds  which  are  known  to  occur  in  mycelium 
and  spores.  Among  these  are  the  proteins.  In  Chap.  4  it  was  noted  that 
the  activity  of  many  enzymes  depends  upon  the  sulfhydryl  or  thiol  group, 
■ — SH.  On  hydrolysis,  fungus  protein  yields  the  following  sulfur-contain- 
ing amino  acids:  cystine,  cysteine,  and  methionine.  Sulfur  is  thus  a 
structural  element.  Another  sulfur-containing  compound  is  the  tripep- 
tide,  glutathione,  which  is  abundant  in  yeast.  The  formula  for  gluta- 
thione is  given  below: 

COOH  CH2SH 

H2N— CH— CH2— CH2— CONH— CH— CONH— CH2— COOH 

This  compound  is  sometimes  represented  by  the  symbol  GSH.  In  spite 
of  intensive  investigation  the  role  of  this  compound  is  not  fully  under- 
stood. Perhaps  one  of  its  functions  is  to  protect  sulfhydryl  enzymes 
from  inactivation. 

The  probable  mechanism  of  the  biosynthesis  of  cystine  has  been  studied 
using  mutants  of  Aspergillus  nidulans  (Hockenhull,  1949).  All  these 
cystine-deficient  mutants  were  able  to  utilize  thiosulfate  sulfur,  methio- 


94  PHYSIOLOGY  OF  THE  FUNGI 

nine,  and  cystine.  It  was  postulated  that  sulfate  sulfur  was  first  reduced 
to  sulfite  and  then  to  sulfoxylate,  which  was  assumed  to  dimerize  to 
thiosulfate.  The  next  reaction  was  believed  to  be  between  serine  and 
thiosulfate  to  form  cysteine  S-sulfonate,  which  is  then  converted  to 
cysteine.     Cysteine  on  being  oxidized  forms  cystine. 

Two  vitamins,  thiamine  and  biotin,  contain  sulfur.  The  role  of  these 
compounds  will  be  considered  in  Chap.  9.  In  addition  to  the  sulfur-con- 
taining amino  acids  and  vitamins  there  is  evidence  that  other  types  of 
organic  sulfur  compounds  are  formed  by  fungi.  Raistrick  and  Vincent 
(1948)  found  that  many  strains  and  species  of  Aspergillus  and  Penicillium 
converted  essentially  all  of  the  sulfate  sulfur  into  organic  sulfur  com- 
pounds, but  not  all  of  these  compounds  were  found  in  the  fungus  proteins. 
Penicillium  chrysogenum  excretes  into  the  medium  various  unidentified 
organic  sulfur  compounds  (Plumlee  and  Pollard,  1949).  The  function  of 
these  compounds  is  unknown. 

The  reactions  whereby  a  fungus  transforms  a  single  source  of  sulfur  into 
these  various  compounds  are  obscure.  When  sulfate  or  other  sources 
containing  oxidized  sulfur  are  utilized,  it  is  necessary  for  the  fungus  to 
reduce  the  sulfur  to  its  lowest  valence.  Schizophyllum  commune  has  been 
shown  to  reduce  sulfate  to  methyl  mercaptan,  CH3SH  (Birkinshaw  et  al., 
1942).  This  substance  contributes  to  the  characteristic  odor  of  this 
fungus. 

PHOSPHORUS 

Raulin  (1869)  found  phosphorus  to  be  an  essential  element  for  Aspergil- 
lus niger.  Omission  of  phosphate  from  his  synthetic  medium  reduced  the 
yield  approximately  50  per  cent.  Phosphorus  is  essential  for  all  forms  of 
life.  Phosphorus  may  be  classified  as  a  structural  element  in  the  sense 
that  definite  compounds  containing  this  element  have  been  isolated  from 
fungi.  Phosphorus  compounds  play  an  important  role  in  the  functions  of 
chemical  transformations  and  energy  transfer. 

Sources  of  phosphorus.  Apparently  phosphorus  is  utilized  only  when 
it  is  in  the  form  of  phosphate.  This  element  is  taken  up  as  phosphate  and 
functions  in  this  form,  mainly  in  the  form  of  phosphate  esters.  It  will  be 
recalled  that  there  are  several  different  phosphates.  The  formulas  for  the 
potassium  salts  are  K3PO4,  potassium  orthophosphate ;  KPO3,  potassium 
metaphosphate ;  and  K4P2O7,  potassium  pyrophosphate.  More  complex 
phosphates  than  pyrophosphate  occur.  Orthophosphoric  acid  may  be 
neutralized  in  three  steps  to  produce  the  following  types  of  salts:  KH2PO4, 
monopotassium  orthophosphate;  K2HPO4,  dipotassium  orthophosphate; 
and  K3PO4,  tripotassium  orthophosphate.  All  these  salts  furnish  utiliz- 
able  phosphate,  but  the  effects  of  these  three  salts  on  the  acidity  of  the 
medium  are  quite  different.     In  addition  to  inorganic  phosphates,  the 


ESSENTIAL  NONMETALLIC  ELEMENTS 


95 


organic  phosphates  (esters)  may  also  be  used  as  sources  of  this  element. 
Dox  (1911-1912)  investigated  the  assimilation  of  various  phosphorus 
compounds  by  Aspergillus  niger  with  the  following  results:  Ortho-, 
meta-,  and  pyrophosphates  supported  excellent  growth,  as  did  such 
organic  compounds  of  phosphorus  as  phytin,  sodium  glycerophosphate, 
sodium  nucleinate,  casein,  and  ovovitellin.  Sodium  hypophosphite 
(NaH2P02-H20)  and  sodium  phosphite  (Na2HP03-5H20)  were  not 
utilized  and  appeared  to  be  toxic. 

Smith  (1949)  studied  the  phosphorus  metabolism  of  MeruUus  lacrymans 
and  Marasmius  chordalis  in  connection  with  the  utilization  of  different 
carbon  sources.  In  glucose  medium  M.  lacrymans  grew  better  when 
supplied  w^ith  inorganic  phosphate,  while  M.  chordalis  grew  miore  rapidly 
when  supplied  with  organic  phosphorus  (adenylic  acid).  On  cellobiose 
medium  M.  lacrymans  grew  faster  when  supplied  with  organic  phosphorus. 

The  role  of  phosphorus.  An  idea  of  the  manifold  ways  in  which  phos- 
phorus enters  into  fungus  metabolism  may  be  gained  from  the  studies  of 
Mann  (1944,  1944a).  Aspergillus  niger  was  grown  on  a  glucose-nitrate 
medium  containing  varying  amounts  of  dipotassium  orthophosphate. 
Some  of  Mann's  data  on  the  effect  of  two  concentrations  of  phosphate  are 
given  in  Table  17. 

Table  17.     The  Effect  of  Two  Concentrations  of  Orthophosphate  upon  the 
Appearance,  Sporulation,  and  Other  Metabolic  Functions  of  Aspergillus  niger 
(Mann,   Biochem.   Jour.   38,   1944.     Published  by  permission  of  the   Cambridge 
University  Press.) 


Characteristics  of 

Grown  in  presence  of 

Grown  in  presence  of 

5-day-old  cultures 

0.02%  K2HPO4 

0.2%  K2HPO4 

MyceUum 

Thin,       white,       smooth. 

Thick,          yellowish.     No 

Conidiophores  present 

conidiophores 

Dry  weight,  mg. 

460 

1,092 

Q02  of  intact  mycellium,  lA 

6.12 

11.4 

Total  N,  xng. 

8.1 

23.7 

Total  P,  mg. 

1.5 

12.1 

Thiamine,  /xg 

3.2 

19.0 

Riboflavin,  ng 

16.1 

78.7 

Nicotinic  acid,  ng 

19.4 

302.0 

Medium 

Colorless 

YeUow 

From  Table  17  it  may  be  seen  that  suboptimal  amounts  of  phosphorus 
affect  the  metabolism  of  A.  niger  in  many  ways  besides  diminishing 
growth.  Nitrogen  utilization  was  affected,  and  the  synthesis  of  three 
vitamins  (thiamine,  riboflavin,  and  nicotinic  acid)  was  greatly  decreased. 
The  ability  of  phosphorus-starved  mycelium  to  utilize  oxygen  was  dimin- 
ished, as  shown  by  the  lower  Qo,-     Mann  also  showed  that  utilization  of 


96  PHYSIOLOGY  OF  THE  FUNGI 

phosphate  by  A.  niger  takes  place  only  in  the  presence  of  oxygen.  The 
utilization  of  phosphorus  by  yeasts,  and  presumably  by  other  fungi  which 
are  capable  of  anaerobic  respiration,  may  take  place  in  the  absence  of 
oxygen.  Various  respiratory  inhibitors  such  as  iodoacetate,  azide,  and 
cyanide  inhibited  both  respiration  and  phosphorus  metabolism.  This 
points  to  an  intimate  connection  between  carbohydrate  and  phosphorus 
metabolism.  By  analysis,  ortho-,  meta-,  and  pyrophosphates  were 
found  in  the  mycelium.  Since  only  orthophosphate  was  supplied  in  the 
medium,  it  is  shown  that  A.  niger  is  capable  of  these  transformations. 

Phosphorus  appears  to  participate  in  almost  every  step  in  the  anaerobic 
dissimilation  of  glucose  into  alcohol  by  yeast.  Some  of  these  steps  may 
be  common  to  other  fungi.  It  is  remarkable  that  the  formation  of  alcohol 
by  yeast  and  lactic  acid  in  muscle  should  follow  almost  the  same  pathways. 
Phosphorus  is  required  in  the  enzymatic  transformation  of  glucose  into 
alcohol  and  carbon  dioxide  (Harden,  1932).  Sumner  and  Somers  (1947) 
and  Tauber  (1949)  have  summarized  the  enzymatic  reactions  involved. 

Either  starch  or  glycogen  may  be  transformed  into  glucose-1-phosphate 
by  enzymatic  esterification.  The  shift  of  the  phosphate  radical  to  the 
other  end  of  the  glucose  molecules  leads  to  glucose-6-phosphate,  which 
may  also  be  formed  by  direct  esterification  of  glucose.  Glucose-6-phos- 
phate  is  transformed  into  fructose-6-phosphate  and  then  into  fructose- 
1,6-diphosphate.  Scission  of  a  molecule  of  fructose-l,6-diphosphate 
yields  dihydroxyacetone-1-phosphate  and  D-1-phosphoglyceric  aldehyde. 
An  equally  long  series  of  transformations  leads  to  pyruvic  acid, 
CHs — CO — COOH,  Avhich  on  decarboxylation  by  the  enzyme  carboxylase 
yields  acetaldehyde,  which  is  enzymatically  reduced  by  DPN-H2  to  ethyl 
alcohol.  Cocarboxylase  and  diphosphopyridine  nucleotide  (DPN)  are 
coenzymes,  both  of  which  contain  phosphorus. 

Gould  et  al.  (1942)  studied  the  formation  of  alcohol  by  Fusarnim 
tricothecioides  and  found  the  limited  production  of  alcohol  by  this  species 
was  due  to  insufficient  synthesis  of  diphosphopyridine  nucleotide.  Alco- 
hol production  was  increased  20-  to  25-fold  by  the  addition  of  either  yeast 
extract  or  DPN  to  the  medium.  The  paper  of  Semeniuk  (1943-1944), 
which  deals  with  the  relation  of  phosphorus  to  glucose  dissimilation  by 
Chaetomium  funicola,  has  an  extensive  bibliography  (117  references). 
Nord  and  Mull  (1945)  have  summarized  a  long  series  of  papers  on  the 
physiology  and  biochemistry  of  Fusarium  lini  and  reached  the  conclusion 
that  fermentation  by  this  fungus  follows  a  pathway  which  does  not 
involve  the  sugar  phosphates.  The  review  of  Barron  (1943)  on  the 
mechanisms  of  carbohydrate  metabolisms  contains  much  information 
about  the  role  of  phosphorus  (219  references)  in  carbohydrate  metabo- 
lism. The  role  of  phosphorus  compounds  in  the  transfer  of  energy  was 
noted  in  Chap.  4. 


ESSENTIAL  NONMETALLIC  ELEMENTS  97 

Phosphorus  enters  into  the  composition  of  the  nucleoproteins,  which  are 
found  in  the  nucleus  and  cytoplasm  of  every  cell.  The  nucleoproteins  are 
conjugated  proteins  which  consist  of  a  protein  moiety  in  combination 
with  purine  or  pyrimidine  nucleotides  (nucleic  acids).  These  nucleotides 
are  important  functional  compounds  and  may  be  classified  according  to 
their  heterocyclic  components. 

The  preliminary  hydrolysis  of  purine  and  pyrimidine  nucleotides 
involves  the  removal  of  phosphoric  acid  and  the  formation  of  nucleosides. 
Nucleosides  on  hydrolysis  yield  sugars,  purines  (adenine,  guanine)  or 
pyrimidines  (cytosine,  thymine,  uracil).  The  nucleotides  are  also  classi- 
fied according  to  the  sugar  moiety,  i.e.,  D-ribose  or  D-desoxyribose. 

The  nucleoproteins  which  contain  D-ribose  are  mainly  found  in  the 

cytoplasm,  while  D-desoxyribose  characterizes  the  nucleoproteins  of  the 

nucleus.     The  Feulgen  stain  is  used  by  cytologists  to  detect  the  presence 

of  D-desoxyribose  nucleic  acid.     Viruses,  chromosomes,  and  genes  consist 

largely  of  nucleoproteins.     For  a  review  of  the  role  of  nucleoproteins  see 

Mirsky  (1943). 

NITROGEN 

This  essential  element  is  used  by  fungi  for  functional  as  well  as  struc- 
tural purposes.  The  cell  wall  of  many  species,  with  the  exception  of  the 
Oomycetes  and  yeasts,  appears  to  be  composed  of  chitin  (Brian,  1949). 
Chitin  is  a  linear  polymer,  similar  to  cellulose,  of  D-glucosamine.  The 
amino  group  of  glucosamine  in  chitin  is  acetylated.  This  substance 
makes  up  the  exoskeleton  of  insects  and  Crustacea.  It  is  interesting  that 
the  chitin  formed  by  fungi,  insects,  and  Crustacea  appears  to  be  the  same 
substance.  Protein,  the  basis  of  protoplasm,  is  composed  of  nitrogenous 
substances.  Purines,  pyrimidines,  and  some  of  the  vitamins  are  also 
nitrogen-containing  compounds. 

Not  all  nitrogen  sources  are  equally  suitable  for  all  fungi.  Fungi  may 
be  specific  in  the  nitrogen  sources  they  utilize.  Our  information  on  this 
subject,  while  extensive,  is  far  from  complete.  The  reports  in  the  litera- 
ture which  indicate  that  specific  fungi  are  able  to  grow  on  a  given  source 
of  nitrogen  may  be  accepted  with  confidence,  but  the  reported  negative 
results  are  to  be  viewed  with  caution.  Failure  of  a  fungus  to  grow  upon  a 
given  nitrogen  source  may  mean  only  that  the  medium  used  did  not  con- 
tain the  necessary  growth  factors,  as  in  the  case  of  Ophioholus  graminis 
(See  Chap.  2). 

Classification  according  to  nitrogen  sources  used.  Robbins  (1937), 
Steinberg  (1939,  1950),  and  others  have  classified  the  fungi  according  to 
their  ability  to  utilize  different  sources  of  nitrogen.  In  the  main  Rob- 
bins's  classification  is  as  follows:  (1)  fungi  able  to  utihze  atmospheric 
nitrogen,  nitrate  nitrogen,  ammonmm  nitrogen,  and  organic  nitrogen;  (2) 
fungi  able  to  utilize  nitrate  nitrogen,  ammonium  nitrogen,  and  organic 


98 


PHYSIOLOGY  OF  THE  FUNGI 


nitrogen  but  not  able  to  utilize  atmospheric  nitrogen;  (3)  fungi  able  to 
utilize  ammonium  and  organic  nitrogen  but  unable  to  utilize  atmospheric 
or  nitrate  nitrogen;  (4)  fungi  which  are  able  to  utilize  only  organic 
nitrogen  and  unable  to  utilize  atmospheric,  nitrate,  or  ammonium  nitro- 
gen. Robbins  recognized  that  the  experimental  conditions  might  affect 
the  classification  of  some  fungi.  In  spite  of  admitted  imperfections  the 
above  classification  is  very  useful  in  preparing  media  and  in  discovering 
the  causes  of  failure  of  some  fungi  to  grow  on  certain  media. 

Nitrogen -fixing  fungi.  It  has  been  shown  to  the  satisfaction  of  all 
competent  investigators  that  various  genera  of  bacteria  {Rhizohium, 
Azotobacter,  Clostridium)  contain  species  which  are  able  to  fix  nitrogen. 

Table  18.     Nitrogen  Fixation  by  Phoma  betae  and  Azotobacter  vinlandii 
(Duggar  and  Davis,  Ann.  Missouri  Botan  Garden  3,  1916.) 


Inoculated  flasks 

Uninoculated  flasks 

Mg.  N  fixed 
per  flask 

Organism 

IVIg.  N  per 
flask 

Ave, 

Mg.  N  per 
flask 

Ave. 

Aspergillus  niger 
(30  days) 

Phoma  betae 
(89  days) 

Azotobacter  vinlandii 
(28  days) 

62.510 
62.545 
63.140 

31.010 
31.360 

46.515 
46.480 
46.445 

62 . 732 

31.185 
46 . 480 

62.510 
62.335 
62.300 

25.585 
25.655 

5.810 
6.405 

62.382 

25.620 
6.108 

0.350 

5.565 
40.372 

No  such  agreement  exists  regarding  fungi.  Much  of  the  early  work  on 
nitrogen  fixation  by  fungi  was  done  without  using  proper  precautions. 
However,  in  several  instances  the  experimental  methods  appear  to  be 
beyond  reproach.  Duggar  and  Davis  (1916)  cultured  Phoma  betae  and 
Aspergillus  niger  in  Kjeldahl  flasks  and  determined  the  nitrogen  content 
after  growth  without  removing  either  the  mycelium  or  medium  prior  to 
digestion.  Two  types  of  controls  were  used.  A  number  of  uninoculated 
flasks  which  had  been  stored  under  the  same  conditions  as  the  inoculated 
flasks  were  analyzed  for  nitrogen  at  the  end  of  the  experiment.  A  culture 
of  Azotobacter  vinlandii  served  as  a  positive  control.  The  data  in  Table  18 
show  that  A.  niger  did  not  fix  nitrogen,  while  P.  betae  and  A.  vinlandii  did. 
However,  the  nitrogen-fixing  power  of  P.  betae  was  slight  compared  with 
that  of  A.  vinlandii.  In  addition,  the  following  fungi  were  tested  for 
ability  to  fix  nitrogen,  with  negative  results:  Macrosporium  commune, 
Penicillium  digitatum,  P.  expansum,  and  Glomerella  gossypii.     For  further 


ESSENTIAL  NONMETALLIC  ELEMENTS  99 

references  to  nitrogen  fixation  by  filamentous  fungi  see  Wolf  and  Wolf 
(1947)  and  Buchanan  and  Fulmer  (1930). 

So  far  as  we  are  aware,  only  one  study  of  nitrogen  fixation  by  fungi 
using  modern  isotopic  techniques  (Tove  et  at.,  1949)  has  been  published. 
Phoma  causarina  was  grown  on  a  sucrose-salts  medium  in  oxygen  and 
nitrogen  enriched  with  N^^.  Growth  was  slow  and  sparse  under  these 
conditions,  but  some  N^^  was  fixed.  These  authors  state  that  the  isotopic 
method  is  about  100  times  more  sensitive  than  the  Kjeldahl  procedure 
used  by  other  investigators. 

While  it  is  probable  that  only  a  relatively  few  fungi  are  able  to  fix  nitro- 
gen, the  importance  of  biological  nitrogen  fixation  is  so  great  that  further 
investigations  with  modern  techniques  are  desirable.  Long  ago  Ternetz 
(1907)  reported  that  five  species  of  Phoma  isolated  from  roots  of  Ericaceae 
fixed  significant  amounts  of  nitrogen.  For  a  discussion  of  nitrogen  fixa- 
tion by  bacteria  see  Wilson  (1940). 

Fungi  utilizing  nitrate  nitrogen.  Nitrates  occur  in  the  soil  and  thus  are 
a  ''natural"  source  of  nitrogen.  A  fungus  which  utilizes  nitrate  nitrogen 
(NOs")  must  be  able  to  reduce  the  nitrogen  to  the  oxidation  level  of 
ammonia.  We  may  assume  that  failure  of  a  fungus  to  utilize  nitrate 
nitrogen  is  coupled  with  inability  to  perform  this  reduction.  According 
to  Robbins  (1937)  no  instances  have  been  recorded  in  the  literature  of  an 
organism  being  able  to  utilize  nitrate  nitrogen  and  unable  to  utihze 
ammonium  nitrogen.  This  does  not  mean  that  fungi  which  are  able  to 
utihze  nitrate  nitrogen  will  grow  at  the  same  rate  on  ammonium  nitrogen, 
or  that  all  sources  of  organic  nitrogen  will  be  as  favorable  as  nitrate  nitro- 
gen.    Yeasts  as  a  rule  do  not  utilize  nitrate  nitrogen. 

The  following  is  a  partial  list  of  fungi  which  have  been  reported  or 
observed  to  utilize  nitrate  nitrogen: 

Armillaria  rnellea  C.  velutipes 

Ascobolus  denudata  Cordyceps  militaris 

A.  leveillei  Dendrophoma  obscurans 
Ascochyta  pisi  Dothidella  quercus 
Aspergillus  spp.  Fusarium  spp. 
Botryotinia  convoluta  Glomerella  cingulata 
Botrytis  allii  Gyrnnoascus  setosus 

B.  cinerea  Helminthosporium  spp. 
Cephalothecium  roseum  Lambertella  corni-maris 
Cercospora  apii  Lentinus  tigrinus 

C.  beticola  Macrosporium  sarcinaeforme 
Chaetomiiim  cochlioides  Marasmius  Julvobidbillosus 
C.  convolutum  Neocosmopara  vasinfecta 

C.  globosum  Ophiobolus  graminis 

Colletotrichum  lagenarium  0.  miyabeanus 

C.  lindeniuthianum  Penidllium  spp. 

Collybia  tuberosa  Phoma  apiicola 


100  PHYSIOLOGY  OF  THE  FUNGI 

P.  betae  S.  sclerotioruni 

Pleurage  curvicolla  Sderotium  bataticola 

Pyronema  confluens  Septoria  nodorum 

Pythiomorpha  gonapodyoides  Sordaria  fimicola 

Pythium  debaryanum  Sphaeroholus  stellatus 

P.  intermedium  Sphaeropsis  malorum 

P.  irregular e  Trichoderma  lignorum 

Rhizodonia  solani  VerticiUium  albo-atrum 

Sderotinia  minor  Xylaria  mali 

Several  of  the  species  in  the  above  Hst  were  reported  by  Young  and 
Bennett  (1922)  and  others  by  Robbins  and  Kavanagh  (1942).  Some 
reports  are  found  in  the  papers  of  various  authors,  while  some  of  the  fungi 
have  been  observed  in  our  laboratory  (see  Fig.  17  for  illustrations). 

Fungi  which  utilize  ammonium  nitrogen.  In  the  nitrogenous  com- 
pounds found  in  fungi  the  nitrogen  is  in  the  same  state  of  oxidation  as  in 
ammonium  compounds.  The  following  is  a  partial  list  of  fungi  which 
have  been  reported  or  observed  to  require  ammonium  or  organic  nitrogen 
and  to  be  unable  to  assimilate  nitrate  nitrogen: 

Absidia  coerulea  M.  putillus 

A.  cylindrospora  M.  ramealis 

A.  diibia  M.  rotula 

A.  glauca  M.  scorodonius 

A.  orchidis  Monilinia  frudicola 

Basidiobolus  ranarum  Mortierella  rhizogena 

Ceratostomella  fimbriata  Mucor  flavus 

C.  ulmi  M.  hiemalis 

Choanephora  cucurbitarum  M.  nodosus 

Cyathus  striatiis  M.  pyriformis 

Endothia  parasitica  M.  saturninus 

Lenzites  trabea  M.  stolonifer 

Marasmius  alliaceus  M.  stridus 

M.  androsaceus  Phycomyces  blakesleeanus 

M.  chordaiis  Pleurotus  ostreatus 

M.  epiphyllus  Rhizophlydis  rosea 

M.  foetidis  Rhizopus  nigricans 

M.  graminum  R.  oryzae 

M.  performis  Sporodina  grandis 

M.  personatus  Zygorrhynchus  moelleri 

A  number  of  fungi  in  the  above  list  were  listed  by  Robbins  (1937). 
The  studies  on  Marasmius  are  reported  by  Lindeberg  (1944).  Others  are 
reported  by  various  authors,  while  some  have  been  observed  in  our 
laboratory.  Obviously,  this  list  is  far  from  complete,  and  numerous 
common  fungi  have  been  omitted  from  both  this  and  the  previous  list 
because  of  lack  of  definite  information  regarding  their  ability  to  utilize 
nitrate  nitrogen. 


ESSENTIAL  NONMETALLIC  ELEMENTS 


101 


Fungi  which  utiUze  only  organic  nitrogen.  Certain  fungi  are  unable  to 
utilize  nitrogen  except  in  the  form  of  amino  acids,  peptides,  and  mixtures 
of  these  compounds  such  as  peptone.  The  use  of  organic  nitrogen  does 
not  extend  to  all  organic  compounds  which  contain  this  element.     Many 


A  B  C  D 

Fig.  17.  Growth  of  two  fungi  on  four  media  differing  in  nitrogen  source.  .4,  no 
nitrogen  added;  B,  potassium  nitrate;  C,  ammonium  tartrate;  D,  asparagine.  Above, 
Helminthosporium  sativum;  below,  Ceratostomella  fimbriata. 

of  the  early  reports  claiming  utilization  of  organic  nitrogen  only  by  various 
species  have  been  found  to  be  in  error. 

All  the  early  work  where  peptone  was  the  nitrogen  source  used  is  to  be 
suspected  because  the  need  for  growth  factors  was  not  recognized.  The 
use  of  other  complex  nitrogen  sources  such  as  proteins  makes  interpreta- 
tion doubtful  for  the  same  reason.  However,  in  the  case  of  amino-acid- 
deficient  fungi  a  portion  of  the  nitrogen  source  must  be  supplied  in  the 


102  PHYSIOLOGY  OF  THE  FUNGI 

form  of  a  particular  amino  acid.  Cantino  (1949)  found  that  Blastocladia 
pringsheimii  is  deficient  for  methionine  and  perhaps  other  amino  acids. 
Presumably  other  amino  acids  are  used  to  supply  a  portion  of  the  metabolic 
nitrogen  of  this  species.  The  same  situation  may  exist  in  the  nitrogen 
utilization  of  amino-acid-deficient  mutants  of  Neurospora.  Leonian  and 
Lilly  (1938)  reported  Coprinus  lagopus  and  Pleurotus  corticatus  to  grow  on 
a  mixture  of  five  amino  acids  and  not  on  ammonium  nitrate  as  a  source  of 
nitrogen. 

Inorganic  sources  of  nitrogen.  The  nitrates  commonly  used  in  prepar- 
ing media  are  potassium  nitrate,  sodium  nitrate,  and  calcium  nitrate. 
These  salts  are  equivalent  in  so  far  as  they  supply  the  same  kind  of  nitro- 
gen. They  are  not  equivalent  in  that  different  cations  are  involved. 
Calcium  ion  may  precipitate  a  varying  amount  of  phosphate,  depending 
upon  the  concentrations  of  the  two  ions  and  the  pH  of  the  medium. 

Some  fungi  utilize  nitrite  (N02~)  nitrogen.  Blakeslea  trispora  makes 
some  growth  on  nitrite  nitrogen  (Leonian  and  Lilly,  1938).  Owing  to  the 
instability  of  nitrites  in  acid  solution  and  the  destructive  effect  of  nitrous 
acid  on  proteins  and  amino  acids,  nitrite  nitrogen  is  little  used  in  making 
media.  Nitrite  is  produced  by  many  fungi  from  nitrate  and  may  accumu- 
late in  the  medium  under  certain  conditions.  The  toxic  effect  is  related 
to  the  pH  of  the  medium,  being  greatest  at  low  pH.  Wirth  and  Nord 
(1942)  attributed  the  accumulation  of  pyruvic  acid  in  the  nitrate  medium 
on  which  Fusarium  lini  grew  to  the  presence  of  the  nitrite,  which  inac- 
tivated thiamine  pyrophosphate  (cocarboxylase). 

Yeasts  utilize  nitrate  nitrogen  poorly  as  a  general  rule.  Pirschle  (1930) 
studied  the  relative  value  of  nitrate  and  ammonium  nitrogen  for  a  yeast 
and  concluded  that  poor  utilization  of  nitrate  nitrogen  was  due  in  part  to 
the  accumulation  of  nitrite  in  the  medium.  This  was  shown  by  the  yields 
of  aerated  and  nonaerated  cultures  on  media  containing  nitrate  and 
ammonium  nitrogen  as  well  as  by  analyses  of  the  culture  medium  for 
nitrite.  Aeration  prevented  the  accumulation  of  toxic  amounts  of  nitrite 
or  its  decomposition  product  nitrogen  trioxide.  In  other  experiments 
Pirschle  showed  that  nitrite  inhibited  the  growth  of  yeast  on  ammonium 
nitrogen.  By  adding  sufficient  nitrite  to  a  medium  containing  ammonium 
sulfate,  growth  was  depressed  below  that  obtained  on  potassium  nitrate. 
How  far  these  conclusions  may  be  applied  to  other  fungi  which  do  not 
utilize  nitrate  nitrogen  is  not  known. 

Inorganic  and  organic  ammonium  salts  are  equivalent  in  that  they 
furnish  inorganic  nitrogen;  i.e.,  ammonium  ion.  The  nitrogen  of  all 
ammonium  salts  is  the  same,  but  the  physiological  effects  of  the  anions 
are  not.  The  ammonium  salts  of  strong  inorganic  acids  generally  tend  to 
make  a  culture  medium  more  strongly  acidic  than  when  an  ammonium 
salt  of  a  weak  acid  is  used.     However,  the  situation  is  far  more  compli- 


ESSENTIAL  NONMETALLIC  ELEMENTS 


103 


cated  than  this  simple  theory  Yv'ould  predict.  It  should  be  emphasized 
that  nitrates  and  ammonium  salts  have  opposite  effects  on  the  acidity  of 
culture  media.  Other  conditions  being  equal,  as  nitrate  ions  are  con- 
sumed, the  culture  medium  becomes  more  alkaline,  while  as  ammonium 
ions  are  utilized,  the  culture  medium  becomes  more  acid. 

Before  considering  the  ammonium  salts  of  the  organic  acids,  the  use  of 
ammonium  nitrate  should  be  mentioned.  Both  ions  contain  nitrogen,  a 
feature  which  has  led  many  investigators  to  use  it  in  media.  If  a  fungus 
is  able  to  utilize  both  kinds  of  nitrogen,  the  pH  of  the  medium  will  be 
somewhat  stabilized.  This  salt  should  not  be  used  if  the  purpose  of  an 
experiment  is  to  determine  whether  a  fungus  can  utilize  either  one  or  the 
other  or  both  forms  of  nitrogen.  Some  fungi  apparently  use  nitrate 
nitrogen  in  preference  to  ammonium  nitrogen  when  both  are  supplied  in 
the  medium.  Fusarium  lini  appears  to  be  such  a  fungus  (Wirth  and 
Nord,  1942). 

Table  19.     The  Effect  of  Various  Organic  Acids  on  the  Growth  of  Four 
Fungi  on  Media  Containing  Ammonium  Nitrate 
Initial  pH  5.5.     Figures  are  milligrams  of  mycelium  produced.     (Leonian  and  LiUy, 
Am.  Jour.  Botany  27,  1940.) 


Organic  acids, 
0.02M 

Mucor  raman- 
nianus 

Phythium 
ascophallon 

Pythiomorpha 
gonapodyoides 

Phycomyces 
blakesleeanus 

Control 

Acetic 

Lactic 

77 
144 
154 
142 
135 
158 
174 
152 

8 

19 

34 

117 

23 

156 

56 

0 

8 

40 

113 

157 

76 

180 

112 

0 

27 

144 
121 

Succinic 

Glutaric 

Fumaric 

165 
144 
189 

Tartaric 

Citric 

149 
171 

M etarrhizium  glutinosum  {Myrothecium  verrucaria)  grew  well  on  nitrate 
nitrogen  alone  and  poorly  on  ammonium  nitrogen  (Brian  et  al.,  1947). 
Ammonium  nitrogen  inhibited  growth  of  this  fungus,  whether  nitrate  was 
present  or  not.  Growth  was  equally  poor  on  ammonium  nitrate  and 
ammonium  sulfate.  Since  this  fungus  grew  well  on  media  containing 
nitrate  as  the  sole  source  of  nitrogen,  these  authors  have  questioned  the 
common  belief  that  all  fungi  which  are  able  to  utilize  nitrate  nitrogen  can 
also  utihze  ammonium  nitrogen.  Most  fungi  appear  to  utihze  ammonium 
nitrogen  before  nitrate  nitrogen  when  both  are  supplied  in  the  medium, 
but  this  is  not  universal.  Rippel  (1931)  found  the  pH  of  the  medium  to 
determine  which  form  of  nitrogen  was  utilized  by  Aspergillus  niger  and 
A.  oryzae.     Additional  examples  are  given  by  Foster  (1949). 

The  utilization  of  ammonium  and  some  forms  of  organic  nitrogen  may 


104 


PHYSIOLOGY  OF  THE  FUNGI 


be  modified  by  the  presence  of  other  compounds  in  the  medium.  Among 
these  the  organic  acids,  especially  the  four-carbon  dicarboxylic  acids,  play 
an  important  role.  This  subject  has  been  studied  by  Leonian  and  Lilly 
(1940),  Burkholder  and  McVeigh  (1940),  Brian  et  at.  (1947),  and  Bernhard 
and  Albrecht  (1947).  The  data  in  Table  19  illustrate  the  effect  of  organic 
acids  on  the  amount  of  growth  of  four  fungi. 

Succinic  and  fumaric  acids  were  most  uniform  in  their  effect  on  nitrogen 
assimilation.     Figure  18  shows  the  reciprocal  effect  of  varying  amounts  of 


200 

~~^^ 

> 

c 

^.-^ 

1 1 

150 

V 

< 

/  \ 

^2.5g.Nh 

'4NO3/I 

100 

/ 

\ 

7.  succinic  ac 

id /I 

C 

50 

< 

\ 

°( 

^ 

D 

1. 

D 

g.  NH4r 

2 
OO3/I 

0 

3.0 

3.0 


2.0 


1.0 


g.  succinic  acid/l 

Fig.  18.  The  reciprocal  effect  of  varying  amounts  of  succinic  acid  (ammonium 
nitrate  constant)  and  ammonium  nitrate  (succinic  acid  constant)  on  the  growth  of 
Phycoviyces  blakesleeanus.  (Drawn  from  data  of  Leonian  and  Lilly,  Am.  Jour.  Botany 
27:  22,  1940.) 

succinic  acid  and  ammonium  nitrogen  on  the  growth  of  Phycomyces 
blakesleeanus,  which  does  not  utilize  nitrate  nitrogen.  The  amount  of 
growth,  within  certain  limits,  is  directly  proportional  to  the  amount  of 
succinic  acid  in  the  medium. 

Brian  et  al.  (1947)  have  suggested  on  the  basis  of  studies  on  Myro- 
thecium  verrucaria  that  a  definite  antagonism  exists  between  the  metabolic 
pathways  involved  in  nitrate  and  ammonium  utilization,  and  in  the 
presence  of  ammonium  nitrogen  the  nitrate  pathway  is  blocked.  Ammo- 
nium nitrogen  is  poorly  utilized  unless  certain  organic  acids  are  present  in 
the  medium.     Malic  acid  has  no  effect  on  utilization  of  nitrate  nitrogen. 


ESSENTIAL  NONMETALLIC  ELEMENTS  105 

These  authors  suggest  that  different  pathways  of  carbohydrate  iitihzation 
may  be  followed,  depending  upon  whether  nitrate  or  ammonium  nitrogen 
is  present. 

Organic  sources  of  nitrogen.  Of  the  vast  number  of  organic  com- 
pounds which  contain  nitrogen  the  ones  of  interest  in  fungus  nutrition  are 
those  which  occur  naturally.  A  few  exceptions  will  be  noted  later.  In 
practice,  this  means  proteins  and  the  products  of  protein  hydrolysis. 
The  following  steps  in  protein  hydrolysis  have  been  recognized :  protein  -^ 
metaprotein  — >  proteoses  — >  peptones  — >  peptides  — >  amino  acids.  Pep- 
tone, which  is  a  complex  mixture  of  peptides  and  amino  acids,  is  frequently 
used  as  a  nitrogen  source  in  media.  According  to  Gortner  (1929), 
peptones  are  neither  coagulated  by  heat  nor  precipitated  by  saturating  a 
solution  with  ammonium  sulfate,  properties  which  distinguish  peptones 
from  proteins,  metaproteins,  and  proteoses.  Since  peptides  having  some 
11  amino-acid  residues  are  precipitated  by  ammonium  sulfate,  it  may  be 
deduced  that  the  peptides  in  peptone  have  on  the  average  10  or  less 
amino-acid  residues.  Peptone  is  a  useful  source  of  nitrogen  when  it  is 
desired  to  culture  a  large  number  of  species  upon  a  single  medium.  A 
part  of  its  virtue  may  be  ascribed  to  its  complex  nature,  for  a  mixture  of 
nitrogen  sources  may  be  better  utilized  than  a  single  source.  Peptone 
also  contains  most  of  the  water-soluble  vitamins  (Stokes  et  al.,  1944). 

Most  of  the  amino  acids  which  have  been  isolated  from  proteins  are 
listed  in  Table  20.  In  addition,  the  amides  of  aspartic  and  glutamic 
acids  are  included.  These  compounds  are  found  free  in  many  plants  and 
are  thus  available  to  the  fungi  in  nature. 

These  amino  acids  are  not  of  equal  value  in  fungus  nutrition.  The 
relative  value  of  24  amino  acids  for  14  fungi  was  tested  by  Leonian  and 
Lilly  (1938)  who  found  no  one  amino  acid  w^as  best  for  all  these  species. 
Steinberg  (1942)  made  an  extensive  study  of  growth  of  Aspergillus  niger 
on  22  amino  acids.  Seven  were  excellent  sources  of  nitrogen  for  A.  niger: 
alanine,  arginine,  aspartic  and  glutamic  acids,  glycine,  proline,  and 
hydroxyproline.  Steinberg  expressed  the  opinion  that  the  seven  amino 
acids  which  supported  the  most  growth  of  A.  niger  are  those  which  are 
synthesized  first  (primary  amino  acids)  by  this  fungus  and  from  which  the 
other  amino  acids  (secondary  amino  acids)  are  normally  formed.  It  is 
assumed  that  the  "primary"  amino  acids  enter  directly  into  the  metabolic 
pathways,  while  the  "secondary"  amino  acids  must  undergo  preliminary 
deamination  before  use.  The  primary  amino  acids  are  probably  not 
the  same  for  all  fungi.  Lilly  and  Leonian  (1942)  investigated  the  effect 
of  nitrogen  source  on  the  growth  of  10  strains  of  Saccharomyces  cerevisiae. 
The  data  in  Table  21  show  clearly  that  different  amino  acids  vary  in 
effectiveness,  and  that  different  strains  of  the  same  organism  respond 
differently  to  the  same  source  of  nitrogen. 


106 


PHYSIOLOGY  OF  THE  FUNGI 


Table  20.     Common  Names  and  Formulas  ok  Some  Alpha-amino  Acius  Isolated 

FROM  Proteins  and  of  Some  Amides  Found  in  Plants 
Monoamine  dicarboxylic  acids: 

Aspartic  acid:  HOOC— CH2— CHCNHo)— COOH 

Glutamic  acid:        HOOC— CHo— CH.— CH(NH2)— COOH 
Amides  of  monoamino  dicarboxylic  acids: 


Asparagine : 
Glutamine: 
Basic  amino  acids: 
Argiiiine*: 
Lysine*: 
Histidine  * : 


NH2OC— CH2— CH(NH2)— COOH 
NH2OC— CH2— CH,— CH(NH2)— COOH 

NHo— C(=NH)— NH— CHo— CH2— CH2— CH(NH2)- 
NH2— CH2— CH2— CH2— CH2— CHCNH,)— COOH 
CH 


-COOH 


N 


^ 


NH 


CH===C— CH2— CH(NH2)— COOH 
Monoamino  monocarboxylic  acids: 


Glycine: 
Alanine : 
Valine  * : 
Leucine*: 
Isoleucine*: 
Phenylalanine  * : 
Serine : 
Threonine*: 


Tryptophane  '* 


Tryosine : 


CHoCNH.)— COOH 
CH3— CH(NH2)— COOH 
(CH  3)  2— CH— CH  (NHo)— COOH 
(CH3)2— CH— CH2— CH(NH2)— COOH 
CH3— CH2— CH(CH3)— CHCNH.)— COOH 
CeHs- CHo— CH  (NH2)— COOH 
CH2(0H)— CH(NH2)— COOH 
CH3— CH(OH)— CH(NH2)— COOH 

C— CH2— CH(NH2)— COOH 

CH 


NH 


HO^^^ 
CH2— CH 


\ 


CH2— CH(NH2)— COOH 


Proline: 


CH2     CH— COOH 


NH 
CHOH 


-CH2 


Hydroxy  proline:      CHo  CH— COOH 

NH 
Sulfur-containing  amino  acids: 

Cysteine:  CHsCSH)— CHCNHo)- COOH 

Cystine:  HOOC— CH(NH2)— CH2— S— S— CH2— CHCNHa)- COOH 

Methionine*:  CH2(SCH3)CH2— CHCNH.)- COOH 

*  The  10  amino  acids  reported  by  Rose  (1938)  as  essential  for  the  nutrition  of  the  white  rat. 

Physiological  specificity  extends  to  the  configuration  as  well  as  the 
composition  of  the  molecule.  Optical  isomers  (enantiomorphs)  usually 
have  different  physiological  properties.  A  mixture  of  amino  acids  may  or 
may  not  be  utilized  better  than  a  single  amino  acid.  The  effect  of  one 
amino  acid  on  the  utilization  of  another  varies  with  the  amino  acids 


ESSENTIAL  XON METALLIC  ELEMENTS 


107 


involved  and  the  specific  fungus  used.  Leonian  and  Lilly  (1940)  tested 
the  growth  of  Phycomyces  blakesleeanus  upon  five  single  amino  acids  and 
upon  a  mixture  of  these  five  amino  acids  with  the  following  results :  mix- 
ture of  five  amino  acids,  214;  asparagine,  209;  DL-alanine,  151;  arginine, 
50;  aspartic  acid,  203;  glycine,  201;  and  glutamic  acid,  189  mg.,  respec- 
tively. Arginine  is  a  poor  nitrogen  source  for  P.  blakesleeanus,  but  the 
presence  of  arginine  in  the  amino-acid  mixture  did  not  depress  growth. 
More  complex  relations  were  found  with  yeast  (Lilly  and  Leonian,  1942). 
Ten  strains  of  yeast  were  grown  upon  media  containing  a  mixture  of  six 
amino  acids  (aspartic  and  glutamic  acids,  arginine,  asparagine,  alanine, 
and  leucine).  Upon  this  mixture  of  amino  acids  two  strains  grew  as  well 
as  or  better  than  upon  the  best  single  amino  acid  (aspartic  acid).     The 

Table  21.     Comparison  of  Various  Soi'rces  of  Nitrogen  for  Six  Strains  of 

Yeast 
Milligrams  of  dry  yeast  cells  produced  in  72  hr.     Each  culture  received  8  mg.  of  X. 
(Lilly  and  Leonian,  Proc.  West  Va.  Acad.  Sci.  16,  1942.) 


Nitrogen  source 


Ammonium  sulfate 

Urea 

L- Aspartic  acid .  .  . 

L-Aspargine 

Glycine 

DL-Norleucine 


Yeast  strain 


18.7 
33.2 
60.7 
49.4 
3.0 
29.3 


21.2 
31.5 
59.9 
45.8 
1.2 
17.6 


22.3 
32.9 
65.6 
50.0 
2.0 
33.0 


17.5 
27.3 
62.0 
49.2 
2.1 
18.4 


21.7 
32.0 
52.4 
47.6 
1.0 
1.2 


23.8 
35.1 
70.6 
35.0 
1.1 
4.2 


amount  of  growth  of  one  strain  was  70.6  mg.  on  aspartic  acid  alone,  while 
on  the  amino-acid  mixture  only  38.6  mg.  was  produced.  Omission  of 
asparagine  from  the  mixture  increased  the  yield  to  52.0  mg.  These 
results  show  that  the  effects  of  multiple  nitrogen  sources  upon  growth,  and 
perhaps  other  functions,  are  complex. 

Organic  acids,  especially  the  four-carbon  dicarboxylic  acids,  affect  the 
utilization  of  some  amino  acids  much  as  they  do  that  of  ammonium  com- 
pounds. Phycomyces  hlakesleeamis  on  a  medium  containing  arginine 
produced  43  mg.  of  mycelium  per  flask.  Addition  of  0.1  per  cent  succinic 
acid  to  the  medium  increased  the  yield  to  192  mg.  (Leonian  and  Lilly, 
1940). 

Nitrogen  utilization  by  the  fungi  has  been  studied  for  almost  a  century, 
but  many  of  the  problems  involved  are  not  yet  solved.  Brenner  (1914) 
has  reviewed  the  early  work  in  this  field,  especially  with  reference  to  the 
divergent  views  of  Raciborski  and  Czapek  on  the  mode  of  utilization  of 
amino  acids.     Raciborski  held  that  amino  acids  were  deaminat^d  before 


108  PHYSIOLOGY  OF  THE  FUNGI 

utilization,  while  Cznpek  believed  that  amino  acids  were  utilized  directly. 
Both  processes  arc  doubtless  involved,  and  only  prolonged  study  of 
specific  fungi  and  various  nitrogen  sources  will  permit  elucidation  of  these 
questions. 

One  of  the  main  uses  of  nitrogen  is  in  the  synthesis  of  proteins.  With 
the  exception  of  certain  amino  acids  (primary  amino  acids)  and  ammonia, 
most  nitrogen  sources  undergo  modification  before  entering  the  synthetic 
metabolic  pathways.  Nitrates,  nitrites,  and  hydroxylamine  are  pre- 
sumably reduced  to  ammonia  before  assimilation.  Those  amino  acids 
(secondary  amino  acids)  which  do  not  enter  directly  into  the  metabolic 
pathways  leading  to  the  synthesis  of  protein  are  probably  deaminated. 
Burk  and  Horner  (1939)  have  listed  the  types  of  deamination  performed 
by  fungi  as  follows: 

1.  Deamination  by  hydrolysis: 

H2O 
R— CH(XH.:)— COOH >  R— CH(OH)— COOH  +  NH3 

2.  Deamination  by  hydrolysis  followed  by  decarboxylation: 

H2O 
R— CHCNHo)— COOH >  R— CH2OH  +  CO2  +  NH3 

3.  Oxidative  deamination: 

MO, 
R— CHCNHa)— COOH >  R— CO— COOH  +  NH, 

The  production  of  higher  alcohols,  "fusel  oil,"  is  due  to  hydrolytic 
deamination  and  decarboxylation  of  various  amino  acids,  especially 
leucine,  which  yields  isoamyl  alcohol.  Various  species  of  filamentous 
fungi,  especially  those  which  produce  alcohol,  are  capable  of  the  same 
reactions.  The  following  amino  acids  are  converted  by  yeasts  into  alco- 
hols having  one  less  carbon  than  the  parent  amino  acid:  leucine,  isoleucine, 
phenylalanine,  trytophane,  and  valine.  Wirth  and  Nord  (1942)  indicate 
that  Fusarium  lini  oxidatively  transforms  alanine  into  pyruvic  acid. 
For  further  information  on  the  process  of  deamination  by  yeast,  see  Thorn 
(1937).  The  process  of  deamination  releases  nitrogen  in  the  form  of 
ammonia,  which  is  utilized  by  most  fungi. 

It  seems  probable  that  the  synthesis  of  amino  acids  is  the  next  step  in 
protein  formation.  The  formation  of  primary  amino  acids  may  result 
from  the  reaction  of  ammonia  with  certain  alpha-keto  acids  (pyruvic, 
oxalacetic,  and  ketoglutaric) ;  this  is  essentially  the  reverse  of  oxidative 
deamination.     This  process  may  be  fo.'mulated  as  follows: 

R— CO— COOH  +  NH3  ->  R— C(=NH)— COOH  +  H.  -^  R— CHCNH.)— COOH 

In  addition,  yeasts  are  able  to  add  ammonia  to  fumaric  acid  to  form 
aspartic  acid  (Haehn  and  Leopold,  1937).  The  role  of  the  four-carbon 
dicarboxylic  acids  in  nitrogen  assimilation  may  be  explained  on  the  basis 
that  these  acids  are  transformed  into  kcto  acids.  Brian  et  al.  (1947)  have 
assumed  that  those  fungi,  such  as  Phycomyces  hlakesleeanus  and  Myro- 


ESSENTIAL  NONMETALUC  ELEMENTS 


109 


thecium  verrucaria,  which  make  hmited  growth  on  ammonium  nitrogen  do 
so  because  they  are  unable  to  synthesize  in  adecjuate  amounts  the  neces- 
sary three-,  four-,  and  five-carbon  keto  acids.  The  interrelation  among 
various  dicarboxylic  acids  is  shown  in  schemes  IV,  VIII,  and  IX. 

The  reactions  discussed  above  account  for  the  synthesis  of  only  a  few  of 
the  20  or  so  amino  acids  found  in  fungus  protein.  Another  type  of  reac- 
tion may  account  for  the  synthesis  of  secondary  amino  acids.  This  is 
called  the  transamination  reaction  and  may  be  represented  as  follows : 

R— CO— COOH  +  R'— CH(NH2)C00H  -^  R— CH(NH2)C00H+R'— CO— COOK 

According  to  Roine  (1947),  Torulopsis  utilis  has  the  necessary  enzymatic 
mechanisms  for  the  synthesis  of  the  following  amino  acids  by  transamina- 


se 


30  40  50 

Time  in  minutes 


60 


70 


Fig.  19.  Amounts  of  soluble  nitrogen  compounds  found  in  the  trichloroacetic  acid 
extract  as  a  function  of  time.  Data  are  based  on  100  ml.  of  yeast  suspension,  or 
about  5  g.  fresh  yeast.  Curve  1  represents  total  soluble  nitrogen,  curve  2  total 
amide  nitrogen,  curve  3  alanine  nitrogen,  and  curve  4  dicarboxylic-amino-acid  nitro- 
gen.    (Courtesy  of  Roine,  Ann.  Acad.  Sci.  Fennicae  26:  63,  1947.) 

tion:  aspartic  acid,  glutamic  acid,  alanine,  valine,  leucine,  and  isoleucine. 
For  a  general  review  of  the  transamination  reaction,  see  Herbst  (1944). 
Roine  (1947)  has  obtained  experimental  evidence  which  indicates  that 
in  Torulopsis  utilis  the  primary  amino  acids  are  formed  first  and  that  the 
secondary  amino  acids  are  then  formed  from  them.  This  evidence  w^as 
obtained  by  analyzing  the  nonprotein  nitrogen  fraction  which  was 
extracted  from  cells  of  various  ages  with  trichloroacetic  acid  (a  protein 
precipitant) .  Nitrogen-starved  cells  of  T.  utilis  were  suspended  in  carbo- 
hydrate-free medium  which  contained  ammonium  nitrogen.  The  culture 
was  aerated.  Every  10  min.  a  portion  of  the  crop  was  harvested,  and  the 
distribution  of  nitrogen  compounds  in  the  trichloroacetic  acid  extract  was 
determined.  Figure  19  show  clearly  that  the  first  stages  of  protein  syn- 
thesis consist  in  the  formation  of  monoamino  dicarboxylic  acids,  their 


no  PHYSIOLOGY  OF  THE  FUNGI 

amides,  and  alanine.  It  may  be  assumed  that  the  amides  of  gkitamic  and 
aspartic  acid  function  in  yeast  as  nitrogen  carriers,  as  they  do  in  green 
plants. 

Preformed  amino  acids  are  probably  used  in  protein  synthesis.  In 
principle  this  process  is  the  reverse  of  hydrolysis.  Many  complex  chem- 
ical reactions  are  involved.  Proteins  vary  in  complexity,  the  simplest 
having  molecular  weights  in  the  neighborhood  of  16,000  to  17,000.  The 
molecular  weight  of  some  proteins  is  said  to  be  greater  than  1,000,000, 
and  tobacco  mosaic  virus  protein  is  estimated  to  have  a  molecular  weight 
of  40,000,000.  In  spite  of  these  enormous  molecular  weights,  a  good 
deal  is  known  about  the  structure  of  proteins.  Fundamentally,  a  pro- 
tein consists  of  amino-acid  residues  joined  together  by  peptide  linkages, 
— CH2 — NH — CO — .  Since  different  proteins  have  highly  specific 
properties  which  depend  upon  the  molecular  structure,  the  synthesis  of 
these  compounds  involves  a  systematic  linking  together  of  amino-acid 
residues  in  a  definite  pattern.  For  reviews  of  protein  structure  the  reader 
is  referred  to  Bull  (1941)  and  Astbu^y  (1943). 

The  general  pathways  of  nitrogen  utilization  by  fungi  are  shown  in 
scheme  III. 

Scheme   III.     Possible  Pathways  of  Protein  Synthesis  from  Various  Sources 

OF  Nitrogen 
Nitrates — ■ >  Ammonia >     Primary  amino  acids 


Secondary  amino  acids 
Ammonia — 


i 
Secondary  amino  acids 


Primary  amino  acids- 


Peptides 

i 
Polypeptides 

i. 
Protems 

OTHER  NONMETALLIC  ELEMENTS 

It  is  not  known  whether  fungi  require  nonmetallic  elements  other  than 
hydrogen,  oxygen,  sulfur,  phosphorus,  and  nitrogen.  Boron  and  iodine 
are  frequently  added  to  culture  media,  but  good  evidence  of  their  essen- 
tiality for  fungi  appears  to  be  lacking.  Sodium  chloride  is  frequently 
added  to  media,  but  neither  sodium  nor  chlorine,  so  far  as  is  known,  is 
essential  for  the  fungi. 

In  nature  fungi  come  in  contact  with  many  nonessential  elements. 
Some  of  these  may  be  metabolized.  Others  may  modify  the  life  processes 
of  the  fungi  by  their  toxic  action  or  by  other  means.  Chlorine  is  found  in 
various  compounds  synthesized  by  fungi,  e.g.,  non-ionic  chlorine  is  found 
in  chloramphenicol,  one  of  the  newer  antibiotics.  Many  species  of  fungi 
metabolize  arsenic.     Penicillium  hrevicaule,  among  other  species,   pro- 


ESSENTIAL  HON  METALLIC  ELEMENTS  111 

duces  a  volatile,  toxic,  organic  arsenic  compound,  trimethylarsine, 
(CH3)3As,  which  has  an  odor  resembling  garlic.  In  the  past  P.  hrevicaule 
has  been  recommended  for  the  detection  of  arsenic  compounds  in  forensic 
medicine.  This  microbiological  test  for  the  presence  of  arsenic  is  said  to 
))e  many  times  as  sensitive  as  the  ]\Iarsh  test.  The  early  work  on  the 
utilization  of  arsenic  compounds  by  fungi  is  reviewed  by  La  Far  (1911) 
and  more  critically  by  Challenger  et  al.  (1933).  P.  hrevicaule  also  pro- 
duces  dimethyl   selenide   from   selenium   compounds    (Challenger   and 

North,  1934). 

SUMMARY 

The  classification  of  essential  elements  as  structural  or  functional  may 
be  misleading  in  that  an  element  usually  plays  many  roles.  This  is 
especially  true  of  the  essential  nonmetallic  elements. 

With  the  exception  of  carbon  dioxide  all  the  organic  compounds  used 
by  or  contained  in  fungi  contain  hydrogen.  One  of  the  most  important 
hydrogen-containing  compounds  is  water.  This  compound  is  associated 
with  proteins  in  the  form  of  bound  water,  and  it  functions  as  a  solvent  in 
which  most  if  not  all  biochemical  reactions  take  place.  Water  enters 
into  many  reactions,  particularly  in  the  hydrolytic  processes  of  ''diges- 
tion."    Apparently  fungi  do  not  utilize  free  hydrogen. 

None  of  the  fungi  appear  to  be  obligate  anaerobes.  Many  are  faculta- 
tive anaerobes,  while  some  appear  to  be  strict  aerobes.  Free  oxygen  is 
used  by  the  fungi  in  respiration,  chiefly  as  an  acceptor  of  hydrogen.  The 
facultative  anaerobes  have  another  mechanism  of  oxidation  which  does 
not  involve  free  oxygen.  This  is  called  anaerobic  respiration,  or  fermen- 
tation. The  rate  and  amount  of  growth  and  sporulation  and  the  meta- 
bolic by-products  of  a  given  fungus  are  affected  by  the  oxygen  supply. 

The  problem  of  specificity  arises  in  connection  with  the  form  of  sulfur 
utilized.  Most  fungi  utilize  sulfate  sulfur,  but  some  require  reduced  sul- 
fur. Other  species  are  unable  to  synthesize  specific  sulfur-containing 
amino  acids,  especially  methionine.  Sulfur  enters  into  the  composition 
of  enzymes  and  other  proteins,  peptides,  and  at  least  two  vitamins. 

The  fungi  utilize  phosphorus  in  the  form  of  phosphate  salts  and  esters. 
Some  specificity  in  the  different  sources  of  phosphate  has  been  found 
Phosphate  esters  enter  into  a  wide  variety  of  enzymatic  reactions,  and 
many  coenzymes  are  phosphate  esters. 

It  is  thought  that  certain  phosphate  esters  act  to  transfer  chemical 
energy  to  certain  enzymatic  reactions.  Phosphorus  enters  into  the  com- 
position of  proteins,  especially  the  nucleoproteins,  which  are  found  in  the 
nucleus  or  cytoplasm  of  every  cell.  Viruses  and  genes  are  thought  to 
consist  largely  of  nucleoproteins. 

Fungi  differ  in  ability  to  utilize  different  forms  of  nitrogen.  A  few 
utilize  atmospheric  nitrogen;  many  utilize  nitrate  nitrogen;  and  a  still 


112  PHYSIOLOGY  OF  THE  FUNGI 

greater  number  utilize  ammonium  nitrogen.  All  species  are  abletoutilize 
some  form  of  organic  nitrogen.  Other  constituents  in  media,  especially 
the  four-carbon  dicarboxylic  acids,  modify  the  availability  of  ammonium 
nitrogen  and  certain  amino  acids.  Not  all  amino  acids  are  of  equal  value 
in  fungus  nutrition.  The  primary  amino  acids  are  those  which  enter 
directly  metabolic  pathways,  while  secondary  amino  acids  are  deaminated 
before  the  nitrogen  is  used. 

Most  of  the  nitrogen  utilized  by  fungi  enters  into  the  synthesis  of  pro- 
teins. The  primary  amino  acids  are  formed  first,  and  the  secondary 
amino  acids  are  formed  from  primary  amino  acids.  Proteins  are  the 
most  complex  compounds  synthesized  by  living  cells.  Many  of  the 
vitamins  and  other  essential  metabolites  also  contain  nitrogen. 

REFERENCES 

AsTBUKY,  W.  T. :  X-rays  and  the  stoichiometry  of  the  proteins,  Advances  in  Enzymol. 
3:  63-108,  1943. 

Barnes,  T.  C:  Textbook  of  General  Physiology,  The  Blakiston  Company,  Phila- 
delphia, 1937. 

Barron,  E.  S.  G.  :  Mechanisms  of  carbohydrate  metabolism.  An  essay  in  compara- 
tive biochemistry,  Advances  in  Enzymol.  3:  149-189,  1943. 

Bernhard,  K.,  and  H.  Albrecht:  Stoffwechselprodukte  des  Mikroorganismus 
Phycomyces  Blakesleeanus  in  glucose  haltiger  Nahrlosung  und  Untersuchungen 
liber  das  Wachstum  dieses  Schimmelpilzes  bei  verschiedenen  Stickstoffquellen, 
Helv.  Chim.  Acta  30 :  627-632,  1947. 

Birkinshaw,  J.  H.,  W.  P.  K.  Findlay,  and  R.  A.  Webb:  Biochemistry  of  wood-rot- 
ting fungi.  3.  The  production  of  methyl  mercaptan  by  Schizophyllum  com- 
mune Fr.,  Biochem.  Jour.  36:  526-529,  1942. 

Bonner,  D.:  Production  of  biochemical  mutants  in  PeniciUium,  Am.  Jour.  Botany 
33:788-791,  1946. 

Brenner,  W.:  Die  Stickstoffnahrung  der  Schimmelpilze,  Cent.  Bakt.,  Abt  II,  49: 
555-647,  1914. 

Brian,  P.  W.:  Studies  on  the  biological  activity  of  griseofulvin,  Ann.  Botany  13: 
59-77,  1949. 

Brian,  P.  W.,  P.  J.  Curtis,  and  H.  G.  Hemming:  Glutinosin:  a  fungistatic  metabolic 
product  of  the  mould  Mctarrhizium  glutinosum  S.  Pope.,  Proc.  Roy.  Soc.  (Lon- 
don), Ser.  B,  135:  106-132,  1947. 

Buchanan,  R.  E.,  and  E.  I.  Fulmer:  Physiology  and  Biochemistry  of  Bacteria, 
Vol.  II,  the  Williams  &  Wilkins  Company,  Baltimore,  1930. 

Bull,  H.  B.  :  Protein  structure.  Advances  in  Enzymol.  1:  1-42,  1941. 

BuRK,  D.,  and  C.  K.  Horner:  Nitrogen  assimilation  by  yeast,  Wallerstein  Labs. 
Communs.  2  :  5-23,  1939. 

BuRKHOLDER,  P.  R.,  and  I.  McVeigh:  Growth  of  Phycomyces  blakesleeanus  in  rela- 
tion to  varied  environmental  conditions,  Am.  Jour.  Botany  27:  634-640,  1940. 
*Cantino,  E.  C:  The  physiology  of  the  aquatic  Phycomycete,  Blastocladia  Pring- 
sheimii,  with  emphasis  on  its  nutrition  and  metabolism.  Am.  Jour.  Botany  36 : 
95-112,  1949. 

Challenger,  F.,  C.  Higginbottom,  and  L.  Ellis:  The  formation  of  organo-metal- 
loidal  compounds  by  microorganisms.  I.  Trimethylarsine  and  dimethylethyl- 
arsine,  Jour.  Chem.  Soc.  1933:  95-101. 


ESSENTIAL  NON METALLIC  ELEMENTS  113 

Challenger,  F.,  and  H.  E.  North:  The  formation  of  organo-metalloidal  compounds 

by  microorganisms.     II.  Dimethylselenide,  Jour.  Chetn.  Soc.  1934:  68-71. 
Dixon,  M.:  Manomctric  Methods  as  Applied  to  the  Measurement  of  Ceil  Respira- 
tion and  Other  Processes,  Cambridge  University  Press,  New  York,  1943. 
DoRRELL,  W.  W. :  The  oxidative  respiration  of  Fusarium  graminearurn,  thesis,  Uni- 
versity of  Wisconsin,  1948. 
Dox,  A.  W. :  The  phosphorus  assimilation  of  Aspergillus  niger,  Jour.  Biol.  Chem.  10 : 

77-80,  1911-1912. 
*DuGGAR,  B.  M.,  and  A.  R.  Davis:  Studies  in  the  physiology  of  the  fungi,     I.  Nitrogen 

fixation,  Ann.  Missouri  Botan.  Garden  3:  413-437,  1916. 
Fellows,  H.:  The  influence  of  oxygen  and  carbon  dioxide  on  the  growth  of  Ophio- 

bolus  graminis  in  pure  culture.  Jour.  Agr.  Research  37:  349-355,  1928. 
Foster,  J.  W.:  Chemical  Activities  of  Fungi,  Academic  Press,  Inc.,  New  York, 

1949. 
Fries,  N.:  X-ray  induced  parathio trophy  in  Ophiostoma,  Svensk  Botan.  Tidskrift  40: 

127-140,  1946. 
Fries,  N.:  Spontaneous  physiological  mutations  in  Ophiostoma,  Hereditas  34:  338- 

350,  1948. 
GoRTNER,  R.  A.:  Outlines  of  Biochemistry,  1st  ed.,  John  Wiley  &  Sons,  Inc.,  New 

York,  1929. 
GoRTNER,  R.  A.:  Outlines  of  Biochemistry,  3d  ed.,  John  Wiley  &  Sons,  Inc.,  New 

York,  1949. 
Gould,  B.  S.,  A.  A.  Tttell,  and  H.  Jaffe:  Biochemistry  of  Fusaria.  The  influence 

of  diphosphopyridine  nucleotide  on  alcoholic  fermentation  {in  vivo),  Jour.  Biol. 

Chem.  146:  219-224,  1942. 
Haehn,  H.,  and  H.  Leopold:  Ueber  eine  Aspartasewirkung  der  Hefe,  Biochem.  Zeit 

292 :  380-387,  1937. 
Harden,  A.:  .Alcoholic  Fermentation,  Longmans,  Roberts  and  Green,  London,  1932. 
Hawker,  L.  E.:  The  effect  of  vitamin  Bi  on  the  utilization  of  glucose  by  Melano- 

spora  destruens  Shear.,  Ann.  Botany  8:  79-90,  1944. 
Herbst,  R.  M.:  The  transamination  reaction,  Advances  in  Enzymol.  4:  75-97,  1944. 
Hevesy,    G.  :   Some   applications   of  radio-active   indicators   in   turnover  studies, 

Advances  in  Enzyinol.  7:  111-214,  1947. 
Hockenhull,  D.  J.  D.:  The  suphur  metabolism  of  mould  fungi:  the  use  of  "bio- 
chemical mutant"  strains  of  Aspergillus  nidulans  in  elucidating  the  biosynthesis 

of  cystine,  Biochim.  et  Biophys.  Acta  3 :  326-335,  1949. 
*HoLLis,  J.  P.:  Oxygen  and  carbon  dioxide  relations  oi Fusarium  oxysporum  Schlecht. 

and  Fusarium  eumartii  Carp.,  Phytopathology  38:  761-775,  1948. 
La  Far,  F.:  Technical  Mycology.     Vol.  II,  Eumycetic  Fermentation  (trans.  C.  T.  C. 

Salter),  Charles  Griffin  &  Co.,  Ltd.,  London,  1911. 
Leonian,  L.  H.,  and  V.  G.  Lilly:  Studies  on  the  nutrition  of  fungi.     I.  Thiamin,  its 

constituents,  and  the  source  of  nitrogen.  Phytopathology  28 :  531-548,  1938. 
Leonian,  L.  H.,  and  V.  G.  Lilly:  Studies  on  the  nutrition  of  fungi.     IV.  Factors 

influencing  the  growth  of  some  thiamin-requiring  fungi.  Am.  Jour.  Botany  27 : 

18-26,  1940. 
Lilly,  V.  G.,  and  L.  H.  Leonian:  Nitrogen  metabolism  in  Saccharomyces  cerevisiae, 

Proc.  West  Va.  Acad.  Sci.  16:  60-70,  1942. 
LiNDEBERG,   G. :   Ucber  die  Physiologic  Ligninabbauender  Bodenhymenomyzeten, 

Symbolae  Botan.  Upsalienses  8(2) :  1-183,  1944. 
Mandels,  G.  R.,  and  R.  G.  H.  Siu:  Rapid  assay  for  growth:  Determination  of 

microbiological  susceptibility  and  fungistatic  activity.  Jour.  Bact.  60 :  249-262, 

1950. 


114  PHYSIOLOGY  OF  THE  FUNGI 

*Mann,  T.:  Studies  in  the  metabolism  of  mould  fungi.     I.  Phosphorus  metabolism 
in  moulds,  Biochem.  Jour.  38:  339-345,  1944. 

Mann,  T.  :  Studies  in  the  metabolism  of  mould  fungi.  II.  Isolation  of  pyrophos- 
phate and  metaphosphate  from  Aspergillus  niger,  Biochem.  Jour.  38:  345-351, 
1944a. 

MiKSKY,  A.  E.:  Chromosomes  and  nucleoproteins,  Advances  in  Enzymol.  3:  1-34, 
1943. 

NicKEKSON,  W.  J.,  and  W.  R.  Carroll:  The  effect  of  culture  history  on  the  metabolic 
activities  of  cells  of  Zygosaccharomyces,  Jour.  Cellular  Comp.  Physiol.  22 :  21-32, 
1943. 

NoRD,  F.  F.,  and  R.  P.  Mull:  Recent  progress  in  the  biochemistry  of  Fusaria, 
Advances  in  Enzymol.  5:  165-205,  1945. 

PiRSCHLE,  K.:  Biologische  Beobachtungcn  liber  Hefewachstum  mit  besonders 
Berlicksichtigen  von  Nitraten  als  Stickstoffquelle,  Biochem.  Zeit.  218:  412-444, 
1930. 

Plumlee,  C.  H.,  and  A.  L.  Pollard:  Studies  on  bromine-oxidizable  sulfur-contain- 
ing compounds  in  mold  metabolism,  Jour.  Bad.  57 :  405-407,  1949. 

Raistrick,  H.,  and  J.  M.  Vincent:  Studies  in  the  biochemistry  of  micro-organisms. 
77.  A  survey  of  fungal  metabolism  of  inorganic  sulphates,  Biochem.  Jour.  43 : 
90-99,  1948. 

Raulin,  J.:  Etudes  chimiques  sur  la  vegetation,  Ann.  sci.  nat.,  Ser.  V,  11:  93-299, 
1869. 

Rippel,  K.:  Quantitative  Untersuchungen  iiber  die  Abhangigkeit  der  Stickstoff as- 
similation  von   der   WasserstofRonenkonzentration   bei   einigen   Pilzen,    Arch. 
Mikrobiol.  2:  72-135,  1931. 
*RoBBiNS,  W.  J.:  The  assimilation  by  plants  of  various  forms  of  nitrogen,  Aj».  Jour. 
Botany  24:  243-250,  1937. 

RoBBiNs,  W.  J.,  and  V.  Kavaxagh:  Vitamin  deficiencies  of  the  filamentous  fungi, 
5otaM.  i^e!;.  8:  411-471,  1942. 
♦■Roine,  p.:  On  the  formation  cf  primary  amino  acids  in  the  protein  synthesis  by 
yeast,  Ann.  Acad.  Sci.  Fennicae  26:  1-83,  1947. 

Rose,  W.  C:  The  nutritive  significance  of  the  amino  acids,  Physiol.  Revs.  18:  109- 
136,  1938. 

ScHEFFER,  T.  C,  and  B.  E.  Livingston:  Relation  of  oxygen  pressure  and  temper- 
ature to  growth  and  carbon  dioxide  production  in  the  fungus  Polystictus  versi- 
color, Am.  Jour.  Botany  24:  109-119,  1937. 

Semeniuk,  G.:  The  dissimilation  of  glucose  by  Chaetomiuin  funicola  Cke.  III. 
Some  phosphorus  relationships  of  Chaetomium  funicola,  Iowa  State  Coll.  Jour. 
Sci.  18 :  325-308,  1943-1944. 

Siu,  R.  G.  II.,  and  G.  R.  Mandels:  Rapid  method  for  determining  mildew  suscepti- 
bility of  materials  and  disinfecting  activity  of  compounds.  Textile  Research 
Jour.  20:  516-518,  1950. 

Smith,  V.  M.:  On  the  mechanism  of  enzyme  action.  XXXIX.  A  comparative 
study  of  the  metabolism  of  carbohydrates,  in  the  presence  of  inorganic  and 
organic  phosphates,  by  Merulius  lacrymans  and  Marasmius  chordalis,  Arch. 
Biochem.  23 :  446-472,  1949. 

Steinberg,  R.  A.:  Effects  of  barium  salts  upon  Aspergillus  niger  and  their  bearing 
upon  the  sulfur  and  zinc  metabolism  of  the  fungus  in  an  optimum  solution, 
Botan.  Gaz.  97:  666-671,  1936. 

Steinberg,  R.  A. :  Growth  of  fungi  in  synthetic  nutrient  solutions,  Botan.  Rev.  5 : 
327-350,  1939. 


ESSENTIAL  NON METALLIC  ELEMENTS  115 

Steinberg,  R.  A.:  Sulfur  and  trace-element    nutrition    of  Aspergillus  niger,  Jour. 

Agr.  Research  63:  109-127,  1941. 
Steinberg,  R.  A. :  The  process  of  amino  acid  formation  from  sugars  by  Aspergillus 

niger,  Jour.  Agr.  Research  64:  615-633,  1942. 
*Steinberg,  R.  a.  :  Growth  of  fungi  in  synthetic  nutrient  solutions.     II.  Botan.  Rev. 

16 :  208-228,  1950. 
Stokes,  J.  L.,  IM.  Gunness,  and  J.  W.  Foster:  Vitamin  content  of  ingredients  of 

microbiological  culture  media,  Jour.  Bad.  47 :  293-299,  1944. 
Sumner,  J.  B.,  and  G.  F.  Somers:  Chemistry  and  Methods  of  Enzymes,  2d  ed., 

Academic  Press,  Inc.,  New  York,  1947. 
Tamiya,    II.:    Atmung,    Giirung   und   die  sich   daran   beteiligenden   Enzyme   von 

Aspergillus,  Advances  in  Enzymol.  2 :  183-238,  1942. 
Tauber,  H.  :  The  Chemistry  and  Technology  of  Enzymes,  John  Wiley  &  Sons,  Inc., 

New  York,  1949. 
Ternetz,    C.  :   Protoplasma  bewegung  und   Fruchtkorperbildung  bei   Ascophanus 

carneus  Pers.,  Jahrb.  wiss.  Botan.  35:  273-312,  1900. 
Ternetz,  C:  Ueber  die  Assimilation  des  atmospharischen  Stickstoff  durch  Pilze, 

Jahrh.  wiss.  Botan.  44 :  353-408,  1907. 
Thorn,  R.  S.  W.  :  The  assimilation  of  nitrogen  from  amino  acids  by  yeast,  Jour.  Inst. 

Brewing  43 :  288-293,  1937. 
TovE,  S.  R.,  H.  F.  Niss,  and  P.  W.  Wilson:  Nitrogen  fixation  by  the  fungus, 

Phoma  causarina,  49th  meeting  Soc.  Am.  Bact.  (Abs.  No.  A-26,  p.  59),  1949. 
Umbreit,  W.  W.,  R.  H.  Burris,  and  J.  F.  Stauffer:  Manometric  Techniques  and 

Related   Methods  for  the  Study  of  Tissue   Metabolism,   Burgess  Publishing 

Co.,  Minneapolis,  1945. 
VoLKONSKY,  M. :  Sur  les  conditions  de  culture  et  le  pouvoir  de  synthese  de  Saprolegnia 

sp.     Etude  qualitative  de  I'alimentation  carbon^e,  azotee  et  sulfuric,   Ann. 

inst.  Pasteur  50 :  703-730,  1933. 
VoLKONSKY,  M.:  Sur  I'assimilation  des  sulfates  par  les  champignons:  euthiotrophie 

et  parathiotrophie,  Compt.  rend.  acad.  sci.  197:  712-714,  1933a. 
YoLKONSKY,  M. :  Sur  la  nutrition  de  quelques  champignons  saprophytes  et  parasites, 

Ann.  inst.  Pasteur  52:  76-101,  1934. 
Wilson,  P.  W. :  The  Biochemistry  of  Symbiotic  Nitrogen  Fixation,  University  of 

Wisconsin  Press,  Madison,  1940. 
WiRTH,  J.   C,  and  F.  F.   Nord:  Essential  steps  in  the  enzymatic  breakdown  of 

hexoses  and  pentoses.     Interaction  between  dehydrogenation  and  fermentation, 

Arch.  Biochem.  1 :  143-163,  1942. 
Wolf,  F.  A.,  and  F.  T.  Wolf:  The  Fungi,  Vol.  II,  John  Wiley  &  Sons,  Inc.,  New 

York,  1947. 
Young,  H.  C,  and  C.  W.  Bennett:  Growth  of  some  parasitic  fungi  in  synthetic 

culture  media,  Am.  Jour.  Botany  9:  459-461,  1922. 


CHAPTER  7 
CARBON  SOURCES  AND  CARBON  UTILIZATION 

Carbon  occupies  a  unique  position  among  the  essential  elements 
required  by  living  organisms.  Almost  half  of  the  dry  weight  of  fungus 
cells  consists  of  carbon.  Protoplasm,  enzymes,  the  cell  wall,  and  reserve 
nutrients  stored  within  the  cells  are  compounds  of  carbon.  Carbon  com- 
pounds are  equally  important  in  fugus  nutrition.  Fungi  secure  energy 
by  oxidizing  organic  compounds.  In  addition  to  being  the  main  struc- 
tural elements,  carbon  compounds  play  an  equally  important  functional 
role.  The  number  of  carbon  compounds  known  far  exceeds  the  total 
of  known  compounds  of  all  the  other  elements,  because  of  the  property  of 
carbon  of  forming  compounds  in  which  carbon  is  linked  to  carbon  in  the 
form  of  chains  and  rings.  Various  other  elements  such  as  nitrogen, 
oxygen,  and  sulfur  may  serve  as  linking  elements.  While  many  carbon 
compounds  are  stable  at  ordinary  temperatures,  others  are  extraordinarily 
sensitive  to  a  wide  range  of  chemical  reagents  and  to  slight  changes  in  the 
physical  environment. 

Organic  compounds  differ  in  composition,  structure,  and  configuration. 
These  are  key  factors  which  must  be  considered  in  relation  to  utilization 
of  organic  compounds  by  fungi.  Since  more  is  known  about  carbohy- 
drates and  related  compounds  as  carbon  sources,  and  about  the  manner 
in  which  they  are  dissimilated  and  assimilated,  than  about  any  other 
class  of  organic  compounds,  most  of  the  discussion  in  this  chapter  will 
be  devoted  to  these  topics.  In  the  main,  only  naturally  occurring  organic 
compounds  will  be  considered. 

MONOSACCHARIDES  AND  RELATED  COMPOUNDS 

The  simple  sugars,  or  monosaccharides,  have  the  general  formula 
C„(H20)n.  The  carbon  chain  is  unbranched  except  in  a  few,  very  rare 
sugars.  The  functional  groups  present  are  primary  ( — CH2OH)  and 
secondary  ( — CHOH — )  alcohol  groups,  and  an  aldehyde  ( — CHO)  or 
ketone  ( — CO — )  group,  actual  or  potential,  is  always  present.  The 
primary  alcohol  and  aldehyde  groups  are  restricted  to  the  end  positions 
of  the  carbon  chain,  while  the  ketone  group  is  usually  on  the  second 

116 


CARBON  SOURCES  117 

carbon  in  the  chain.  Sugars  having  an  aldehyde  group  are  called  aldoses, 
those  having  ketone  group,  ketoses;  the  ending  -ose  denotes  a  sugar.  In 
addition,  the  sugars  are  further  classified  according  to  the  number  of 
carbon  atoms  in  the  chain,  e.g.,  pentoses,  hexoses,  or  more  specifically 
as  aldopentoses,  ketohexoses,  etc.  While  it  will  be  necessary  in  the 
discussion  to  follow  to  include  some  information  about  the  chemistry 
and  structure  of  the  sugars,  the  reader  is  advised  to  consult  suitable 
texts  for  further  information.  Those  of  Oilman  (1943)  and  Pigman  and 
Goepp  (1948)  are  recommended. 

Compounds  which  have  the  same  composition  and  the  same  molecular 
weight  are  called  isomers.  There  are  16  aldohexoses  (32,  if  the  alpha 
and  beta  forms  are  considered),  which  have  the  same  percentage  com- 
position and  the  same  functional  groups  as  glucose  (dextrose).  There 
are  eight  possible  ketohexoses  isomeric  with  fructose.  Two  kinds  of 
isomers  exist  among  the  sugars:  First,  there  are  those  which  have  the 
same  physical  properties  but  differ  in  the  direction  in  which  they  rotate 
plane-polarized  light  (enantiomorphs) .  Isomers  of  this  kind  occur  in 
pairs,  and  the  configuration  of  the  functional  groups  of  one  isomer  is  the 
mirror  image  of  the  configuration  of  the  other.  Enantiomorphs  usually 
differ  physiologically.  One  such  isomer  may  be  utilized  and  the  other 
not,  or  one  may  be  utilized  much  more  rapidly  than  the  other.  Pasteur 
(1860)  was  the  first  to  demonstrate  that  fungi  are  able  to  distinguish 
between  such  isomers.  Penicillium  glaucum  utilized  c?-tartrate  more 
rapidly  than  Z-tartrate  {d  and  /  refer  to  optical  rotation).  Second,  there 
are  those  isomers  which,  although  they  have  the  same  functional  groups, 
have  these  groups  arranged  in  a  different  order,  so  that  one  isomer  is  not 
the  mirror  image  of  the  other  (diastereoisomers).  It  is  usually  safe  to 
assume  that  one  member  of  a  pair  of  enantiomorphs  will  be  better 
utilized  than  the  other,  but  such  an  assumption  about  utilization  of 
diastereoisomers  is  not  possible. 

Since  not  all  sugars  of  a  group  such  as  the  aldohexoses  are  utilized  by 
fungi,  it  is  of  interest  to  compare  chemical  structure  or  configuration  with 
utilization.  Not  all  fungi  are  able  to  utilize  exactly  the  same  sugars 
(Fig.  20).  Whether  a  sugar  is  utilized  or  not  depends  upon  both  the 
configuration  of  the  sugar  and  the  particular  abihties  of  the  specific 
fungus.  By  configuration  is  meant  the  spatial  arrangement  of  the 
hydrogen  and  hydroxyl  groups.  The  long  history  of  chemical  investi- 
gation which  established  the  configuration  of  the  simple  sugars  must  be 
passed  by.  Inasmuch  as  glucose  is  the  key  compound  in  sugar  chemistry, 
as  well  as  in  physiology,  particular  emphasis  will  be  devoted  to  this 
aldose. 

The  structures  of  the  glucose  enantiomorphs  are  given  at  the  top  of 
page  119. 


118 


PHYSIOLOGY  OF  THE  FUNGI 


A  B  CD 

Fig.  20.  Growth  of  three  fungi  on  four  sugars.  A,  gkicose;  B,  fructose;  C,  sucrose; 
D,  maltose.  Top  row,  Monilinia  Jructicola  (8  days);  middle,  Mucor  ramannianus 
(8  days);  bottom,  Ustilago  striiformis,  fragmenting  strain  (20  days). 


CARBOS  SOURCES 


119 


1. 

2. 


CHO 

H— C— OH 


3.  HO— C— H 

I 

4.  H— C— OH 


5. 
6. 


H— C— OH 

I 

CHoOH 
D-Glucose 


CHO 

I 
HO— C— H 

I 
H— C— OH 

I 

HO— C— H 

HO— C— H 

I 

CH.OH 
L-Ghicosc 


The  letters  d  and  l  indicate  that  these  sugars  belong  to  different  series; 
they  do  not  indicate  optical  rotation.  The  small  letters  d  and  /  have 
been  used  in  the  past  to  express  two  separate  ideas,  optical  rotation  or 
configuration.  The  use  of  d  and  I  in  the  old  literature  makes  it  difficult 
at  times  to  discover  which  enantiomorph  was  meant.  The  configuration 
of  the  secondary  hydroxyl  group  farthest  from  the  carbonyl  group  deter- 
mines to  which  series  a  sugar  belongs.  D-Glucose  is  the  form  which 
occurs  naturally  and  is  meant  when  glucose  is  used  without  qualification. 
Not  all  naturally  occurring  sugars  belong  to  the  d  series;  e.g.,  L-arabinose. 
For  the  sake  of  clearness  and  accuracy,  the  series  designation  should 
always  be  used  where  there  is  any  chance  of  confusion  and  misinterpre- 
tation. Pigman  and  Goepp  (1948)  point  out  that  only  sugars  of  the 
galactose  type  occur  naturally  as  both  enantiomorphs.  D-Galactose  is 
fermented  by  some  yeasts,  while  L-galactose  is  not. 

Hexoses.  The  following  hexoses  occur  naturally:  D-glucose,  D-man- 
nose,  D-galactose,  L-galactose,  D-fructose,  and  L-sorbose.  It  is  doubtful 
if  L-sorbose  occurs  in  green  plants,  but  it  is  formed  from  sorbitol  by 
bacterial  {Acetobacter  suboxydans)  oxidation  (Bertrand,  1904). 

CHO  CHO  CHoOH 


H— C— OH 


HO— C— H 


C=0 


HO— C— H 

1 

HO— C— H 

HO— C— H 

H— C— OH 

H— C— OH 

H— C— OH 

H— C— OH 

1 

H— C— OH 

H— C— OH 

CH2OH 

D-Glucose 

CH2OH 
D-Mannose 

CH2OH 

D-Fructose 

CHO 

1 

CHO 

CH2OH 

H— C— OH 

HO— C— H 

C=0 

1 

HO— C— H 

1 

H— C— OH 

HO— C— H 

HO— C— H 

1 

H— C— OH 

H— C— OH 

H— C— OH 

HO— C— H 

HO— C— H 

t 

CH2OH 
D-Galactose 

CH2OH 

ly-Galactose 

CH2OH 
L-Sorbose 

120 


PHYSIOLOGY  OF  THE  FUNGI 


The  configuration  of  glucose,  mannose,  and  fructose  is  the  same  for 
carbons  3  to  6.  In  the  presence  of  dilute  alkali  these  sugars  undergo 
enolization  to  produce  the  same  enol  form. 


D-Glucose 
CHO 

H— C— OH 


R 


D-Mannose 
CHO 

HO— C— H 

I 
R 

it 
CHOH 

II 
COH 


D-Fructose 
CH,OH 

C=0 

I 
R 


R    ±: 


Common  enol  form 

Other  effects  of  alkali  and  heat  on  sugars  were  noted  in  Chap.  2. 

Many  fungi  will  utilize  these  three  sugars  if  configuration  is  important 
in  determining  availability.  However,  these  sugars  are  not  equivalent 
for  all  fungi.  The  fact  that  galactose  is  not  utilized  by  all  fungi  which 
utilize  the  three  closely  related  sugars  is  illustrated  by  the  data  in  Table 
22. 

Glucose  is  utilized  by  more  fungi  than  any  other  sugar  and  is  nearly  a 
universal  carbon  source.  In  attempting  to  culture  fungi  of  unknown 
nutritional  requirements  on  synthetic  or  semisynthetic  media,  glucose 
should  be  the  first  carbon  source  used.  However,  there  are  a  few  fungi 
which  are  unable  to  utilize  glucose,  or  any  sugar,  as  a  carbon  source. 
Leptomitus  lacteus  (Schade,  1940;  Scliade  and  Thimann,  1940)  is  unable  to 
utilize  glucose,  fructose,  galactose,  or  sucrose.  Skoog  and  Lindegren 
(1947)  have  reported  the  behavior  of  12  strains  of  Saccharomyces  cere- 
visiae  which  did  not  utilize  glucose  when  first  isolated.  These  strains 
became  adapted  to  glucose  on  sufficiently  long  exposure  to  this  sugar. 
Cheo  (1949)  found  certain  isolates  of  Ustilago  striiformis  to  be  unable  to 
grow  on  glucose  when  freshly  transferred  from  a  sucrose  medium.  After 
2  to  4  weeks  these  isolates  began  to  grow.  This  behavior  suggests  the 
formation  of  an  adaptive  enzyme  which  was  not  formed  when  these 
isolates  were  cultured  on  sucrose  medium.  Some  fungi,  such  as  L.  lacteus, 
apparently  lack  the  ability  to  form  adaptive  enzymes  for  glucose  utili- 
zation and  must  be  classed  as  absolutely  incapable  of  glucose  utilization, 
while  the  yeasts  of  Skoog  and  Lindegren  and  the  isolates  of  U.  striiformis 
are  facultatively  able  to  utilize  glucose.  The  differences  among  these 
fungi  probably  lie  in  the  ability  to  form  adaptive  enzymes. 

No  carbon  source  can  be  utilized  if  the  medium  is  lacking  in  any 
essential  element  or  compound.  Kinsel  (1937)  and  Stevens  and  Larsh 
(1939)  reported  that  Diplodia  macrospora  would  grow  only  on  disacchar- 
ides  and  not  on  media  containing  glucose  or  other  monosaccharides. 


CARBON  SOURCES  121 

The  explanation  of  this  anomalous  situation  was  given  by  Margolin 
(1940)  and  confirmed  by  Wilson  (1942),  who  found  that  D.  macrospora 
was  deficient  for  biotin.  It  is  probable  that  other  vitamin-deficient 
fungi  have  been  reported  in  the  past  as  unable  to  utilize  certain  sugars 
owing  to  the  absence  of  specific  growth  factors.  Negative  results 
reported  in  the  literature  are  therefore  to  be  viewed  with  caution. 

Wolf  and  Shoup  (1943)  studied  the  oxidation  of  carbohydrates  by 
Allomyces  arhuscula,  A.  javanicus,  A.  moniliformis,  and  A.  cystogenus. 
All  four  species  oxidized  dextrin  (degraded  starch),  while  A.  arhuscula 
oxidized  maltose  and  sucrose  in  addition.  The  other  common  naturally 
occurring  sugars,  including  glucose  and  fructose,  were  not  oxidized.  It 
has  since  been  shown  that  A.  arhuscula  is  deficient  for  methionine  and 
thiamine  (Yaw,  1950). 

While  there  is  an  immense  amount  of  information  scattered  throughout 
the  literature  to  the  effect  that  a  certain  sugar  is  utilized  by  various 
species,  much  of  this  information  deals  with  relatively  few  sugars. 
Critical  studies  on  the  utilization  of  the  sugars  are  rare.  Margolin 
(1942)  studied  the  amount  of  growth  of  21  fungi  on  four  hexoses.  These 
data  (Table  22)  were  obtained  under  uniform  conditions.  A  mixed 
nitrogen  source  (ammonium  nitrate  and  amino  acids)  was  used,  and 
vitamins  were  supplied  to  the  deficient  fungi.  The  time  chosen  for 
harvest  in  this  study  was  the  time  maximum  weight  was  attained  on 
glucose.  This  work  suffers  from  the  common  defect  that  the  yields  are 
compared  on  the  basis  of  a  fixed  time  of  harvest.  The  ideal  way  of 
determining  the  value  of  different  sugars  for  fungi  would  be  to  study 
both  the  rate  and  amount  of  growth  as  a  function  of  time  of  incubation. 

The  following  generalizations  about  utilization  of  the  common  hexoses 
may  be  drawn  from  the  data  in  Table  22:  (1)  There  is  no  single  sugar 
which  supports  the  maximum  amount  of  growth  for  all  of  these  fungi. 
(2)  All  of  these  fungi  utilize  glucose,  although  the  maximum  amount  of 
growth  was  not  always  attained  on  this  sugar.  (3)  The  more  closely  the 
configuration  of  another  sugar  approaches  that  of  glucose,  the  more 
fungi  utilize  it.  It  is  believed  that  these  generalizations  are  valid  for  all 
fungi  which  utilize  sugars. 

Steinberg  (1939)  found  D-glucose,  D-fructose,  D-mannose,  L-sorbose, 
and  sucrose  to  be  equally  effective  in  the  nutrition  of  Aspergillus  niger 
while  D-galactose,  lactose,  glycerol,  and  mannitol  were  poor  sources  of 
carbon  for  this  fungus.  Herrick  (1940)  reported  that  two  isolates  of 
Stereum  gausapatum  grew  on  glucose,  fructose,  mannose,  and  galactose. 
One  isolate  made  significantly  better  growth  on  fructose;  the  other  grew 
equally  well  on  all  four  sugars.  This  indicates  that  not  all  isolates  of  a 
species  are  alike  in  ability  to  utilize  a  given  sugar.  The  utilization  of 
different  carbon  sources  by  A.  oryzae  was  investigated  in  detail  by  Tamiya 


122 


PHYSIOLOGY  OF  THE  FUNGI 


(1932).  This  paper  should  he  consulted  for  the  experimental  details 
and  references  to  the  literature.  One  hundred  twenty-three  carbon 
compounds  were  investigated,  and  of  the  hexoses,  mannose  supported 

Table  22.     Milligrams  of  Mycelium  Produced  by  21  Fungi  Grown  on  Media 

Containing  Different  Sugars 
AL  the  sugars  were  used  at  a  rate  which  supphed  8  g.  of  carbon  per  liter.     Each 
125-ral.  flask  contained  20  ml.  of  medium.     Cultures  were  incubated  at  25°C.     Each 
weight  in  the  table  is  the  average  of  12  cultures.      (Margolin,  thesis,  West  Virginia 
University,  1942.) 


Fungus 


Blakeslea  trispora 

Diplodia  macrospora 

D.  natalensis 

Fusarium  lycopersici 

Helicostylum  pyriforme 

Helmintiwsporum  sativum. . . 

Mucor  ramannianus 

Pilaira  moreaui 

Phycomyces  blakesleeanus .  .  . 

Phytophthora  cactorum 

P.  erythroseptica 

P.fagopyri 

Pythiomorpha  gonapodyoides 

Pythium  ascophaUon 

Rhizopus  nigricans 

R.  suinus 

Rosellinia  arcuata 

Sordaria  fimicola 

Syncephalastrum  racemosum. 

Thielavia  basicola 

Typhula  variabilis 


Days  of 
incu- 
bation 


6 

15 

8 

6 

5 

8 

8 

7 

7 

14 

12 

6 

6 

6 

4 

6 

6 

6 

5 

10 

12 


Mg.  mycelium 


D- 

Glu- 
cose 


91 

83 

199 

108 

126 

75 

89 

40 

138 

119 

79 

89 

152 

85 

121 

130 

73 

121 

131 

60 

181 


D- 

Fruc- 
tose 


94 

55 
154 
101 

81 
128 
118 

32 
130 

40 

20 

51 
122 

56 
114 
128 

58 
162 
141 

54 
122 


D- 

Man- 
nose 


98 

71 

89 

100 

126 

83 

115 

45 

139 

16 

81 

19 

79 

84 

117 

136 

49 

147 

126 

55 

113 


D- 

Ga- 
lac- 
tose 


123 
55 
50 

126 
99 
46 

116 
44 
74 
11 
17 
11 
14 
10 

121 

135 
33 
28 

140 
78 
23 


Mal- 
tose 


113 

94 

190 

119 

102 

96 

128 

44 

101 

157 

114 

20 

76 

111 

121 

30 

63 

127 

132 

57 

202 


Suc- 
rose 


10 
58 

199 
74 
11 

100 
12 
11 

111 
77 
93 

130 

142 

116 
7 
12 
38 
16 
15 
61 

126 


Lac- 
tose 


7 

21 

17 

18 

40 

40 

124 

44 

6 

4 

10 

13 

12 

27 

5 

8 

34 

52 

13 

6 

15 


the  most  growth.  Quantitative  data  on  the  utilization  of  L-sorbose  by 
fungi  is  less  abundant  than  for  the  other  hexoses.  Observations  in  this 
laboratoiy  indicate  that  many  fungi  either  do  not  utilize  sorbose  or  do  so 
slowly. 

Pentoses.  The  pentoses  shown  below  occur  naturally,  mostly  in  the 
form  of  polysaccharides  or  other  complex  compounds.  L-arabinose  and 
D-xylose  are  the  most  easily  available  and  have  been  more  extensively  used 
than  the  other  pentoses.  The  formulas  for  the  naturally  occurring 
pentoses  are  given  below: 


CARBON  SOURCES 

CHO 

CHO 

1 

CHO 

HO— C— H 

H— C— OH 

H— C— OH 

1 

H— C— OH 

HO— C— H 

H— C— OH 

1 

H— C— OH 

1 

HO— C— H 

H— C— OH 

CH2OH 

D-Arabinose 

CH2OH 

L-Arabinose 

CH2OH 
D-Ribose 

CHO 

CH2OH 

1 

CHO 

H     C     OH 

1 
C— 0 

H— C— OH 

HO— C— H 

H— C— OH 

1 

H— C— OH 

H— C— OH 

HO— C— H 

HO— C— H 

CH2OH 
D-Xylose 

CH2OH 

L-Xylulose 

CH2OH 
ir-Lyxose 

123 


Aspergillus  niger  utilizes  D-xylose  and  L-arabinose  but  not  their 
enantiomorphs,  as  is  shown  in  Table  23.  Many  of  the  pentoses  listed 
in  Table  23  are  difficult  to  obtain  in  quantity,  which  accounts  for  the 
varied  amounts  used  per  culture. 

Table  23.     The  Amount  of  Growth  and  Sporulation  of  Aspergillus  niger  on 

Various  Pentoses 
Time  of  incubation,  4  days.     Cultures  incubated  at  35°C.     (Steinberg,  Jour.  Agr. 
Research  64,  1942.) 


Pentose 

Pentose,  g. 
per  culture 

Mg.  mycelium 

Sporulation 

D-Lyxose 

D- Xylose 

L-Xylose 

D-Arabinose 

L-Arabinose 

D-Ribose 

L-Ribose 

1.0 

2.0 

0.5 

2.0 

2.0 

0.25 

0.25 

0.2 
860.2 

6.2 

0 
205.1 

0 

5.2 

0 
10 

1 
0 
6 
0 
0 

Herrick  (1940)  found  Stereum  gausapatum  to  utilize  xylose  better  than 
arabinose,  while  Aspergillus  oryzae  utilizes  arabinose  better  than  xylose, 
(Tamiya,  1932).  Lentinus  lepideus  utilizes  xylose  (Nord  and  Vitucci, 
1947).  A  comparative  study  of  five  fungi  on  xylose  and  arabinose  indi- 
cated that  xylose  was  utilized  either  more  completely  or  more  rapidly 
than  arabinose  (Margolin,  1942).  The  data  of  Margolin  are  given  in 
Table  24;  for  comparable  growth  of  these  species  on  other  sugars,  see 
Table  22. 


124 


PHYSIOLOGY  OF  THE  FUNGI 


Methylpentoses.  D-Isorhamnose,  L-fucose,  and  L-rhamnose  are  related 
to  D-glucose,  L-galactose,  and  L-mannose  in  that  carbon  G  with  its  primary 
alcohol  group  in  these  hexoses  has  been  replaced  by  a  methyl  group. 
These  methylpentoses  have  not  been  thoroughly  investigated  in  nutri- 
tional studies  involving  many  fungi.  Aspergillus  niger  utilizes  L-rham- 
nose to  some  extent,  but  L-fucose  is  not  utilized  (Steinberg,  1942).  A. 
oryzae  makes  much  poorer  growth  on  L-rhamnose  than  on  D-xylose  or 

Table  24.     Milligrams  of  IMycelium  Produced  by  Five  Fungi  Grown  upon 

Xylose  and  Arabinose 
These  sugars  were  used  at  concentrations  which  suppHed  8  g.  of  carbon  per  liter. 
Each  125-ml.  flask  contained  20  ml.  cf  medium.     Cultures  were  incubated  at  25 °C. 
(Margolin,  thesis,  West  Virginia  University,  1942.) 


Fungus 


Blakeslea  trispora 

Mucor  ramannianus 

Phycomyces  blakesleeanus . .  . . 
Phytophihora  erythroseptica .  . 
Pythiomorpha  gonapodyoides . 


Days  of 
incubation 


6 
8 
7 
12 
6 


D-Xylose ' 


77 
77 
126 
15 
33 


L-Arabinose 


49 
74 
85 
7 
18 


*  This  sugar  was  called  Z-xylose  in  the  earlier  literature. 

L-arabinose  (Tamiya,  1932).  Of  the  five  fungi  listed  in  Table  24  only 
Mucor  ramannianus  utilizes  L-rhamnose  (Margolin,  1942).  Stereum 
gausapatum  utilizes  rhamnose  about  as  well  as  arabinose  (Herrick,  1940), 
Sugar  alcohols.  Reduction  of  the  aldehyde  or  keto  group  of  the 
simple  sugars  converts  them  into  alcohols.  Several  sugar  alcohols  are 
widely  distributed  in  nature.  Only  the  formulas  for  three  of  the  natu- 
rally occurring  sugar  alcohols  will  be  given. 


CHoOH 

CHoOH 

1 

CH2OH 

H— C— OH 

HO— C— H 

H— C— OH 

[0— C— H 

HO     C     H 

HO— C—H 

1 

H— C— OH 

H— C— OH 

HO— C—H 

H— C— OH 

H— C— OH 

1 

H— C— OH 

CH2OH 

Sorbitol 

CH2OH 
Mannitol 

CHoOH 
Galactitol  (Dulcitol) 

Most  fungi  appear  to  utilize  the  corresponding  sugars  with  greater 
facility  than  the  sugar  alcohols.  Data  for  the  comparative  growth  of 
five  fungi  on  these  sugar  alcohols  and  the  parent  sugars  are  given  in 
Table  25. 


CARBON  SOURCES 


125 


Sugar  acids.  Three  types  of  sugar  acids  may  be  produced  from  aldoses 
by  oxidizing  the  terminal  groups.  Oxidation  of  the  aldehyde  group 
yields  aldonic  acids,  such  as  D-gluconic  acid  from  glucose,  while  oxidation 
of  the  primary  alcohol  group  yields  uronic  acids,  such  as  D-galacturonic 
acid  from  D-galactose.  Oxidation  of  both  the  aldehyde  and  primary 
alcohol  groups  yields  saccharic  acids.  The  uronic  acids  are  widely 
distributed  in  natural  polysaccharides  such  as  plant  gums  and  mucilages 
and  in  pectin.  The  fungi  in  nature  must  frequently  come  in  contact 
with  uronic  acids,  but  data  on  utilization  of  these  and  other  sugar  acids 
are  rare.     Steinberg  (1942)  cultured  Aspergillus  niger  on  media  which  con- 

Table  25.     Milligrams  op  Mycelium  Produced  by  FrvE  Fungi  Grown  upon 
Glucose,  Mannose,  and  Galactose  and  the  Corresponding  Sugar  Alcohols 
These  compounds  were  used  at  a  rate  which  supplied  8  g.  of  carbon  per  liter.    Each 
125-ml.  flask  contained  20  ml.  of  medium.     Cultures  were  incubated  at  25°C.     (Mar- 
golin, thesis,  West  Virginia  University,  1942.) 


Fungus 

D-Glu- 
cose 

Sor- 
bitol 

D- Man- 
nose 

Man- 
nitol 

D-Ga- 

lactose 

Galac- 
titol 

Blakeslea  trispora 

Mucor  ramannianus 

90 

89 
138 

79 
152 

12 
93 
59 
10 
13 

98 
115 
139 

81 
79 

9 
149 
108 

8 
9 

123 

116 

74 

17 

14 

10 
6 

Phycomyces  blakesleeanus 

Phytophthora  erythroseptica 

Pythiomorpha  gonapodyoides .  .  . 

6 

8 

10 

tained  1  g.  of  the  calcium  salts  of  the  following  sugar  acids  per  culture 
(the  weight  of  mycelium  in  milligrams  is  given  in  parentheses) :  2-keto-D- 
gluconic  (201),  5-keto-D-gluconic  (25),  D-gluconic  (32),  D-glucuronic  (206), 
and  mucic  (102).  Tamiya  (1932)  reports  that  A.  oryzae  utilizes 
D-gluconic  acid.  While  such  compounds  as  the  sugar  acids  are  little 
used  in  making  media,  they  are  of  interest  in  attempting  to  discover  the 
relation  between  structure  and  configuration  on  the  one  hand  and 
utilization  on  the  other. 

Mixed  carbon  sources.  In  nature  the  fungi  usually  come  in  contact 
with  mixed  carbon  sources  rather  than  a  single  source  of  carbon.  Certain 
fungi  make  more  growth  when  supplied  with  a  mixture  of  carbon  sources. 
This  increased  utilization  may  be  expected  only  if  one  or  both  carbon 
sources  are  poorly  utilized.  Horr  (1936)  investigated  the  growth  of 
Aspergillus  niger  upon  mixtures  of  glucose  and  galactose.  Some  of  these 
data  are  given  in  Table  26.  If  these  two  carbon  sources  were  utilized 
independently,  and  without  one  affecting  the  utilization  of  the  other,  the 
weight  of  mycelium  produced  on  the  combination  of  18  g.  of  galactose 
and  2  g.  of  glucose  should  be  42.4  -\-  145.6,  or  188  mg.  The  actual 
yield  was  577.4  mg.     The  experiment  indicates  that  A.  niger  is  able  to 


126  PHYSIOLOGY  OF  THE  FUNGI 

utilize  galactose  to  better  advantage  in  the  presence  of  glucose.  The 
experiments  of  Steinberg  (1939)  on  the  effect  of  two  poor  carbon  sources 
on  the  growth  of  A.  niger  were  made  at  35°C.  Some  combinations  of 
poor  carbon  sources  supported  more  growth  than  when  these  sources 
were  used  singly.  Thus,  the  calculated  weight  of  mycelium  for  the 
combination,  D-mannitol-lactose  was  21.4;  the  actual  yield  was  233.6  mg. 
Some  combinations  of  poor  carbon  sources  resulted  in  a  decrease  in 
amount  of  mycelium  formed  (glycerol-D-galactose :  calculated  yield, 
243.7  mg.;  actual  yield,  154.7  mg.).  The  effect  of  mixed  carbon  sources 
in  the  amount  of  growth  of  Phy corny ces  blakesleeanus  and  Pythiomorpha 
gonapodyoides  appeared  to  be  purely  additive  (Margolin,  1942). 

Table  26.     The  Effect  of  Galactose  and  Glucose,  Singly  and  in  Combination, 
UPON  the  Amount  of  Growth  of  Aspergillus  niger 
Cultures  incubated  7  days  at  20°C.      (Horr,  Plant  Physiol.  11,  1936.) 
Grams  of  Sugars  Used  per  Liter  Yield,  Mg.  per  Culture 

10  galactose 45 . 1 

18  galactose 42.4 

20  galactose 44.3 

2  glucose 145 . 6 

10  glucose 411 .0 

18  galactose  +  2  glucose 577.4 

10  galactose  +  10  glucose 1,151.6 

All  these  results  indicate  that  the  effect  of  mixed  carbon  sources  is 
highly  specific.  A  mixture  of  poor  carbon  sources  may  or  may  not 
result  in  increased  growth,  depending  on  the  carbon  sources  involved  as 
well  as  the  fungus  concerned. 

The  favorable  effects  of  mixtures  of  poor  carbon  sources  on  the  rate 
and  amount  of  growth  have  been  ascribed  to  the  ease  with  which  a  fungus 
is  able  to  synthesize  certain  key  intermediates.  If  the  synthesis  of 
intermediate  X  from  carbon  source  A  is  slow  and  difficult,  and  the  syn- 
thesis of  X  is  rapid  from  carbon  source  B,  it  is  probable  that  growth  will 
be  more  rapid  on  media  which  contain  both  carbon  sources. 

ORGANIC  ACIDS 

An  organic  acid  is  characterized  by  having  one  or  more  carboxyl 
( — COOH)  groups.  Some  organic  acids  are  utilized  as  sources  of  carbon 
and  in  other  ways.  Two  series  of  organic  acids  are  especially  interesting 
from  the  standpoint  of  physiology.  The  fatty  acids  are  monocarboxylic 
acids;  the  higher  members,  when  esterified  with  glycerol,  form  fats. 
The  dicarboxylic  acids,  especially  those  which  contain  four  carbon  atoms, 
enter  into  the  metabolic  pathways  of  the  fungi  in  various  ways;  e.g., 
utilization  of  ammonium  nitrogen  (Chap.  6). 


I 


CARBON  SOURCES  127 

The  form  in  which  an  organic  acid  exists  (free  acid  or  salt)  is  a  function 
of  the  pH  of  the  medium  or  cells.  The  free  acid  is  the  predominant  form 
at  low  pH  values.  The  terms  for  an  acid  and  its  salt  {e.g.,  fumaric  acid, 
fumarate)  are  used  in  the  literature  somewhat  loosely.  The  effect  of  a 
free  acid  and  its  anion  may  be  different  (Chap.  8). 

Leptomitus  ladeus,  w^hich  does  not  utilize  sugars,  grows  on  various 
fatty  acids — acetic,  butyric  to  capric — but  not  on  formic  or  propionic 
acids  (Schade,  1940).  Apodachlya  hrachynema  utilizes  the  same  fatty 
acids  as  L.  ladeus and  also,  fumarate,  succinate  and  malate.  Aspergillus 
niger,  according  to  Steinberg  (1942),  makes  some  growth  on  acetate, 
lactate,  tartrate,  malate,  and  fumarate.  Growth  was  very  poor  com- 
pared with  that  on  sucrose.  Dulaney  (1949)  reported  that  little  strepto- 
mycin was  produced  when  organic  acids  were  used  by  Streptomyces 
griseus.  Yeasts  use  acetate  to  synthesize  fat  (White  and  Werkman, 
1947).  Tamiya  (1932)  investigated  the  utilization  of  many  organic  acids 
by  Aspergillus  oryzae.  Growth  w^as  poor  on  most  of  these  compounds 
except  quinic  acid.  While  an  organic  acid  may  serve  as  the  sole  source 
of  carbon  for  fungi,  in  general  acids  do  not  allow  as  much  or  as  rapid 
growth  as  carbohydrates. 

An  amino  acid  may  serve  as  a  source  of  both  nitrogen  and  carbon. 
Peptone  may  serve  as  a  source  of  carbon  and  nitrogen  for  many  fungi. 
Aspergillus  niger,  when  grown  on  peptone  as  the  sole  source  of  carbon, 
deaminates  the  peptides  and  amino  acids  and  releases  ammonia  in 
quantities  greater  than  the  fungus  can  use.  The  utilization  of  amino 
acids  as  carbon  sources  by  A.  niger  w^as  investigated  by  Steinberg  (1942a), 
who  found  certain  combinations  of  "primary"  amino  acids  to  be  utilized 
about  three-fourths  as  efficiently  as  sucrose. 

The  utilization  of  individual  amino  acids  by  Penicillium  roqueforti  and 
Fusarium  oxysporum  var.  lycopersici  was  studied  by  Gottlieb  (1946). 
Not  all  the  naturally  occurring  amino  acids  were  utilized  as  carbon  sources 
by  these  fungi.  The  six-carbon  straight-chain  amino  acids  norleucine  and 
lysine  and  the  sulfur-containing  amino  acids  cysteine  and  methionine 
were  not  utilized  as  carbon  sources.  Glycine  and  valine  were  poor 
carbon  sources  for  P.  roqueforti,  while  F.  oxysporum  var.  lycopersici  grew 
well  on  these  amino  acids.  Alternaria  solani,  Helminthosporium  sativum, 
Rhizoctonia  solani,  Fusarium  moniliforme,  Chaetomium  globosum,  and 
Aspergillus  niger  were  unable  to  utilize  the  naturally  occurring  sulfur- 
containing  amino  acids  as  a  source  of  carbon. 

Yeasts  differ  in  ability  to  utilize  different  amino  acids  as  the  sole 
source  of  carbon  (Schultz  et  al.,  1949).  Glutamic  acid  and  proline  were 
available  to  more  species  than  other  amino  acids.  It  is  characteristic  of 
fungi  cultivated  on  amino-acid  media  as  the  sole  source  of  carbon  that 
the  medium  becomes  alkaline.     This  is  probably  due  to  accumulation  of 


128  PHYSIOLOGY  OF  THE  FUNGI 

ammonia  which  results  from  deamination.     In  general,  the  amino  acids 
appear  to  be  poor  sources  of  carbon. 

GLYCOSIDES 

The  carbon  sources  to  be  discussed  in  this  and  the  next  two  sections 
differ  from  those  previously  considered  in  that  they  undergo  hydrolysis. 
The  complex  carbohydrates  and  carbohydrate-like  compounds  yield 
simple  sugars  w^hen  hydrolyzed.  In  some  instances,  other  compounds 
are  also  formed.  In  most  instances,  fungi  utilize  these  compounds  only 
after  hydrolysis.  Therefore,  utilization  will  be  dependent  upon  the  pro- 
duction of  the  necessary  hydrolytic  enzymes.  If  a  fungus  is  unable  to 
perform  this  preliminary  "digestion,"  such  complex  carbohydrates  will 
be  unavailable. 

Many  of  the  compounds  to  be  considered  in  this  section  are  isomers. 
The  simple  sugars  exist  mainly  in  the  form  of  ring  structures,  rather  than 
the  open-chain  forms  which  were  depicted  in  the  previous  sections  of  this 
chapter.  The  chemical  evidence  may  be  reviewed  in  Pigman  and 
Goepp  (1948)  or  other  text  dealing  with  the  sugars.  Glucose  exists  in 
aqueous  solution  as  an  equilibrium  mixture  of  a-D-glucose  and  /3-D-glu- 
cose.  These  formulas  contain  a  six-membered  ring  of  w^hich  one  atom  is 
oxygen  (pyranose).  Some  sugars,  however,  contain  a  five-membered 
ring  (furanose). 

The  formulas  for  these  two  forms  of  glucose  are  given  below: 

H  OH  HO  H 

\   /  \    / 

c ,  c- 


H— C— OH 
HO— C— H 
H— C— OH 
H— C— O— 


H— C— OH 

I 
HO— C— H 

I 
H— C— OH 

I 
H— C— O— 


CH2OH  CH2OH 

a-D-Glucose  /3-D-Glucose 

The  simple  glycosides  are  a  widely  distributed  group  of  naturally 
occurring  compounds  which  contain  a  sugar  moiety  and  an  alcohol  or 
phenol  moiety.  The  form  glucoside  was  formerly  used  to  designate  com- 
pounds of  this  type  irrespective  of  the  sugar  moiety.  Specific  glycosides 
are  designated  by  adding  the  ending  oside  to  the  name  of  the  sugar 
involved;  e.g.,  glucoside,  mannoside,  etc. 

Two  glucosides  are  formed  when  glucose  is  treated  with  methanol 
under  appropriate  conditions.     The  formulas  are  given  below: 


CARBON  SOURCES  129 

H           OCH3  H3CO  H 

C ,  C- 


H— C— OH 

I 
HO— C— H 

I 
H— C— OH 

I 
H— C— O— 


H— C— OH 
HO— C— H 

H— C— OH 

I 
H— C— O— 


CH,OH  CH2OH 

a-Methyl-D-glucoside  /3-Methyl-D-glucoside 

These  formulas  correspond  to  the  alpha  and  beta  isomers  of  glucose. 
The  proof  that  a-methylglucoside  and  a-glucose  have  the  same  structure 
was  furnished  by  Armstrong  (1903),  who  followed  the  enzymatic 
hydi'olyses  of  these  glucosides  polarimetrically. 

Our  interest  in  the  glycosides  is  not  in  the  chemical  structure  per  se, 
but  in  the  fact  that  utilization  of  these  and  other  compounds  having  the 
same  type  of  glycoside  linkage  is  dependent  upon  configuration.  Differ- 
ent enzymes  are  required  for  the  hydrolysis  of  the  a-  and  /3-glycoside 
linkages.  Some  fungi  possess  both  types  of  hydrolytic  enzymes,  others 
but  one,  and  some  fungi  appear  to  lack  both.  Thus,  certain  yeasts 
ferment  a-methylglucoside  but  not  /3-methylglucoside.  These  yeasts 
have  an  enzyme  or  enzymes  which  catalyze  the  hydrolysis  of  the  a-gly- 
coside  linkage  but  not  the  /3-glycoside  linkage  (lactose-fermenting  yeasts 
are  able  to  hydrolyze  ;S-glycosides). 

The  use  of  the  methylglucosides  is  not  always  a  safe  guide  in  pre- 
dicting which  complex  sugars  will  be  utilized  by  fungi.  Aspergillus  niger 
utilizes  /3-methylglucoside  rapidly  and  completely,  while  a-methyl- 
glucoside  is  poorly  utilized  (Dox  and  Neidig,  1912).  Attempts  to  adapt 
A.  niger  to  utilize  a-methylglucoside  as  a  sole  source  of  carbon  were 
without  much  success,  although  the  fungus  apparently  utilized  this 
compound  in  the  presence  of  sucrose  (Dox  and  Roark,  1920).  A.  niger 
utilizes  lactose  poorly.  Tamiya  (1932)  found  A.  oryzae  made  only  a 
trace  of  growth  on  a-methylglucoside.  /3-Methylglucoside  was  not 
tested.  This  fungus  grows  well  on  maltose.  These  results  are,  perhaps, 
not  unexpected  in  view  of  the  specificity  of  enzymes.  The  utilization  of 
the  naturally  occurring  simple  glycosides  by  fungi  has  been  investigated 
but  slightly. 

OLIGOSACCHARIDES 

These  sugars  are  derived  from  two,  three,  or  four  hexose  sugars  by  the 
elimination  of  water.  On  hydrolysis,  the  individual  sugars  are  regener- 
ated. Five  factors  which  determine  the  structure  of  the  oligosaccharides 
are  (1)  the  component  sugars;  (2)  the  component  sugar  which  functions 


130 


PHYSIOLOGY  OF  THE  FUNGI 


as  the  alcohol;  (3)  the  stereochemical  nature  of  the  glycoside  linkage;  (4) 
the  carbon  of  the  alcohol  moiety  which  forms  the  glycoside  linkage;  and 
(5)  the  ring  structure  of  the  component  sugars  (see  Oilman,  1943), 

Maltose.  It  is  doubtful  whether  this  disaccharide  occurs  free  in 
nature.  It  is  formed  when  starch  is  enzymatically  hydrolyzed;  on 
further  hydrolysis  two  molecules  of  glucose  are  formed.  This  disac- 
charide is  utilized  by  many  fungi.  The  glycoside  linkage  is  alpha  in 
maltose. 


H         OH 

C 


H-C-OH 

I 
HO-C-H 

I 
H-C-0 

I 
H-C-0 

I 

CH2OH 


Maltose 

Cellobiose.  The  occurrence  of  this  sugar  as  a  repeating  unit  in  cellu- 
lose makes  it  important.  Cellobiose  differs  from  maltose  only  in  the 
nature  of  the  glycoside  linkage.  With  few  exceptions  only  fungi  which 
produce  enzymes  which  attack  the  /3-glycoside  linkage  will  utilize  this 
sugar. 


HO-C-H 


CH2OH 


CH2OH 


Cellobiose 


Since  cellobiose  and  maltose  differ  only  in  the  nature  of  the  glycoside 
linkage,  it  would  be  interesting  to  compare  the  utilization  of  these  two 
sugars  by  a  large  number  of  fungi.  Cellobiose  has  been  studied  so  infre- 
quently that  the  necessary  data  are  lacking. 


CARBON  SOURCES 


131 


Lactose.  This  sugar  is  probably  present  in  the  milk  of  all  animals. 
Hydrolysis  of  lactose  by  acids  or  lactase  yields  a  molecule  each  of  glucose 
and  galactose.  This  sugar  is  hydrolyzed  by  emulsin  and  is  therefore  a 
/3-glycoside. 


CH2OH 


Lactose 

Sucrose.  This  sugar  is  of  common  occurrence  in  plants.  On  hydroly- 
sis one  molecule  of  glucose  and  one  of  fructose  are  formed;  a  mixed  a-  and 
j8-glycoside  linkage  unites  the  sugar  moieties.  Sucrose  apparently  is 
utilized  by  fewer  fungi  than  maltose,  but  more  extensively  than  lactose 
(see  Table  22). 


CH2OH 


CH2OH 


Sucrose 

In  addition  to  the  three  common  disaccharides  (maltose,  lactose, 
and  sucrose),  many  other  oligosaccharides  are  known.  Owing  to  cost 
and  relative  unavailability,  these  sugars  have  not  been  studied  inten- 
sively. Some  of  these  "rare"  sugars  are  used  in  differential  media  in 
bacteriology.  Brief  mention  will  be  made  here  of  some  of  these  sugars. 
The  nonreducing  disaccharide  trehalose  (mushroom  sugar)  is  syn- 
thesized by  various  fungi  and  is  fermented  by  many  yeasts.  Trehalose 
on  hydrolysis  yields  glucose ;  it  differs  from  maltose  in  the  position  of  the 


132 


PHYSIOLOGY  OF  THE  FUNGI 


glycoside  linkage.  Tamiya  (1932)  reported  that  Aspergillus  oryzae 
utilized  trehalose  and  raffinose.  The  trisaccharide  raffinose  is  obtained 
as  a  by-product  of  beet-sugar  manufacture.  On  complete  hydrolysis 
galactose,  glucose,  and  fructose  are  formed  in  equivalent  amounts.  The 
structure  for  raffinose  is  given  below. 

Volkonsky  (1934)  found  raffinose  to  be  utilized  readily  by  Pythmm 
deharyanum  and  a  species  of  Sporotrichum.  One  isolate  of  Phytophthora 
parasitica  utilized  raffinose  rapidly,  while  another  isolate  utilized  this 
sugar  slowly.  Phytophthora  cactorum  and  P.  palmivora  utilized  this 
sugar  slowly.  The  great  majority  of  fungi  tested  by  Volkonsky  did  not 
utilize  raffinose. 


CH2OH 
O— C 


^ 


HO— C— H 

H— C— OH 

I 
H— C— O— 


H- 

H— C— OH 

HO— C— H 

I 
H— C— OH 

H— C— O— 


CHoOH 

D-Fructose 


CH2O- 

D-Glucose 


H— C 


H— C— OH 

I 
HO— C— H 

I 
HO— C— H 

H— C— O— 


CH2OH 

u-Galactose 


Raffinose 


Oligosaccharides  and  polysaccharides  are  utilized  by  fewer  fungi  than 
is  glucose.  All  microorganisms  which  can  utilize  a  given  polysaccharide 
are  also  able  to  utilize  its  hydrolytic  products  (Van  Niel,  1944).  Not  all 
polysaccharides  yield  glucose  on  hydrolysis,  but  the  majority  of  them  do. 

While  the  evidence  at  hand  does  not  exclude  the  direct  utilization 
of  disaccharides  by  some  fungi,  it  is  probable  that  these  sugars  are 
hydrolyzed  before  utilization  in  most  instances.  Smith  (1949)  suggests 
that  Marasmius  chordalis  attacks  cellobiose  by  a  route  that  involves 
neither  preliminary  hydrolysis  nor  phosphorylation. 

The  failure  of  a  fungus  to  utilize  an  oligosaccharide  may  be  due  either  to 
the  lack  of  the  necessary  hydrolytic  enzyme  or  to  inability  to  utilize  the 
component  sugars.  Failure  to  synthesize  the  necessary  hydrolytic 
enzymes  appears  to  be  by  far  the  most  common  cause  of  nonutilization. 
This  is  borne  out  by  the  data  in  Table  22.  Of  the  21  fungi  studied  by 
Margolin,  two  failed  to  grow  on  maltose,  while  eight  did  not  utilize 
sucrose.  Since  all  these  fungi  grew  well  on  glucose  and  fructose,  it  is 
evident  that  failure  to  utilize  maltose  and  sucrose  was  due  to  the  fact  that 
these  fungi  could  not  hydrolyze  these  sugars.  The  nonutilization  of 
lactose  by  Syncephala strum  racemosum  is  evidently  due  to  the  failure  of 


CARBON  SOURCES  133 

this  fungus  to  synthesize  lactase,  for  this  fungus  makes  good  growth  on 
either  glucose  or  galactose.  The  same  argument  applies  to  Blakeslea 
trispora,  Fusarium  lycopersici,  Rhizopus  nigricans,  and  R.  suinus.  Non- 
utilization  of  a  complex  carbohydrate  is  usually  due  to  the  lack  of  the 
necessary  hydrolytic  enzymes. 

The  hydrolysis  of  oligosaccharides  by  fungi  is  easily  demonstrated. 
Phycomyccs  blakesleeanus  utilizes  sucrose  while  Mucor  ramannianus  does 
not.  If  the  mycelium  of  P.  blakesleeanus  is  removed  from  a  flask  of 
sucrose  medium  after  several  days'  incubation  and  the  flask  reinoculated 
with  M.  ramannianus,  the  latter  fungus  will  grow.  P.  blakesleeanus 
excretes  sucrase,  which  catalyses  the  hydrolysis  of  sucrose  to  D-glucose 
and  D-fructose,  both  of  which  are  utilized  by  M.  ramannianus. 

A  complex  carbohydrate  and  its  hydrolytic  products  are  not  necessarily 
equivalent  in  all  respects.  Hawker  (1947)  reported  that  the  amount  of 
mycelium  produced  by  Melanospora  destruens  was  different  when  this 
fungus  was  grown  on  equivalent  amounts  of  glucose,  fructose,  mixtures  of 
glucose  and  fructose,  and  sucrose.  More  mycelium  was  produced  from 
glucose  than  from  an  equivalent  amount  of  sucrose,  and  this  was  true 
whether  the  concentrations  of  these  sugars  were  low  or  high.  On  the 
other  hand,  perithecia  were  produced  more  abundantly  on  sucrose  than 
on  glucose  media.  Indeed,  hydrolysis  of  the  same  lot  of  sucrose  to 
glucose  and  fructose  allowed  the  production  of  no  more  perithecia  than 
other  samples  of  these  sugars.  The  conclusion  seems  inescapable  that 
the  particular  structure  of  sucrose  was  in  some  way  favorable  for  the  pro- 
duction of  perithecia.  While  a  fungus  may  utilize  an  oligosaccharide  and 
its  hydrolytic  products,  it  is  not  safe  to  assume  that  both  are  used  with 
the  same  efficiency  for  all  purposes. 

POLYSACCHARIDES 

The  chemistry  of  the  polysaccharides  resembles  that  of  the  oligosac- 
charides except  that  the  number  of  sugar  residues  is  much  larger.  These 
substances  constitute  the  bulk  of  carbohydrate  materials  synthesized  by 
plants  and  animals.  The  most  important  polysaccharides  are  cellulose, 
starch,  and  glycogen.  On  hydrolysis  simple  sugars  are  formed.  The 
molecular  weights  of  polysaccharides  may  be  very  large;  cellulose  from 
different  sources  may  have  a  molecular  weight  ranging  from  200,000  to 
400,000.  The  molecular  weights  of  many  polysaccharides  are  much  less 
than  that  of  cellulose.  In  general,  polysaccharides  are  insoluble  or  only 
colloidally  soluble.  The  utilization  of  these  substances  by  fungi  is 
dependent  upon  the  excretion  of  the  necessary  hydrolytic  enzymes.  Pig- 
man  and  Goepp  (1948)  classify  polysaccharides  on  the  basis  of  the 
hydrolytic  products  as  homopolysaccharides,  which  yield  only  one  mono- 
saccharide on  hydrolysis,  and  heteropolysaccharides,  which  yield  two  or 


134  PHYSIOLOGY  OF  THE  FUNGI 

more  monosaccharides  or  related  compounds  on  hj^drolysis.  Cellulose, 
starch,  and  glycogen  are  members  of  the  first  class  and  yield  glucose  on 
hydrolysis.  Polysaccharides  are  frequently  named  b}^  replacing  the  end- 
ing -ose  of  the  parent  monosaccharide  by  -an.  Fructan  (le\ailan)  desig- 
nates a  polysaccharide  which  yields  fructose  on  hydrolysis.  A  hexosan 
is  a  polysaccharide  which  yields  hexose  sugars  on  hydrolysis,  and  a  pen- 
tosan yields  pentoses.     Pectins  are  polymers  of  galacturonic  acid. 

The  heteropolysaccharides  occur  in  lesser  amounts  than  the  homopoly- 
saccharides.  Among  them  are  the  hemicelluloses,  which  on  hydrolysis 
yield  D-xylose  as  the  principal  sugar,  the  plant  gums,  and  agar. 

Cellulose.  Chemically,  cellulose  is  a  linear  polymer  of  o-glucose.  The 
glucose  residues  are  joined  together  through  /3-glycoside  linkages  as  in 
cellobiose,  and  cellulose  may  be  thought  of  as  consisting  of  repeating 
cellobiose  units.  Norman  and  Fuller  (1942)  postulate  that  the  majority 
of  fungi  are  able  to  utilize  cellulose.  In  spite  of  the  importance  of  cellu- 
lose utilization  by  fungi  in  the  economy  of  nature  much  remains  to  be 
learned  about  this  process. 

It  is  commonly  accepted  that  the  first  stage  in  utilization  of  cellulose  is 
hydrolysis,  although  Campbell  (1932)  has  suggested  oxidation.  The 
hydrolysis  of  cellulose  may  be  expressed  schematically  as  follows:  cellu- 
lose — >  cellodextrins  -^  cellotetrose  -^  cellobiose  -^  D-glucose.  Fungus 
cellulases  appear  to  have  been  infrequently  studied.  Grassmann  et  al. 
(1933)  separated  cellulase  and  cellobiase  from  Aspergillus  oryzae.  This 
cellulase  was  inactive  in  hydrolyzing  cellulose  degradation  products  hav- 
ing a  molecular  weight  less  than  1,000  (six  glucose  residues),  while  the 
cellobiase  hydrolyzed  cellulose  fragments  containing  from  two  to  six 
glucose  residues. 

Fungi  differ  widely  in  ability  to  utilize  cellulose.  In  general,  the  rate  of 
utilization  of  cellulose  is  less  than  that  of  glucose.  This  is  probabty  due 
to  the  insolubility  of  cellulose,  which  limits  the  action  of  cellulase  to  the 
surface,  or  to  an  inadequate  rate  of  enzyme  synthesis. 

The  principal  source  of  cellulose  available  to  fungi  in  nature  is  wood  and 
other  plant  remains,  ^^^lile  cellulose  is  the  chief  constituent  in  such 
materials,  hemicelluloses,  gums,  tannins,  and  lignin  are  also  present. 
The  wood-rotting  fungi  have  been  classified  according  to  whether  they 
cause  white  or  brown  rots.  The  fungi  which  cause  brown  rots  attack 
cellulose  in  preference  to  lignin.  The  fungi  which  preferentially  attack 
the  noncellulosic  constituents  of  wood  cause  white  rots.  The  latter 
species  are  apparently  more  numerous  than  those  which  cause  brown  rots. 
The  following  are  some  of  the  fungi  listed  by  Nobles  (1948)  as  causing 
white  rots:  Armillaria  mellea,  Ganoderma  lobaturn,  Lenzites  hetulinus, 
Pleurotus  ostreatus,  Polyporus  ahietinus,  P.  cinnabarinus,  P.  pargamenus. 
A  few  fungi  causing  bro^^^l  rots  are  Daedalea  quercina,  Lentinus  lepideus, 


CAR BOX  SOURCES 


135 


Lenzites  trabea,  Merulius  lacrymans,  Polyporus  betulinus,  and  Trametes 
americana. 

The  effect  of  a  typical  fungus  causing  white  rot  on  the  composition  of 
wood  is  given  in  Table  27.  Polyporus  pargamenus  was  allowed  to  act  on 
blocks  of  aspen  wood  for  20  months.  At  the  end  of  this  time  the  wood 
block  showed  three  degrees  of  attack.  The  tan-colored  portion  was 
altered  least.  The  pink-colored  portion  was  intermediate,  while  the 
white  portion  had  lost  the  most  lignin.  P.  pargamenus  also  degraded  the 
cellulose  somewhat,  as  shown  by  lower  degree  of  polymerization. 

T.\BLE  27.     The  Effect  of  Polyporus  pargamenus  ix  Altering  the  Composition  of 

Aspen  Wood 
Time  of  incubation  20  months.      (Heuser  et  al.,  Arch.  Biochem.  21, 1949.     Published 
by  permission  of  Academic  Press,  Inc.) 


Portion  of 
wood  block 

Lignin,  % 

Pentosans,  % 

CeUulose,  % 
(calculated) 

Original 

Tan 

Pink 

\Miite 

17.5 

10.4 

4.5 

3.4 

19.3 

12.8 

8.3 

8.4 

GO.  68 
73.84 
84.20 
85.32 

The  effect  of  fungi  causing  brown  rots  on  the  composition  of  coniferous 
woods  has  been  studied  by  Schubert  and  Xord  (1950).  Lenzites  saepiaria 
in  13  months  caused  a  decrease  in  cellulose  in  pine  sawdust  from  45.5  to 
18.5  per  cent.  During  this  period  the  apparent  lignin  content  increased 
from  33.9  to  50.1  per  cent.  Similar  results  were  obtained  with  Lentinus 
lepideus  and  Poria  vaillantii.  For  a  recent  review  of  the  microbiological 
degradation  of  cellulose  see  Nord  and  Vitucci  (1948). 

Starch.  Like  cellulose,  starch  is  a  polymer  of  D-glucose.  The  glucose 
residues  are  j  oined  through  a-glycoside  linkages,  and  starch  (and  glycogen) 
may  be  thought  of  as  consisting  of  repeating  units  of  maltose.  Starch 
consists  of  two  tj-pes  of  molecules.  The  linear  portion  of  starch  is  called 
amylose,  while  the  branched-chain  fraction  is  known  as  amylopectin. 
Starch  is  sjTithesized  by  green  plants,  while  glycogen  is  formed  by 
animals  and  fungi.  The  enzymes  which  catatyze  the  hydrolysis  of  starch 
are  known  as  amylases  and  were  discussed  in  Chap.  4.  The  enzymatic 
hj'drolysis  of  starch  may  be  represented  schematically  as  follows :  starch  -^ 
dextrins— ^  maltose  -^  D-glucose.  The  branched-chain  dextrins  are  incom- 
pletely hydrolyzed  by  amylase,  while  the  straight-chain  dextrins  are 
completely  converted  to  maltose  (]\Iyrback,  1948). 

Starch  is  insoluble  in  water.  Only  those  furgi  which  produce  amylase 
are  able  to  utilize  starch.  This  ability  is  common  among  fungi  but  not 
universal.     Volkonsky    (1934)    found    26    isolates    and    species    of    the 


136  PHYSIOLOGY  OF  THE  FUNGI 

Saproliginales  to  utilize  starch  and  its  hydrolytic  products  (dextrin, 
maltose,  and  glucose).  Thirteen  other  carbon  sources,  including  fructose, 
were  not  utilized.  Margolin  (1942)  found  that  19  out  of  21  fungi  which 
utilized  maltose  also  utilized  dextrin. 

The  nonutilization  of  starch  by  Sclerotinia  libertiana  has  been  suggested 
as  the  basis  of  a  method  of  preparing  potato  starch  (Kakeura,  1946). 
Few  yeasts  utilize  starch,  although  maltose  and  glucose  are  readily 
utilized. 

All  the  fungi  listed  in  Table  22  except  Pythiwn  ascophallon  and  Phy- 
tophthora  jagopijri  utilized  dextrin.  A  comparison  of  the  ability  of  fungi 
to  utilize  glycogen  and  starch  has  not  been  investigated  thoroughly. 
Tamiya  (1932)  found  the  yield  of  mycelium  of  Aspergillus  oryzae  to  be 
greater  on  glycogen  than  on  dextrin.  Dextrin  was  a  better  carbon  source 
than  starch. 

The  role  of  the  pectin-destroying  enzymes  in  parasitism  and  the  rotting 
of  fruits  and  vegetables  is  discussed  in  Chap.  17.  Presumably  these  fungi 
utilize  some  or  all  of  the  hydrolytic  products  of  pectin  (n-galacturonic  acid 
and  methyl  alcohol).  None  of  the  fungi,  in  so  far  as  is  known,  utilize  agar 
as  a  source  of  carbon.  A .  niger  utilizes  the  arabo-galactan  from  western 
larch  as  a  source  of  carbon  (Ratajak  and  Owens,  1942). 

HETEROTROPHIC  UTILIZATION  OF  CARBON  DIOXIDE 

The  assimilation  of  carbon  dioxide  is  not  restricted  to  green  plants. 
Carbon  dioxide  fixation  has  been  demonstrated  in  bacteria,  fungi, 
protozoa,  liver  slices,  barley  roots,  and  intact  green  plants  in  the  absence 
of  light.  The  basis  for  classifying  organisms  according  to  the  way  they 
utilize  carbon  dioxide  is  discussed  by  Werkman  and  Wood  (1942),  By 
the  use  of  carbon  isotopes  an  elegant  method  is  available  for  demonstrating 
carbon  dioxide  assimilation.  In  addition,  the  mechanism  of  fixation  can 
be  studied.  This  involves  isolation  and  degradation  studies  of  the  com- 
pounds synthesized  Avhile  the  organisms  were  exposed  to  isotopic  carbon 
dioxide.  Either  stable  or  radioactive  carbon  isotopes  may  be  used.  The 
finding  of  isotopic  carbon  in  compounds  synthesized  is  proof  of  assimilation. 

Aspergillus  niger  and  Rhizopus  nigricans  were  shown  to  assimilate  car- 
bon dioxide  (Foster  et  al.,  1941).  Radioactive  carbon  dioxide  (C^i02) 
was  used  in  these  experiments.  Mycelium  of  R.  nigricans  was  suspended 
in  5  per  cent  glucose  solution  and  agitated  in  a  closed  system  containing 
isotopic  carbon  dioxide.  At  the  end  of  the  experiment  the  mycelium  and 
the  medium  were  analyzed  for  radioactivity.  More  than  one-third  of 
the  carbon  dioxide  assimilated  was  incorporated  into  cell  constituents 
which  were  not  decomposed  by  boiling  for  1  hr.  with  2M  hydrochloric 
acid.  Carbon  dioxide  was  assimilated  under  aerobic  and  anaerobic 
conditions.     The  data  of  such  an  experiment  are  given  in  Table  28. 


CARBON  SOURCES 


137 


Table  28.     Distribution  of  Radioactive  Carbon  (C^O  in  the  Culture  Medium 
AND  Mycelium  of  Rhizopus  nigricans  Exposed  to  C^'02  in  the  Gas  Phase  for 

30  Minutes 
Results  are  expressed  as  percentage  of  C^Oo  assimilated.      (Foster  et  ah,  Proc. 

Natl.  Acad.  Sci.,  U.S.  27,  1941.) 


Substance  tested 


Total  C*  in  supernatant  solution  after  removing  cells. 

Fumaric  *  acid  in  this  solution 

C*  in  neutral  volatile  distillate 

Total  C*  in  acid  extract  of  cells 

Fumaric  *  acid  in  this  solution 

C*  remaining  in  cells  after  acid  extraction 


Aerobic 

19 

5 

8 

0 

0 

1 

44 

0 

6 

5 

5 

Anaerobic 


29.0 
25.0 
0.2 
30.0 
12.0 
41.0 


*  Designates  radioactive  carbon. 

It  is  probable  that  carbon  dioxide  enters  into  various  metabolic  proc- 
esses. Foster  and  Davis  (1948)  postulate  that  strains  of  Rhizopus 
nigricans  which  produce  fumaric  acid  anaerobically  do  so  according  to 
scheme  IV.     Cantino  (1949),  in  studying  the  metabolism  of  Blastocladia 

Scheme  IV.     A  Scheme  for  the  Anaerobic  Transformation  of  Glucose  into 

Fumaric  Acid  by  Rhizopus  nigricans* 

nC6Hi206 


CH3CHOHCOOH 
Lactic  acid 


I 


i 


+2H  yf-nAU 

nCHsCOCOOH 

Pyruvic  acid 


+  CO2 


HOOCCH2COCOOH 
Oxalacetic  acid 

>- 


■^  CO2+  CH3CHO 
Acetaldehyde 


+  2H 


CH3CH2OH 

Ethyl  alcohol 


+  2H 


HOOCCHgCHOHCOOH 
Malic  acid 


I 


*  Courtesy  of  Foster  and  Davis,  Jour.  Bad.,  56  :  335,  1948, 
&  Wilkins  Company. 


-HOH 

HOOCCH-.CHCOOH 
Fumaric  acid 

Published  by  permission  of  The  Williams 


pringsheimii,  found  that,  by  increasing  the  carbon  dioxide  in  the  gaseous 
phase,  the  formation  of  lactic  acid  was  decreased,  while  the  amount  of 


138 


PHYSIOLOGY  OF  THE  FUNGI 


succinic  acid  was  increased.     It  was  surmised  that  this  fungus  utilizes 
carbon  dioxide,  since  none  was  set  free. 

The  formation  of  oxalacetic  acid  by  the  reaction  between  pyruvic  acid 
and  carbon  dioxide  suggests  that  heterotrophic  carbon  dioxide  fixation 
may  play  a  role  in  amino-acid  synthesis.  Support  of  this  hypothesis  may 
be  found  in  the  work  of  Ajl  and  Werkman  (1949),  who  found  the  carbon 
dioxide  requirement  of  Aerohacter  aerogenes  could  be  replaced  by  oxal- 
acetic, a-ketoglutaric,  fumaric,  or  aspartic  acid.  For  further  information 
on  carbon  dioxide  utilization  by  fungi  see  Foster  (1949). 

UTILIZATION  OF  CARBON 

Carbon  compounds  are  utilized  by  fungi  for  two  general  purposes,  as  a 
source  of  energy  and  as  a  source  of  the  chief  structural  element.  These 
two  processes  may  be  the  same  until  a  number  of  chemical  transforma- 
tions have  taken  place  but  may  then  diverge  after  certain  intermediate 
compounds  are  formed.  The  over-all  use  of  carbon  is  quite  easily 
determined,  but  it  is  a  problem  of  a  different  order  to  trace  all  the  chemical 
transformations  which  occur  when  a  compound  is  utilized. 

Table  29.     The  Distribution  of  Carbon  from  Arabinose  among  the  Products 

OF  Metabolism  of  Fusarium  lini 
(White  and   Willaman,   Biochem.  Jour.   22,   1928.     Published  by  permission  of 
Cambridge  University  Press.) 


Age  of 

Mycelium, 

CO2, 

Alcohol, 

Lead  pre- 

Sugar, 

Total 

culture,  days 

% 

% 

% 

cipitate,  % 

% 

carbon,  % 

5 

0.8 

0.6 

7.6 



90.6 

99.6 

10 

3.4 

4.4 

7.6 

0.6 

85.2 

101.2 

15 

4.6 

6.1 

6.6 

1.0 

80.4 

98.7 

25 

4.0 

9.4 

3.3 

1.5 

81.2 

99.4 

40 

10.4 

20.8 

9.9 

1.7 

55.2 

98.3 

Carbon  balances.  A  general  idea  of  the  way  a  carbon  source  is  utilized 
may  be  gained  by  following  the  amounts  of  mycelium  synthesized,  carbon 
dioxide  evolved,  and  other  metabolic  products  formed.  If  the  initial 
amount  of  carbon  is  known,  its  distribution  can  be  followed  by  analysis. 
From  95  to  99  per  cent  of  the  carbon  is  usually  accounted  for  in  such 
experiments.  The  accompanying  data  from  White  and  Willaman  (1928) 
illustrate  this  distribution  of  carbon  from  arabinose  by  Fusarium  lini 
(Table  29). 

While  the  analytical  difficulties  in  experiments  of  this  kind  are  con- 
siderable, chemical  analysis  of  the  mycelium  and  the  other  metabolic 
products  reveals  how  the  carbon  originally  present  in  the  carbon  source  is 
distributed.     Such  analyses  are  useful  in  detecting  the  major  metabolic 


CARBON  SOURCES  139 

products.  Carbon  balances  are  especiall}^  useful  in  determining  the  effi- 
ciency with  which  a  fungus  produces  metabolic  products  of  value,  such  as 
alcohol  and  citric  acid.     For  further  examples  see  Raistrick  ct  al.  (1931). 

Utilization  ratios.  The  relations  of  the  amounts  of  fungus  metabolic 
products  to  the  amount  of  carbon  soiu'ce  (or  other  substance)  used  are 
frequently  expressed  as  ratios.  However,  these  ratios  are  valid  only  for 
the  fungi  and  the  experimental  conditions  used.  These  ratios  should  be 
considered  as  absolute  values  only  for  the  conditions  under  which  they 
were  obtained.  The  various  utilization  ratios  are  of  less  value  than 
complete  carbon  balances,  but  the  analytical  determinations  are  fewer. 
To  be  of  most  value,  these  ratios  should  be  determined  at  various  intervals 
during  incubation,  because  these  ratios  change  with  age. 

The  most  useful  of  these  ratios  is  the  economic  coefficient,  which  is 
obtained  by  dividing  the  weight  of  mycelium  and  spores  by  the  weight  of 
sugar  or  other  carbon  source  used.  The  residual  carbon  source  in  the 
medium  must  be  determined  at  the  end  of  an  experiment.  In  general  an 
efficient  fungus  will  convert  half  the  weight  of  sugar  supplied  in  the 
medium  into  cellular  material.  The  efficiency  of  most  fungi  when  grown 
on  laboratory  media  is  much  less.  This  is  due  in  part  to  the  use  of 
unbalanced  media  and  to  the  type  of  carbon  metabolism  taking  place. 
The  carbon  which  is  not  utilized  for  the  synthesis  of  cellular  material 
appears  either  as  carbon  dioxide  or  as  intermediate  metabolic  products, 
such  as  alcohol  and  organic  acids.  In  industrial  applications  it  is  desir- 
able to  employ  cultural  conditions  which  divert  a  large  part  of  the  carbon 
used  into  the  desired  intermediate  products,  rather  than  into  the  produc- 
tion of  mycelium  and  carbon  dioxide. 

The  economic  coefficient  of  Fusarium  sambucinum  under  various 
cultural  conditions  has  been  studied  by  Holzapfel  (1925).  This  fungus 
utilized  sucrose  (0.33)  and  fructose  (0.36)  more  efficiently  than  glucose 
(0.24).  The  economic  coefficient  varied  with  the  concentration  of  the 
carbon  source  and  w^ith  the  source  of  nitrogen,  as  w'ell  as  with  the  age  of 
the  cultures. 

For  a  discussion  of  other  utilization  ratios  and  examples,  see  Steinberg 
(1942),  Peterson  et  al.  (1922),  White  and  Willaman  (1928),  and  Fries 
(1938). 

Intermediary  metabolism.  The  problem,  to  be  considered  here  is  the 
way  fungi  utilize  the  various  sources  of  carbon  available  to  them.  From 
the  data  and  discussion  in  the  earlier  part  of  this  chapter  it  is  clear  that 
structure  and  configuration  play  an  important  role  in  determining  which 
compounds  may  serve  as  a  source  of  carbon  for  a  given  fungus.  The 
availability  of  complex  natural  compounds,  such  as  the  carbohydrates, 
was  found  to  depend  upon  the  production  of  the  necessary  extracellular 
hydrolytic  enzymes.     The  utilization  of  simple  compounds,  such  as  the 


140  PHYSIOLOGY  OF  THE  FUNGI 

monosaccharides,  is  likewise  an  enzymatically  catalyzed  chain  of  meta- 
bolic processes.  It  may  be  assumed  that  the  chemical  composition  of  the 
fungus  will  be  about  the  same,  irrespective  of  the  carbon  source  utilized. 
Therefore,  at  some  place  along  the  path  of  synthesis  the  initial  carbon 
sources  are  converted  into  the  same  compounds.  It  is  probable  that  the 
original  compounds  are  converted  into  the  same  intermediate  compounds 
before  synthesis.  Thus,  galactose  is  apparently  transformed  by  Sac- 
charomyces  fragilis  into  galactose- 1-phosphate,  which  is  then  converted 
into  glucose- 1-phosphate  (Caputto  et  at.,  1949).  These  intermediate 
compounds  then  enter  the  various  metabolic  reaction  chains  which  lead 
to  the  production  of  materials  which  make  up  the  fungus.  We  may 
suppose  that  the  first  steps  in  utilization  are  those  which  transform  a 
carbon  source  into  key  intermediates. 

The  intermediate  metabolic  products  should  also  serve  as  a  source  of 
carbon  for  the  fungus  in  question.  If  a  fungus  transforms  compound  A 
into  compound  B,  then  compound  B  should  serve  as  a  source  of  carbon. 
Nonutilization  of  compound  B  indicates  that  this  compound  is  not  part  of 
the  metabolic  pathway.  This  simple  hypothesis  neglects  two  important 
considerations :  compound  B  may  not  enter  the  fungus  cells  with  the  same 
facility  as  compound  A,  or  compound  B  may  be  toxic  in  the  concentra- 
tions present.  As  an  example  of  this  approach,  the  work  of  Steinberg 
(1942)  may  be  consulted.  Since  Aspergillus  niger  made  only  a  trace  of 
growth  on  D-gluconic  acid,  it  seems  probable  that  the  first  step  in  the 
utilization  of  glucose  by  this  fungus  is  not  the  oxidation  of  the  aldehyde 
group.  The  isolated  enzymes  from  a  fungus  may  also  be  studied  to 
determine  the  reactions  catalyzed,  or  the  effect  of  specific  enzyme 
inhibitors  on  the  intact  fungus  may  be  studied. 

In  some  instances  intermediates  of  sugar  dissimilation  are  excreted  into 
the  medium  and  may  be  isolated.  Thus,  the  production  of  acetaldehyde 
may  be  demonstrated  by  adding  bisulfite  to  the  medium.  Acetaldehyde 
forms  an  insoluble  addition  product  with  this  reagent.  The  excretion 
of  intermediate  metabolites  may  be  due  to  slowness  of  the  next  step  in  the 
metabolic  process.  These  products  are  usually  utilized  in  the  course  of 
time.  Among  such  intermediates  which  have  been  identified  are  acet- 
aldehyde, ethyl  alcohol,  and  pyruvic  acid. 

On  the  basis  of  the  evidence  now  available  we  may  not  assume  that  all 
fungi  utilize  a  sugar  or  other  carbon  source  in  exactly  the  same  way,  or 
that  a  fungus  has  only  one  metabolic  pathway  for  the  utilization  of  a 
sugar.  Nord  and  Mull  (1945)  consider  that  species  of  Fusarium  dis- 
similate  carbohydrates  by  oxidation,  by  splitting  the  carbon  chain,  and 
by  a  phosphorylation  mechanism.  The  relative  importance  of  these 
three  methods  of  attack  depends  upon  the  fungus  involved  and  upon  the 
environmental  conditions.     Identity  of  a  metabolic  product  formed  by 


CARBOX  SOURCES 


141 


two  fungi  is  not  proof  that  the  reaction  mechanism  is  the  same  in  both 
instances.  Yeast  and  certain  species  of  Fusarium  produce  alcohol,  but 
the  pathways  from  glucose  to  alcohol  appear  to  be  different.  The 
mechanism  of  carbohydrate  dissimilation  by  Fusarium  lini,  when  grown 
upon  a  nitrate  medium,  is  believed  to  take  place  as  shown  in  scheme  V. 
An  essential  feature  of  this  scheme  is  the  formation  of  pyruvic  acid  from 
both  pentoses  and  hexoses.  The  intermediate  steps  in  this  biosynthesis 
by  Fusarium  lini  have  not  been  elucidated.  A  portion  of  the  hydrogen 
derived  from  the  dissimilation  of  carbohydrate  is  enzymatically  trans- 
ferred and  used  for  the  reduction  of  nitrate  ion  which  acts  as  a  hydrogen 
acceptor.  The  nitrite  produced  inhibits  the  carboxylase  enzyme  system 
which  transforms  pyruvic  acid  into  carbon  dioxide  and  acetaldehyde. 
Pyruvic  acid  does  not  accumulate  in  the  culture  medium  when  ammonium 
nitrogen  is  used. 

Scheme  V.     The  Pathway  of  Hexose  and  Pentose  Utilization  by  Fusarium  lini 

Grown  on  Nitrate  Medium 
(Courtesy  of  Wirth  and  Nord,  Arch.  Biochem.  \ :  155,  1942.     Published  by  permis- 
sion of  Academic  Press,  Inc.) 

Acceptor 
(Nitrate) 

Nitrite 

I 

Carboxylase  system 

Hydroxylamine 
+ 
->-   Pyruvic  acid    ^-  Alcohol 

X  reduction 
Amino  acid 

Utilization 


Hexoses 

Pentoses 


Pyruvic  acid  is  the  key  intermediate  compound  formed  in  the  dis- 
similation of  hexoses  and  pentoses  by  F.  lini.  The  transformation  of 
pyruvic  acid  into  alcohol  by  F.  lini  and  yeasts  appears  to  follow  the  same 
pathway  and  to  require  the  same  coenzymes,  cocarboxylase  and  code- 
hydrogenase  I. 

The  anaerobic  dissimilation  (fermentation)  of  glucose  by  yeast  and 
the  comparable  process  in  muscle  (glycolysis)  have  been  intensively 
studied.  These  are  perhaps  the  best  understood  of  all  metabolic  proc- 
esses. Although  it  does  not  function  in  glucose  dissimilation  by  F.  lini  in 
the  same  way  as  in  yeast,  phosphate  plays  a  role  in  all  these  transforma- 
tions until  pyruvic  acid  is  formed.     Many  investigators  have  contributed 


142 


PHYSIOLOGY  OF  THE  FLi.XGI 


to  the  scheme  of  gUicose  dissimilation  presented  in  schemie  VI  (Meyerhof, 
1938,  1949).  Further  information  about  these  reactions  may  be  found  in 
Sumner  and  Soraers  (1947),  Tauber   (1949),  and  Prescott  and  Dunn 

(1949). 

Scheme  VI.     The  Pathway  of  Glucose  Dissimiliation  by  Yeast  (Alcoholic 
Fermentation)  and  Muscle  (Glycolysis) 
(Courtesy  of  Meyerhof,  WaUerstein  Labs.  Communs.  12 :  256,  1949.     Published  by 
permission  of  WaUerstein  Laboratories.) 
Glycogen,  starch  D-Glucose 


±  H,P04 


II 


+  H3PO4 


Glucose- 1-phosphate   (Cori  ester)^=±GIucose-()-phosphate  (Robison  ester) 


Fructose-6-phosphate  (Neuberg  ester) 

+  H3PO4 

'4' 

Fructose-ljG-diphosphate  (Harden- Young  ester) 


Dioxyacetone  phosphate* 


+  H2 


CHaOHCiOCHrOPOsHa 


/-Q^Glycero  phosphate 

[CHsOHCHOH-CHaOPOsHs 

Glycerol  +  H3PO4 


Aceta 
Ethanol 


dehyde  +  C02< 


Kd)  3-GIyceraIdehyde  phosphate  (Fischer-Baer 

ester) 
tCHOCHOHCH2-0-P03H2 

+  H2 


+  H3PO4 
(d)   1,3-Diphosphoglyceric  jicid 
+  H3PO4 


'CO-O-POsHsCHOH-CHa-O-POsH, 


(d)  3-Phosphoglvceric  acid 

TCOOH-CHOH-CHa-O-POsHs 

(d)   2-Phosphoglvceric  acid 

*  CH2dHCH(OP03H2)COOH 


+  H2O 

(Enol)-Phosphopvruvic  acid 

TCH2:C(OP03Ho)(COOH) 

-Pyruvic  acid  +  H3PO4   . 

±H2 

Lactic  acid 


Pyruvic  acid  is  a  key  intermediate  compound  in  metabolism.  Pyruvic 
acid  serves  as  a  source  of  carbon  for  many  fungi,  although  the  rate  of 
growth  on  this  substance  is  frequently  slow.  This  is  in  accord  with  the 
hypothesis  that  intermediate  metabolites  are  able  to  replace  the  original 
carbon  source.  The  accumulation  of  this  compound  in  the  culture 
medium  may  be  demonstrated  by  the  formation  of  iodoform  in  the  cold 
by  adding  a  solution  of  iodine  and  making  the  medium  strongly  alkaline. 
The  sensitive  color  test  of  Lu  (2,4-dinitrophenylhydrazine)  may  also  be 
used  [see  Friedemann  and  Haugen  (1943)  for  details].  Acetaldehyde  also 
yields  iodoform  under  these  conditions,  but  gentle  heating  will  drive  off 


CARBON  SOURCES 


143 


this  substance.  AVe  have  noted  in  this  laboratory  that  pyruvic  acid 
ordinarily  disappears  from  culture  medium  as  the  time  of  incubation  is 
increased.  The  disappearance  of  the  pyruvic  acid  in  the  culture  medium 
is  usually  correlated  with  a  rise  in  pH.  Some  typical  reactions  of  pyruvic 
acid  are  shown  in  scheme  VII.  For  a  review  of  pyruvate  metabolism  see 
Stotz  (1945). 

Scheme  VIIo     Some  Typical  Transformations  of  Pyruvic  Acid 

CH3-CO— COOH 

H  ^^  Pl7ri11M^  rtmA  ^\    NH3+H 


CH3-CHOH-COOH 

Lactic  acid 


0 


Pyruvic  acid 

-CO2 

CH3— CHO 
Acetaldehyde 


CH3-CH(NH2)-COOH 

ot  -  Alanine 


H 


CH3-COOH 
Acetic  acid 


CH3-CH2OH 


Ethyl  alcohol 

It  is  probable  that  most  intermediates  used  in  the  synthesis  of  proto- 
plasm are  synthesized  from  low-molecular-weight  compounds.  Acetate 
is  used  by  yeasts  for  the  synthesis  of  fats  and  other  cellular  constituents. 
Weinhouse  and  Millington  (1947)  studied  the  metabolism  of  isotopic 
acetate  by  yeast  depleted  of  endogenous  nutrients.  Acetate  was  rapidly 
utilized.     The  distribution  of  the  carbon  from  the  isotopic  acetate  was 


Scheme  VIII.     Oxidation  of  Acetate  by  Yeast  by  Means  of  the  Krebs  Citric 

Acid  Cycle* 

COOH 

Hooc— CH2— c— CHj— coon 


CH3— COOH -I   HOOC— CHr— CO— COOH  - 

T 


2CH3— COOH 


HOOC— CH2— CHO  H— COOH 

T 

HOOC— CH=CH— COOH 

T 

HOOC— CH2—C Ho— COOH  <— 


HOOC— CH: 


-CO2 


—CH— CHO  H— COOH 

COOH 
CO2 


HOOC— CH2—CH2— CO— COOH 


*  Original  scheme  modified  according  to  Weinhouse.     Courtesy  of  Weinhouse  and  Millington,  Jour. 
Am.  Chem.  Soc.  69:  3093,  1947.      Published  by  permission  of  the  American  Chemical  Society. 

determined  by  analysis.  A  portion  of  the  acetate  was  oxidized;  another 
portion  was  found  in  the  lipide  fraction  and  cell  residue;  some  was  con- 
verted to  citric  acid.  It  was  calculated  that  from  one-fourth  to  one-third 
of  the  lipides  found  in  the  yeast  cells  at  the  end  of  the  experiment  (a  7-hr. 
period)  were  newly  synthesized  from  acetate.  The  cell  residue  (after 
extraction  of  the  fats)  contained  only  a  little  isotopic  carbon.  This  is  not 
surprising,  since  nitrogen  was  not  furnished  during  these  experiments. 


144 


PHYSIOLOGY  OF  THE  FUNGI 


The  mechanism  of  acetate  oxidation  by  yeast  is  postulated  by  these 
authors  to  follow  a  modified  Krebs  citric  acid  cycle  (scheme  VIII).  The 
oxidation  of  acetate  is  thus  the  result  of  a  rather  complex  cyclic  process. 
While  the  four-carbon  dicarboxylic  acids  of  the  Krebs  cycle  are  poor 
sources  of  carbon  for  most  fungi,  they  are  important  in  intermediary 
metabolism.  These  acids  are  readily  interconvertible.  The  role  of  the 
keto  acids  in  amino-acid  synthesis  was  noted  in  Chap.  6.  Lewis  (1948) 
studied  the  metabolism  of  mutants  of  Neurospora  crassa  which  were 
unable  to  synthesize  either  aspartic  or  glutamic  acids.     These  amino 

Scheme  IX.     A  Generalized  Krebs  Isocitric  Acid  Cycle  Proposed  to 

Illustrate  the  Pathways  of  Conversion  of  Carbohydrate  into 

Aspartic  and  Glutamic  Acids  by  Neurospora* 

Carbohydrate 

It 

Pyruvate  -^ 


(CHg-CO-R) 


Cis-aconitate 
n 


Isocitrate 
A 
A 


oi  -Ketoglutarate 


Oxalacetate 


Succinate 


*  Courtesy  of  Lewis,  Am.  Jour.  Botany  36:  294,  1948. 


acids  could  be  replaced  by  a-ketoglutaric,  succinic,  malic,  or  fumaric 
acids.  A  generalized  Krebs  cycle  was  proposed  by  Lewis  which  indicates 
the  pathway  of  synthesis  of  aspartic  and  glutamic  acids  from  glucose 
(scheme  IX).  Compounds  utilized  by  the  Neurospora  mutants  are 
printed  in  italics.  The  probable  location  of  the  genetic  block  which 
prevents  the  biosynthesis  of  aspartic  and  glutamic  acids  is  indicated  by  A. 

SUMMARY 

Organic  compounds  are  utilized  by  fungi  for  the  synthesis  of  structural 
and  functional  compounds  and  as  sources  of  energy.  The  fungi  utilize  a 
wide  range  of  natural  organic  compounds  including  those  of  great  com- 


CARBON  SOURCES  145 

plexity.  Not  all  fungi  utilize  all  natural  organic  compounds,  nor  do  all 
species  utilize  a  given  compound  with  the  same  facility.  The  composi- 
tion, structure,  and  configuration  of  organic  compounds  affect  utilization, 
but  the  effect  of  these  factors  may  be  different  for  different  fungi. 

The  carbohydrates  are  the  most  common  and  important  sources  of 
carbon  for  the  fungi.  Sugars  (and  other  compounds)  having  the  same 
structure,  but  with  mirror-image  configuration,  differ  physiologically. 
Usually  only  one  enantiomorph  is  utilized,  or  one  is  utilized  much  more 
rapidly  than  the  other.  Glucose  is  utilized  by  more  fungi  than  any  other 
sugar.  Few  fungi  are  unable  to  utilize  glucose.  A  few  species  are  appar- 
ently unable  to  utilize  any  sugar;  e.g.,  Leptomitus  lacteus.  The  species 
that  utilize  the  pentoses,  sugar  alcohols,  acids,  and  other  simple  organic 
compounds  are  fewer  in  number  than  those  which  utilize  glucose. 

The  oligo-  and  polysaccharides  are  utilized  by  fewer  species  than  is 
glucose.  The  nature  of  the  glycoside  linkage  as  well  as  the  sugar  residues 
is  important  in  determining  whether  these  compounds  are  utilized  by  a 
given  fungus.  It  is  probable  that  most  fungi  hydrolyze  oligosaccharides 
before  utilization  occurs.  This  does  not  exclude  direct  utilization  in  some 
instances.  An  oligosaccharide  and  its  hydrolytic  products  are  not  always 
physiologically  equivalent.  The  general  order  of  availability  of  the  three 
common  disaccharides  appears  to  be  maltose,  sucrose,  and  lactose. 
Among  the  polysaccharides,  cellulose  and  starch  are  the  most  abundant. 
These  compounds  are  insoluble  and  must  be  hydrolyzed  or  otherwise 
degraded  to  low-molecular-weight  compounds  before  utilization.  Only 
those  fungi  which  form  cellulase  and  amylase  are  able  to  utilize  these 
compounds.  This  "digestion"  is  accomplished  by  enzymes.  Ability  to 
utilize  other  polysaccharides  is  also  dependent  upon  possession  of  the 
necessary  hydrolytic  enzymes. 

Some  fungi  utilize  carbon  dioxide,  but  not  as  a  sole  source  of  carbon. 
It  is  postulated  that  carbon  dioxide  combines  with  pyruvic  acid  and  other 
keto  acids  to  form  key  intermediate  products  which  are  necessary  for  the 
formation  of  amino  acids. 

The  fate  of  the  carbon  supplied  to  a  fungus  is  best  determined  by 
carbon-balance  studies;  i.e.,  by  complete  chemical  analyses  of  the  myce- 
livim  and  other  metabolic  products,  including  the  carbon  dioxide  produced. 

The  first  step  in  utilization  of  sugars  and  other  carbon  sources  is  the 
formation  of  certain  key  intermediate  metabolic  compounds.  These  key 
compounds  are  in  part  utilized  for  synthesis  and  in  part  oxidized  to  pro- 
vide energy.  The  metabolic  pathways  leading  to  the  formation  of  key 
intermediates  differ,  depending  upon  the  environmental  conditions  and 
the  fungus  involved.  Among  the  key  intermediates  pyruvic  acid  is 
especially  noteworthy.  Reduction  of  this  compound  yields  lactic  acid, 
while  amination  and  reduction  leads  to  alanine.     Decarboxylation  pro- 


146  PHYSIOLOGY  OF  THE  FUNGI 

duces  acetaldehyde,  which  in  turn  may  yield  either  ethyl  alcohol  or  acetic 
acid.  Acetate  is  utilized  by  yeast  and  other  fungi  for  the  synthesis  of 
fats  and  other  cellular  constituents.  A  fungus  utilizes  a  compound  by  a 
series  of  step-by-step  transformation.  Among  the  best  understood  of 
these  metabolic  activities  is  the  transformation  of  glucose  into  alcohol  by 
yeasts. 

REFERENCES 

Ajl,  S.  J.,  and  C.  II.  Werkman:  On  the  mechanism  of  carbon  dioxide  replacement 

in  heterotrophic  metaboHsm,  Jour.  Bad.  57 :  579-593,  1949. 
Armstrong,  E.  F.  :  Studies  in  enzyme  action.     I.  The  correlation  of  the  stereoisomeric 

alpha  and  beta  glucosides  with  the  corresponding  glucoses.  Jour.  Chem.  Soc.  83 : 

1305-1313,  1903. 
Bertrand,  G.:  Etude  biochemique  de  la  bacteria  du  sorbose,  Ann.  chim.  et  phys.  3: 

181-288,  1904. 
Campbell,  W.  G.:  The  chemistry  of  the  white  rots  of  wood.     III.  The  effect  on 

wood   substance   of  Ganoderma   applanatum    (Pers.)   Pat.,   Fomes   fomentarius 

(Linn.)    Fr.,  Polyporus  adustus    (Willd.)   Fr.,  Pleurotus  ostreatus    (Jacq.)    Fr., 

ArmiUaria  meUea  (Vahl.)  Fr.,  Trametes  pit^i  (Brot.)  Ft.,  Sind  Polystictus  abietinus 

(Dicks.)  Fr.,  Biochem.  Jour.  26:  1829-1838,  1932. 
Canting,  E.  C:  The  phj^siology  of  the  aquatic  Phycomycete,  Blastocladia  Pring- 

sheimii,  with  emphasis  on  its  nutrition  u,nd  metabolism,  Am.  Jour.  Botany  36: 

95-112,  1949. 
Caputto,  R.,  L.  F.  Leloir,  R.  E.  Trucco,  C.  E.  Cardini,  and  A.  C.  Paladini: 

The  enzymatic  transformation  of  galactose  into  glucose  derivatives,  Jour.  Biol. 

Chem.  179 :  497-198,  1949. 
Cheo,  p.  C:  Stripe  smut  of  blue  grass  (Ustilago  striiformis  forina  Poae-pratensis): 

Spore  germniation,  artificial  inoculation,  pathological  histology  and  growth  in 

artificial  media,  thesis.  West  Virginia  University,  1949. 
Dox,   A.   W.,  and  R.  E.  Neidig:  Spaltung  von  a-  und  /3-Methylglucosid  durch 

Aspergillus  nigcr,  Biochem.  Zeit.  46:  397-402,  1912. 
Dox,  A.  ^Y.,  and  G.  W.  Roark,  Jr.:  The  utilization  of  a/p/ia-methylglucoside  by 

Aspergillus  nigcr,  Jour.  Biol.  Chem.  41:  475-481,  1920. 
Dulaney,  E.  L.  :  Observations  on  Streptomyces  griseus.     III.   Carbon  sources  for 

growth  and  streptomycin  production,  Mycologia  41:  1-10,  1949. 
Foster,  J.  W. :  Chemical  Activities  of  Fungi,  Academic  Press,  Inc.,  New  York,  1949. 
Foster,  J.  W.,  S.  F.  Carson,  S.  Rl'ben,  and  IM.  D.  Kamen:  Radioactive  carbon 

as  an  indicator  of  carbon  dioxide  utilization.     VII.  The  assimilation  of  carbon 

dioxide  by  molds,  Proc.  Natl.  Acad.  Sci.  U.S.  27:  590-596,  1941. 
Foster,  J.  W.,  and  J.  B.  D.wis:  Anaerobic  formation  of  fumaric  acid  by  the  mold 

Rhizopus  nigricans.  Jour.  Bad.  56 :  329-338,  1948. 
Friedemann,  T.  E.,  and  G.  E.  Haugen:  Pyruvic  acid.     II.  The  determination  of 

keto  acids  in  blood  and  urine.  Jour.  Biol.  Chem.  147:  415-442,  1943. 
Fries,  N.:  Ucber  die  Bedeutung  von  Wuchsstoffen  flir  das  Wachstum  verscheidener 

Pilze,  Symbolae  Botan.  Upsalienses  3 :  2,  1938. 
GiLMAN,  II.:  Organic  Chemistry,  an  Advanced  Treatise,  Vol.  II,  John  Wiley  &  Sons, 

Inc.,  New  York,  1943. 
Gottlieb,  D.  :  The  utilization  of  amino  acids  as  a  source  of  carbon  by  fungi,  Arch. 

Biochem.  9:  341-351,  1946. 
Grassmann,  W.,  L.  Zeichmeister,  G.  Toth,  and  R.  Stadder:  Ueber  den  enzyma- 

tischen   Abbau   der   Cellulose   und   ihrer   Spaltprodukte.     2.   Mitteilung   iiber 

enzymatische  Spaltung  von  Polysacchariden,  J.  Leibigs  Ann.  d.  Chem.  503: 

167-179,  1933. 


CARBOX  SOURCES  147 

■*Ha\vker,    L.    E.:    Further   experiments   on   growth   and   fruiting   of   Melanospora 

destruens  Shear  in  the  presence  of  various  carbohydrates,  with  special  reference 

to  the  effects  of  gkicose  and  of  sucrose,  Ann.  Botany  (N.S.)  11:  245-260,  1947. 
Herrick,  J.  A.:  The  carbon  and  nitrogen  metabolism  of  Stereurn  gausapatum  Fries., 

Ohio  Jour.  Sci.  40:  123-129,  1940. 
*Heuser,  E.,  B.  F.  Shema,  W.  Shockley,  J.  W.  Appling,  and  J.  F.  McCoy:  The 

effect  of  lignin-destroying  fungi  upon  the  carbohydrate  fraction  of  wood,  Arch. 

Biochem.  21 :  343-350,  1949. 
HoLZAPFEL,  II.  H.:  Untersuchung  iiber  die  C-und  N-Quellen  einiger  Fusarien,  Cent. 

Bakt.,  Abt.  II.,  64:  174-222,  1925. 
■*HoRR,  W.  II.:  Utilization  of  galactose  by  Aspergillus  niger  and  Pcnicillium  glaucum, 

Plant  Physiol.  11:  81-99,  1936. 
Kakeura,  M.:  Starches  by  using  parasitic  fungi,  Japanese  patent  172,381,  Mar.  6, 

1946.     Chem.  Abs.  43:  6850,  1949. 
KiNSEL,  K.:  Carbohydrate  utilization  by  the  corn  Diplodias,  Phytopathology  27: 

1119-1120,  1937. 
Lewis,  R.  W.  :  Mutants  of  Neurospora  crassa  requiring  succinic  acid  or  a  biochemi- 
cally related  acid  for  growth,  Am.  Jour.  Botany  35 :  292-295,  1948. 
Margolin,  A.  S.:  The  carbohydrate  requirements  of  Diplodia  macrospora,  Proc. 

West  Va.  Acad.  Sci.  14 :  56-59,  1940. 
Margolin,  A.  S. :  The  effect  of  various  carbohydrates  upon  the  growth  of  some 

fungi,  thesis.  West  Virginia  University,  1942. 
Meyerhof,  O:  The  intermediary  reactions  of  fermentation.  Nature  141:  855-858, 

1938. 
Meyerhof,  O.:  Glycolysis  of  animal  tissue  extracts  compared  with  the  cell-free 

fermentation  of  yeast,  Wallerstein  Labs.  Communs.  12  :  255-264,  1949. 
Myrbach,  K.:  The  structure  of  starch,  Wallerstein  Labs.  Communs.  11:  209-219, 

1948. 
Nobles,   M.  K.:  Studies  in  forest  pathology.     VI.  Identification  of  cultures  of 

wood-rotting  fungi,  Can.  Jour.  Research,  Sec.  C,  26:  281-431,  1948. 
NoRD,  F.  F.,  and  R.  P.  Mull:  Recent  progress  in  the  biochemistry  of  Fusaria, 

Advances  in  Enzymol.  5:  165-205,  1945. 
NoRD,  F.  F.,  and  J.  C.  Vitucci:  On  the  mechanism  of  enzyme  action.     XXX.  The 

formation  of  methyl-p-methoxycinnamate  by  the  action  of  Lentinus  lepideus 

on  glucose  and  xylose.  Arch.  Biochem.  14:  243-247,  1947. 
NoRD,  F.  F.,  and  J.  C.  Vitucci:  Certain  aspects  of  the  microbiclegical  degradation 

of  cellulose.  Advances  in  Enzymol.  8:  253-298,  1948. 
Norman,  A.  G.,  and  W.  H.  Fuller:  Cellulose  decomposition  by  microrganisms. 

Advances  in  Enzymol.  2 :  239-264,  1942. 
Pasteur,  L.:  Note  relative  au  Penicillium  glaucum  et  a  la  dissymetrie  mcleculaire 

des  produits  organiques  naturels,  Compt.  rend.  acad.  sci.  51:  298-299,  1860. 
Peterson,  W.  H.,  E.  B.  Fred,  and  E.  G.  Schmidt:  The  fermentation  of  pentoses  by 

molds,  Jour.  Biol.  Chem.  54:  19-34,  1922. 
PiGMAN,  W.  W.,  and  R.  M.  Goepp:  Chemistry  of  the  Carbohydrates,  Academic 

Press,  Inc.,  New  York,  1948. 
Prescott,  S.  C,  and  C.  G.  Dunn:  Industrial  Microbiology,  2d  ed.,  McGraw-Hill 

Book  Company,  Inc.,  New  York,  1949. 
Raistrick,  H.,  J.  H.  BiRKiNSHAW,  J.  H.  V.  Charles,  P.  W.  Clutterbuck,  F.  P. 

Coyne,  A.  C.  Hetherington,  C.  H.  Lilly,  M.  L.  Rintoul,  W.  Rintoul,  R. 

Robinson,  J.  A.  R.  Stoyle,  C.  Thom,  and  W.  Young:  Studies  in  the  bio- 
chemistry of  micro-organisms.  Trans.  Roy.  Soc.  (London),  Ser.  B,  220:  1-367, 

1931. 
Ratajak,  E.  J.,  and  H.  S.  Owens:  Optimal  conditions  for  the  hydrolysis  of  arabo 

galactan  by  Aspergillus  niger,  Botan.  Gaz.  104:  329-337,  1942. 


148  PHYSIOLOGY  OF  THE  FUNGI 

ScHADE,  A.  L.;  The  nutrition  of  Leptomitus,  Am.  Jour.  Botany  27:  370-384,  1940. 
ScHADE,  A.  L.,  and  K.  V.  Thimann:  The  metabolism  of  the  water  mold  Leptomitus 

lacteus,  Am.  Jour.  Botamj  27:  659-670,  1940. 
Schubert,  W.  J.,  and  F.  F.  Nord:  Investigations  on  lignin  and  lignification.     I. 

Studies  on  softwood  lignin,  Jour.  Am.  Chem.  Soc.  72  :  977-981,  1950. 
ScHULTZ,  A.  S.,  D.  K.  Mc  INI  ANUS,  and  S.  Pomper:  Amino  acids  as  carbon  source  for 

the  growth  of  yeasts,  Arch.  Biochem.  22 :  412-419,  1949. 
Skoog,  F.  K.,  and  C.  C.  Lindegern:  Adaptation  in  a  yeast  unable  to  ferment  glucose 

Jour.  Bact.  53:  729-742,  1947. 
Smith,  V.  M.:  On  the  mechanism  of  enzyme  action.     XXXIX.  A  comparative 

study  of  the  metabolism  of  carbohydratqs,  in  the  presence  of  inorganic  and 

organic  phosphates,  by  MeruHus  lacrymans  and  Marasmius  chordalis,  Arch. 

Biochem.  23 :  446-472,  1949. 
Steinberg,  R.  A.:  Relation  of  carbon  nutrition  to  trace-element  and  accessory 

requirements  of  Aspergillus  niger,  Jour.  Agr.  Research  59  :  749-763,  1939. 
Steinberg,  R.  A.:  The  process  of  amino  acid  formation  from  sugars  in  Aspergillus 

niger,  Jour.  Agr.  Research  64:  615-633,  1942. 
Steinberg,  R.  A.:  Effects  of  trace  elements  on  growth  of  Aspergillus  niger  with 

amino  acids.  Jour.  Agr.  Research  64:  455-475,  1942o. 
Stevens,  N.  E.,  and  H.  W.  Larsh:  Carbohydrate  utilization  by  Diplodia  macro- 

spora,  Trans.  Illinois  State  Acad.  Sci.  32 :  82,  1939. 
*Stotz,  E.:  Pyruvate  metabolism.  Advances  in  Enzymol.  5:  129-164,  1945. 
Sumner,  J.  B.,  and  G.  F.  Somers:  Chemistry  and  Methods  of  Enzymes,  Academic 

Press,  Inc.,  New  York,  1947. 
Tamiya,    H.:    Ueber   die    Verwendbarkeit   von   verschiedenen    Kohlenstoffverbin- 

dungen  im  Bau-  und  Beitreibsstoffwechsel  der  Schimmelpilze.     Studien  iiber 

die    Stoffwechselphysiologie    von    Aspergillus    oryzae.     IV.   Acta    Phytochim. 

(Japan)  6 :  1-129,  1932. 
Tauber,  H.:  The  Chemistry  and  Technology  of  Enzymes,  John  Wiley  &  Sons,  Inc., 

New  York,  1949. 
Van  Niel,  C.  B.:  Recent  advances  in  our  knowledge  of  the  physiology  of  micro- 
organisms, Bact.  Revs.  8 :  225-234,  1944. 
VoLKONSKY,    M.:    Sur    la    nutrition    de    quelques    champignons   saprophytes   et 

parasites,  Ann.  inst.  Pasteur  52:  76-101,  1934. 
Weinhouse,  S.,  and  R.  H.  Millington:  Acetate  metabolism  in  yeast,  studied  with 

isotopic  carbon,  Jour.  Am.  Chem.  Soc.  69:  3089-3093,  1947. 
■*Werkman,  C.  H.,  and  IL  G.  Wood:  Heterotropic  assimilation  of  carbon  dioxide, 

Advances  in  Enzymol.  2 :  135-182,  1942. 
White,  A.  G.  C,  and  C.  H.  Werkman:  Assimilation  of  acetate  by  yeast.  Arch. 

Biochem.  13:27-32,  1947. 
White,  M.  G.,  and  J.  J.  Willaman:  Biochemistry  of  plant  diseases.     X.  Fermenta- 
tion of  pentoses  by  Fusarium  lini,  Biochem.  Jour.  22:  583-591,  1928. 
*WiLSON,  W.  E.:  Physiological  studies  on  two  species  of  Diplodia  parasitic  on   corn. 

Phytopathology  32 :  130-140,  1942. 
WiRTH,  J.  C,  and  F.  F.  Nord:  Essential  steps  in  the  enzymatic  breakdown  of 

hexoses  and  pentoses.     Interaction  between  dehydrogenation  and  fermentation, 

Arch.  Biochem.  1:  143-163,  1942. 
Wolf,  F.  T.,  and  C.  S.  Shoup  :  The  effects  of  certain  sugars  and  amino  acids  upon 

the  respiration  of  Allomyces,  Mycologia  35 :  192-200,  1943. 
Yaw",  K.  :  Personal  communication,  1950, 


CHAPTER  8 
HYDROGEN-ION  CONCENTRATION 

The  growth  of  fungi  and  bacteria  may  be  inhibited  or  prevented  by 
media  which  are  too  acidic  or  too  alkaline.  A  completely  satisfactory 
medium  may  be  made  useless  by  the  addition  of  relatively  small  amounts 
of  strong  acids  or  bases  but  may  have  its  former  usefulness  restored  if  the 
excess  acid  or  base  is  neutralized.  This  suggests  that  the  ions  which 
characterize  acids  and  bases  are  particularly  active  in  life  processes.  It  is 
necessary  to  understand  certain  fundamental  ideas  about  acidity  and 
ways  of  measuring  concentration  of  these  ions  before  discussing  in  detail 
the  effects  of  acids  and  bases  on  the  activities  of  the  fungi. 

IONIZATION  OF  COMPOUNDS 

Since  water  is  the  universal  solvent  for  all  life  processes,  our  discussion 
will  be  confined  to  aqueous  solutions.  The  chemical  compounds  which 
comprise  natural  and  synthetic  media  may  be  divided  into  two  classes, 
those  which  form  ions  in  solution  (acids,  bases,  and  salts),  and  those 
which  do  not  ionize  (organic  compounds  in  general,  except  organic  acids 
and  bases).  Water  is  a  compound  of  the  first  class,  although  it  forms  ions 
to  a  very  slight  degree.  The  ionization  of  water  may  be  represented  by 
the  following  equation: 

(1)  HOH  ^  H+  +  OH-        or         HOH  +  HOH  ^  H3O+  +  OH" 

For  each  molecule  of  water  ionized  one  hydrogen  and  one  hydroxyl  ion 
are  formed. 

In  any  aqueous  solution  the  product  of  the  concentrations  of  the 
hydrogen  and  the  hydroxyl  ions  (in  moles)  is  equal  to  a  constant  (K^), 
Water  is  a  neutral  compound,  i.e.,  the  concentrations  of  hydrogen  and 
hydroxyl  ions  are  equal.  A  solution  which  contains  a  greater  concentra 
tion  of  hydrogen  than  of  hydroxyl  ions  is  acidic ;  a  solution  which  contains 
a  greater  concentration  of  hydroxyl  ions  than  of  hydrogen  ions  is  basic,  or 
alkaline.  Since  all  aqueous  solutions  contain  hydrogen  and  hydroxy] 
ions,  the  deleterious  effects  of  these  ions  must  be  due  to  their  relative 
concentrations.  At  room  temperature  (23  to  25°C.)  the  concentration  of 
hydrogen  and  hydroxyl  ions  in  water  is  1  X  10"^  mole  per  liter,  or  1  mole 
each  of  these  ions  in  10  million  liters.  The  degree  of  ionization  of  water 
increases  with  temperature.     However,  water  is  a  neutral  substance  at 

149 


150  PHTSIOLOGT  OF  71- 1  FVXGI 

all  lempersTures  because  equal  number?  of  hydrogen  and  hydrox\-l  ions 
are  pi^sent.  The  value  of  K^  at  23  to  2o^C.  is  obtained  by  multiphing 
the  conceniraticais  irf  hydrogwi  and  hydroxyl  ic«is  present. 

(3j  K,  =  :  >  :>--      1  X  icr-  =  1  X  lO-^* 

For  ihe  parpoee  of  this  discussicsa.  we  will  consider  an  acid  to  be  a  com- 
pound. whcKe  aqueous  scJuticHi  contains  a  greater  concentration  of  hydro- 
gen rhxn  at  hydroxyl  ioais.  A  t-o^  is  a  compound  whotse  aqueous  solution 
contains  a  greater  concentraticai  of  hydro^l  ions  than  of  hydrc^n  itms. 
Tbese  deaniiioas  vdll  include  a  number  ci  compounds  which  are  ordinarily 
con^doed  as  salts.  A  solmion  of  an  acid  c«Mitains  a  greater  concentra- 
tion of  hydrogen  ions  th.--^-  pure  water  by  Tirtue  of  the  ionization  of  the 
acid.     Hydrochloric  acid  ionises  as  follow?: 

C3)  Ha  ^  H-  -  a- 

In  a  sofaxtica  of  hydrochloric  acid  there  are  two  sources  <rf  hydrogen  ions. 
the  acid  and  'irater.  Enou^  of  the  hydrogen  icais  will  combine  with  the 
hvdrasyi  ions  to  reduce  the  concentration  erf  this  ion  so  that  Eq.  (2)  will 
b*  satined.     The  ionization  o£  a  base  may  be  represented  as  follows: 

,  i  XaOH  =  Xa-  -  OH" 


-  ^  c-       -  -  --  es^ctly  the  same  as  gi"r«i  above  fca*  an  acid  except  that 
._-  ,  _  _  :>ii  C'f  hy<irogen  ions  is  less  in  a  solution  of  a  base  than  in 

pure 

A  s~.:   i-^      :  i-  soch  as  Lyir  >;h-oric.  is  considered  to  be  completely  itm- 

ixed  even  lu   ;  rated  solutions-     A  weak  acid,  such  as  acetic,  in 

IX  s-i-u  :  I-    -  uly  sli^tly    about  1  per  cent  .     The  perc-entage 

of  iooixi"    :     :  i  ^  increases  as  the  dHuticm  increases.     The  c<»- 

ccntri "          :  _          liS  in.  equal  volumes  of  normal  hydrochloric  and 

acetic  -    -  -^  — r  chief  differraice  between  these  acids.     Thus  the 

stoaiiT            r  '  ~   ~  '  -  -^pressed  in  tw^o  ways.  (1  <  the  total  acidity 

A*^^        ■  .  \ includes  both  the  ionized  and  nonionized 

mofecuks  CH  iL  i.e..  the  titratab'ie  acidity,  and  (2 )  the  actwal  acidity 
at  any  instant.  -  :_  ^i  is  a  functicHi  of  the  cc«icentration  of  hydrogen  ions 
jwesBQt.  T:  -  -tration  c^  hydrogen  ions  is  a  nincrion  of  the  con- 
centre I.  -  :  i-izatitHi  of  the  acid  involved.  It  is  the  actual 
acadity  ~  -  esses.  It  is  also  necessary  to  consider 
the  phy-  .  riieii-  :  ;  z^  or  cations  which  are  associated 
wz- ^-     -              :      -".  i=  impossibie  to  add  just  hydrogen 

7  HZ   MIAXC^G  OF  pH 

T:  -     ::.     I.  :       UL     :    -    _  ^-:i  :  ui  in  a  soluticai  can  be  expressed  in 
Ti  —ays.     A  derived  unit  pH  is  most  used  in  biological  work.     The 


HYDROGES-IOS  COSCESTRA  TIGS' 


151 


calculated  c<Hic«itrati<Mis  of  hydrogen  and  hydroxyi  ions  in  sohition;  <rf 
hydrochloric  acid  and  potasdum  hydroxide  are  given  in  Tat4e  30.  C<«n- 
plete  ionization  was  assumed  in  these  falmlatioPiS. 


Table  30.     Calcxxated  Coxcest^atioks  or  Htoso^cix  axd  Htdeoiti.  loss  rs 
Prsz  Watee  axd  cf  SoLmoxs  of  Htdeoceuckic  Acid  axd  PotasctctiC 


Vonaaiit. 

'?*;c   -r^'r ;."._- j_ 

-    JL ;  -:i.*ji-_:_ 

Ha 

1 

1.0 

1  X10» 

0 

1  X  10-^ 

0.1 

1  X10-- 

X 

1  X  10-^ 

0.01 

1  xio-= 

2 

1  X  10-- 

0.001 

1  xio-» 

3 

1  X  10-^ 

0.0001 

1  X  10-* 

~ 

1  X  io-» 

0.00001 

1  X10-* 

0 

i  X  10-* 

0.000001 

1  xio-^ 

6 

1  X  10-* 

H20 

1  X10-- 

7 

1  X10-- 

KOH 

0.000001 

1  xio-« 

8 

1  X10-* 

0.00001 

1  X  io-» 

9 

1  X  10-* 

0.0001 

1  X  10--' 

10 

1  X  10-* 

0.001 

1  X  10-- 

11 

1  xia-» 

0.01 

1  X  lO-i^ 

12 

1  X  10-s 

0.1 

1  X  10--^ 

13 

1  X  10-^ 

1.0 

1  X  10-  * 

-IX 

I  ^  Iff 

S<$rensen  devised  a  logariiiunic  ~     ^  -     ~-^: 
hydrc^en  icHis  in  solutions.     The  ij^:..         _  .  -  . 
tial  of  hydrogen.'"  which  is  abreviated  to  pH. 
and  hydrc^en-ion  concentration  is  grven  bdow. 


-  -V:.c  -^1=  ;c 


in-e  "I  r.i-- 
i^nseei  pH 


5) 


pH=logi 


•^H="] 


In  words  this  equation  means  pH  is  equal  t :  _  -     . '     of 

the  reciprocal  (rf  the  hydrogaMon  :  r.    r  _  :   :_  x^wk  oO  li  £ 

seen  that  the  hydrogen-itm  ctMicentrtiiioii  ci  •j.l,'^»juu'UV  z.ydrorhloric  add 
is  1  X  10~*  mole  per  liter.     Suhstitutiiiz  "his  value  in  Eq.  ^5),  "we  hxve 


(6) 


1  .X.  l\f~* 


-.  =  lie  10*  =  6 


One  obvious  advantage  <rf  the  pH  scale  lies  in  the  fact  that  hydrogen-ion 
concentrations  are  expressed  as  positive  numbers  rather  than  as  decimal 


152  PHYSIOLOGY  OF  THE  FUNGI 

fractions.  If  a  solution  contains  4.23  X  10""  mole  per  liter  of  hydrogen 
ion,  this  concentration  of  hydrogen  ions  may  be  expressed  in  terms  of  pH 
by  use  of  Eq.  (5). 

(7)  pH  =  log  ^^3  ^  ^Q_,  =  log  1  +  log  10«  -  log  4.23  =  5.37 

By  reversing  the  above  calculations,  the  hydrogen-ion  concentration  may 
be  calculated  if  the  pH  value  is  known. 

The  pH  scale  has  three  features  which  may  be  perplexing:  (1)  alkaline 
solutions  are  designated  on  the  same  scale  as  acidic  solutions;  (2)  increas- 
ing acidity  is  expressed  by  smaller  pH  values;  and  (3)  the  logarithmic 
nature  of  the  scale.  A  difference  of  one  pH  unit  indicates  a  tenfold 
difference  in  hydrogen-ion  concentration,  while  a  difference  of  0.3  pH  unit 
means  a  twofold  difference.  A  solution  having  a  pH  value  of  4  contains 
10,000  times  as  many  hydrogen  ions  per  unit  volume  as  a  solution  having 
a  pH  of  8.  A  solution  having  a  pH  value  of  6.2  has  twice  the  hydrogen- 
ion  concentration  of  a  solution  having  a  pH  value  of  6.5.  The  student 
should  remember  that  pH  is  the  name  of  a  logarithmic  scale  used  to 
measure  hydrogen-ion  concentration  and  not  an  entity  in  itself. 

BUFFERS  AND  BUFFER  CAPACITY 

A  medium  having  pH  values  between  5  and  6  at  the  time  of  inoculation 
is  suitable  for  most  fungi,  but  more  acidic  or  more  alkaline  media  are  used 
at  times.  It  is  often  important  that  the  pH  of  the  culture  medium  does 
not  change  too  greatly  as  the  result  of  metabolic  activities  of  the  organism. 
The  ease  with  which  the  pH  of  a  medium  is  modified  depends  upon  the 
composition  of  the  medium. 

Substances  which  tend  to  maintain  the  pH  of  a  solution  relatively 
constant  when  an  acid  or  base  is  added,  or  when  the  solution  is  diluted, 
are  called  buffers.  In  general,  the  kinds  of  compounds  which  act  as 
buffers  are  mixtures  of  weak  acids  or  bases  and  their  salts,  acid  salts  of 
polybasic  acids,  basic  salts  of  polyacidic  bases,  and  amphoteric  compounds 
such  as  amino  acids  and  peptides.  As  an  example  of  a  buffer  we  may 
consider  a  solution  which  contains  equivalent  amounts  of  acetic  acid  and 
sodium  acetate.  If  a  strong  acid  such  as  hydrochloric  is  added  to  this 
buffer  solution,  a  reaction  will  occur  between  the  sodium  acetate  and  the 
hydrochloric  acid. 

(8)  CH3— COONa  +  HCl  -^  CH3— COOH  +  NaCl 

The  net  effect  of  adding  hydrochloric  acid  is  the  formation  of  an  equiv- 
alent amount  of  acetic  acid.  Inasmuch  as  acetic  acid  is  only  slightly 
ionized,  no  great  change  in  hydrogen-ion  concentration  will  take  place 
until  most  of  the  sodium  acetate  has  been  converted  into  acetic  acid.     If 


HYDROGEN-ION  CONCENTRATION  153 

sodium  hydroxide  is  added  to  an  acetate-acetic  acid  buffer,  the  sodium 
hydroxide  will  react  with  the  acetic  acid. 

(9)  CH3— COOH  +  NaOH  -*  CHsCOONa  +  H2O 

This  reaction  will  convert  sodium  hydroxide,  which  is  a  strong  base, 
into  a  salt  of  a  weak  acid  and  water.  A  slight  increase  in  pH  will  occur, 
owing  to  the  hydrolysis  of  the  sodium  acetate  formed.  After  most  of  the 
acetic  acid  has  reacted  with  sodium  hydroxide,  the  addition  of  more  of 
the  base  will  result  in  a  rapid  increase  in  the  pH  value  of  the  solution. 

In  culturing  fungi,  it  is  important  to  choose  buffers  which  retain  the 
pH  of  the  medium  in  the  desired  range.  The  effective  yH  range  of  buffers 
prepared  from  weak  acids  and  their  salts  is  related  to  the  degree  of  ion- 
ization of  the  acids.  The  more  highly  an  acid  ionizes,  the  lower  will  be 
the  pH  range  of  a  buffer  prepared  from  it  and  one  of  its  salts.  The  degree 
of  ionization  of  weak  acids  is  designated  by  a  term  called  the  ionization 
constant  {Kg).  Mixtures  of  weak  bases  and  their  salts  are  also  buffers. 
A  few  ionization  constants  of  w^eak  acids  are  acetic,  1.8  X  10~^;  carbonic 
(first  hydrogen),  3.5  X  10"^;  phosphoric  (first  hydrogen),  1.1  X  10"^; 
phosphoric  (second  hydrogen),  7.5  X  10^^.  Extensive  data  of  this  kind 
may  be  found  in  various  handbooks  of  chemistry. 

The  ionization  constants  of  weak  acids  may  be  used  to  calculate  the 
effective  pH  range  of  buffers  prepared  from  these  compounds  and  their 
salts  by  means  of  the  following  relation: 

[salt] 


(10)  pH  =  p2v„  +  log 


[acid] 


The  symbol  p/C  is  equivalent  to  log  {l/Ka).  When  the  mole  concen- 
trations of  the  weak  acid  and  its  salt  are  equal,  Eq.  (10)  becomes: 

(11)  pH  =  log  -^  =  p/Va 

The  p/va  value  of  a  weak  acid  is  thus  the  pH  of  a  buffer  which  contains 
equivalent  quantities  of  a  weak  acid  and  one  of  its  soluble  salts.  The 
pH  of  an  acetate  buffer  containing  equivalent  amounts  of  acetic  acid  and 
an  acetate  may  be  calculated  using  Eq.  (11). 

(12)  pH  =  log  ^  g  ^  ^Q_,  =  log  1  +  log  105  -  log  1.8  =  4.74 

In  an  analogous  manner  it  can  be  shown  that  the  pH  of  a  buffer  com- 
posed of  equivalent  amounts  of  a  weak  base  and  one  of  its  salts  is  related 
to  the  ionization  constant  of  the  base  {Kh)  by  the  following  equation: 

(13)  pH  =  14  -  log  ^  =  pA^t, 

For  a  derivation  of  the  formulas  relating  pH  and  ionization  constants 


154 


PHYSIOLOGY  OF  THE  FUNGI 


see  Umbreit  et  al.  (1945).  The  useful  range  of  a  buffer  extends  one  pH 
unit  above  and  below  the  p/v^  (or  p/v^)  value.  The  data  in  Table  31 
giving  the  pH  range  of  a  number  of  buffer  systems  of  biological  interest 
were  calculated  using  Eq.  (11)  and  (13).  The  information  in  Table  31  is 
useful  in  selecting  buffers  which  are  active  in  certain  pH  ranges.  Direc- 
tions for  preparing  buffer  solutions  of  definite  pH  are  given  by  Gortner 
(1949).  In  practice,  media  are  diluted  with  these  buffers  (Lindeberg, 
1944). 

Table  31.     The  Useful  pH  Range  of  a  Number  of  Buffer  Systems  of  Biological 
Interest  Calculated  by  the  Use  of  Eqs.  (11)  and  (13) 


Acid  or  base 

Equilibrium  reaction 

pH  range 

pA'a  or  pA'a 

Acetic 

CH3— COOH  ^  CH3— COO- 

3.7-  5.7 

4.7 

Phosphoric 

H3PO4  ^  H2PO4- 

1.0-  3.0 

2.0 

H2PO4-  ^  HPO4- 

5.7-  7.7 

6.7 

Carbonic 

H2CO3  ^  HCO3- 

5.5-  7.5 

6.5 

Fumaric 

HOOC— CH^CH— COOH  ^ 

HOOC     CH— CH     COO- 

2.0-  4.0 

3.0 

HOOC— CH=CH— COO-  ?=i 

3.5-  5.5 

4.5 

-OOC— CH=CH— COO- 

Malic 

HOOC— CH2— CHOH— COOH  ^ 

HOOC— CHo— CHOH— COO- 

2.4-  4.4 

3.4 

HOOC—CH2— CHOH— COO-  <=± 

4.1-  6.1 

5.1 

-OOC— CHo— CHOH— COO- 

Ammonia 

NH4OH  ^  NH4+  +  OH- 

8.3-10.3 

9.3 

A  number  of  other  considerations  enter  into  the  choice  of  buffers. 
They  must  be  nontoxic,  but  even  a  buffer  composed  of  an  essential 
nutrient  such  as  phosphate  may  be  inhibitory  if  used  in  high  concentra- 
tions. The  use  of  calcium  compounds  in  media  was  found  to  reduce  the 
toxicity  of  phosphate  and  citrate  buffers  to  species  of  Marasmius  (Linde- 
berg,  1944).  The  amount  of  a  buffer  required  to  achieve  a  given  degree 
of  immobilization  of  pH  during  growth  cannot  be  specified  without  con- 
sidering the  fungus  involved  and  the  medium  used.  In  experiments 
where  it  is  desired  to  maintain  the  pH  of  the  culture  medium  essentially 
constant,  the  concentration  of  nutrients,  especially  the  sugar,  in  the 
medium  should  be  low,  and  an  organic  source  of  nitrogen  should  be  used 
in  preference  to  nitrates  and  ammonium  salts.  Lindeberg  (1944)  used 
Af /25  phosphate  buffer  in  a  medium  which  contained  10  g.  of  glucose  and 
1  g.  of  asparagine,  with  satisfactory  results. 

In  general,  the  usefulness  of  highly  buffered  media  is  restricted  to  deter- 
mining pH  limits  and  for  special  problems.  Many  fungi  would  not 
develop  "normally"  in  media  having  a  fixed  pH,  especially  in  certain  pH 
ranges. 


H YDROGEN-ION  CONCENT RA TION 


loo 


If  a  medium  contains  several  buffer  systems,  as  is  frequently  the  case, 
each  buffer  system  will  play  its  role  over  the  pH  range  in  which  it  is  active 
alone.  If  two  buffers  with  overlapping  pH  ranges  are  present,  it  will 
require  more  acid  or  alkali  to  effect  a  change  of  unit  pH  than  if  one  buffer 
were  present.  The  effect  of  buffer  concentration  has  not  been  considered 
in  the  above  discussion.  ^Vhile  the  pH  of  a  buffer  depends  solely  upon 
the  ratio  of  the  concentrations  of  the  weak  acid  and  its  salt,  it  is  obvious 
that  the  amount  of  an  acid  or  base  required  to  change  the  pH  value  of  a 


7.0 


I 

a. 


6.0 


5.0 


^4.0 


I 


•  o"  \ 

5                 f 

> 

.*• " 

t 
1 

1 

/ 

/ 

i 

1 

1 

1 

A 

V 

^ 

•^4--'fT 

y 

1.0  0.8        0.4 

Ml  O.IN  HCl 


0.4 


2.0 


0.8  1.2  1.6 

Ml. O.IN  NAOH 

Fig.  21.  Buffer-capacity  curve.s  of  two  media.  The  dotted  line  was  obtained  by 
titrating  20  ml.  of  glucose-asparagine  medium  with  O.IjV  hydrochloric  acid  and  O.liV 
sodium  hydroxide.  The  pH  was  determined  after  each  addition  of  acid  or  base.  The 
solid  line  was  obtained  in  the  same  way  on  the  above  medium  to  which  10  mg.  of 
neutralized  glutamic  acid  had  been  added,  (Courtesy  of  Robbins  and  Schmitt,  Am. 
Jour.  Botany  32 :  324,  1945.) 


buffer  one  unit  wdll  depend  upon  the  concentrations  of  the  buffer  acid  and 
salt  present.  The  hufer  capacity  of  a  medium  is  measured  by  titrating  a 
definite  volume  of  medium  (usually  100  ml.)  with  standard  acid  and 
alkali.  The  pH  is  measured  after  each  addition  of  acid  or  alkali.  From 
the  curve  drawn  from  these  data  the  buffer  capacity  for  any  range  of  pH 
A^alues  may  be  obtained.  The  curves  in  Fig.  21  illustrate  the  buffering 
capacity  of  two  media  (Robbins  and  Schmitt,  1945).  These  media 
differed  in  the  buffers  present.  The  unsymmetrical  nature  of  the  curves 
is  due  to  the  presence  of  overlapping  buffers. 

The  pH  of  culture  media  may  be  controlled  within  desirable  limits,  in 
some  instances,  by  adding  calcium  carbonate  to  the  medium.  Calcium 
carbonate  is  essentially  insoluble  in  neutral  and  alkaline  media  but  acts 
as  a  neutralizing  agent  for  acids.  The  calcium  carbonate  is  used  up  as 
acid  is  produced  by  a  fungus.  The  degree  of  neutralisation  achieved 
depends  upon  the  amount  of  calcium  carbonate  added  and  whether  the 


156 


PHYSIOLOGY  OF  THE  FUNGI 


cultures  are  agitated.     See  Foster  (1949)  for  a  discussion  of  the  use  of 
calcium  carbonate  in  industrial  microbiological  processes. 

For  fungi  which  have  an  extremely  narrow  pH  range,  the  special  cul- 
ture flask  devised  by  Cantino  (1949)  for  culturing  Blastodadia  pring- 
sheimii  may  be  used  (Fig.  22).     A  base  (or  acid)  is  placed  in  the  side  arm 

and  an  internal  indicator  of  the  desired  pH 
range  is  added  to  the  medium.  A  little  of  the 
base  is  tipped  into  the  culture  flask  as  desired. 
Flasks  with  two  side  arms  may  be  used  so  that 
either  acid  or  base  maybe  added  to  the  culture 
medium. 


Fig.  22.  Flask  designed  for 
the  study  of  glucose  dis- 
similation   by    Blastodadia. 

A,  the  side  arm  containing 
NaOH    for    neutralization; 

B,  sintered-glass  aerator;  C, 
inlet  for  aeration  with  differ- 
ent gas  mixtures;  D,  the 
outlet  for  removal  of  media. 
(Courtesy  of  Cantino,  Am. 
Jour.  Botany  36 :  100,  1949.) 


METHODS  OF  DETERMINING  pH  VALUES 

Only  two  general  methods  of  measuring  pH 
values  will  be  discussed.  The  colorimetric 
method  is  simple,  inexpensive,  and  sufficiently 
accurate  for  most  purposes,  but  it  cannot  be 
used  with  highly  colored  or  turbid  media. 
The  potentiometric  method  using  the  glass 
electrode  is  more  accurate  and  is  often  the 
preferred  method. 

Colorimetric  methods.  The  use  of  indi- 
cators w^hich  change  color  in  response  to 
varying  concentrations  of  hydrogen  ion  is 
the  basis  of  this  method.  Indicators  may 
be  considered  as  weak  acids  or  bases,  and  as 
such  they  act  as  buffers,  but  the  amounts  used  are  so  small  they  do  not 
affect  the  accuracy  of  a  determination.  For  methods  of  measuring  the 
pH  of  unbuffered  solutions  see  Snell  and  Snell  (1948).  The  property  of 
these  indicator  buffers  which  distinguishes  them  from  other  buffers  is 
that  the  colors  of  the  salts  and  free  acids  or  bases  (nondissociated)  are 
different.  Within  the  usable  pH  range,  the  color  of  the  indicator  is  a 
function  of  the  hydrogen-ion  concentration  of  the  medium.  For  exam- 
ple, bromocresol  purple  (p/va,  6.3)  is  yellow  in  solutions  having  pH  values 
of  5.2  or  less  and  purple  at  pH  6.8  or  more.  Within  the  pH  range  5.2 
to  6.8  the  color  changes  from  yellow  to  purple.  To  determine  the  pH 
value  of  an  unknown  solution  within  this  range,  the  indicator  is  added  to 
equal  amounts  of  standard  buffers  and  the  unknown  solution,  and  from 
the  color  of  the  standard  buffers  of  known  pH,  the  pH  value  of  the 
unknown  may  be  estimated  to  within  0.1  pH  unit.  By  a  suitable  choice 
of  indicators  the  pH  range  of  interest  may  be  covered.  A  few  indicators 
with  their  pH  ranges  are  listed  in  Table  32. 

Two  methods  of  color  comparison  are  in  general  use.     The  first  involves 


HYDROGEN -ION  CONCENTRATION 


157 


the  use  of  the  familiar  comparator  block.  A  slight  color  or  turbidity 
of  the  medium  may  be  compensated  for  by  the  use  of  suitable  blanks.  A 
porcelain  spot  plate  may  be  used  instead  of  a  comparator  block  with 
considerable  saving  of  time  and  materials,  although  the  accuracy  is  some- 
what less.  Drops  of  the  indicator  are  added  to  the  depressions  in  the 
spot  plate.  A  drop  of  the  medium  is  added  to  one  depression,  and  drops 
of  standard  buffers  to  the  other  depressions.  The  pH  of  the  medium 
is  estimated  from  the  pH  of  the  buffer  which  yields  a  color  matching  that 
developed  in  the  medium. 

Table  32.     The  pH  Range  and  Color  Changes  of  Various  Indicators 

(Courtesy  of  Eastman  Kodak  Company.) 


Indicator 


Bromophenol  blue . 
Bromocresol  green. 
Chloro phenol  red . . 
Bromocresol  purple 
Bromothymol  blue . 
Phenol  red 


pH  range 


3.0-4.7 
3.8-5.4 
4.8-6.8 
5.2-6.8 
6.0-7.6 
6.8-8.4 


Color  change 


Yellow-blue 

Yellow-blue 

Yellow-red 

Yellow-purple 

Yellow-blue 

Yellow-red 


All  colorimetric  methods  of  measuring  pH  require  the  use  of  standard 
buffers  (buffers  of  known  pH)  or  permanent  standards.  Buffers  may  be 
prepared  in  the  laboratory  or  purchased  from  laboratory  supply  houses. 
It  is  convenient  to  use  prepared  buffer  tablets,  which  need  only  to  be 
dissolved  in  a  measured  amount  of  water  before  use.  Potentiometric 
pH  meters  also  require  the  use  of  a  standard  buffer  for  calibration. 
The  easiest  of  these  to  prepare  is  a  saturated  solution  of  potassium  hydro- 
gen tartrate  (pH  3.57).  The  use  of  this  buffer  was  recommended  by 
Lingane  (1947).  It  is  simple  to  prepare,  and  temperature  affects  the  pH 
very  little. 

From  Table  32  it  will  be  noted  that  the  pH  range  of  a  single  indicator 
is  less  than  two  pH  units.  Much  time  can  be  saved  in  pH  determinations 
by  the  use  of  a  wide-range  indicator  to  determine  the  approximate  pH 
before  using  a  single  indicator  for  the  final  measurement.  Wide-range 
indicators  (pH  range  2  to  10)  may  be  purchased  or  prepared  by  mixing 
suitable  indicators  (Snell  and  Snell,  1948).  The  pH  value  of  a  medium 
may  easily  be  determined  within  0.5  pH  unit  by  the  use  of  a  wide-range 
indicator.  Either  the  comparator  block  or  the  spot-plate  method  may 
be  used.  For  detailed  information  about  indicators,  see  Kolthoff  and 
Rosenblum  (1937). 

Potentiometric  methods.  The  potential  difference  which  develops 
between  certain  pairs  of  electrodes  when  they  are  dipped  into  a  solution 
is  a  function  of  the  hydrogen-ion  concentration.     Solutions  which  give 


158  PHYSIOLOGY  OF  THE  FUNGI 

rise  to  the  same  potential  difference  have  the  same  pH  value.  Modern 
pH  meters  are  calibrated  in  pH  units  so  that  direct  readings  are  obtained. 
Color  or  turbidity  does  not  affect  potentiometric  measurement  of  pH. 

The  glass  electrode  in  conjunction  with  the  calomel  half  cell  is  the  most 
commonly  used  for  liquids  of  biological  interest.  The  glass  electrode 
consists  of  a  bulb  blown  from  a  special  glass.  The  bulb  is  filled  with 
O.IA'^  hydrochloric  acid.  A  potential  difference  develops  between  the 
inside  and  the  outside  of  the  electrode;  the  magnitude  of  this  potential 
difference  depends  upon  the  hydrogen-ion  concentration  of  the  liquid 
in  which  the  bulb  is  dipped.  Measuring  the  potential  difference  which 
develops  between  the  glass  electrode  and  the  calomel  half  cell  is  equivalent 
to  determining  the  pH  value  of  the  unknown  solution.  Sensitive  auxiliary 
electrical  equipment  is  required  to  measure  this  potential  difference. 
For  a  discussion  of  the  glass  electrode,  see  Dole  (1941). 

Many  suitable  pH  meters  are  available.  The  trend  appears  to  be 
toward  instruments  which  use  alternating  current  rather  than  batteries 
as  a  source  of  power.  Since  the  details  of  operation  are  somewhat  dif- 
ferent for  the  various  makes,  the  directions  of  the  manufacturer  should 
be  consulted. 

The  pH  of  media  should  be  determined  before  autoclaving  and  the 
reaction  adjusted  by  the  addition  of  acid  or  alkali  if  necessary.  The  pH 
of  a  sample  of  a  medium  should  also  be  determined  after  autoclaving  and 
before  inoculation.  The  pH  value  at  this  time  is  known  as  the  initial 
pH.  Alkaline  media  absorb  carbon  dioxide  from  the  atmosphere,  causing 
a  slow  decrease  in  pH,  Pritham  and  Anderson  (1937)  reported  that  the 
pH  of  uninoculated  alkaline  media  may  decrease  as  much  as  two  units 
during  the  course  of  an  experiment.  This  factor  is  of  particular  impor- 
tance when  upper  pH  limits  are  being  investigated.  For  methods  of 
adjusting  pH,  see  Suggested  Laboratory  Exercises. 

EFFECTS  ON  FUNGI 

Hydrogen  and  hydroxyl  ions  are  present  in  all  media  and  in  substrates 
upon  which  fungi  grow  in  nature.  The  pH  of  the  medium  exerts  a 
decided  effect  upon  the  rate  and  amount  of  growth  and  many  other  life 
processes.  A  medium  may  have  a  pH  which  is  favorable  for  growth  and 
unfavorable  for  sporulation  or  other  processes.  The  production  of  pig- 
ments, vitamins,  and  antibiotics  may  be  influenced  by  the  pH  of  the 
medium.  As  a  result  of  metabolic  activity  a  fungus  ordinarily  changes 
the  pH  of  the  medium  upon  which  it  grows. 

pH  limits.  The  upper  and  lower  pH  values  between  which  a  fungus 
grows  form  the  pH  range  of  that  species.  The  pH  values  which  inhibit 
growth  vary  with  the  species.  Between  the  limiting  pH  values  there  is 
a  pH  range  which  allows  optimum  growth.     An  initial  pH  of  5  to  6  is 


HYDROGEN-ION  CONCENTRATION 


159 


satisfactory  (not  necessarily  optimum)  for  the  majority  of  the  fungi. 
The  optimum  pH  ranges  for  Blastocladia  pringsheimii,  Allomyces  arhus- 
cula,  and  Blastocladiella  simplex  are  rather  narrow  (Emerson  and  Cantino, 
1948)  (see  Fig.  23).  Most  of  the  pH  optima  reported  in  the  literature 
are  less  than  7.  Meacham  (1918)  reported  pH  3  to  be  optimum  for 
Lenzites  saepiaria,  Fomes  roseus,  Merulius  lacrymans,  and  Coniophora 
cerebella.  Wolpert  (1924)  found  the  pH  optimum  of  various  Basidio- 
mycetes  to  be  in  the  neighborhood  of  5.5.  Johnson  (1923)  reported  that 
the  upper  pH  limit  of  Penicillium  varidbile  is  10.1  to  11.1,  which  is  con- 


Blastocladfa 


Allomyces 


Blastocladiella 


Fig.  23.  Relation  of  pH  of  the  medium  to  growth  of  Blastocladia  -pringsheimii, 
Allomyces  arbuscula,  and  Blastocladiella  simplex.  (Courtesy  of  Emerson  and  Cantino. 
Am.  Jour.  Botany  35:  162,  1948.) 

siderably  higher  than  that  of  most  fungi.  The  lower  pH  limits  reported 
vary  from  5.3  for  B.  simplex  (Emerson  and  Cantino,  1948)  to  0.5  for 
Acontium  velatum  and  an  unidentified  imperfect  fungus  (Starkey  and 
Waksman,  1943). 

The  method  used  to  determine  the  pH  limits  of  a  fungus  is  to  inoculate 
a  series  of  nutrient  solutions  having  pH  values  spaced  0.2  to  0.4  unit 
apart.  Growth  may  be  observed  visually,  or  the  mycelium  may  be 
weighed.  Such  media  should  be  well  buffered.  The  pH  limits  for  a 
given  fungus  as  determined  in  different  laboratories  are  frequently  at 
variance.  This  is  not  unexpected,  since  the  composition  of  the  medium 
and  the  nature  of  the  buffer  influence  the  tolerance  of  fungi  to  hydrogen 
and  hydroxyl  ions.  The  behavior  of  Marasmius  graminum  is  revealing 
(Lindeberg,  1944).  Calcium  ion  was  effective  in  overcoming  the  toxic 
effect  of  an  initial  pH  of  3.3.  The  weight  of  M.  graminum  after  12  days 
was  0.4  mg.,  but  when  calcium  sulfate  was  added  to  the  medium,  the 
yield  was  8.0  mg.  Tamiya  (1928)  also  found  calcium  ion  to  protect 
Aspergillus  oryzae  to  some  extent  against  high  concentrations  of  hydrogen 
ion.  The  optimum  pH  for  Gibber ella  saubinetti  is  lower  when  calcium  is 
present  in  the  medium  (Lundegardh,  1924).     Wolpert  (1S24)  also  found 


160 


PHYSIOLOGY  OF  THE  FUNGI 


the  pH  range  of  many  fungi  to  vary  on  different  media  and  concluded 
that  the  widest  pH  range  was  obtained  on  favorable  media. 

The  temperature  of  incubation  may  influence  the  optimum  pH  as  well 
as  the  pH  range  of  a  fungus.  The  optimum  pH  for  Phacidium  infestans 
is  4.5  at  5°C.,  5.0  at  10°C.,  5.5  at  15°C.  and  6.0  at  20°C.  (Pehrson,  1948). 
The  pH  range  of  Armillaria  mellea  on  a  sucrose-peptone  medium  was 
reported  to  be  2.5  to  7.5  at  15°C.,  2.0  to  7.8  at  25°C.,  and  2.5  to  7.4  at 
35°C.  (Wolpert,  1924). 


21  0 


180 


150 


.120 


■o 

.^    90 

o 


60 


30 


.y^""^* 

^^^0--^ 

— 

1 

^  -.• 

/ 

,.-'-' 

^/-'- 

■'"/ 

ucose-). 

oofossiui 

77  nitrate 

\ 

7 

^  —  ^Gi 

\ 

\ 
\ 
\ 

/ 

> 

^x 

ST  — 

\ 

um  sulfa 

1          X 

/^ 

lucose-ammoni 

fe 

£. xJ 

90 


8.0 


7.0 


60 


5,0 


4.0 


3.0 


20 


25 


0  5  10  15 

Doys  of  incubation 

Fig.  24.  Rate  of  mycelial  growth  of  Sordaria  fimicola  and  accompanying  changes 
in  pH  of  two  media.  Media  contained  biotin  but  no  thiamine.  Sohd  hnes  indicate 
growth,  and  the  broken  lines  represent  pH  values. 

Two  pH  optima  have  been  reported  for  a  number  of  fungi.  Rhizopus 
nigricans,  when  grown  on  potato-glucose  liquid  medium,  has  two  optimum 
pH  ranges,  one  on  either  side  of  the  isoelectric  point  of  the  mycelium, 
which  was  about  pH  5.5  (Robbins,  1924).  Scott  (1924)  reported 
Fusarium  lycopersici  to  have  two  optimum  pH  ranges  for  growth  on 
glucose-nitrate  medium:  pH  4.5  to  5.3  and  5.8  to  6.8.  Mathur  et  al. 
(1950)  obtained  evidence  that  there  are  two  optimum  pH  ranges  for  the 
sporulation  of  Colletotrichum  Undemuthianuni. 

In  addition  to  the  use  of  media  having  low  initial  pH  values,  the  lower 
pH  limit  may  be  determined  in  some  instances  by  choosing  a  medium 
in  which  the  fungus  produces  sufficient  acid  to  inhibit  growth  completely. 
This  is  illustrated  by  the  pH  and  growth  curves  of  Sordaria  fimicola  in 
Fig.  24.  This  fungus  was  grown  upon  a  glucose-ammonium  sulfate 
medium  having  initial  pH  6.0;  after  a  few  days  the  pH  of  the  culture 
medium  fell  to  3.3  and  remained  there  for  5  weeks.  More  difficulty  may 
be  experienced  in  determining  the  upper  pH  limit.     If  a  fungus  is  able 


HYDROGEN-ION  CONCENTRATION 


161 


to  make  a  trace  of  growth  in  an  alkaline  medium,  the  carbon  dioxide 
produced  will  lower  the  pH.  Organic  acids  may  also  be  produced.  Car- 
bon dioxide  from  the  air  will  be  absorbed  by  alkaline  media. 

0.08 


c 

"e 

t  0.06 

0) 


0) 

a. 


■I  0.04 

_>i 

2 

•o 
>« 

o 

%  0.02 

•a 
m 

a. 


' ' 

Urease 

^ 

\ 

\ 

■J 

7 

/ 

.f 

N 

7 
pH 


Fig.  25. 

of  urea 


25.     The  shift  of  optimum  pH  for  urease  activity  due  to  change  in  concentration 
sa.      (Courtesy  of  \'an  Slyke,  Advances  in  Enzymol.  2:  41,  1942.     Pubhshed  by 


permission  of  Interscience  Publishers,  Inc.) 


pH 
Pig.  26.     The  effect  of  pH  on  the  rate  of  linear  growth  of  Neurospora  crmsKi. 
tesy  of  Ryan,  Beadle,  and  Tatum,  Am.  Jour.  Botany  30:  790,  1943.) 


(0,11. 1  V 


It  was  pointed  out  in  Chap.  4  that  pH  affects  the  activity  of  a.i/jymes. 
In  general,  there  is  a  striking  correlation  between  the  optimum  pH  range 
for  most  enzymes  and  the  optimum  pH  range  for  gro^vth.  In  a  survey 
of  the  literature  Haldane  (1930)  found  all  but  9  of  105  enzymes  to  have 
optima  between  pH  4  and  8.     Most  fungi  have  pH  optima  for  growth 


ief2 


PHYSIOLOGY  OF  THE  FUNGI 


between  these  limits.  The  effects  of  pH  upon  the  activity  of  urease 
(Van  Slyke,  1942)  and  upon  the  rate  of  growth  of  Neurospora  crassa 
(Ryan  et  al.,  1943)  are  shown  in  Figs.  25  and  2G.  From  the  general 
similarity  of  these  two  curves  it  appears  probable  that  pH  affects  the  rate 
of  growth  of  fungi,  at  least  in  part,  by  modifying  the  rate  of  certain 
enzymatic  reactions. 

pH  changes  in  media  during  growth.  Fungi,  as  a  result  of  their 
metabolic  activities,  ordinarily  change  the  pH  of  the  media  in  which 
they  grow.  These  changes  cannot  be  studied  by  making  a  single  deter- 
8.0r 


8 


10 


2  3  4  5  6  7 

Doys  of  incubation 

Fig.  27.  Changes  in  pH  during  incubation  of  Sordaria  fimicola  in  different  volumes  of 
liquid  glucose-casein  hydrolysate  medium  at  25°C. 

mination  of  pH  at  any  fixed  time.  Just  as  it  is  necessary  to  study  growth 
as  a  function  of  time  of  incubation  (growth  curve),  it  is  necessary  to  deter- 
mine the  pH  changes  in  an  inoculated  medium  day  after  day  to  obtain  a 
complete  representation  of  these  changes  (pH  curve).  The  pH  of  the 
medium  should  be  followed  in  connection  with  the  other  functions  being 
studied.  Since  fungi  differ  in  their  metabolic  activity  and  rate  of  growth, 
the  pH  changes  produced  in  the  culture  medium  will  differ.  The  pat- 
terns of  pH  changes  for  the  same  fungus  will  depend  upon  the  composition 
and  concentration  of  the  media  used. 

As  an  illustration  of  the  effect  of  the  composition  of  the  medium  upon 
the  pH  changes,  some  of  our  data  for  Sordaria  fimicola  are  given  in  Fig. 
24.  The  correlation  of  the  pH  changes  with  the  rate  and  amount  of 
growth  of  this  fungus  may  be  obtained  by  comparing  the  growth  curves 
obtained  at  the  same  time.     From  Fig.  27  it  is  evident  that  the  hydrogen- 


HYDROGEN-ION  CONCENTRATION  163 

ion  concentration  of  a  nutrient  solution  may  change  10,000-fold  during  a 
few  days  as  a  result  of  the  metabolic  activities  of  a  fungus.  These 
changes  in  pH  are  due  to  changes  in  the  relative  amounts  of  acids  and 
bases  formed  or  withdrawn  and  to  the  ionization  constants  of  these 
compounds.  Some  of  the  metabolic  processes  which  result  in  a  change  in 
pH  of  a  nutrient  solution  are  discussed  below. 

The  utilization  of  cations,  such  as  ammonium  ion,  for  the  synthesis  of 
protoplasm  or  for  any  other  purpose  whereby  essentially  non-ionic  com- 
pounds are  formed,  leaves  an  equivalent  number  of  anions  in  the  nutrient 
solution.  Since  solutions  are  electrically  neutral,  an  equivalent  number 
of  both  cations  and  anions  must  be  present.  Thus,  when  an  equivalent 
of  ammonium  ion  is  transformed  into  non-ionic  compounds,  an  equivalent 
of  some  other  cation  or  cations  will  be  formed  in  the  nutrient  solution. 
These  ''new"  cations  are  usually  hydrogen  ions,  which  are  formed  from 
water.  If  it  is  assumed  that  both  cations  and  anions  are  adsorbed  on 
the  protoplasmic  membrane,  the  process  may  be  thought  of  as  replace- 
ment. The  production  of  acid  would  result  from  the  utilization  of  other 
cations  as  well. 

The  utilization  of  nitrate  ion  or  other  anion  such  as  phosphate  or 
sulfate  for  the  formation  of  non-ionized  compounds  has  the  effect  of 
increasing  the  hydroxyl-ion  concentration  of  the  medium.  We  may 
assume  the  same  type  of  mechanism  as  before,  except  that  the  anion 
released  to  the  nutrient  solution  is  the  hydroxyl  ion. 

Fungi  produce  acids  from  nonacidic  nutrients  such  as  carbohydrates. 
Among  these  acids  are  carbon  dioxide  and  various  organic  acids  such  as 
pyruvic,  citric,  and  succinic  acids.  Carbon  dioxide  combines  with  water 
to  form  carbonic  acid,  which  is  unstable  in  the  presence  of  stronger  acids 
and  decomposes  to  set  free  carbon  dioxide.  Under  alkaline  conditions 
carbonic  acid  reacts  with  bases  to  form  bicarbonates.  Pyruvic  acid 
accumulates  in  the  nutrient  solution  in  which  many  fungi  are  grown,  and 
in  some  instances  the  formation  of  this  acid  accounts  for  a  considerable 
part  of  the  early  depression  of  pH.  The  eventual  utilization  of  pyruvic 
acid  causes  the  pH  of  the  nutrient  solution  to  rise.  Other  metabolizable 
acids  behave  similarly.  Ammonia  is,  perhaps,  the  most  common  basic 
substance  produced  by  fungi.  Piricularia  oryzae  produces  ammonia  in 
considerable  amounts  (Henry  and  Andersen,  1948).  The  production  of 
ammonia  results  from  the  deamination  of  amino  acids  and  proteins.  The 
processes  discussed  above  may  occur  simultaneously.  Whether  a  culture 
solution  becomes  more  acid  or  alkaline  depends  upon  the  extent  of  these 
various  processes.  In  general,  the  processes  which  produce  acid  pre- 
dominate during  early  growth,  especially  when  ammonium  nitrogen  is 
used. 

The  importance  of  the  composition  of  the  medium  in  determining  what 


164  PHYSIOLOGY  OF  THE  FUNGI 

changes  in  pH  will  take  place  during  growth  is  illustrated  by  the  work  of 
Dimond  and  Peltier  (1945),  who  studied  the  pH  changes  produced  by 
Penicillium  notatum  as  a  function  of  the  carbon  and  nitrogen  nutrition. 
When  the  initial  pH  was  6.0  and  sodium  nitrate  was  the  nitrogen  source, 
the  lowest  pH  values  attained  on  different  sugars  were  glucose,  5.1; 
sucrose,  4.0;  lactose,  3.2;  maltose,  4.8;  and  galactose,  4.8.  These  were 
the  lowest  pH  values  attained  under  these  conditions.  In  another  experi- 
ment a  mixture  of  tryptophane,  asparagine,  and  cystine  w^as  used  as  the 
nitrogen  source.  The  pH  again  varied  with  the  sugar  used  in  the  medium 
The  lowest  pH  attained  with  fructose  was  5.3;  glucose  3.5;  sucrose,  4.0; 
and  an  equimolecular  amount  of  fructose  and  glucose,  3.5.  When  lactose 
w^as  used  in  combination  with  these  amino  acids,  the  pH  of  the  culture 
medium  remained  essentially  constant  at  7.0 

Any  changes  in  environmental  factors  which  affect  the  rate  of  growth 
of  a  fungus  may  also  affect  the  changes  in  pH  of  the  culture  medium. 
Robbins  and  Schmitt  (1945)  found  that  the  time  required  for  Phycomyces 
blakesleeanus  to  lower  the  pH  of  a  glucose-asparagine  medium  to  a  given 
level  was  a  function  of  temperature  of  incubation.  Growth  and  the 
production  of  acid  were  more  rapid  at  26°C  than  at  20°C.  The  rate  at 
which  the  pH  of  a  culture  medium  is  changed  by  a  fungus  is  also  depend- 
ent upon  the  volume  of  medium  used  in  flasks  of  the  same  size.  Some  of 
our  data  which  illustrate  this  for  Sordaria  fimicola  are  shown  in  Fig.  27. 
The  time  required  for  these  cultures  to  attain  maximum  weight  and  to 
produce  perithecia  correlated  with  the  changes  in  pH. 

Effect  of  acidity  on  media.  The  composition  of  a  medium  may  be 
changed  as  a  result  of  changing  the  pH.  The  various  cations  and  anions 
may  combine  to  form  insoluble  compounds  at  certain  pH  values.  Mag- 
nesium and  phosphate  ions  are  compatible  in  acidic  solutions,  but  as  the 
concentration  of  hydrogen  ion  is  decreased,  these  ions  combine  to  form 
an  insoluble  compound,  the  solubility  of  which  becomes  less  as  the  pH  is 
increased.  Calcium  phosphate  is  likewise  less  soluble  in  alakline  solu- 
tions. Ferric  iron  may  be  largely  removed  from  media  as  either  the 
hydroxide  or  the  phosphate,  by  making  the  media  alkaline.  If  an  alkaline 
medium  is  filtered,  certain  constituents  will  be  removed  to  a  greater  or 
lesser  extent.  Lilly  and  Leonian  (1945)  found  that  by  making  a  medium 
alkaline  to  pH  8  and  filtering,  the  iron  concentration  w^as  lowered  to  such 
levels  that  Rhizohium  trifolii  made  about  one-fifth  as  much  growth  as 
when  250  jug  of  iron  per  liter  was  added  to  the  medium.  If  a  precipitate 
is  not  removed  by  filtration,  the  situation  is  different.  Any  insoluble 
precipitate  is  in  equilibrium  with  the  dissolved  compound,  as  indicated 
below. 

FeP04  ^  FeP04  ^  Fe+  +  +  +  PO4" 
solid     in  solution  ionized 


HYDROGEN-ION  CONCENTRATION  1C5 

As  the  ions  are  utilized,  more  and  more  of  the  precipitate  will  dissolve. 
The  effect  of  a  change  in  pH  of  the  solution  as  a  result  of  the  metabolic 
activities  of  the  fungus  must  be  considered.  An  acid  reaction  will  hasten 
solution  of  the  precipitate,  while  an  increase  in  alkalinity  will  delay  the 
process.  It  is  possible  that  the  harmful  effects  sometimes  noted  in 
alkaline  media  may  be  due,  in  part,  to  an  induced  iron  deficiency. 

The  influence  of  pH  on  the  solubility  of  certain  ions  may  be  modified 
by  the  presence  of  other  compounds,  especially  those  which  form  com- 
plexes. The  solubility  of  iron  in  alkaline  solutions  is  greatly  increased  in 
the  presence  of  hydroxy  organic  acids  such  as  citric,  tartaric,  and  malic 
acids.  Ammonia  and  amino  acids  also  form  complexes  with  certain  ions, 
e.g.,  copper.  The  presence  of  any  complex-forming  compound  may 
modify  the  availability  of  the  ions  with  which  it  forms  complexes.  The 
chemical  changes  in  media  due  to  alteration  of  pH,  whether  imposed 
from  the  outside  or  caused  by  the  fungus,  affect  metabolic  processes. 
The  pH  of  a  culture  medium  changes  during  the  growth  of  a  fungus,  and 
these  changes  may  affect  the  composition  of  the  medium  and  thus  the 
response  of  the  fungus. 

pH  and  oxygen  supply.  The  solubility  of  oxygen  in  water  is  slight, 
being  less  than  10  mg.  per  liter  at  20°C.  The  rate  of  diffusion  of  oxygen 
into  media  is  dependent  upon  the  composition  and  the  pH.  Rahn  and 
Richardson  (1941)  have  described  a  simple  and  elegant  method  of  measur- 
ing the  rate  of  diffusion  of  oxygen  into  agar  media.  Methylene  blue,  an 
organic  dye  which  is  colorless  when  reduced  and  blue  when  oxidized,  w^as 
used  as  an  indicator.  When  this  dye  (1/200,000)  is  autoclaved  with 
media  which  contain  easily  oxidized  constituents  such  as  glucose,  the  dye 
is  reduced  to  the  leuco,  or  colorless,  form.  As  oxygen  diffuses  into  the 
medium,  the  reduced  dye  is  oxidized,  and  the  rate  at  which  the  blue  zone 
advances  into  the  medium  is  a  measure  of  the  rate  of  oxygen  diffusion. 
The  pH  of  the  medium  also  affects  the  ease  with  which  certain  constitu- 
ents are  oxidized.  Some  data  of  Rahn  and  Richardson  (1941)  on  the  rate 
of  oxygen  diffusion  into  a  peptone  medium  are  shown  in  Fig.  28.  The 
amount  of  oxygen  available  to  submerged  mycelium  is  greater  in  acidic 
than  in  alkaline  media. 

Effect  of  pH  on  utilization  of  nutrients.  Before  any  substance  (ion 
or  molecule)  is  utilized,  it  must  first  pass  through  the  cell  wall  and  the 
protoplasmic  membrane.  The  cell  wall  is  nonliving  and  consists  of 
polysaccharide-like  compounds.  For  a  discussion  of  the  nature  of  the 
cell  wall  and  literature  citations,  see  Brian  (1949).  The  protoplasmic 
membrane  appears  to  be  composed  of  proteins  and  lipoid-protein  com- 
plexes. Proteins  are  colloidal  amphoteric  compounds.  An  amphoteric 
compound  possesses  both  acidic  and  basic  properties  and  may  form  salts 
with  either  acids  or  bases.     The  protoplasmic  membrane  has  acidic 


166 


PHYSIOLOGY  OF  THE  FUNGI 


properties  due  to  carboxyl  and  sulfhydryl  groups  and  basic  properties  by 
virtue  of  having  amino  and  other  basic  groups.  The  protophismic  mem- 
brane, therefore,  should  form  salt-hke  compounds  with  both  cations  and 
anions. 

Bacteria  are  considered  by  McCalla  (1940)  to  act  as  ion-exchange  sub- 
stances, and  fungus  spores  have  been  shown  to  act  in  the  same  manner. 
McCalla  investigated  ion  replacement  by  saturating  cells  of  Escherichia 


60 
Time  in   hours 

Fig.  28.  The  effect  of  hydrogen-ion  concentration  on  the  rate  of  diffusion  of  oxygen 
into  1  per  cent  peptone  in  phosphate  buffer.  Leucomethylcne  bkie  was  used  as  an 
indicator.  The  rate  of  penetration  of  oxygen  with  time  was  followed  by  measuring  the 
depth  of  the  blue  zone.  (Drawn  from  the  data  of  Rahn  and  Richardson,  Jour.  Bad. 
41 :  240,  1941.     By  permission  of  The  Williams  &  Wilkins  Company.) 

coli  with  magnesium  ion  and  tested  the  replacing  effects  of  other  cations. 
Sodium  and  potassium  ions  replaced  only  a  little  magnesium,  while  hydro- 
gen and  calcium  ions  were  much  more  effective. 

From  this  viewpoint  the  relative  amounts  of  the  various  cations 
adsorbed  from  a  medium  would  be  a  function  of  the  concentration  of  the 
ions  present  and  the  relative  affinity  of  the  membrane  proteins  for  the 
different  cations.  The  concentrations  of  the  hydrogen  and  hydroxyl  ions 
in  a  culture  medium  change  during  growth  and  may  act  to  regulate  the 
adsorption  of  other  ions.  The  pH  of  the  culture  medium  may  alter  the 
relative  adsorption  of  other  ions  which  are  essential  to  nutrition  or  which 
are  toxic.  At  the  lower  pH  limit  the  protoplasmic  membrane  may  be  so 
thoroughly  saturated  with  hydrogen  ions  that  the  essential  cations  are 
unable  to  enter  the  cell  in  adequate  amounts.     The  same  situation  would 


HYDROGEN-ION  CONCENTRATION  1G7 

exist  at  the  limiting  alkaline  pH  values,  except  that  it  is  the  adsorption  of 
essential  anions  which  would  be  hindered  by  hydroxyl  ions. 

A  satisfactory  explanation  of  all  the  phenomena  involved  in  cell  perme- 
ability is  lacking.  It  is  known  that  the  external  pll  affects  the  absorption 
of  various  compounds,  particularly  those  which  ionize.  The  mycelium 
of  Aspergillus  niger  takes  up  acid  dyes,  such  as  light  green  and  methyl 
orange,  when  the  external  pH  is  3.1  or  less.  Basic  dyes  such  as  methylene 
blue  and  neutral  red  are  absorbed  only  when  the  external  pH  is  greater 
than  3.1  (Biinning,  193G).  These  dyes  escaped  from  the  cells  only  when 
the  external  pH  was  in  the  same  range  in  which  these  dyes  were  absorbed. 

Wyss  ct  al.  (1944)  found  the  utilization  of  p-aminobenzoic  acid  by  a 
deficient  mutant  of  Neurospora  crassa  to  be  greatly  increased  in  acidic 
media.  The  ionization  constant  of  p-aminobenzoic  acid  is  about 
2  X  10'"''  {pKa,  about  4.8).  Therefore,  at  pH  3.8  about  90  per  cent  of 
the  metabolite  would  exist  in  the  form  of  the  free  acid,  and  at  pH  5.8  only 
10  per  cent  would  be  in  this  form.  It  was  found  that  about  eight  times 
as  much  of  this  vitamin  was  required  at  pH  6.0  as  at  pH  4.0  to  support  the 
same  amount  of  growth  (see  Fig.  41).  On  theoretical  grounds,  it  is 
probable  that  the  pH  of  the  medium  would  affect  the  utilization  of  other 
vitamins  which  are  weak  acids  (biotin,  pantothenic  and  nicotinic  acids). 

The  external  pH  of  the  medium  has  been  shown  to  affect  the  internal 
pH  of  fungus  cells.  By  changing  the  external  pH  and  by  using  indica- 
tors, Biinning  (1936)  found  the  internal  pH  of  Aspergillus  niger  cells  could 
be  changed  between  4.2  and  5.0  without  injuring  the  cells.  Greater 
changes  in  internal  pH  were  possible,  but  injury  and  death  ensued. 
Armstrong  (1929)  crushed  the  fruit  bodies  of  a  number  of  fleshy  fungi 
and  measured  the  pH  of  the  expressed  juice.  The  pH  range  of  these 
liquids  was  5.9  to  6.2.     At  best  these  are  but  average  values. 

It  is  well  known  that  the  external  pH  may  affect  certain  processes 
within  the  fungus  cells.  For  example,  growth  of  Sordaria  fimicola  in 
glucose-casein  hydrolysate  medium  was  slow  when  the  initial  pH  of  the 
medium  was  4.0,  but  when  the  initial  pH  of  the  medium  was  3.6  to  3.8, 
normal  development  did  not  occur  (Lilly  and  Barnett,  1947).  This 
failure  to  grow  was  traced  to  a  thiamine  deficiency,  for  when  thiamine 
was  added  to  the  medium  (initial  pH  3.6  to  3.8),  normal  growth  and 
perithecial  formation  took  place.  It  appears  possible  that  the  low  exter- 
nal pH  may  have  lowered  the  internal  pH  to  such  an  extent  that  the 
synthesis  of  thiamine  was  prevented  (this  fungus  is  self-sufficient  for 
thiamine  when  the  pH  of  the  medium  is  4  or  greater).  These  effects  are 
shown  in  Figs.  38  and  40.  Additional  evidence  indicated  that  these 
conclusions  are  correct,  for  pyruvic  acid  accumulated  in  the  culture 
medium  when  the  initial  pH  was  3.6  to  3.8.  On  the  addition  of  thiamine 
this  acid  disappeared  from  the  culture  medium. 


168  PHYSIOLOGY  OF  THE  FUNGI 


SUMMARY 


All  aqueous  solutions  contain  hj^drogen  and  hydroxyl  ions.  Hydrogen- 
ion  concentration  is  most  often  expressed  in  terms  of  S0rensen's  scale  of 
pH.  The  pH  scale  is  logarithmic.  Acidity  and  alkalinity  are  expressed 
on  the  same  scale.  A  pH  of  7  indicates  equivalent  concentrations  of 
hydrogen  and  hydroxyl  ions.  Values  of  less  than  7  indicate  acidity,  and 
pH  values  greater  than  7  indicate  alkalinity.  The  smaller  the  pH  values, 
the  greater  the  concentration  of  hydrogen  ions. 

Buffers  are  substances  which  tend  to  maintain  the  pH  of  a  solution 
constant  when  either  strong  acid  or  strong  alkali  is  added  or  when  the 
solution  is  diluted  with  water.  Mixtures  of  weak  acids  or  bases  and 
their  soluble  salts,  and  amphoteric  compounds  such  as  amino  acids  and 
proteins  act  as  buffers.  The  pH  range  over  which  a  given  buffer  is  effec- 
tive is  a  function  of  the  ionization  constant  of  the  weak  acid  (Ka)  or  base 
(Kb)  from  which  the  buffer  is  made.  The  effective  pH  range  of  a  simple 
buffer  is  two  pH  units. 

The  upper  and  lower  pH  values  between  which  a  fungus  is  able  to  grow 
is  called  the  pH  range.  The  pH  ranges  of  various  species  are  different. 
Fungi  generally  tolerate  more  acid  than  alkali.  The  optimum  pH  range 
may  be  different  for  growth  and  sporulation.  The  pH  of  a  medium  in 
which  a  fungus  is  growing  may  change.  High  buffer  concentration  and 
limited  growth  may  keep  the  changes  in  pH  at  a  minimum.  To  follow 
the  changes  in  pH  of  a  culture  medium,  frequent  determinations  should 
be  made. 

Four  metabolic  processes  operate  to  change  the  pH  of  a  culture  medium : 
(1)  utilization  of  cations,  (2)  utilization  of  anions,  (3)  formation  of  acids 
from  neutral  metabolites  (especially  carbohydrates),  and  (4)  formation 
of  bases  (especially  ammonia)  from  amino  acids  and  proteins.  The  net 
change  in  pH  is  the  result  of  the  interaction  of  all  of  these  processes, 

REFERENCES 

Armstrong,  J.  I.:  Hydrogen-ion  phenomena  in  plants.     Hydrion  concentration  and 

buffers  in  the  fungi,  Protoplasma  8:  222-260,  1929. 
Brian,  P.  W.:  Studies  on  the  biological  activity  of  griseofulvin,  Ann.  Botany  (N.S.) 

13:59-77,  1949. 
BiJNNiNG,  E, :  Ueber  Farbstoff-  und  Nitrataufnahme  bei  Aspergillus  niger,  Flora 

131:87-112,  1930. 
Canting,  E.  C:  The  physiology  of  the  aquatic  Phycomycete,  Blastocladia  Pring- 

sheimii,  with  emphasis  on  its  nutrition  and  metabolism,  Am.  Jour.  Botany  36: 

95-112,  1949. 
*DiMOND,  A.  E.,  and  G.  L.  Peltier:  ControUing  the  pH  of  cultures  of  PenicilUum 

notatum  through  its  carbon  and  nitrogen  nutrition.  Am.  Jour.  Botany  32  :  46-50, 

1945. 
Dole,  M.:  The  Glass  Electrode.     Methods,  Applications,  and  Theory,  John  Wiley  & 

Sons,  Inc.,  New  York,  1941. 


HYDRUUEN-ION  CONCENTRATION  169 

^Emerson,  R.,  and  E.  C.  Canting:  The  isolation,  growth,  and  metabohsm  of  Blasto- 
cladia  in  pure  culture.  Am.  Jour.  Botany  35:  157-171,  1948. 
Foster,  J.  W.:  Chemical  Activities  of  Fungi,  Academic  Press,  Inc.,  New  York,  1949. 
GoRTNER,  R.  A.:  Outhnes  of  Biochemistry,  3d  ed.,  John  Wiley  &  Sons,  Inc.,  New- 
York,  1949. 
Haldane,  J.  B.  S.r  Enzymes,  Longmans,  Roberts  and  Green,  London,  1930. 
Henry,  B.  W.,  and  A.  L.  Andersen:  Sporulation  by  Piricularia  oryzae,  Phytopathol- 
ogy 38:  265-278,  1948. 
Johnson,  H.  W.:  Some  relationships  between  hydrogen  ion,  hydroxyl  ion  and  salt 

concentration  and  the  growth  of  seven  soil  molds,  loiva  State  Coll.  Ayr.  Mech, 
Arts  Research  Bull.  76,  1923. 
KoLTHOFF,  I.  M.,  and  C.  Rosenblum:  Acid-base  Indicators,  The  Macmillan  Com- 
pany, New  York,  1937. 
*LiLLY,  V.  G.,  and  11.  L.  Barxett:  The  influence  of  pH  and  certain  growth  factors 

on  mycelial  growth  and  perithecial  formation  by  Sordaria  fimicola,  Am.  Jour. 

Botany  34:  131-138,  1947. 
Lilly,  V  G.,  and  L.  LI.  Leonian:  The  interrelationship  of  iron  and  certain  accessory 

factors  in  the  growth  of  Rhizobium  trifolii  strain  205,  Jour.  Bad.  50:  383-395, 

1945. 
Lindeberg,   G.:  Ueber  die   Physiologie  Ligninabbauender  Bodenhymenomyzeten, 

Symbolae  Botan.  Upsalienses  8(2):  1-183,  1944. 
LiNGANE,  J.  J. :  Saturated  potassium  hydrogen  tartrate  solutions  as  a  pH  standard. 

Anal.  Chem.  19:  810-811,  1947. 
LuNDEGARDH,   H. :  Dcr  Einfluss  der  WasserstofRonenkonzentration  in  Gegenwart 

von  Salzen  auf  das  Wachstum  von  Gibherella  Saubinetti,  Biochem.  Zeit.  146: 

564-572,  1924. 
McCalla,  T.  M.:  Cation  adsorption  by  bacteria.  Jour.  Bad.  40:  23-32,  1940. 
Mathur,  R.  S.,  II.  L.  Barnett,  and  V.  G.  Lilly:  Sporulation  of  Colletotrichum 

Imdemuthianum  in  cultvire.  Phytopathology  40:  104-114,  1950. 
Meacham,  M.  R.:  Note  upon  the  hydrogen  ion  concentration  necessary  to  inhibit 

the  growth  of  four  wood-destroying  fungi,  Science  48:  499-500,  1918. 
Pehrson,  S.  O.:  Studies  on  the  growth  physiology  of  Phacidium  infestans  Karst., 

Physiologia  Plantar um  1:  38-56,  1948. 
Pritham,  G.  H.,  and  A.  K.  Anderson:  The  carbon  metabolism  of  Fusarium  lyco- 

persici  on  glucose,  Jour.  Agr.  Research  55 :  937-949,  1937. 
Rahn,  O.,  and  G.  L.  Richardson:  Oxygen  demand  and  oxygen  supply.  Jour.  Bad. 

41:  225-249,  1941. 
RoBBixs,  W.  J.:  Isoelectric  points  for  the  mycelium  of  fungi,  Jour.  Gen.  Physiol. 

6:  259-271,  1924. 
*RoBBiNS,  W.  J.,  and  M.  B.  Schmitt:  Factor  Z2  and  gametic  reproduction  by  Phy- 

comyces,  Am.  Jour.  Botany  32 :  320-326,  1945. 
Ryan,  J.  F.,  G.  W.  Beadle,  and  E.  L.  Tatum:  The  tube  method  of  measuring  the 

growth  rate  of  Neurospora,  Am.  Jour.  Botany  30 :  784-799,  1943. 
*Scott,  I.  T.:  The  influence  of  hydrogen-ion  concentration  on  the  growth  of  Fusarium 

lycopersici  and  on  tomato  wilt,  Missouri  Coll.  Agr.  Research  Bull.  64,  1924. 
Snell,  F.  D.,  and  C.  T.  Snell:  Colorimetric  INIethods  of  Analysis,  3d  ed.,  D.  Van 

Nostrand  Company,  Inc.,  New  York,  1948. 
Starkey,  R.  L.,  and  S.  A.  Waksman:  Fuiigi  tolerant  to  extreme  acidity  and  high 

concentrations  of  copper  sulfate,  Jour.  Bud.  4C :  509-519,  1943. 
Tamiya,   H.:  Studi^n  liber  die  Stoffwechseiphysiologie  von  Aspergillus  oryzae.  II. 

Acta  Phyiochim.  {Japan)  4:  77-213,  1928. 
Umbreit,  W.  W.,  R.  IL  Burris,  and  -J.  F.  Staufj^es?  Manomptrjc  Techniques  and 


170  PHYSIOLOGY  OF  THE  FUNGI 

Related  Methods  for  the  Study  of  Tissue  MetaboUsm,  Burgess  Publishing  Co, 

Minneapolis,  1945.  ,    ,    .   ,        .  .1    j 

Van  Slyke,  D.  D.:  The  kinetics  of  hydrolytic  enzymes  and  their  bearmg  on  methods 

of  measuring  enzyme  activity,  Advances  in  Enzymol.  2  :  33-47,  1942. 
WoLPERT    F.  S.:  Studies  in  the  physiology  of  the  fungi.     XVII.  The  growth  of 

certain  wood-destroying  fungi  in  relation  to  the  H-ion  concentration  of  the 

media,  Ann.  Missouri  Botan.  Garden  11:  43-97,  1924. 
Wyss,  O.,  V.  G.  Lilly,  and  L.  H.  Leonian:  The  effect  of  pH  on  the  availability  of 

p-am'inobenzoic  acid  to  Neurospora  crassa,  Science,  99 :  18-19,  1944. 


CHAPTER  9 
VITAMINS  AND  GROWTH  FACTORS 

It  is  kno^\^l  that,  for  normal  growth  and  development,  animals  and 
man  require  in  their  diet  minute  amounts  of  certain  organic  compounds, 
in  addition  to  those  which  yield  energy  or  are  used  for  structural  purposes. 
Similarly,  certain  fungi  must  obtain  from  the  substrate  some  of  the  same 
substances  for  growth,  reproduction,  and  other  vital  functions.  Other 
fungi  are  able  to  synthesize  these  compounds,  which  are  called  groivth 
factors,  or  vitamins.  Both  terms  have  often  been  applied  to  the  same 
compounds,  although  the  terms  are  not  always  synonymous.  Originally, 
the  term  vitamin  was  applied  to  the  accessory  factors  in  animal  nutrition, 
and  some  workers  would  restrict  its  use  to  animals  and  man.  The  term 
growth  factor  has  a  somewhat  broader  connotation  than  vitamin.  It 
includes  the  components  and  derivatives  of  some  vitamins,  as  well  as 
other  compounds  which  cannot  be  classified  otherwise  at  present.  The 
chemical  names  of  the  vitamins  also  may  be  used. 

GENERAL  CONSIDERATIONS 

A  number  of  vitamins,  such  as  thiamine  and  biotin,  have  been  shown 
to  perform  definite  functions  in  fungi  as  well  as  in  animals,  and  there  is 
no  reason  to  assume  that  the  fundamental  functions  in  the  two  groups 
of  organisms  are  essentially  different.  The  characteristic  features  of  a 
growth  factor  (vitamin)  include  the  following:  (1)  its  organic  nature;  (2) 
its  activity  in  minute  quantities;  (3)  its  catalytic  action;  (4)  the  specificity 
of  its  action.  It  is  known  that  some  vitamins  are  components  of  enzyme 
systems,  and  it  may  be  assumed  that  all  act  in  this  way. 

In  the  fungi  the  relative  effects  of  the  presence  of  vitamins  in  the 
medium  usually  are  measured  by  the  resultant  vegetative  growth, 
although  vitamins  are  known  to  affect  reproduction  and  other  processes. 
Needless  to  say,  studies  of  vitamin  deficiencies  must  be  carried  out  under 
carefully  controlled  conditions,  using  clean  glassware,  purified  chemicals, 
and  precaution  against  contamination.  Despite  all  precautions  possible, 
variable  results  often  occur,  and  tests  may  need  to  be  repeated  several 
times  before  the  vitamin  deficiencies  of  some  fungi  can  be  definitely 
determined. 

SYNTHESIS  OF  VITAMINS  BY  FUNGI 

Many  fungi  are  able  to  grow  and  develop  normally  on  a  substrate  con- 
taining no  vitamins.     For  example,  Aspergilhis  niger  grows  well  on  a 

171 


172 


PHYSIOLOGY  OF  THE  FUNGI 


synthetic  medium  composed  of  pure  chemicals  (glucose,  asparagine,  salts, 
and  micro  elements).  Phycomyces  hlakesleeaniis  makes  no  growth  on 
this  medium  unless  thiamine  is  added.  We  may  conclude  that  A.  niger 
either  does  not  need  thiamine  in  its  metabolism  or  is  capable  of  synthesiz- 
ing from  the  compounds  of  the  medium  all  vitamins  in  sufficient  quantities 
to  meet  its  needs.  The  growth  of  P.  blakesleeanus  on  the  culture  filtrate 
of  A.  niger  is  proof  that  thiamine  is  synthesized  by  the  latter  species. 
Thus,  A.  niger  may  be  called  a  self-sufficient  fungus  with  respect  to 
vitamins.  Schopfer  (1943)  has  applied  the  term  autotrophic  with  respect 
to  vitamins  to  this  group  of  organisms.  The  detection  of  self-sufficient 
fungi  in  the  laboratory  is  dependent  upon  their  ability  to  grow  on  vitamin- 
free  synthetic  media  containing  suitable  sources  of  carbon  and  nitrogen. 
A  discussion  of  the  economic  importance  of  certain  vitamins  as  metabolic 
products  of  fungi  is  given  in  Chap.  13. 

Some  fungi  which  have  been  reported  to  be  self-sufficient  with  respect 
to  vitamins  are  listed  below: 


Aspergillus  (most  species  tested) 
Basisporium  gallarum 
Botrytis  allii 
Cercospora  apii 
C.  beticola 

Chaetomium  globosum 
Cordyceps  militaris 
Daldinia  concentrica 
Fusarium  (most  species  tested) 
Glomerella  cingulata 
Helminthosporium  gramineum 


H.  victoriae 

Monascus  purpurea 

Monilinia  fructicola  (some  isolates) 

Neocosmopara  vasinfecta 

Penicillium  (most  species  tested) 

Phoma  betae 

Rhizopus  nigricans 

Sclerotinia  sclerotiorum 

Septoria  nodorum 

Sphaeropsis  malorum 

Ustilago  striiformis 


Growth  curves  of  Chaetomium  globosum  are  presented  in  Fig.  29,  as  an 
example  of  a  self-sufficient  fungus.  It  is  evident  that  good  mycelial 
growth  was  made  in  the  vitamin-free  medium  and  that  the  addition  of 
four  vitamins  caused  no  significant  increase  in  the  rate  of  growth  at  any 
time. 

VITAMIN  DEFICIENCIES  IN  FUNGI 

As  pointed  out  above,  some  fungi  do  not  grow  on  synthetic  media 
composed  of  pure  chemicals,  because  they  are  unable  to  synthesize  certain 
vitamins.  These  fungi  have  been  called  variously  vitamin-deficient, 
vitaminless,  or  heterotrophic  with  respect  to  one  or  more  specific  vitamins. 
We  prefer  to  use  the  term  vitamin-deficient,  following  Robbins  and 
Kavanagh  (1942).  Vitamin  deficiencies  among  the  fungi  have  been 
detected  only  for  certain  members  of  the  water-soluble  B-complex  group. 
The  most  common  vitamins  involved  are  thiamine,  biotin,  inositol, 
pyridoxine,  nicotinic  acid,  and  pantothenic  acid.  Vitamin  deficiencies 
can  be  detected  accurately  only  on  synthetic  media  which,  other  than 


VITAMINS 


173 


E 

O 


-^^■"^^ 

^ 

^^^— — 1 

250 

•'^ 

^"^ 

y^'* 

4  vitamins—  _^ 
present 

>^ 

^ 

200 

A 

r/^ 

^^ 

150 

> 
/ 

(/ 

// 

1 

//" 

1 

~ Without  vitamins 

100 

// 

// 

> 

/ 

50 

// 

// 

n 

^ 
^ 

10 


12 


14 


0  2  4  6  8 

Days  of  incubation 

Fig.  29.  Growth  curves  of  Chaetomiurn  globosum,  a  self-sufficient  fungus,  in  25  ml.  of 
liquid  glucose-casein  hydrolysate  medium  in  the  absence  of  vitamins  and  when 
thiamine,  biotin,  inositol,  and  pyridoxine  were  added. 


Fig.  30.  Mutualistic  symbiosis  with  regard  to  vitamins.  Phycomyces  blakesleeanus, 
thiamine-deficient,  inoculated  on  the  right  and  Sordaria  fimicola,  biotin-deficient, 
inoculated  on  the  left.  Both  fungi  made  only  slight  growth  until  the  two  colonies  met. 
Note  the  perithecia  of  Sordaria  on  the  right  and  the  sporangiophores  of  Phycomyces  on 
the  left.  Each  fungus  excreted  into  the  medium  the  vitamin  which  the  other  could  not 
synthesize. 


174  I'HYSIOLOGY  OF  THE  FUNGI 

vitamins,  meet  all  the  requirements  for  normal  growth  and  development 
of  the  fungus  under  study.  The  effect  of  one  deficient  fungus  on  another 
is  shown  in  Fig.  30. 

Methods  of  detecting  vitamin  deficiencies.  Tests  for  vitamin  deficiencies 
of  fungi  are  not  difficult  to  perform,  but  they  do  require  clean  glassware 
and  careful  preparation  of  media.  It  is  convenient  to  conduct  pre- 
liminary experiments  using  only  the  four  vitamins  (thiamine,  biotin, 
inositol,  pyridoxine)  for  which  fungi  are  most  frequently  deficient.  A 
greater  number  of  vitamins  may  be  included  in  subsequent  tests  if  a 
fungus  does  not  grow  well  on  any  of  the  media  first  used.  Either  agar 
or  liquid  media  may  be  used,  and  the  visual  measure  of  growth  is  satis- 
factory for  the  screening  tests.  A  simple  and  convenient  method  for 
preliminary  tests  for  deficiencies  in  filamentous  fungi  isolated  from  nature 
is  by  the  use  of  agar  media  in  test  tubes,  as  shown  in  Fig.  31.  Slight 
growth  on  agar  media  without  added  vitamins  may  be  due  to  impurities 
in  the  medium.  A  high  percentage  of  the  deficiencies  will  be  detected 
by  this  method,  since  deficiencies  for  only  one  or  two  vitamins  are  com- 
mon among  the  filamentous  fungi.  After  the  deficiencies  have  been 
identified  by  preliminary  experiments,  it  is  then  highly  desirable  to  grow 
a  fungus  in  liquid  media,  so  that  the  mycelium  may  be  harvested  and 
dry  weights  determined  (see  Suggested  Laboratory  Exercises  for  direc- 
tions). The  casein  hydrolysate-glucose  medium,  given  in  Chap.  10,  has 
proved  quite  satisfactory  for  accurate  vitamin  studies.  From  the  dry 
weights  of  cultures  determined  at  intervals  throughout  the  growth  period 
of  a  fungus,  growth  curves  may  be  plotted.  Such  curves  are  necessary 
for  accurate  interpretations  of  the  effects  of  vitamins  in  the  medium. 

A  somewhat  different  method  is  used  by  Burkholder  (1943)  for  defi- 
ciency studies  of  yeasts,  where  deficiencies  for  more  than  two  vitamins 
are  common.  This  method  is  illustrated  in  Fig.  32.  A  deficiency  is 
detected  by  the  inability  to  grow  in  a  medium  which  is  complete  except 
for  one  vitamin.  Failure  to  grow  in  a  medium  indicates  a  deficiency  for 
the  vitamin  omitted.  Liquid  media  in  test  tubes  are  used  for  yeasts, 
so  that  growth  may  be  measured  by  photoelectric  colorimeter. 

Total  and  partial  deficiencies.  Phycomyces  hlakesleeanus  was  widely 
used  in  the  early  studies  of  thiamine.  Schopfer  established  the  deficiency 
for  thiamine  and  determined  the  requirements  for  this  vitamin.  Schopfer's 
graph  (Fig.  33)  shows  the  growth  curves  of  the  fungus  over  a  period  when 
different  amounts  of  thiamine  were  added  to  the  basal  culture  medium 
(Schopfer,  1943).  The  fact  that  no  growth  occurred  in  the  medium 
lacking  thiamine  is  not  shown  by  the  graph.  An  increase  in  both  the 
rate  and  the  total  amount  of  growth,  as  the  amount  of  thiamine  is 
increased,  is  clearly  shown  between  the  fifth  and  seventh  days.  Thus 
P.  hlakeslseanus  is  unable  to  synthesize  thiamine,  which  it  must  obtain 


VITAMINS 


175 


A  B  C  D  E 

Fig.  31.  Method  of  detecting  common  vitamin  deficiencies  of  filamentous  fungi. 
Deficiencies  are  evident  by  failure  to  grow  on  media  lacking  the  necessary  vitamin  or 
vitamins.  The  above  media  contained:  A,  no  vitamins;  B,  thiamine;  C,  bio  tin;  D, 
thiamine  and  biotin;  E,  thiamine,  biotin,  inositol,  and  pyridoxine.  The  fungi,  from 
top  to  bottom,  are  Ceratostomella  fimbriata,  Sordaria  fimicola,  Pleurage  curvicolla,  and 
C.  ulmi. 


1 


176 


PHYSIOLOGY  OF  THE  FUNGI 


Fig.  32.  Method  of  detecting  multiple  vitamin  deficiencies  of  yeasts.  Failure  to 
grow  in  the  absence  of  a  particular  vitamin  indicates  a  deficiency  if  the  culture  grew  in 
a  medium  supplied  with  a  combination  of  vitamins.  Growth  of  a  strain  of  Saccharo- 
myces  cerevisiae  (above)  and  Mycoderma  valida  (below)  after  5  days  at  25°C.  From 
left  to  right  the  vitamin  supplements  were:  Tube  1,  none;  2,  less  thiamine;  3,  less 
pantothenic  acid;  4,  less  pyridoxine;  5,"  less  inositol;  6,  less  bio  tin;  7,  less  nicotinic 
acid ;  8,  all  six  vitamins. 


VITAMINS 


177 


from  its  substrate.  Figure  33  emphasizes  two  important  features  which 
must  be  considered  in  vitamin  studies:  (1)  the  effects  of  different  amounts 
of  the  vitamin  in  the  medium,  and  (2)  the  response  of  the  fungus  over  a 
period  of  time  sufficiently  long  to  allow  maximum  growth.  The  three- 
dimensional  graph  permits  one  to  plot  dry  weight  against  both  variables. 
The  failure  of  a  fungus  to  make  an  appreciable  amount  of  growth  even 
after  an  extended  period  of  incubation  on  a  medium  essentially  free  of  a 


Asparagine  0.1  % 


r,         9       ^ 


Mg. 
^90 


Fig.  33.  Three-dimensional  graph  showing  growth  of  Phycomyces  blakesleeanus  on  a, 
synthetic  medium  as  a  function  of  thiamine  concentration  and  time.  (Courtesy  of 
Schopfer,  Protoplasma  28:  383,  1937;  also  from  the  book  "Plants  and  Vitamins," 
p.  102,  1943.     PubUshed  by  permission  of  Chronica  Botanica  Co.) 


particular  vitamin,  like  the  case  illustrated  by  P.  hlakesleeanus  and 
thiamine,  indicates  that  the  deficiency  is  total;  i.e.,  the  synthesis  of  that 
vitamin  is  zero.  Vitamin  deficiencies  of  many  fungi  are  only  partial,  as 
shown  by  a  slower  rate  of  growth  in  a  vitamin-free  medium  than  in  the 
presence  of  added  vitamins.  The  degree  of  partial  deficiency  may  vary 
M^dely,  from  slight  to  nearly  total.  Partial  deficiencies  may  be  easily 
overlooked  by  terminating  an  experiment  too  soon.  An  incubation 
period  of  1  or  2  months  is  often  required  to  distinguish  between  partial 
and  total  deficiencies  of  some  fungi. 

An  example  of  partial  thiamine  deficiency  is  illustrated  by  Lenzites 
trabea  (Fig.  34).     In  a  medium  containing  thiamine,  maximum  Aveight 


178 


PHYSIOLOGY  OF  THE  FUNGI 


was  attained  in  20  days,  while,  in  medium  lacking  thiamine,  the  fungus 
required  approximately  40  days  to  reach  the  maximum  weight.  This  is 
attributed  to  the  slow  rate  of  synthesis  of  thiamine.  Other  isolates  of 
L.  trahea  showed  varying  degrees  of  partial  deficiency  (Lilly  and  Barnett, 
1948). 

Single  and  multiple  deficiencies.  The  above  discussion  has  dealt  with 
examples  of  single  deficiencies  (for  a  single  vitamin).  For  example, 
Sordaria  fimicola  is  deficient  only  for  biotin  (Fig.  31),  Lenzites  trahea. 


EO  30  40 

Incubation  (days) 

Fig.  34.  Growth  of  a  haploid  isolate  of  Lenzites  trahea  and  change  in  pH  of  liquid 
glucose-casein  hydrolysate  medium  at  25°C.,  with  and  witliout  the  addition  of 
thiamine.  These  curves  illustrate  a  partial  deficiency  for  thiamine.  (After  Lilly  and 
Barnett,  Jour.  Agr.  Research  77:  290,  1948.) 

Ceratostomella  Hmhriata  (Fig.  31),  and  Phycomyces  blakesleeanus  for 
thiamine  only.  On  the  other  hand,  some  fungi  have  multiple  deficiencies 
(for  two  or  more  vitamins).  These  may  be  total  or  partial.  An  illustra- 
tion of  multiple  deficiency  is  furnished  by  Sclerotinia  camelliae  (Fig.  35). 
Little  or  no  growth  occurred  on  vitamin-free  medium  or  that  containing 
either  thiamine  or  biotin  alone;  the  fungus  grew  well  only  in  media  con- 
taining both  thiamine  and  biotin.  When  inositol  also  was  added,  growth 
was  consistently  better  than  in  the  presence  of  the  two  vitamins.  This 
indicates  a  partial  deficiency  for  inositol,  in  addition  to  the  total,  or  near 
total,  deficiencies  for  thiamine  and  biotin.  Pyridoxine,  when  added  to 
the  other  three  vitamins,  had  little  or  no  effect  on  growth  under  these 
conditions. 

Other  examples  of  multiple  vitamin  deficiencies  are  common.     Pleurage 
curvicolla  (Fig.  31),  Chaetomium  convolutum,  Coemansia  interrupta,  and 


VITAMINS 


179 


400 


10  12  14  16 

Days  of  incubation 

Fig.  35.  Growth  of  Sclerotinia  camelliae  in  25  ml.  of  liquid  glucose-casein  hydrolysate 
medium  at  25°C.  Note  the  nearly  total  deficiency  for  biotin  and  the  partial  deficiency 
for  inositol.  Failure  to  grow  in  thiamine  alone  and  in  the  absence  of  vitamins  indicates 
total  deficiency  for  thiamine. 


6  12  18  24 

Days  of  incubation 

Fig.  36.  Growth  of  Lambertella  pruni  in  25  ml.  of  liquid  glucose-casein  hydrolysate 
medium  containing  various  vitamins.  Partial  deficiencies  for  both  thiamine  and 
biotin  are  evident,  being  greater  for  thiamine.  Note  that  the  addition  of  inositol 
and  pyridoxine  to  media  containing  thiamine  and  biotin  depressed  growth. 


180 


PHYSIOLOGY  OF  THE  FUNGI 


Ophiobolus  graminis  are  highly  or  totally  deficient  for  both  thiamine  and 
biotin.  Partial  deficiencies  for  both  thiamine  and  biotin  are  illustrated 
by  Lambertella  pruni  (Fig.  36).  Slight  growth  in  the  control  and  excellent 
growth  only  in  media  containing  both  thiamine  and  biotin  identify  the 
deficiencies.  Intermediate  growth  in  thiamine  alone  and  in  biotin  alone 
shows  that  the  deficiencies  are  partial.  The  synthetic  capacity  is  rela- 
tively greater  for  biotin  than  for  thiamine.  The  deficiencies  of  Endothia 
parasitica  are  similar  to  those  of  L.  pruni.  Blastodadia  pringsheimii  was 
30 


-^   20 
E 
o> 

E 
^^ 

4* 

.5* 

i 

A    10 


» 

^* 

» 

• 

Biotin 

•              < 
Thiamin 

weight  (mgms) 
D            o            c 

♦' 

i 

^^* ■ — 1 

y^     Nicotinamide 

/ 

1 

..  1 1 

Q 

— 1 — 1 — t 

1 

• 

1 

• 

i_ 

05             1.0           1.5 
Micrograms  per  75  cc. 

'       1        1        1       1 

2.0 

0 


0.002 


0.015 


0.005  0.01 

Micrograms  per  75  cc. 

Fig.  37.  The  effect  of  concentration  of  essential  vitamins  on  dry  weight  of  Blasto- 
dadia pringsheimii.     (Courtesy  of  Cantino,  Am.  Jour.  Botany  35:  241,  1948.) 

reported  (Cantino,  1948)  to  be  partially  deficient  for  thiamine  and  biotin 
and  nearly  totally  deficient  for  nicotinic  acid  (Fig.  37).  Cerotostomella 
ips  No.  255  was  shown  to  be  completely  deficient  for  thiamine,  biotin,  and 
pyridoxine  (Ptobbins  and  Ma,  1942a). 

Multiple  vitamin  deficiencies  are  more  common  among  the  yeasts  than 
among  the  filamentous  fungi,  and  some  yeasts  show  deficiencies  not 
known  to  exist  in  filamentous  fungi  isolated  from  nature.  For  these 
reasons  the  yeasts  as  a  group  have  received  much  attention  in  vitamin 
investigations.  The  vitamin  requirements  of  38  species  and  strains  of 
yeast  were  reported  by  Burkholder  (1943),  and  for  110  additional  named 
species  and  varieties  by  Burkholder  et  at.  (1944).  A  summary  of  the 
deficiencies  reported  in  these  two  papers  is  as  follows:  biotin,  114;  thia- 
mine, 48;  pantothenic  acid,  44;  inositol,  19;  nicotinic  acid,  19;  pyridoxine, 
19.  No  deficiency  for  riboflavin  was  found.  Several  isolates  were 
deficient  for  three  or  more  vitamins.  Saccharomyces  oviformis  was 
deficient  for  biotin,  pantothenic  acid,  and  pyridoxine,  while  S.  mace- 
doniensis  Y-91  showed  complete  or  partial  deficiencies  for  thiamine, 


VITAMINS 


181 


pantothenic  acid,  nicotinic  acid,  and  biotin.  S.  ludwigii  Y-974  and 
Kloeckera  brevis  were  totally  or  partially  deficient  for  six  vitamins  (thia- 
mine, biotin,  inositol,  pyridoxine,  nicotinic  acid,  and  pantothenic  acid). 
Growth  of  all  of  the  38  yeasts  reported  by  Burkholder  (1943)  was  increased 
by  the  addition  of  liver  extract  to  the  medium  containing  the  seven 
vitamins. 

The  preceding  discussion  of  the  effects  of  added  vitamins  on  the  growth 
of  fungi  has  been  based  on  the  assumption  that  near-optimum  amounts 
of  the  vitamins  were  present  in  the  media.     However,  the  optimum 


20  28 

Days  of  incubation 

Fig.  38.  The  effect  of  concentration  of  biotin  on  the  rate  and  amount  of  growth  ot 
Sordaria  fimicola  in  25  ml.  liquid  glucose-casein  hydrolysate  medium,  initial  pH  4.4. 
Growth  in  this  medium  containing  biotin  but  no  thiamine  is  evidence  that  this  fungus 
can  synthesize  thiamine  under  these  conditions.  (After  Lilly  and  Barnett,  Am.  Jour. 
Botany  34:  134,  1947.) 

amount  of  a  vitamin  may  vary  with  changes  in  other  conditions  and  may 
be  different  for  different  fungi.  We  have  found  that  the  following 
amounts  per  liter  of  the  four  commonly  needed  vitamins  are  near  optimum 
for  many  filamentous  fungi:  thiamine,  100  Mg;  pyridoxine,  100  jug;  biotin, 
5  /ig;  inositol,  5  mg. 

The  effects  of  biotin  concentration  on  the  growth  of  Sordaria  fimicola 
are  illustrated  in  Fig.  38,  which  shows  a  decided  increase  in  growth  rate 
with  greater  amounts  of  biotin.  Growth  was  most  rapid  in  a  medium 
containing  6.4  jug  biotin  per  liter  (0.16  /zg  per  flask),  but  a  steady  slow 
increase  in  dry  weight  is  evident  in  as  low  as  0.1  ^g  biotin  per  Hter. 

Absolute  and  conditioned  deficiencies.  According  to  Robbins  and 
Kavanagh  (1942),  the  deficiency  of  a  fungus  for  a  specific  vitamin  may  be 
absolute   or   conditioned.     Phycomyces   blakesleeanus   and    Ceratostomella 


182 


PHYSIOLOGY  OF  THE  FUNGI 


fimbriata  show  absolute  total  deficiencies  for  thiamine.  No  environ- 
mental condition  is  known  to  allow  the  synthesis  of  this  vitamin  by  these 
fungi.  In  the  case  of  a  conditioned  deficiency,  the  synthesis  of  the 
vitamin  may  be  influenced  by  certain  environmental  conditions,  such  as 
temperature,  composition,  concentration,  and  pH  of  the  medium. 

Pythium  hutleri  failed  to  grow  in  a  mineral  salts-asparagine  medium 
containing  16.4  g.  of  salts  per  liter  unless  thiamine  was  added  (Robbins 
and  Kavanagh,  1938).  When  the  salt  concentration  was  reduced  to  1.64 
g.  per  liter,  this  species  grew  without  the  addition  of  thiamine.     A  defi- 


30  35  40  45 

Temperature 
Fig.  39.  Growth-temperature  relations  for  wild-type  Neurospora  and  a  temperature- 
sensitive  mutant  deficient  for  riboflavin.  Amounts  of  riboflavin  are  indicated  on  the 
curves  in  micrograms  per  20  ml.  of  medium.  Below  25°C.  growth  was  good  without 
riboflavin,  while  no  growth  occurred  above  28°C.  without  added  riboflavin.  (Courtesy 
of  Mitchell  and  Houlahan.     Am.  Jour.  Botany  33:  31,  1946.) 

ciency  for  riboflavin  conditioned  by  temperature  was  reported  by  Mitchell 
and  Houlahan  (1946)  for  a  mutant  of  Neurospora  (Fig.  39).  Growth  was 
poor  or  none  at  temperatures  above  25°C.  unless  riboflavin  was  added. 
Below  25°C.  the  fungus  was  able  to  synthesize  riboflavin.  The  partial 
deficiency  of  Sclerotinia  camelUae  for  inositol  was  influenced  by  tempera- 
ture, particularly  in  the  above-optimum  range  (Barnett  and  Lilly,  1948). 
Low  pH  of  the  medium  resulted  in  partial  thiamine  deficiency  of 
Sordaria  fimicola,  while  no  deficiency  for  thiamine  was  apparent  at  initial 
pH  4.0  or  above  (Lilly  and  Barnett,  1947).  Within  the  range  of  3.8  to 
3.4,  growth  was  quite  slow,  but  the  addition  of  thiamine  overcame  the 
inhibition  due  to  the  high  acidity  (Fig.  40).  These  results  indicate  that 
pH  3.8  or  lower  inhibits  the  synthesis  of  thiamine  by  S.  fimicola.  In  a 
similar  way,  the  availability  of  p-aminobenzoic  acid  to  a  mutant  of 


VITAMINS 


183 


10  14 

Days  of  incubation 
Fig.  40.     Effect  of  concentration  of  thiamine  on  the  rate  and  amount  of  growth  of 
Sordaria  fiynicola  in  25  ml.  Hquid  glucose-casein  hydrolysate  medium,  initial  pH  3.8. 
Compare  with  Fig.  38.      (After  Lilly  and  Barnett,  Am.  Jour.  Botany  34:  134,  1947.) 


40 


30 


T5 

O20 


10 


PH4.0 

/ 

^ 

< 

/ 

pHj^ 

— •■ 

:— 

— 

T * 

H 

^ 

.^              f 

( 

,^ 

pH  6.0 

, -< 

^ 

/ 

/ 

y 

// 

/ 

/ 

/ 

pHTCL 

^ 

z' 

/ 

C^ 

i — 

0.05        0.1  0.2  0.3  0.4 

Microgroms  p-aminobenzoic  ocid  per  25  ml.  medium 


0.5 


Fig.  41.  Effect  of  concentration  of  p-aminobenzoic  acid  at  different  pH  values  on  the 
growth  of  a  mutant  of  N eurosj}ora  crassa  deficient  for  this  vitamin.  The  fungus  was 
grown  on  liquid  glucose-casein  hydrolysate  medium  for  3  days.  (Drawn  from  the 
data  of  Wyss,  Lilly,  and  Leonian,  Science  99:  18,  1944.) 


184  PHYSIOLOGY  OF  THE  FUNGJ 

Neurospora  crassa  deficient  for  this  vitamin  was  found  to  be  influenced 
by  the  pH  of  the  medium  (Wyss  et  al,  1944)  (Fig.  41). 

The  abiHty  of  a  mutant  of  Neurospora  sitophila  to  synthesize  pyridoxine 
was  shown  to  be  dependent  not  only  on  the  pH  of  the  medium  but  also 
on  the  source  of  nitrogen  (Stokes  et  al.,  1943).  When  nitrate,  amino, 
amide,  or  certain  other  nitrogen  compounds  served  as  the  nitrogen  source, 
no  growth  occurred  without  the  addition  of  pyridoxine.  However,  in 
the  presence  of  ammonium  salts,  growth  occurred  at  an  initial  pH  range 
of  5.6  to  7.3,  without  added  pyridoxine.  In  this  pH  range,  free  ammonia 
is  formed.  In  the  absence  of  free  ammonia,  the  pyridoxine  synthesized 
is  unavailable  to  this  mutant  (Strauss,  1951). 

According  to  Fromageot  and  Tschang  (1938),  the  red  yeast,  Rhodotorula 
sanniei,  requires  thiamine  when  the  carbon  source  is  glucose,  but  when 
redistilled  glycerol  replaces  glucose,  thiamine  is  not  needed.  It  is  inter- 
esting to  speculate  whether  this  fungus  is  better  able  to  synthesize  thia- 
mine in  a  glycerol  medium  or  whether  much  less  thiamine  is  required  to 
metabolize  glycerol  than  glucose. 

The  concentration  of  the  micro  essential  elements  has  also  been  shown 
to  influence  the  synthesis  of  vitamins  by  microorganisms  (see  Chap.  13 
for  specific  information). 

INHIBITORY  EFFECTS  OF  VITAMINS 

In  certain  cases  vitamins  may  have  an  inhibitory  effect  on  growth, 
particularly  when  present  in  excessive  dosages.  The  interrelated  effects 
of  temperature  and  amount  of  inositol  were  described  (Barnett  and  Lilly, 
1948)  for  Sclerotinia  camelliae.  At  a  temperature  below  26°C.  the  partial 
deficiency  was  overcome  by  adding  5  mg.  inositol  to  the  medium.  Above 
26°C.  the  fungus  was  highly  sensitive  to  small  changes  in  temperature 
and  in  amounts  of  inositol  in  the  medium  (Fig.  42).  The  same  amount 
of  inositol  which  stimulated  grow^th  at  or  below  26°C.  was  strongly  inhibi- 
tory at  27°C.  Increased  amounts  of  inositol  caused  greater  inhibition 
of  growth.  Since  the  maximum  temperature  for  growth  is  slightly  above 
27°C.,  it  is  believed  that,  as  the  temperature  approaches  this  point,  the 
fungus  becomes  highly  sensitive  to  the  increased  amounts  of  inositol  in 
the  medium. 

Some  vitamins  are  known  to  have  a  depressing  effect  on  growth  of 
certain  fungi  not  deficient  for  these  particular  vitamins.  For  example, 
Fig.  36  shows  that  Lambertella  pruni  produces  more  dry  weight  in  the 
presence  of  both  thiamine  and  biotin  than  when  inositol  and  pyridoxine 
are  also  added  to  the  medium.  Similarly,  it  is  reported  (Elliott,  1949) 
that,  for  a  self-sufficient  isolate  of  Fusarium  avenaceum,  both  the  rate  of 
growth  and  maximum  amount  of  mycelium  were  greater  in  vitamin-free 
medium  than  when  vitamins  were  added.     The  presence  of  thiamine  also 


VITAMINS 


185 


Fig.  42.  The  interrelated  inhibitory  effects  of  high  concentrations  of  inositol  and 
near-maximum  temperatures  on  the  growth  of  Sclerotinia  camdliae,  which  is  partially 
deficient  for  inositol  below  26°C.  Cultures  19  days  old.  Thiamine  and  biotin  were 
added  to  all  media.  Temperatures  ±0.3°C.  Left  to  right:  26°C.,  26.6°C.,  27°C. 
Amounts  of  inositol  added  per  liter  were:  A,  none;  B,  1  mg.;  C,  10  mg.;  D,  100  mg. 


186  PHYSIOLOGY  OF  THE  FUNGI 

has  been  reported  to  depress  the  growth  of  several  fungi,  including 
Collctotrichum  lindemuthianum  (Mathur  et  al.,  1950),  Rhizopus  suinus 
(fechopfer  and  Guilloud,  1945),  and  Fusarium  lini  (Wirth  and  Nord, 
1942).  Other  cases  have  been  observed  in  our  laboratory.  In  the  case 
of  Rhizopus  suinus,  the  addition  of  inositol  overcame  the  inhibitory  effects 
of  thiamine,  and  we  believe  it  to  be  effective  with  certain  other  fungi. 
On  the  basis  of  these  reports,  it  would  seem  unwise  to  add  vitamins 
indiscriminately  to  media  used  for  the  study  of  fungi  which  are  self- 
sufficient  for  these  vitamins. 

VITAMERS 

Certain  microorganisms  are  less  specific  in  their  vitamin  requirements 
than  are  animals,  owing  apparently  to  their  greater  synthetic  ability. 
Some  vitamin-deficient  fungi  may  respond  well  to  one  of  the  vitamin 
moieties,  as  in  the  case  of  thiamine,  or  to  a  compound  similar  to  the 
vitamin.  The  term  vitamer  w^as  suggested  by  Burk  et  al.  (1944)  to  denote 
a  compound  having  vitamin  activity  but  differing  in  molecular  structure 
from  the  true  vitamin.  Usually  the  structure  of  a  vitamer  is  closely 
related  to  that  of  the  vitamin.  More  specifically,  these  compounds  are 
known  as  thiamine  vitamers,  biotin  vitamers,  etc.  In  general,  a  vitamer 
is  active  for  fewer  fungi  than  is  the  vitamin.  Some  vitamers  are  anti- 
vitamins.     This  topic  is  discussed  in  Chap.  11. 

UNIDENTIFIED  GROWTH  FACTORS 

It  is  quite  probable  that  some  fungi  wdll  be  discovered  which  are  defi- 
cient for  vitamins  or  other  growth  factors  w^hich  are  at  present  unknown. 
Fungi  which  fail  to  grow  in  synthetic  media  to  which  all  the  known  growth 
factors  have  been  added  offer  a  challenge  and  an  opportunity  to  the 
investigator.  Burkholder  and  Moyer  (1943)  reported  that  Candida 
albicans  475  and  Mycoderma  vini  939  did  not  grow  unless  liver  extract 
was  added  to  glucose-asparagine  medium  containing  six  vitamins.  One 
may  speculate  that  the  effect  of  liver  extract  was  due  to  some  amino  acid 
or  to  an  undetermined  growth  factor,  possibly  vitamin  B12,  which  is 
known  to  be  present  in  liver.  In  view  of  the  common  experience  regard- 
ing the  stimulating  effect  of  natural  substances  on  growth  of  fungi,  it  is 
evident  that  much  more  investigation  on  this  phase  of  nutrition  is  needed. 

SPECIFIC  VITAMINS 

In  the  first  portion  of  this  chapter  the  general  aspects  of  vitamins  and 
growth  factors  were  considered.  Different  types  of  vitamin  deficiencies 
and  the  methods  of  detecting  deficiencies  were  discussed.  The  second 
portion  deals  with  the  specific  vitamins,  their  characteristics  and 
functions. 


VITAMINS 


187 


THIAMINE  AND  ITS  MOIETIES 

Thiamine  (vitamin  Bi,  aneurine)  was  the  first  vitamin  shown  to  be 
required  by  a  filamentous  fungus.  Thiamine  deficiency  in  man  is  known 
as  beriberi.  Certain  fungi  and  other  microorganisms  resemble  man  in 
that  they  are  unable  to  synthesize  this  vitamin.  It  is  probably  required 
in  the  metabolism  of  all  forms  of  life,  and  its  function,  to  a  large  extent,  is 
believed  to  be  the  same  in  all  organisms. 

Schopfer  (1934)  demonstrated  that  Phycomyces  blakesleeanus  failed  to 
grow  in  a  synthetic  medium  unless  thiamine  was  added.  This  was  a 
stimulus  for  numerous  studies  on  vitamin  deficiencies  of  fungi.  The 
chemical  synthesis  of  thiamine,  in  1936,  was  another  important  step  in 
vitamin  research.  The  student  is  referred  to  Williams  and  Spies  (1938), 
Rosenberg  (1942),  and  Schopfer  (1943)  for  information  on  the  history, 
synthesis,  and  natural  occurrence  of  thiamine. 

The  structural  formula  for  thiamine  is 


N=C— NHrHCl 


CH,— C     C 


-CH,— N 


/ 


CHs 

c==c— 


CH2— CH,OH 


Cl        CH — S 
N— CH 

Thiamine  chloride  hydrochloride 

The  thiamine  molecule  contains  two  ring  structures,  a  substituted 
pyrimidine  and  a  substituted  thiazole.  The  pyrimidine  moiety  has  the 
following  formula: 

N====C— NH2 


CH, 


N 


C— CH2X 


CH 


Thiamine  pyrimidine 
2-Methyl-4-amino-5-methylpyrimidine 

X  in  the  substituted  methyl  group  on  C5  may  be  hydroxyl,  chlorine, 
bromine,  etc.     The  thiazole  moiety  has  the  following  formula: 

CH3 


C= 


=C— CH2— CH2OH 


N 


/ 

V 


X 


CH- 


Thiamine  thiazole 
4-Methyl-5-/3-hydroxyethylthiazole 


188  PHYSIOLOGY  OF  THE  FUNGI 

These  moieties  are  referred  to  in  the  Hterature  as  the  thiamine  pyrimidine 
and  thiamine  thiazole,  respectively. 

Thiamine  is  somewhat  unstable  when  exposed  to  alkali  and  heat,  but 
at  pH  3.5  it  is  unaffected  by  autoclaving.  Sulfur  dioxide  and  sulfites 
are  destructive  at  pH  5  to  6.  These  factors  must  be  taken  into  considera- 
tion, and  it  is  sometimes  desirable  to  sterilize  thiamine  separately,  either 
by  filtration  or  by  autoclaving  in  an  acidified  solution.  For  most  investi- 
gations, however,  it  is  permissible  to  autoclave  thiamine  with  the  medium. 
For  most  fungi  100  fxg  of  thiamine  per  liter  of  medium  is  near  optimum 
for  growth  and  sporulation.  However,  the  optimum  varies  with  the 
amount  of  sugar  in  the  medium  and  with  other  conditions. 

Soon  after  pure  thiamine  became  available,  it  was  discovered  that 
certain  treatments  destroyed  its  activity  for  animals  but  did  not  greatly 
affect  the  potency  when  certain  fungi  were  used  as  test  organisms.  The 
solution  to  this  problem  was  reached  when  it  became  known  that  thia- 
mine, when  autoclaved  in  the  presence  of  alkali,  was  broken  down  into 
thiamine  pyrimidine  and  thiamine  thiazole. 

Thiamine-deficient  fungi  differ  in  their  ability  to  utilize  or  synthesize 
the  moieties  of  thiamine.  These  fungi  may  be  classified  into  four  groups 
on  this  basis:  (1)  The  intact  molecule  of  thiamine  is  required  by  some 
fungi  which  are  unable  to  synthesize  either  moiety  or  to  complete  the 
synthesis  of  thiamine,  even  when  both  moieties  are  supplied.  Examples 
of  the  group  are  species  of  Phytophthora.  (2)  Some  other  fungi,  such  as 
Phycomyces  hlakesleeanus,  are  capable  of  utilizing  thiamine,  or  of  syn- 
thesizing thiamine  when  furnished  with  a  mixture  of  the  two  thiamine 
moieties.  (3)  The  addition  of  thiamine  or  thiamine  pyrimidine  satisfies 
the  need  of  those  fungi  which  are  able  to  synthesize  the  thiazole  moiety 
and  combine  it  with  the  pyrimidine  moiety  to  make  thiamine.  Examples 
are  Parasitella  simplex  and  Rhodotorula  rubra.  (4)  Other  fungi  are  able 
to  synthesize  only  the  thiamine  pyrimidine  and  complete  the  synthesis 
of  thiamine  when  furnished  with  the  thiazole  moiety.  Mucor  raman- 
nianus  and  Stereum  frustulosum  are  examples. 

In  the  above  discussion  it  was  assumed  that  in  every  case  the  intact 
molecule  was  the  active  product  and  that  neither  moiety  nor  the  presence 
of  the  two  had  any  activity  until  thiamine  was  synthesized.  Leonian 
and  Lilly  (1940)  found  this  hypothesis  to  be  correct.  The  following  fungi 
Avere  grown  in  a  basal  medium  to  which  had  been  added  the  minimal 
growth  factor:  Fusarium  niveum  (none),  Pythiomorpha  gonapodyoides 
(pyrimidine),  Mucor  ramannianus  (thiazole),  Phycomyces  hlakesleeanus 
(both  moieties  of  thiamine),  and  Phytophthora  erythroseptica  (thiamine). 
After  growth,  the  mycelium  and  the  medium  were  tested  for  thiamine 
and  its  moieties  by  growing  fungi  of  known  thiamine  or  thiamine-moiety 
requirements  upon  media  containing  the  mycelium  extract  and  the 
medium.     Some  of  these  data  are  collected  in  Table  33. 


VITAMINS 


189 


Table   33.     Assay   for   Thiamine   and   Thiamine    Moieties   in    Mycelium   and 
Medium  Extracts  of  Some  Fungi  after  Growth  on  Media  Containing 

THE  Minimum  Growth-factor  Requirements 
Numbers  refer  to  relative  growth  on  the  scale  of  10.      (Leonian  and  Lilly,  Plant 
Physiol.  15,  1940.) 


Test  fungi  and  substance  tested  for 

Fungi  tested  and 

minimum  vitamin 

requirements 

Pythium 
ascophallon 
(thiamine) 

Phycomyces 

blakesleeanus 

(both 

moieties) 

Pythiomorpha 

gonapodyoides 

(pyrimidine) 

Mucor 

ramannianus 

(thiazole) 

Fusarium  niveum 
(none) 

Pythiomorpha      gonapody- 
oides 
(pyrimidine) 

Mucor  ramannianus 
(thiazole) 

Phycomyces  blakesleeanus 
(both  moieties) 

Phytophthora  erythroseptica 
(thiamine) 

10* 
0* 

10 

1 

1 

0 

10 
0 

10 

1 

10 
2 

10 

4 

3 
2 

10 
8 

8 
6 

10 

4 

10 

10 

5 
3 

10 
8 

10 
6 

10 
10 

8 

8 

8 

7 

10 
10 

10 
10 

*  Upper  figures  refer  to  extract  of  mycelium,  lower  figures  to  extract  of  medium. 

It  is  evident  that  Fusarium  niveum  was  able  to  synthesize  thiamine 
from  the  basal  medium  because  two  test  fungi  which  require  thiamine 
per  se  grew  on  extracts  prepared  from  the  hyphae.  The  same  type  of 
proof  shows  that  Pythiomorpha  gonapodyoides  synthesized  thiamine  when 
thiamine  pyrimidine  was  added  to  the  basal  medium.  Mucor  raman- 
nianus synthesized  thiamine  when  thiamine  thiazole  was  added,  and 
Phycomyces  blakesleeanus  synthesized  thiamine  when  both  moieties  were 
added. 

In  all  cases  the  greater  portion  of  thiamine  was  stored  within  the 
mycelium,  and  only  small  amounts  were  present  in  the  medium.  The 
medium  extract  from  three  fungi  contained  no  thiamine,  although  appre- 
ciable quantities  of  the  pyrimidine  and  thiazole  moieties  were  present 
in  all  media.  This  shows  that  Phytophthora  erythroseptica,  for  example, 
had  broken  down  the  thiamine  molecule  into  its  moieties,  which  diffused 
into  the  medium  and  were  later  utilized  by  certain  fungi,  such  as  Phyco- 
myces blakesleeanus.  This  suggests  that  in  the  process  of  its  utilization 
thiamine  is  slowly  destroyed.  The  moieties  may  be  recombined  by 
certain  organisms  but  not  by  those  which  require  the  entire  thiamine 


190  PHYSIOLOGY  OF  THE  FUNGI 

molecule.  Robbins  and  Kavanagh  (1941)  showed  that  P.  blakesleeanus 
destroyed  the  thiazole  more  rapidly  than  it  did  the  pyrimidine  moiety. 
Thus,  an  excess  of  thiazole  in  the  mixture  of  the  two  moieties  was  more 
effective  than  equal  quantities.     They  termed  this  the  thiazole  effect. 

Some  thiamine -deficient  fungi.  A  deficiency  for  thiamine  is  by  far 
the  most  common  vitamin  deficiency  among  filamentous  fungi  isolated 
from  nature.  Fries  (1948)  states  than  over  200  fungi  are  known  to  be 
partially  or  totally  deficient  for  thiamine.  No  doubt  this  is  a  modest 
estimate.  Deficiency  for  this  vitamin  is  more  common  among  certain 
groups  of  fungi  than  others.  For  example,  all  species  of  Phytophthora 
studied  have  been  found  to  require  the  intact  molecule  of  thiamine. 
Only  a  few  species  of  the  true  Basidiomycetes  have  been  reported  to  be 
self-sufficient  for  thiamine.  Many  of  these  fungi  show  only  partial 
deficiencies,  while  some  are  totally  deficient.  In  most  cases,  however, 
there  seems  to  be  little  or  no  correlation  between  thiamine  deficiency  and 
taxonomic  relationship. 

Some  common  filamentous  fungi  (other  than  Basidiomycetes)  which 
have  been  reported  to  be  totally  or  partially  deficient  for  thiamine  or  its 
moieties,  with  other  deficiencies  (if  any)  indicated  in  parentheses,  are 
as  follows :  Blakeslea  trispora,  Ceratostomella  fimbriata,  C.  ips  (biotin  and 
pyridoxine),  C.  montium  (biotin  and  pyridoxine),  C.  pini  (biotin),  Chae- 
tomium  convolutum  (biotin),  Choanephora  cucurbitarum,  Coemansia  inter- 
rupta  (biotin),  Dendrophoma  obscurans,  Endothia  parasitica  (biotin), 
Hypoxylon  pruinatum  (biotin),  Lambertella  pruni  (biotin),  Lophodermium 
pinastri  (biotin  and  inositol),  Melanconium  betulinum  (biotin  and  inositol), 
Melanospora  destruens  (biotin),  Mucor  ramaymianus,  Nectria  coccinia, 
Ophiobolus  graminis  (biotin),  Phycomyces  blakesleeanus,  Phytophthora 
spp.,  Piricularia  oryzae  (biotin),  Pleurage  curvicolla  (biotin),  Podospora 
curvida  (biotin),  Pythiomorpha  gonapody aides,  Pythium  arrhenomanes ,  P. 
ascophallon,  P.  butleri,  P.  oligandrum,  Sclerotinia  camelliae  (biotin  and 
inositol),  S.  minor,  Sordaria  fimicola,  certain  isolates  only  (biotin), 
Thielaviopsis  basicola,  Valsa  pini  (biotin  and  inositol),  and  Xylaria 
hypoxylon. 

Reports  of  deficiencies  for  most  of  the  above-named  fungi  may  be  found 
in  the  references  for  this  chapter.  Some  few  of  these  fungi  have  been 
studied  in  our  laboratory  and  have  not  been  previously  reported  as  being 
deficient  for  thiamine.  For  thiamine-deficient  yeasts  see  the  reports  of 
Burkholder  (1943)  and  Burkholder  et  al.  (1944). 

Mode  of  action.  One  of  the  primary  uses  of  thiamine  in  plants  and 
animals  is  for  the  regulation  of  carbohydrate  metabolism.  It  is  also 
probable  that  thiamine  may  be  involved  in  other  processes.  A  vitamin 
which  constitutes  a  part  of  an  enzyme  system  is  known  as  a  coenzyme. 
Generally  a  vitamin  must  be  combined  with  organic  or  inorganic  com- 


VITAMINS  191 

pounds  (or  both)  before  it  combines  with  the  protein  portion  (apoenzyme) 
of  the  enzyme  system.  The  pyrophosphoric  ester  of  thiamine  is  known 
as  cocarboxylase,  or  as  thiamine  pyrophosphate.  This  compound  is  the 
coenzyme  of  carboxylase. 

CH,  O  O 

N=C— NH2  I  II  II 

I  C=C— CH2— CH2— O— P~0— P— OH 


CH3— C     C— CH2— N 


o 

H  H 


i 


CI         CH— S 

N— CH 

Thiamine  pyrophosphate  (cocarboxylase) 

This  substance  is  as  active  as  thiamine  (mole  for  mole) .  Lilly  and  Leonian 
(1940)  compared  the  action  of  thiamine  and  thiamine  pyrophosphate  on 
several  thiamine-deficient  fungi.  No  significant  differences  were  found 
in  the  maximum  weights  of  mycelium  formed  in  the  presence  of  equivalent 
quantities  of  these  two  growth  factors.  The  rate  of  early  growth  was 
greater  with  thiamine  pyrophosphate  than  with  thiamine  for  Phyco- 
myces  hlakesleeanus  and  less  for  Mucor  ramannianus  and  Phytophthora 
erythroseptica. 

Pyruvic  acid,  one  of  the  key  intermediate  products  of  carbohydrate 
metabolism,  is  transformed  into  carbon  dioxide  and  acetaldehyde  by  the 
action  of  the  enzyme  carboxylase.  Pyruvic  acid  accumulates  in  the 
culture  media  of  many  thiamine-deficient  fungi  when  insufficient  thiamine 
is  present.  Haag  and  Dalphin  (1940)  found  that  the  maximum  accumu- 
lation in  Phycomyces  hlakesleeanus  cultures  occurred  when  about  one- 
twentieth  of  the  optimum  amount  of  thiamine  was  added.  Wirth  and 
Nord  (1942)  studied  the  effect  of  added  thiamine  upon  the  accumulation 
of  pyruvic  acid  in  cultures  of  Fusarium  lini,  a  self-sufficient  fungus  with 
respect  to  thiamine.     Some  of  the  data  are  presented  in  Table  34. 

The  accumulation  of  pyruvic  acid  in  the  culture  medium  is  common, 
especially  during  the  early  period  of  growth.  Pyruvic  acid  may  be 
detected  qualitatively  by  adding  of  iodine  solution  (KI3)  to  the  culture 
filtrate  and  making  the  solution  strongly  alkaline  with  sodium  hydroxide. 
Iodoform  is  produced  instantly  without  heating.  Acetaldehyde,  which  is 
very  volatile,  also  reacts  with  iodine  and  alkali  in  the  cold  to  produce 
iodoform.  Sordaria  fimicola,  Lenzites  trahea,  or  other  fungi  which  produce 
acid  during  the  early  stages  of  growth  may  be  used  to  demonstrate  the 
production  of  pyruvic  acid. 

Specificity.  So  far  as  is  known,  thiamine  which  occurs  in  nature  has 
the  structure  given  in  the  formula.  This  vitamin  has  been  isolated  from 
only  a  few  substances  such  as  wheat  germ  and  rice  polish.  The  ethyl 
homologue  (ethyl  in  place  of  methyl  in  position  2)  of  thiamine  is  slightly 
more  active  for  certain  fungi  than  ordinary  thiamine.     Higher  homologues 


192 


PHYSIOLOGY  OF  THE  FUNGI 


have  been  reported  to  be  less  active  or  inhibitory.  Whether  ethyl  thia- 
mine occurs  in  nature  is  not  known.  The  student  is  referred  to  Schopfer 
(1943)  for  further  information  on  thiamine  specificity. 

Table  34.     The  Effect  of  Added  Thiamine  upon  the  Accumulation  of  Pyruvic 
Acid  in  the  Culture  Filtrate  of  Fusarium  lint  Grown  on  Glucose- 
Nitrate  Medium 
(Wirth  and  Nord,  Arch.  Biochem.  1,  1942.     Published  by  permission  of  Academic 
Press,  Inc.) 


Days  of 

Glucose  fermented, 
g.  per  liter 

Pyruvic  acid  accumulated, 
mg.  per  liter 

Mycelium  produced, 
mg.  per  50  ml. 

incuba- 
tion 

Thiamine  added,  ^g  per  liter 

0 

500 

0 

500 

0 

500 

2 
4 
6 
8 

1.5 
17.3 
34.7 
40.9 

2.5 
13.5 
33.9 
40.7 

50 
1,590 
1,710 
1,550 

Trace 

80 
260 
Trace 

132 

259 

91 
135 

BIOTIN 

Biotin  (vitamin  H)  was  originally  isolated  as  a  grow^th  factor  for  yeast. 
It  is  known  to  be  the  factor  which  prevents  raw-egg-white  injury  to 
animals  and  is  the  respiratory  coenzyme  (coenzyme  R)  for  species  of 
Rhizohium.  Biotin  is  active  at  greater  dilutions  than  are  the  other 
vitamins.  Pure  biotin  methyl  ester  was  first  isolated  by  Kogl  and  Tonnis 
(1936)  w^ho  obtained  1.1  mg.  of  this  substance  from  250  kg.  of  dried  duck 
eggs.  The  structure  of  biotin  was  determined  by  Du  Vigneaud  et  al. 
(1942a)  and  confirmed  by  the  synthesis  of  this  compound  (Harris  et  al., 
1943).     The  structure  of  the  biotin  molecule  is  as  follows: 

CO 

NH        NH 

I  I 

CH CH 

CH2       CH— (CH2)4— COOH 

\/ 

Biotin 

Some  fungi  deficient  for  biotin.  Biotin  deficiency  appears  to  be 
characteristic  of  most  yeasts  (Burkholder,  1943 ;  Burkholder  and  Moyer, 
1943;  Leonian  and  Lilly,  1942).  Numerous  filamentous  fungi  have  been 
reported  to  be  deficient  for  biotin,  but  this  number  is  not  so  great  as  that 
for  thiamine.  Frequently  biotin  deficiency  accompanies  thiamine 
deficiency. 


VITAMINS  193 

Among  the  first  investigators  to  test  the  action  of  biotin  on  filamentous 
fungi  were  Kogl  and  Fries  (1937),  who  showed  that  Nematospora  gossypii 
did  not  grow  in  the  absence  of  biotin.  As  httle  as  0.4  ^g  of  biotin  per 
liter  permitted  almost  as  much  growth  as  did  ten  times  that  amount.  For 
most  filamentous  fungi  5  /xg  of  biotin  per  liter  is  adequate.  The  effects 
of  biotin  deficiency  on  the  development  of  the  ascospores  of  Sordaria 
fimicola  are  shown  in  Fig.  G8. 

Some  filamentous  fungi  reported  as  being  partially  or  totally  deficient 
for  biotin,  with  other  deficiencies  (if  any)  given  in  parentheses,  are  as 
follows:  Chaetomium  convolutum  (thiamine),  Coemansia  interrupta  (thia- 
mine), Diplodia  macrospora,  Endothia  parasitica  (thiamine),  Hypoxylon 
pruinatum  (thiamine),  Lamhertella  pruni  (thiamine),  Melanospora  destru- 
ens (thiamine),  Memnoniella  echinata,  Neurospora  spp.,  Ophioholus 
graminis  (thiamine),  0.  oryzinus,  Ophiostoma  catonianum,  Penicillium 
digitatum  (thiamine,  pyridoxine,  pantothenate),  Piricularia  oryzae  (thia- 
mine), Pleurage  curvicolla  (thiamine),  Podospora  curvula  (thiamine), 
Pseudopeziza  ribis,  Rosellinia  arcuata,  Sclerotinia  camelliae  (thiamine, 
inositol),  Sordaria  fimicola,  Sporormia  intermedia,  Stachybotrys  atra, 
Thraustotheca  clavata. 

Specificity.  The  biotin  molecule  is  not  separable  into  moieties  as  is 
thiamine.  One  of  the  first  related  compounds  to  be  studied  was  desthio- 
biotin.  As  the  name  indicates,  the  molecule  no  longer  contains  sulfur. 
The  structure  of  the  desthiobiotin  molecule  is  as  follows: 

CO 

/   \ 
NH        NH 

I  I 

CH3— CH CH— (CH2)6— COOH 

Desthiobiotin 

The  removal  of  sulfur  from  the  biotin  molecule  destroyed  the  tetrahydro- 
thiophene  ring  and  introduced  a  methyl  group.  In  addition,  the  acidic 
chain  of  desthiobiotin  contains  one  more  methylene  group  than  does  that 
of  biotin.  Stokes  and  Gunness  (1945)  tested  the  growth  of  some  biotin- 
deficient  microorganisms  on  desthiobiotin  and  found  that  this  compound 
was  utilized  by  Neurospora  sitophila  and  three  strains  of  Saccharomyces 
cerevisiae,  but  Rhizobium  trifolii  209,  Lactobacillus  casei,  and  L.  arabinosus 
17-5  were  unable  to  utilize  desthiobiotin.  From  further  experiments  it 
was  concluded  that  the  yeast  synthesized  biotin,  or  some  other  compound 
which  replaced  it,  from  desthiobiotin,  rather  than  utilizing  desthiobiotin 
directly.  The  source  of  sulfur  in  the  medium  was  found  to  influence  the 
amount  of  desthiobiotin  converted  into  biotin,  with  methionine  and 
sodium  sulfate  being  better  sources  than  cystine,  sulfanilamide,  or 
thiamine  thiazole. 

Lilly  and  Leonian   (1944)   studied  the  effect  of  desthiobiotin  on  45 


194  PHYSIOLOGY  OF  THE  FUNGI 

biotin-deficient  microorganisms  and  found  that  it  replaced  biotin  for 
some  fungi,  while  it  acted  as  an  antibiotin  for  some  few  others.  Desthio- 
biotin  replaced  biotin  quantitatively  for  Ceratostomella  ips.  Goldberg 
et  al.  (1947)  found  some  homologues  of  biotin  to  inhibit  growth  of  yeast 
139  and  Lactobacillus  casei.  Whether  any  of  these  biotin  homologues 
will  replace  biotin  for  other  microorganisms  must  await  further  testing. 
These  preliminary  results  indicate  that  the  length  of  the  acidic  side  chain 
of  the  biotin  molecule  is  of  great  importance  in  biological  activity. 

Oxybiotin  is  also  known  as  0-heterobiotin  and  has  the  same  structure 
as  biotin  except  that  the  sulfur  in  the  tetrahydrothiophene  ring  has  been 
replaced  by  oxygen.  Pilgrim  et  al.  (1945)  found  oxybiotin  to  be  active 
for  Lactobacillus  casei,  L.  arabinosus,  and  a  strain  of  Saccharomyces  cere- 
visiae.  Oxybiotin  is  apparently  used  as  such  and  is  not  converted  into 
biotin  by  the  organism  (Axelrod  et  al.,  1947).  This  is  the  only  instance 
that  has  come  to  our  attention  where  a  vitamer  is  used  directly  instead 
of  being  converted  into  the  vitamin.  Rubin  et  al.  (1945)  had  previously 
reported  that  oxybiotin  was  converted  into  biotin.  The  cause  of  this 
disagreement  is  unknown. 

Pimelic  acid  is  a  growth  factor  for  certain  strains  of  the  diphtheria 
bacterium  (Mueller,  1937).  It  is  reported  (Du  Vigneaud  et  al.,  1942) 
that  pimelic  acid  replaced  biotin  and  was  probably  the  precursor  in  the 
synthesis  of  biotin  by  a  strain  of  the  diphtheria  organism.  The  higher 
and  lower  homologs  of  pimelic  acid  were  ineffective.  The  formula  for 
pimelic  acid  is  HOOC— CH2— CH2— CHa— CH2— CH2— COOH.  At 
present  there  is  no  evidence  that  pimelic  acid  replaces  biotin  for  the 
fungi.  This  observation  is  supported  by  the  findings  of  Robbins  and  Ma 
(1942),  who  studied  13  biotin-deficient  fungi.  A  favorable  effect  of  the 
presence  of  pimelic  acid  was  reported  by  Eakin  and  Eakin  (1942),  who 
found  that  Aspergillus  niger  synthesizes  much  more  biotin  in  the  presence 
of  pimelic  acid  than  in  its  absence.  Cysteine  and  also  cystine  increase 
the  synthesis  of  biotin.  The  lower  homologues  of  pimelic  acid  (adipic, 
glutaric,  and  succinic)  were  without  effect,  while  the  higher  homologs 
(suberic  and  azelaic)  were  as  effective  as  pimelic  acid.  This  is  interesting, 
inasmuch  as  homobiotin  and  bishomobiotin  are  reported  inactive  for 
yeast  growth  (Goldberg  et  al.,  1947). 

Mode  of  action.  It  has  been  assumed  that  biotin  acts  as  a  coenzyme 
for  various  enzyme  systems,  but  definite  proof  seems  to  be  lacking, 
Winzler  et  al.  (1944)  found  that,  when  biotin  was  added  to  a  biotin- 
starved  yeast,  some  time  elapsed  before  any  effect  was  noted.  The  order 
of  response  was  fermentation,  respiration,  and  growth.  The  assimilation 
of  ammonia  did  not  take  place  unless  biotin  was  added. 

The  presence  of  aspartic  acid  in  the  culture  medium  has  been  shown 
to  reduce  the  amount  of  biotin  required  by  Torula  cremoris  (Koser  et  al., 


VITAMINS  195 

1942)  and  by  Memnoniella  echinata  and  Stachyhotrys  atra  (Perlman, 
1948).  There  is  also  evidence  (Stokes  et  al.,  1947)  that  biotin  plays  a 
role  in  the  synthesis  of  aspartic  acid  by  certain  bacteria.  Thus,  it  appears 
probable  that  one  of  the  functions  of  biotin  is  connected  with  the  synthesis 
of  aspartic  acid.  When  aspartic  acid  is  added  to  the  medium,  it  is 
unnecessary  for  the  organism  to  perform  this  synthesis  and  the  need  for 
biotin  is  greatly  reduced.  However,  it  should  be  noted  that,  although 
the  absolute  amount  of  biotin  needed  is  reduced,  exogenous  biotin  is  still 
required  by  these  biotin-deficient  organisms.  From  this  it  may  be 
deduced  that  biotin  has  a  multiple  role  in  the  cell. 

INOSITOL 

meso-Inositol  (also  known  as  inactive  inositol,  isoinositol,  inosite,  or 
dambose)  is  widely  distributed  in  both  plants  and  animals.  It  was  first 
isolated  in  1850.  It  was  not  until  1928  that  Eastcott  (1928)  showed  that 
it  was  a  growth  factor  for  a  strain  of  yeast.  Later,  Woolley  (1940) 
recognized  it  as  a  vitamin  for  animals.  meso-Inositol  is  a  hexahydroxy- 
cyclohexane.     It  has  the  following  configuration: 

H        H 

Q Q 

OH/i  i\    H 

1/     OH     0H\| 

c  c 

:\     H        OH/1 
H     \|  1/     OH 

C C 

OH     li 
meso-Inositol 

There  are  seven  different  cis-trans  isomers,  which  are  optically  inactive, 
and  a  pair  of  optically  active  d  and  I  forms.  The  available  evidence 
indicates  that  the  stereochemical  configuration  of  weso-inositol  is  specific 
for  vitamin  activity.  Some  of  the  isomers  have  only  slight  activity. 
Inositol  is  active  only  in  high  concentrations  as  compared  to  the  other 
vitamins.     The  usual  amount  added  is  around  5  mg.  per  hter  of  medium. 

Fungi  deficient  for  inositol.  Many  strains  of  yeast  are  deficient  for 
this  vitamin,  while  others  are  not.  In  most  cases  the  deficiency  appar- 
ently is  partial  rather  than  total.  Partial  deficiencies  for  various  yeasts 
are  reported  by  Leonian  and  Lilly  (1942),  Burkholder  (1943),  and  Burk- 
holder  and  Moyer  (1943).  In  the  last  two  references  total  deficiencies 
for  inositol  are  reported  for  Saccharomyces  uvarum  Y  969  and  Schizosac- 
charomyces  pomhe. 

Kogl  and  Fries  (1937)  were  apparently  the  first  to  investigate  the 
action  of  inositol  on  various  filamentous  fungi.  They  found  that  Nemato- 
spora  gossypii  was  totally  deficient  and  that  Lophodermium  pinastri  was 


196  »  PHYSIOLOGY  OF  THE  FUNGI 

partially  deficient  for  this  vitamin.  The  partial  deficiency  of  Sclerotinia 
camelliae  is  shown  in  Fig.  35.  Deficiencies  for  inositol  are  commonly 
accompanied  by  deficiencies  for  thiamine  and  biotin.  Trichophyton 
discoides  is  reported  as  being  totally  deficient  for  inositol,  pyridoxine,  and 
thiamine  (Robbins  et  at.,  1942).  Totally  deficient  mutants  of  Neurospora 
crassa  have  been  developed.  Their  use  in  bioassays  for  inositol  was 
described  by  Beadle  (1944)  and  by  Leonian  and  Lilly  (1945). 

Some  filamentous  fungi  reported  to  be  partially  or  totally  deficient  for 
inositol,  with  other  deficiencies  given  in  parentheses,  are  as  follows: 
Colletotrichum  lindemuthianum  (certain  strains  only),  Epichloe  typhina 
(thiamine),  Lophodermium  pinastri  (thiamine,  biotin),  Melanconium 
hetulinum  (thiamine,  biotin),  Nematospora  gossypii  (thiamine,  biotin), 
Sclerotinia  camelliae  (thiamine,  biotin),  Trichophyton  discoides  (thiamine, 
pyridoxine),  Valsa  pini  (thiamine,  biotin). 

The  effects  of  temperature  upon  the  synthesis  of  inositol  by  Sclerotinia 
camelliae  and  upon  the  toxicity  of  high  concentrations  of  inositol  at  high 
temperatures  were  described  by  Barnett  and  Lilly  (1948)  and  are  illus- 
trated in  Fig.  42. 

Mode  of  action.  The  addition  of  inositol  overcame  the  inhibition  of 
growth  of  Rhizopus  suinus  due  to  excess  thiamine  (Schopfer  and  Guilloud, 
1945).  In  part,  the  inhibition  was  due  to  an  increased  production 
of  alcohol  (pyruvate  ^  acetaldehyde—^  alcohol).  Similarly,  we  have 
observed  in  our  laboratory  the  same  favorable  effect  of  inositol  on  growth 
of  certain  fungi  which  are  inhibited  by  the  presence  of  excess  thiamine. 

NICOTINIC  ACID 

A  deficiency  for  nicotinic  acid,  or  nicotinic  acid  amide,  leads  to  pellagra 
in  man  and  blacktongue  in  dogs.  The  structural  formulas  of  these  com- 
pounds follow: 


/\^r.f^c^vi  /\ 


-CONH2 


Nicotinic  acid  Nicotinic  acid  amide 

Nicotinic  acid  was  obtained  by  the  oxidation  of  nicotine  in  1867.  Knight 
(1937)  and  Mueller  (1937o)  recognized  that  nicotinic  acid  amide  was  a 
growth  factor  for  certain  bacteria.  So  far  as  is  known,  the  amide  is  the 
form  utilized  by  organisms.  Some  microorganisms  can  convert  nicotinic 
acid  into  its  amide  with  ease,  others  with  difficulty;  still  others  are  unable 
to  use  nicotinic  acid  but  require  either  nicotinic  acid  amide  or  a  coenzyme 
containing  the  amide. 

Fungi  deficient  for  nicotinic  acid.     Rogosa  (1943)  tested  114  strains  of 
yeast  that  ferment  lactose  and  found  that  all  of  them  required  an  exoge- 


VITAMINS  197 

nous  supply  of  nicotinic  acid  for  growth.  Rogosa  used  the  technique 
of  serial  passage  in  a  medium  devoid  of  nicotinic  acid.  It  is  possible  to 
overlook  a  vitamin  deficiency  by  failure  to  observe  this  precaution. 
Yeasts  found  to  be  deficient  for  this  vitamin  include  Saccharomyces 
anamensis  154,  S.  lactis  131,  S.  fragilis  15,  Zy  go  saccharomyces  lactis  (two 
strains),  Torida  lactosa  168,  T.  sphaerica  13,  T.  cremoris  2,  Torulopsis 
kefyr  149,  Mycotorula  lactis  130.  Strains  of  Saccharomyces  cerevisiae 
failed  to  show  deficiency  for  nicotinic  acid  (Rogosa,  1943;  Leonian  and 
Lilly,  1942;  Burkholder,  1943). 

Until  recently,  nicotinic  acid  deficiency  among  filamentous  fungi  iso- 
lated from  nature  was  unknown.  Cantino  (1948)  has  shown  that  Blasto- 
cladia  pringsheimii  is  completely  deficient  for  nicotinamide  and  partially 
deficient  for  thiamine  and  biotin.  Some  of  Cantino's  results  are  pre- 
sented in  Fig.  37.  A  second  filamentous  fungus,  a  strain  of  Microsporum, 
audouini,  is  reported  as  deficient  for  nicotinic  acid  (Area  Leao  and  Cury, 
1949).  Mutants  deficient  for  this  vitamin  have  been  developed  in 
Neurospora  by  Bonner  and  Beadle  (1946)  and  in  Penicillium  by  Bonner 
(1946). 

Specificity.  In  so  far  as  the  fungi  are  concerned,  nicotinic  acid  replaces 
nicotinic  acid  amide,  but  few  critical  studies  in  this  connection  have  been 
made.  Various  studies  have  been  made  of  the  specificity  for  bacteria 
of  the  compounds  related  to  nicotinic  acid.  Bovarnick  (1943)  reported 
that  heating  asparagine  and  glutamic  acid  together  produced  a  compound 
which  replaced  nicotinic  acid  or  its  amide  for  various  species  of  bacteria. 
This  author  later  showed  that  this  substance  was  nicotinic  acid  amide. 
This  is  an  unsuspected  way  of  adding  a  vitamin  to  a  basal  medium. 

Mode  of  action.  Nicotinic  acid  amide  is  a  constituent  of  two  or  more 
coenzymes.  Codehydrogenase  I  on  hydrolysis  yields  adenine,  nicotinic 
acid  amide,  and  two  molecules  of  D-ribosephosphoric  acid.  Codehydro- 
genase II  yields  the  same  products  as  codehydrogenase  I  except  that  three 
molecules  of  phosphoric  acid,  instead  of  two,  are  produced.  In  the 
literature  codehydrogenase  I  is  often  referred  to  as  DPN  (diphospho- 
pyridine  nucleotide)  and  codehydrogenase  II  as  TPN  (triphosphopyridine 
nucleotide).  These  coenzymes  in  combination  with  specific  proteins 
form  enzyme  systems  which  transfer  hydrogen  (oxidation-reduction). 
Apparently  the  amide  of  nicotinic  acid  is  reversibly  oxidized  and  reduced 
in  the  process. 

One  organism.  Hemophilus  parainfluenzae,  requires  codehydrogenase 
I  as  a  growth  factor.  This  organism  is  unable  to  form  the  coenzyme 
u'hen  furnished  with  the  moieties,  nicotinic  acid  amide,  adenine,  D-ribose, 
and  phosphate.  DPN  is  also  known  as  factor  V  (Gingrich  and  Schlenk, 
1944).  Other  bacteria  are  known  which  require  preformed  coenzymes 
as  growth  factors.     While  no  fungus  isolated  from  nature  has  yet  been 


198  PHYSIOLOGY  OF  THE  FUNGI 

shown  to  require  such  growth  factors,  it  is  possible  that  some  do  exist. 
Such  requirements  may  be  found  among  the  artificially  induced  mutants. 

PANTOTHENIC  ACID 

Pantothenic  acid  was  first  discovered  (Williams  et  al.,  1932)  as  a  growth 
factor  for  the  Gebriide  Mayer  strain  of  Saccharomyces  cerevisiae.  The 
isolation,  identification,  and  synthesis  of  this  compound  was  complete  by 
1940.  It  was  later  shown  to  be  a  vitamin  for  animals.  Pantothenic 
acid  consists  of  two  moieties  joined  together  by  means  of  an  amide  link- 
age.    The  chemical  formula  for  this  vitamin  is  given  below: 

CH3 

I 
HO— CH2— C— CHOH— CO— NH— CH2— CH2— COOH 

I 
CH3 

Pantothenic  acid 

Pantothenic  acid  may  be  hydrolyzed  to  form  /3-alanine  (jS-amino- 
propionic  acid),  the  formula  of  which  is  H2N — CH2 — CH2 — COOH,  and 
a,7-dihydroxy-/3,;5-dimethylbutyric  acid,  a  substituted  butyric  acid  that 
forms  a  lactone  by  elimination  of  one  molecule  of  w^ater  between  the 
carboxyl  and  the  gamma  hydroxyl  (pantoyl  lactone).  Pantothenic  acid 
is  thus  analogous  to  thiamine,  in  that  the  molecule  may  be  split  into  two 
moieties.  We  might  expect  to  find  different  pantothenic  acid-deficient 
organisms  which  require  the  intact  molecule  or  one  or  both  moieties.  It 
was  found  that  the  Gebriide  Mayer  strain  of  Saccharomyces  cerevisae  was 
stimulated  by  /3-alanine  and  that  this  yeast  completed  the  synthesis  of 
pantothenic  acid  when  furnished  with  )8-alanine  in  the  medium  (Wein- 
stock  et  al.,  1939).  Most  yeasts  deficient  for  pantothenic  acid  are  unable 
to  synthesize  the  j8-alanine  moiety  of  this  vitamin.  In  general,  this 
moiety  is  not  used  so  efficiently  as  pantothenic  acid,  and  more  than  an 
equivalent  amount  is  required  to  support  the  same  amount  of  growth. 
The  composition  of  the  medium  affects  utilization,  since,  in  the  presence 
of  sufficient  asparagine,  /3-alanine  is  not  utilized  (Atkin  et  al.,  1944). 

So  far  as  is  known,  none  of  the  fungi  require  pantoyl  lactone  as  a  growth 
factor,  but  this  compound  was  found  (Ryan  et  al.,  1945)  to  replace  panto- 
thenic acid  for  Clostridium  septicum.  It  was  shown  by  microbiological 
tests  that  this  bacterium  completed  the  synthesis  of  pantothenic  acid. 

Fungi  deficient  for  pantothenic  acid.  Of  the  10  strains  of  Saccharo- 
myces cerevisiae  tested  for  vitamin  deficiencies  by  Leonian  and  Lilly 
(1942),  9  w^ere  highly  deficient  for  this  vitamin.  Burkholder  (1943)  found 
14  of  the  38  species  and  strains  tested  to  be  deficient  for  pantothenic  acid, 
9  of  these  being  species  of  Saccharomyces.  It  appears  that  deficiency 
for  this  vitamin  is  more  common  in  Saccharomyces  than  in  other  genera 
of  yeasts.     Varying  degrees  of  pantothenic  acid  deficiency  were  found  in 


VITAMINS  199 

species  of  Zygosaccharomyces  (Lockhead  andLanderkin,  1942).  /S-Alanine 
could  be  used  in  place  of  pantothenic  acid  for  the  deficient  species  of 
Zygosaccharomyces. 

Growth  of  Penicillium  digitatum  is  reported  (Wooster  and  Cheldehn, 
1945)  to  be  stimulated  by  pantothenate,  as  well  as  by  pyridoxine,  biotin, 
and  thiamine.  To  our  knowledge  this  is  the  only  report  of  a  filamentous 
fungus  isolated  from  nature  being  stimulated  by  the  presence  of  this 
vitamin.  Tatum  and  Beadle  (1945)  reported  a  mutant  of  Neurospora 
which  was  deficient  for  pantothenic  acid. 

Specificity.  As  in  the  case  of  thiamine  and  inositol,  the  structure  of 
pantothenic  acid  is  almost  specific  for  activity.  A  hydroxypantothenic 
acid  synthesized  by  Mitchell  et  al.  (1940)  has  a  varying  ability  to  replace 
pantothenic  acid  for  some  organisms.  The  activity  of  this  compound 
for  the  Gebriide  Mayer  yeast  was  low  as  compared  with  pantothenic  acid. 

Mode  of  action.  Pantothenic  acid  was  found  to  favor  the  accumula- 
tion of  glycogen  by  yeasts  (Williams  et  al,  1936),  and  to  increase  markedly 
the  rate  of  carbon  dioxide  evolution  by  the  pantothenic  acid-deficient 
Gebriide  Mayer  yeast  (Pratt  and  Williams,  1939).  More  recent  work 
(Novelli  and  Lipmann,  1947)  has  show^n  that  pantothenic  acid  is  phos- 
phorylated  and  acts  as  a  coenzyme.  This  enzyme  system  performs  vari- 
ous oxidations  and  acetylations  in  the  cell. 

PYRIDOXINE 

Pyridoxine  is  also  known  as  adermin  or  as  vitamin  Be.  While  inositol 
and  pantothenic  acid  were  first  investigated  as  growth  factors  for  micro- 
organisms, pyridoxine  was  discovered  as  a  result  of  animal  research. 
This  vitamin  was  isolated  independently  by  five  groups  of  workers  in 
1938  and  was  synthesized  the  next  year.  The  structural  formula  is  given 
below : 


HO- 
CHs- 


■^    n— CH2OH 


Pyridoxine 

Pyridoxine  is  quite  soluble  in  water  and  is  stable  to  acid  and  alkah  but  is 
destroyed  by  light. 

Fungi  deficient  for  pyridoxine.  Partial  or  total  deficiencies  for  this 
vitamin  have  been  reported  for  various  species  and  strains  of  yeasts 
(Eakin  and  Williams,  1939;  Burkholder,  1943).  Among  these  are  Sac- 
charomyces  carlsbergensis  var.  mandshuricus  Y-379,  S.  chodati  Y-140,  S. 
oviformis,  S.  ludwigii,  and  Mycoderma  valida  Y-7. 

Leonian  and  Lilly  (1942)  found  that  the  omission  of  either  thiamine  or 
pyridoxine  alone  from  the  medium  was  without  effect  on  9  of  the  10  strains 


200  PHYSIOLOGY  OF  THE  FUNGI 

of  yeast  tested.  However,  the  omission  of  both  pyridoxine  and  thiamine 
caused  a  decrease  in  the  growth  of  two  of  these  strains.  Apparently 
these  two  yeasts  were  capable  of  synthesizing  either  thiamine  or  pyri- 
doxine, provided  that  the  other  vitamin  w^as  present.  This  is  a  common 
effect  among  fungi  partially  deficient  for  two  or  more  vitamins.  The 
presence  of  one  vitamin  for  which  a  fungus  is  partially  deficient  may 
enable  the  fungus  to  synthesize  other  vitamins  with  greater  ease. 

Among  the  filamentous  fungi,  deficiency  for  pyridoxine  seems  to  be 
characteristic  of  certain  species  of  Ceratostomella  and  a  few  other  fungi 
(Robbins  and  Ma,  1942o,  19426).  Some  species  reported  to  be  deficient 
for  pyridoxine,  with  other  deficiencies  given  in  parentheses,  are  Cerato- 
stomella ulmi,  C.  ips  (thiamine,  biotin),  C.  pseudotsugae  (thiamine),  C. 
piceaperda  240  (biotin),  C.  pini  (thiamine,  biotin),  C  montium  (thiamine, 
biotin),  C.  pilifera,  C.  multiannulata  (thiamine),  C.  pluriannulata  (thia- 
mine), C.  microspora  (thiamine,  biotin),  Ophiostoma  catonianum  (thia- 
mine), Trichophyton  discoides  (thiamine,  inositol). 

Specificity.  One  of  the  important  uses  of  vitamin-deficient  organisms 
is  for  the  purpose  of  vitamin  assay.  Certain  vitamin-deficient  fungi 
and  bacteria  are  used  to  determine  the  vitamin  content  of  foodstuffs  and 
other  natural  products.  For  such  tests  to  be  of  any  value,  it  is  necessary 
to  know  if  the  organism  used  responds  to  substances  other  than  the 
vitamin  itself.  Snell  et  at.  (1942)  found  that  Streptococcus  faecalis  gave 
much  greater  apparent  yields  of  pyridoxine  when  used  for  assay  than  did 
yeast.  It  was  then  discovered  (Snell,  1942)  that  autoclaving  pyridoxine 
with  the  basal  medium  for  20  min.  increased  the  activity  of  pyridoxine 
forty-one  times,  and  that  this  change  in  activity  for  certain  organisms 
was  correlated  with  oxidation  and  heating  with  certain  amino  acids. 

Snell  (1944)  then  postulated  that  vitamers  of  pyridoxine  were  formed 
by  these  treatments.  When  this  problem  was  under  investigation,  these 
vitamers  of  unknown  structure  were  called  "pseudopyridoxine,"  which 
was  later  found  to  consist  of  either  one  or  both  of  the  following  compounds : 


HO- 
CH3 


^^— CHoOH  HO— r^^— CH2OH 

Pyridoxal  Pyridoxamine 


These  two  compounds  were  synthesized  by  Harris  et  al.  (1944)  and  tested 
by  Snell. 

It  was  concluded  that  this  vitamin  consists  of  three  or  more  closely 
related  compounds.  Saccharomyces  carlsbergensis  responds  about  equally 
to  the  three  compounds,  while  the  reaction  of  certain  bacteria  is  much 
greater  to  pyridoxal  and  pyridoxamine  than  to  pyridoxine,     Ceratosto- 


VITAMINS  201 

mella  ulmi  grew  at  a  more  rapid  rate  with  pyridoxamine  than  vvith  pvri- 
doxal  or  pyridoxine  (Snell  and  Rannefelt,  1945).  All  three  forms  of  this 
vitamin  occur  in  natural  products.  Assays  for  this  vitamin  are  discussed 
in  Chap.  10. 

Mode  of  action.  One  of  the  earliest  clues  to  the  action  of  pyridoxine 
was  discovered  by  Snell  and  Guirard  (1943)  in  the  interrelationship 
among  glycine,  alanine,  and  pyridoxine  and  growth  of  Streptococcus 
faecalis  R.  They  found  that  alanine  could  replace  pyridoxine  for  this 
organism  and  that  glycine  caused  inhibition  which  was  overcome  by 
the  addition  of  either  alanine  or  pyridoxine.  It  was  also  found  that 
/3-alanine,  serine,  and  threonine  inhibited  growth.  It  is  possible  that 
alanine  serves  as  a  precursor  for  pyridoxine  in  this  organism,  or  that  one 
function  of  pyridoxine  is  the  synthesis  of  alanine.  At  any  rate  the  action 
of  pyridoxine  appears  to  be  connected  with  either  amino-acid  synthesis 
or  amino-acid  utilization,  or  both.  Like  other  vitamins,  pyridoxine  (or 
its  conversion  products)  has  been  assumed  to  function  in  the  cell  as  a 
part  of  a  coenzyme. 

Pyridoxal  is  phosphorylated  before  it  functions  in  enzyme  systems. 
In  this  it  is  like  thiamine  and  pantothenic  acid.  Pyridoxal  phosphate 
is  said  to  function  as  a  coenzyme  in  the  transformation  of  tryptophane 
into  indole  by  Escherichia  coli  (Wood  et  at.,  1947).  We  may  assume  that 
the  function  of  this  vitamin  is  the  same  in  the  fungi  as  in  the  bacteria. 

p-AMINOBENZOIC   ACID 
p-Aminobenzoic  acid  has  the  following  structure: 

COOH 


NH2 
p-Aminobenzoic  acid 

Rubo  and  Gillespie  (1940)  found  p-aminobenzoic  acid  to  be  a  growth 
factor  for  nine  strains  of  Clostridium  acetohutylicum.  Most  of  the  interest 
in  this  compound  centers  in  its  antagonistic  action  to  sulfonamides.  A 
discussion  of  this  subject  is  presented  in  Chapter  11. 

Fungi  deficient  for  p-aminobenzoic  acid.  Robbins  and  Ma  (1944) 
reported  Rhodotorula  aurantica  to  be  deficient  for  p-aminobenzoic  acid 
and  thiamine.  Concentrations  of  as  low  as  0.03  /xg  per  liter  had  a  positive 
effect  on  the  growth  of  this  yeast,  while  maximum  growth  was  attained 
in  the  presence  of  3  /zg  per  liter.  The  intensity  of  the  pink  color  developed 
by  this  yeast  was  a  function  of  the  p-aminobenzoic  acid  content  of  the 
medium. 


202  PHYSIOLOGY  OF  THE  FUNGI 

In  so  far  as  we  are  aware,  no  filamentous  fungus  isolated  from  nature 
has  been  shown  to  be  deficient  for  p-aminobenzoic  acid.  Tatum  and 
Beadle  (1942)  described  a  mutant  of  Neurospora  which  was  unable  to 
synthesize  this  vitamin.  Wyss  et  at.  (1944)  found  that  the  availability 
of  p-aminobenzoic  acid  to  the  deficient  mutant  of  Neurospora  crassa  was 
a  function  of  the  pH  of  the  medium  (see  Fig.  41). 

Mode  of  action.  The  functions  of  p-aminobenzoic  acid  are  unknown. 
We  may  assume,  on  the  basis  of  the  behavior  of  other  vitamins,  that  it 
functions  as  a  coenzyme,  or  as  a  part  of  a  coenzyme.  Recent  work 
indicates  that  p-aminobenzoic  acid  is  a  constituent  part  of  folic  acid. 

RIBOFLAVIN 
The  structure  of  riboflavin  is  given  below: 

CH2OH 

I 
HC— OH 

I 
HC— OH 

I 
HC— OH 

CH2 

I 

N  N 

\c^  \c=o 

N  C 


s 


Riboflavin 

Many  bacteria,  especially  species  of  Lactobacillus  are  unable  to  synthe- 
size riboflavin  (Peterson  and  Peterson,  1945).  So  far  as  we  are  able  to 
determine,  none  of  the  fungi  isolated  from  nature  have  been  found  to  be 
deficient  for  riboflavin.  This  vitamin  is  synthesized  by  the  fungi. 
Mitchell  and  Houlahan  (1946)  described  a  mutant  of  Neurospora  which 
required  the  addition  of  riboflavin  to  the  medium  for  growth  at  tempera- 
tures above  28°C.  Between  15  and  25°C.  the  growth  rate  of  the  mutant 
without  added  riboflavin  was  equal  to  that  of  the  wild  type.  The  rate 
of  growth  decreased  rapidly  as  the  temperature  increased  from  25  to 

28°C. 

SUMMARY 

It  is  assumed  that  all  living  organisms  require  a  number  of  vitamins,  or 
growth  factors,  for  normal  growth,  reproduction,  and  other  vital  proc- 
esses. However,  organisms  differ  widely  in  their  synthetic  capacities 
for  the  various  vitamins.  Some  fungi  are  self-sufficient  with  respect  to 
vitamins,  being  able  to  synthesize  their  vitamins  from  pure  chemicals 
of  a  synthetic  medium.     Others  lack  the  ability  to  synthesize  sufficient 


VITAMINS  203 

quantities  of  one  or  more  vitamins  and  are  called  vitamin-deficient  fungi. 
The  deficiency  may  be  single  or  multiple,  complete  or  partial.  Partial 
deficiency  may  vary  from  nearly  complete  to  nearly  self-suflficient  and 
is  more  pronounced  during  the  early  stages  of  growth. 

A  single  deficiency  for  thiamine  has  been  more  commonly  reported 
among  filamentous  fungi  than  any  other  type.  Biotin  deficiency  is  like- 
wise commonly  found,  often  in  combination  with  thiamine  deficiency. 
Deficiencies  for  inositol  and  pyridoxine  are  less  common.  Two  filamen- 
tous fungi  isolated  from  nature  are  reported  to  be  deficient  for  nicotinic 
acid.  Numerous  other  deficiencies  have  been  induced  in  mutants  by  ir- 
radiation. Some  yeasts  show  complete  or  partial  multiple  deficiencies 
for  three  to  six  vitamins,  while  relatively  few  filamentous  fungi  are 
deficient  for  as  many  as  three  vitamins. 

Absolute  deficiencies  are  not  known  to  be  influenced  by  the  environ- 
ment, while  conditioned  deficiencies  may  be  affected  either  by  nutritional 
factors  or  by  factors  of  the  physical  environment.  Among  these,  tem- 
perature and  the  composition  and  pH  of  the  medium  seem  to  be  the  most 
important. 

Methods  of  detecting  vitamin  deficiencies  are  exact,  and  accurate 
determination  depends  on  the  ability  or  inability  of  a  fungus  to  grow  on 
a  synthetic  medium  composed  of  pure  chemicals,  to  which  known  amounts 
of  the  various  vitamins  to  be  tested  are  added.  Vegetative  growth, 
measured  by  dry  weight,  is  apparently  the  most  useful  criterion  of  the 
utilization  of  vitamins,  although  reproduction  and  other  processes  are 
likewise  affected. 

Compounds  having  vitamin  activity  but  differing  in  molecular  struc- 
ture are  called  vitamers.  In  general,  only  compounds  of  closely  related 
structure  have  vitamin  activity. 

The  inhibitory  effects  of  vitamins  in  excess  quantities  are  apparently 
common.  They  are  usually  evident  by  slight  reduction  in  rate  or  maxi- 
mum amount  of  growth  and  are  more  common  with  self-sufficient  fungi 
than  with  those  deficient  for  the  vitamin  in  question.  Thiamine  is  more 
commonly  reported  as  a  growth  depressor  than  other  vitamins.  One 
instance  of  severe  inhibition  due  to  excess  inositol  and  temperatures  near 
maximum  for  growth  is  discussed. 

The  known  effects  of  vitamins  on  the  growth  of  fungi  emphasize  the 
important  fact  that  growth  is  a  result  of  a  number  of  interacting  factors, 
among  which  are  the  vitamins.  A  proper  balance  between  the  different 
vitamins  and  with  the  other  nutritional  and  environmental  factors  must 
exist  if  maximum  rate  of  growth  is  to  take  place. 

REFERENCES 

Ar^a  Leao,  a.  E.,  and  A.  Cury:  Sobre  a  exigencia  de  ^cido  nicotlnico  observada  em 
cogumelo  filamentoso;  {Microsporum  audouini),  O  Hospital  (Rio  de  Janiero) 
35:  347-351,  1949. 


204  PHYSIOLOGY  OF  THE  FUNGI 

Atkin,  L.,  W.  L.  Williams,  A.  S.  Schultz,  and  C.  N.  Frey:  Yeast  microbiological 

methods  for  determination  of  vitamins.     Pantothenic  acid,  Ind.  Eng.  Chern., 

Anal.  Ed.  16:  67-71,  1944. 
AxELROD,  A.  E.,  B.  C.  Flinn,  and  K.  Hofmann:  The  metabolism  of  oxybiotin  in 

yeast,  Jour.  Biol.  Chem.  169 :  195-202,  1947. 
Barnett,  H.  L.,  and  V.  G.  Lilly:  The  interrelated  effects  of  vitamins,  temperature 

and  pH  upon  vegetative  growth  of  Sclerotinia  camelliae,  Am.  Jour.  Botany  35: 

297-302,  1948. 
Beadle,  G.  W.  :  An  inositolless  mutant  strain  of  Neurospora  and  its  use  in  bioassays, 

Jour.  Biol.  Chem.  156 :  683-689,  1944. 
Bonner,    D.:    Production    of   biochemical   mutations   in   Penicillium,    Ann.    Jour. 

Botany  33:  788-791,  1946. 
Bonner,  D.,  and  G.  W.  Beadle:  Mutant  strains  of  Neurospora  requiring  nicotin- 
amide or  related  compounds  for  growth,  Arch.  Biochem.  11 :  319-328,  1946. 
Bovarnick,    M.    R.  :    Substitution    of   heated    asparagine-glutamate    mixture    for 

nicotinamide  as  a  growth  factor  for  Bacterium  dysenteriae  and  other  micro- 
organisms. Jour.  Biol.  Chem.  148:  151-161,  1943. 
BuBK,  D.,  M.  L.  Hesselback,  D.  F.  MacNeary,  and  R.  J.  Winzler:  The  vitamer 

concept.  Trans.  N.Y.  Acad.  Sci.  6:  275-283,  1944. 
*BuRKHOLDER,  P.  R. :  Vitamin  deficiencies  in  yeasts.  Am.  Jour.  Botany  30:  206-211, 

1943. 
BuRKHOLDER,  P.  R.,  I.  McVeigh,  and  D.  Moyer:  Studies  on  some  growth  factors 

of  yeasts.  Jour.  Bad.  48:  385-391,  1944. 
Burkholder,  p.  R.,  and  D.  Moyer:  Vitamin  deficiencies  of  fifty  yeasts  and  molds, 

Bull.  Torrey  Botan.  Club  70 :  372-377,  1943. 
Canting,  E.  C.:  The  vitamin  nutrition  of  an  isolate  of  Blastocladia  pringsheimii, 

Am.  Jour.  Botany  35 :  238-242,  1948. 
DU  Vigneaud,  v.,  K.  Dittmer,  E.  Hague,  and  B.  Long:  The  growth-stimulating 

effect  of  biotin  for  the  Diphtheria  bacillus  in  the  absence  of  pimelic  acid,  Science 

96:  186-187,  1942. 
DU  Vigneaud,  V.,  K.  Hofmann,  and  D.  B.  Melville:  On  the  structure  of  biotin. 

Jour.  Am.  Chem.  Soc.  64 :  188-189,  1942a. 
Eakin,  R.  E.,  and  E.  A.  Eakin:  A  biosynthesis  of  biotin.  Science  96:  187-188,  1942. 
Eakin,  R.  E.,  and  R.  J.  Williams:  Vitamin  Be  as  a  yeast  nutrilite.  Jour.  Am.  Chem. 

Soc.  61:  1932,  1939. 
Eastcott,  E.   v.:   Wildiers'  Bios.     The  isolation  and  identification  of  "Bios   1," 

Jour.  Phys.  Chem.  32:  1094-1111,  1928. 
Elliott,  E.  S.  :  Effect  of  vitamins  on  growth  of  some  graminicolous  species  of 

Helminthospori um  and  Fusarium,  Proc.  West  Va.  Acad.  Sci.  20:  65-68,  1949. 
*Fries,  N.:  The  nutrition  of  fungi  from  the  aspect  of  growth  factor  requirements, 

Trans.  Brit.  Mycol.  Soc.  30:  118-134,  1948. 
Fromageot,  C.,  and  J.  L.  Tschang:  Sur  la  synthese  des  pigments  carotenoides  par 

Rhodotorula  sanniei,  .Arch.  Mikrobiol.  9 :  434-448,  1938. 
Gingrich,  W.,  and  F.  Schlenk:  Codehydrogenase  I  and  other  pyridinium  com- 
pounds as   V-factor  for  Hemophilus  influenzae  and  H.  parainfluenzae,   Jour. 

Bad.  47 :  535-550,  1944. 
Goldberg,  M.  W.,  L.  H.  Sternbach,  S.  Kaiser,  S.  D.  Heineman,  J.  Scheiner, 

and  S.  H.  Rubin:  Antibiotin  effect  of  some  biotin  homologs,  Arch.  Biochem. 

14:480-482,  1947. 
Haag,  E.,  and  C.  Dalphin:  Une  Reaction  biochimique  de  I'aneurine  d'une  grande 

sensibility,  Compt.  rend.  soc.  phys.  hist.  nat.  Geneve  57  :  76-77,  1940. 


VITAMINS  205 

Harris,  S.  A.,  D.  Heyi,  and  K.  Folkers:  The  structure  and  synthesis  of  pyridox- 

amine  and  pyridoxal,  Jour.  Biol.  Chem.  154:  315-316,  1944. 
Harris,  S.  A.,  D.  E.  Wolf,  R.  Mozingo,  and  K.  Folkers:  Synthetic  biotin,  Science 

97:447-448,  1943. 
Knight,  B.  C.  J.  G.:  The  nutrition  of  Staphylococcus  aureus:  Nicotinic  acid  and 

vitamin  Bi,  Biochem.  Jour.  31:  731-737,  1937. 
KoGL,  F.,  and  N.  Fries:  Ueber  den  Einfiuss  von  Biotin,  .Aneurin  und  Meso-Inosit 

auf  das  Wachstum  verschiedener  Pilzarten,  Zeit.  physiol.  Chem.  249:  93-110, 

1937. 
KoGL,    F.,   and  B.  Tonnis:   Ueber  das  Bios-Problem.     Darstelking  von  krystal- 

lisiertem  Biotin  aus  Eigelb,  Zeit.  physiol.  Chem.  242 :  43-73,  1936. 
KosER,  S.  A.,  M.  H.  Wright,  and  A.  Dorfman:  Aspartic  acid  as  a  partial  substitute 

for  the  growth  stimulating  effect  of  biotin  on  Torula  cremoris,  Proc.  Soc.  Exptl. 

Biol.  Med.  51 :  204-205,  1942. 
Leonian,  L.  H.,  and  V.  G.  Lilly:  Auxithals  synthesized  by  some  filamentous  fungi, 

Plant  Physiol.  15 :  515-525,  1940. 
Leonian,  L.  H.,  and  V.  G.  Lilly:  The  effects  of  vitamins  on  ten  strains  of  Sac- 

charomyces  cerevisiae,  Am.  Jour.  Botany  29:  459-464,  1942. 
Leonian,  L.  H.,  and  V.  G.  Lilly:  The  comparative  value  of  different  test  organisms 

in  the  microbiological  assaj^  of  B  vitamins,  West  Va.  Agr.  Expt.  Sta.  Bull.  319, 

1945. 
Lilly,  V.  G.,  and  H.  L.  Barnett:  The  influence  of  pH  and  certain  growth  factors 

on  mycelial  growth  and  perithecial  formation  by  Sordaria  fimicola,  Amer.  Jour. 

Botany  34:  131-138,  1947. 
Lilly,  V.  G.,  and  H.  L.  Barnett:  The  inheritance  of  partial  thiamine  deficiency  in 

Lenzites  trabea,  Jour.  Agr.  Research  77:  287-300,  1948. 
Lilly,  V.  G.,  and  L.  H.  Leonian:  The  growth  rate  of  some  fungi  in  the  presence  of 

co-carboxylase,  and  the  moieties  of  thiamin,  Proc.  West  Va.  Acad.  Set.  14 :  44-49, 

1940. 
Lilly,  V.  G.,  and  L.  H.  Leonian:  The  anti-biotin  effect  of  desthiobiotin,  Science  99: 

205-206,  1944. 
LocKHEAD,  A.   G.,  and  G.   B.  Landerkin:  Nutrilite  requirements  of  osmophilic 

yeasts.  Jour.  Bad.  44:  343-351,  1942. 
Mathur,  R.  S.,  H.  L.  Barnett,  and  V.  G.  Lilly:  Sporulation   of  Colletotrichum. 

lindemuthianum  in  culture.  Phytopathology  40:  104-114,  1950. 
Mitchell,  H.  K.,  and  M.  B.  Houlahan:  Neurospora.     IV.  A  temperature-sensi- 
tive riboflavinless  mutant,  Am.  Jour.  Botany  33:  31-35,  1946. 
Mitchell,  H.  K.,  E.  E.  Snell,  and  R.  J.  Williams:  Pantothenic  acid.     IX.  The 

biological  activity  of  hydroxy-pantothenic  acid,  Jour.   Am.   Chem.   Soc.   62 : 

1791-1792,  1940. 
Mueller,  J.  H.:  Studies  on  cultural  requirements  of  bacteria.     X.  Pimclic  acid 

as  a  growth  stimulant  for  C.  diphtheriae,  Jour.  Bad.  34:  163-178,  1937. 
Mueller,  J.  H. :  Nicotinic  acid  as  a  growth  accessory  substance  for  the  diphtheria 

bacillus.  Jour.  Bad.  34:  439-441,  1937a. 
NovELLi,  G.  D.,  and  F.  Lipmann:  Bacterial  conversion  of  pantothenic  acid  into 

co-enzyme  A  (acetylation)  and  its  relation  to  pyruvic  oxidation.  Arch.  Biochem. 

14:23-27,  1947. 
Perlman,  D.:  On  the  nutrition  of  Memnoniella  echinata  and  Stachybotrys  atra,  Am. 

Jour.  Botany  35:  36-41,  1948. 
Peterson,  W.  IL,  and  M.  S.  Peterson:  Relation  of  bacteria  to  \dtamins  and  other 

growth  factors,  Bact.  Revs.  9 :  49-109,  1945. 


206  PHYSIOLOGY  OF  THE  FUNGI 

Pii.GRiM,  F.  J.,  A.  E.  AxELROD,  T.  WiNNicK,  and  K.  Hofmann:  The  microbiological 

activity  of  an  oxygen  analog  of  biotin,  Science  102  :  35-36,  1945. 
Pratt,  E.  F.,  and  R.  J.  Williams:  The  effects  of  pantothenic  acid  on  respiratory 

activity.  Jour.  Gen,  Physiol.  22 :  637-647,  1939. 
RoBBiNS,  W.  J.,  and  F.  Kavanagh:  Thiamin  and  growth  of  Pythium  butleri,  Bull. 

Torrey  Botan.  Club  65:  453-461,  1938. 
RoBBiNS,  W.  J.,  and  F.  Kavanagh:  Thiazole  effect  on  Phycomyces,  Proc.  Natl. 

Acad.  Sci.  U.S.  27:  423-427,  1941. 
♦RoBBiNS,  W.  J.,  and  V.  Kavanagh:  Vitamin  deficiencies  of  the  filamentous  fungi, 

Botan  Rev.  8:  411-471,  1942. 
RoBBiNS,  W.  J.,  and  R.  Ma:  Pimelic  acid,  biotin  and  certain  fungi,  Science  96: 

406-407,  1942. 
*RoBBiNS,  W.  J.,  and  R.  Ma:  Vitamin  deficiencies  of  Ceratostomella  and  related  fungi, 

Am.  Jour.  Botany  29:  835-843,  1942a. 
RoBBiNS,  W.  J.,  and  R.  Ma:  Vitamin  deficiencies  of  Ceratostomella,  Bull.  Torrey 

Botan.  Club  69:  184-203,  19426. 
RoBBiNS,  W.  J.,  and  R.  Ma:  A  Rhodotorula  deficient  for  para-aminobenzoic  acid, 

Science  100 :  85-86,  1944. 
RoBBiNS,  W.  J.,  J.  E.  MacKinnon,  and  R.  Ma:  Vitamin  deficiencies  of  Trichophyton 

discoides,  Bull.  Torrey  Botan.  Club  69:  509-521,  1942. 
RoGOSA,  M.:  Nicotinic  acid  requirements  of  certain  yeasts,  Jour.  Bact.  46:  435-440, 

1943. 
Rosenberg,  H.  R.  :  Chemistry  and  Physiology  of  the  Vitamins,  Interscience  Pub- 
lishers, Inc.,  New  York,  1942. 
Rubin,  S.  H.,  D.  Flowers,  F.  Rosen,  and  L.  Drekter:  The  biological  activity  of 

0-heterobiotin,  Arch.  Biochem.  8:  79-90,  1945. 
RuBBO,  S.  D.,  and  J.  M.  Gillespie:  Para-aminobenzoic  acid  as  a  bacterial  growth 

factor,  Nature  146 :  838-839,  1940. 
Ryan,  F.  J.,  R.  Ballentine,  E.  Stolong,  M.  E.  Corson,  and  L.  K.  Schneider: 

The  biosynthesis  of  pantothenic  acid,  Jour.  Am.  Chein.  Soc.  67 :  1857-1858,  1945. 
ScHOPFER,   W.  H.:  Ueber  die  Wirkung  von  reinen  kristallisierten  Vitaminen  auf 

Phycomyces,  Ber.  d.  deut.  botan.  Ges.  52:  308-311,  1934. 
ScHOPFER,  W.  H.:  Recherches  sur  le  metabolisme  de  I'azote  d"un  microorganisme 

acellulaire  (Phycomyces  blakesleeanus  Bgf.).     Le  role  des  facteurs  de  croissance, 

Protoplasma  28:  381-434,  1937. 
*ScHOPFER,  W.  H.:  Plants  and  Vitamins,  Chronica  Botanica  Co.,  Waltham,  1943. 
ScHOPFER,  W.  H.,  and  M.  Guilloud:  Recherches  experimentales  sur  les  facteurs  de 

croissance  et  le  pourvoir  de  synthese  de  Rhizopus  cohnii  Berl.  et  de  Toni  (Rhizo- 

pus  suinus  Neilson),  Zeit.  fur  Vitaminforsch.  16:  181-296,  1945. 
Snell,  E.  E.  :  Effect  of  heat  sterilization  on  growth-promoting  activity  of  pyridoxine 

for  Streptococcus  lactis  R.,  Proc.  Soc.  Exptl.  Biol.  Med.  51:  356-358,  1942. 
Snell,  E.  E.:  The  vitamin  activities  of  "pyridoxal"  and  "pyridoxamine,"  Jour. 

Biol.  Chem.  154:  313-314,  1944. 
Snell,  E.  E.,  and  B.  M.  Guirard:  Some  interrelationships  of  pyridoxine,  alanine 

and  glycine  in  their  effect  on  certain  lactic  acid  bacteria,  Proc.  Natl.  Acad.  Sci. 

U.S.  29:  66-73,  1943. 
Snell,  E.  E.,  B.  M.  Guirard,  and  R.  J.  Williams:  Occurrence  in  natural  products 

of  a  physiologically  active  metabolite  of  pyridoxine,   Jour.  Biol.  Chem.    143 : 

519-530,  1942. 
Snell,  E.  E.,  and  A.  N.  Rannefelt:  The  vitamin  Be  group.     III.  The  vitamin 

activity  of  pyridoxal  and  pyridoxamine  for  various  organisms,  Jour.  Biol.  Chem. 

157 :  475-489,  1945. 


VITAMINS  207 

Stokes,  J.  L.,  J.  W.  Foster,  and  H.  B.  Woodward:  Synthesis  of  pyridoxin  by  a 

"pyridoxin-less"   x-ray   mutant   of   Neurospora   sitophila,    Arch.    Biochem.    2: 

235-245,  1943. 
Stokes,  J.  L.,  and  M.  Gunness:  Microbiological  activity  of  synthetic  biotin,  its 

optical  isomers  and  related  compounds,  Jour.  Biol.  Chem.  157:  121-126,  1945. 
Stokes,  J.  L.,  A.  Larsen,  and  M.  Gunness:  Biotin  and  the  synthesis  of  aspartic 

acid  by  microorganisms.  Jour.  Biol.  Chem.  167:  613-614,  1947. 
Strauss,  B.  S.:  Studies  on  the  Be-requiring,  pH-sensitive  mutants  of  Neurospora 

crassa,  Arch.  Biochem.  30:  292-305,  1951. 
Tatum,  E.  L.,  and  G.  W.  Beadle:  Genetic  control  of  biochemical  reactions  in 

Neurospora:  an  "aminobenzoic-less"  mutant,  Proc.  Natl.  Acad.  Sci.  U.S.  28: 

234-243,  1942. 
Tatum,  E.  L.,  and  G.  W.  Beadle:  Biochemical  genetics  of  Neurospora,  Ann.  Mis- 
souri Botan.  Garden  32  :  125-129,  1945. 
Weinstock,  H.  H.,  H.  K.  Mitchell,  E.  F.  Pratt,  and  R.  J.  Williams:  Pantothenic 

acid.     IV.  Formation  of  beta-sAanme  by  cleavage.  Jour.  Am.  Chem.  Soc.  61: 

1421-1425,  1939. 
Williams,  R.  J.,  C.  M.  Lyman,  G.  H.  Goodyear,  and  J.  H.  Truesdail:  Is  the 

nutrilite  for  Gebriide  Mayer  yeast  of  universal  biological  importance?,  Jour. 

Am.  Chem.  Soc.  54:  3462-3463,  1932. 
Williams,  R.  J.,  W.  A.  Mosher,  and  E.  Rohrman:  The  importance  of  "panto- 
thenic acid"  in  fermentation,  respiration  and  glycogen  storage,  Biochem.  Jour. 

30:  2036-2039,  1936. 
Williams,  R.  R.,  and  T.  D.  Spies:  Vitamin  Bi  (Thiamin)  and  Its  Use  in  Medicine, 

The  Macmillan  Company,  New  York,  1938. 
WiNZLER,  R.  J.,  D.  Burk,  and  V.  du  Vigneaud:  Biotin  in  fermentation,  respiration 

and  nitrogen  assimilation  by  yeast.  Arch.  Biochem.  6 :  25-47,  1944. 
WiRTH,  J.  C,  and  F.  F.  Nord:  Essential  steps  in  the  enzymatic  breakdown  of 

hexoses  and  pentoses.     Interaction  between  dehydrogenation  and  fermentation, 

Arch.  Biochem.  1:  143-163,  1942. 
Wood,  W.  A.,  I.  C.  Gunsalus,  and  W.  W.  Umbreit:  Function  of  pyridoxal  phos- 
phate :  Resolution  and  purification  of  the  tryptophanase  enzyme  of  Escherichia 

coli,  Jour.  Biol.  Chem.  170:  313-321,  1947. 
Woolley,  D.  W.:  The  nature  of  the  antialopecia  factor,  Science  92:  384-385,  1940. 
Wooster,  R.  C.,  and  V.  H.  Cheldelin:  Growth  requirements  of  Penicillium  digita- 

tum,  Arch.  Biochem.  8:  311-320,  1945. 
Wyss,  O.,  V.  G.  Lilly,  and  L.  H.  Leonian:  The  effect  of  pH  on  the  availability  of 

p-aminobenzoic  acid  to  Neurospora  crassa,  Science  99 :  18-19,  1944. 


CHAPTER  10 
FUNGI  AS  TEST  ORGANISMS 

Numerous  physiological  problems  are  accessible  to  investigation 
through  the  use  of  microorganisms.  By  the  proper  choice  of  deficient 
organisms,  it  is  feasible  to  detect  minute  amounts  of  physiologically  active 
compounds  such  as  the  vitamins  and  amino  acids.  Knowledge  has  been 
gained  of  the  way  vitamins  and  amino  acids  are  synthesized  and  destroyed 
by  various  organisms.  The  amino-acid  composition  of  proteins  and  the 
availability  of  certain  essential  elements  in  soil  may  be  determined  by  the 
use  of  fungi  and  bacteria.  These  highly  practical  studies  are  based  upon 
a  knowledge  of  the  compounds  and  elements  essential  for  the  nutrition 
of  microorganisms.  Since  these  are,  in  general,  the  same  elements  and 
compounds  needed  by  animals,  there  is  a  very  close  relation  between 
fungus  and  animal  physiology  in  nutritional  problems.  Foodstuffs  for 
man  and  animals  are  the  most  common  materials  analyzed  in  routine 
assays. 

Some  of  the  advantages  which  have  contributed  to  the  widespread  use 
of  microorganisms  for  assay  purposes  are  simple  technique  and  apparatus, 
sensitivity,  specificity,  and  the  short  time  required.  Perhaps  the  most 
important  single  factor  is  the  small  sample  needed  and  the  fact  that  little 
or  no  purification  or  concentration  of  the  active  material  is  required. 
These  advantages  are  to  be  compared  with  chemical  methods  or  the  use 
of  animals  for  obtaining  the  same  information.  All  analytical  methods 
have  advantages  and  disadvantages.  A  knowledge  of  the  limitations  of 
any  method  is  essential  for  valid  results. 

Most  microbiological  assays  depend  upon  the  proportional  response  of 
deficient  test  organisms  to  the  substances  for  which  they  are  deficient. 
This  proportional  response  occurs  only  for  a  limited  range  of  concentra- 
tions. The  usable  range  of  concentration  depends  upon  the  substance 
being  assayed,  the  test  organism,  and  the  basal  medium.  In  theoiy, 
any  organism  may  be  used  to  assay  any  substance  for  which  it  is  deficient, 
but  in  practice  not  all  organisms  having  the  same  deficiency  are  equally 
suitable.  For  example,  Rhizohium  trijolii  205  is  about  100  times  as  sen- 
sitive to  biotin  as  Sordaria  fimicola. 

The  following  are  essential  to  any  quantitative  microbiological  assay: 
(1)  a  suitable  test  organism;  (2)  the  preparation  of  a  basal  medium  ade- 
quate in  all  respects,  but  essentially  free  from  the  substance  to  be  assayed ; 
(3)  liberation,  from  the  material  to  be  analyzed,  of  the  substance  to  be 

208 


FUNGI  AS  TEST  ORGANISMS  209 

assayed,  in  a  water-soluble  condition;  (4)  a  standard  sample  of  the  sub- 
stance to  be  analyzed ;  (5)  preparation  of  a  range  of  concentrations  of  the 
known  and  unknown  substances  in  the  basal  medium ;  (6)  uniform  inocula- 
tion; (7)  incubation  under  uniform  conditions;  (8)  measuring  the  response 
of  the  test  organism;  (9)  construction  of  the  standard  curve  from  the 
response  of  the  test  organism  to  kno\\Ti  amounts  of  the  substance  under 
test;  (10)  calculating  the  content  of  the  substance  contained  in  the  sample. 
The  above  discussion  assumes  the  use  of  pure  compounds  in  obtaining 
standard  curves.  The  utility  of  microbiological  assay  methods  is  not 
confined  to  the  assay  of  known  compounds.  They  are  of  great  utihty 
in  studies  of  methods  of  isolation  of  new  growth  factors  and  other  active 
compounds.  These  occur  in  complex  natural  products  and,  before  they 
are  isolated,  are  known  only  by  the  physiological  effects  they  produce  in 
living  organisms.  Given  a  deficient  fungus,  or  other  organism,  it  is  pos- 
sible to  follow  the  efficiency  of  the  various  steps  in  an  isolation  procedure. 
The  isolation  of  many  of  the  water-soluble  vitamins  has  been  facilitated 
by  the  use  of  test  fungi.  The  use  of  a  biotin-deficient  yeast  enabled 
Kogl  and  Tonnis  (1936)  to  isolate  biotin  for  the  first  time  as  a  pure 
compound. 

GENERAL  PROCEDURES 

The  following  discussion  of  the  steps  involved  in  microbiological  assay 
may  serve  also  as  a  guide  to  the  quantitative  study  of  the  physiology  of 
fungi.  Such  studies  are  the  surest  way  to  gain  knowledge  and  under- 
standing of  the  physiology  of  the  fungi. 

Selection  of  test  organisms.  The  first  requirement  of  a  test  organism 
is  specificity  for  the  compound  under  assay.  A  fungus  which  responds 
to  either  or  both  moieties  of  thiamine  is  less  suitable  than  one  which 
requires  the  intact  thiamine  molecule.  Other  considerations  may  out- 
weigh the  advantages  of  strict  specificity,  but  the  response  of  the  test 
organism  to  moieties,  vitamers,  and  related  compounds  must  be  known. 

Other  considerations  besides  specificity  enter  into  the  selection  of  test 
organisms.  Test  organisms  should  be  easily  maintained  in  culture, 
easily  handled  in  the  laboratory,  and  have  stable  biochemical  character- 
istics. Rapid  and  uniform  growth  is  desirable.  The  habit  of  growth  is 
important.  A  fungus  which  forms  mucilaginous  colonies  which  adhere 
to  the  walls  of  the  flasks  is  difficult  to  harvest,  and  yeasts  which  clump 
are  difficult  to  determine  by  turbidimetric  methods. 

The  basal  medium.  Except  for  the  compound  or  element  under 
investigation  the  basal  medium  should  be  complete  and  balanced.  If  a 
test  organism  is  deficient  for  more  than  one  factor,  all  the  factors  except 
the  one  under  investigation  should  be  present  in  optimum  amounts. 
Other  requirements  are  easily  available  sources  of  carbon  and  nitrogen 
and  a  medium  which  is  adequately  buffered  in  the  optimum  pH  range. 


210  PHYSIOLOGY  OF  THE  FUNGI 

The  basal  medium  should  be  essentially  free  from  the  vitamin  or  other 
factor  under  test.  The  response  of  the  test  organism  to  the  basal 
medium  should  be  slight;  this  value  is  known  as  the  blank,  or  control. 
The  size  of  the  blank  depends  upon  the  residual  concentration  of  the 
factor  in  the  basal  medium  and  the  amount  and  kind  of  inoculum  used. 
The  degree  to  which  a  basal  medium  should  be  freed  of  the  substance 
under  test  depends  upon  the  sensitivity  of  the  test  organism. 

The  best  basal  medium  for  any  test  organism  can  be  determined  only 
after  a  prolonged  investigation  of  the  nutritional  requirements  of  the 
organism.  This  arduous  task  is  too  infrequently  attempted.  Fre- 
quently, it  is  desirable  to  use  some  natural  material  in  the  medium.  A 
complex  medium  which  supplies  several  sources  of  carbon  and  nitrogen 
as  well  as  other  organic  compounds  may  support  more  rapid  growth  than 
a  simple  minimal  medium. 

The  sample  being  analyzed  may  contain  compounds  which  stimulate 
or  depress  growth.  Stimulation  of  growth  due  to  the  presence  of  acces- 
sory factors,  is  perhaps  more  often  encountered  than  growth  depression. 
The  adequacy  of  the  basal  medium  may  be  tested  by  comparing  the 
growth  curve  obtained  on  the  sample  with  the  standard  curve.  If  the 
response  of  the  test  organism  to  the  sample  is  due  solely  to  the  factor  con- 
tained in  the  sample,  the  two  curves  will  be  identical.  The  presence  of 
inhibiting  substances  in  the  sample  is  detected  when  the  sample  curve 
falls  below  the  standard  curve.  Stimulating  substances  are  revealed  by 
an  upward  drift  of  the  sample  curve. 

If  biologically  pure  compounds  were  available,  the  preparation  of  basal 
media  for  assay  purposes  would  be  greatly  simplified.  No  general 
method  of  purification  is  useful  for  all  purposes.  Riboflavin  is  destroyed 
by  light,  and  media  can  be  freed  of  this  vitamin  by  exposure  to  strong 
illumination.  Activated  charcoal  (Norit  or  Darco)  is  very  useful  in 
adsorbing  residual  traces  of  many  vitamins.  Recrystallization  of  sugars, 
asparagine,  and  mineral  salts  is  helpful  in  some  instances.  Casein  is 
extracted  with  hot  alcohol  to  remove  vitamins.  The  essential  micro 
elements  may  be  removed  in  the  ways  discussed  in  Chap.  5.  Frequently 
reagents  made  by  one  manufacturer  are  purer  in  certain  respects  than 
those  of  another. 

Three  basal  media  which  have  been  used  for  fungi  in  microbiological 
assays  are  given  below. 

Glucose-Asparagine 

Glucose 30  e, 

Asparagine 1  g. 

MgS04-7HoO 0.5  g. 

KH2PO4 1.5  g. 

Distilled  water  to  make 1  liter 


FUNGI  AS  TEST  ORGANISMS  211 

This  medium  was  used  for  thiamine  assay  using  Phycomyces  blake- 
sleeanus  as  the  test  fungus  (Schopfer,  1945). 

Sucrose-Ammonium  Tartrate-Ammonium  Nitrate 

Sucrose 20  g. 

KH2PO4 1  g. 

MgS04-7H20 0.5  g. 

Ammonium  tartrate 5 . 0  g. 

NH4NO3 1.0  g. 

NaCl 0.1  g. 

CaCli 0.1  g. 

B 0.01  mg. 

Mo 0 .  02  mg. 

Fe 0.2  mg., 

Cu 0.1  mg. 

Mn 0.02  mg. 

Zn 2.0  mg. 

Bio  tin 5  Mg 

Distilled  water  to  make 1,000  ml. 

This  medium  was  used  by  Horowitz  and  Beadle  (1943)  and  by  Beadle 
(1944)  for  the  assay  of  choline  and  inositol  by  biochemical  mutants  of 
Neurospora  crassa. 

Glucose-Casein  Hydrolysate 

Glucose 25  g. 

Casein  hydrolysate  equivalent  to 2  g.  casein 

MgS04-7H20 0.5  g. 

KH2PO4 l.Og. 

Fumaric  acid 1 .  32  g. 

NaaCOs 1.12  g. 

Fe+  +  +  as  sulfate 0.2  mg. 

Zn+  +  as  sulfate 0.2  mg. 

Mn+  +  as  sulfate 0.1  mg. 

Distilled  water  to  make 1  liter 

This  medium  was  used  by  Leonian  and  Lilly  (1945)  for  the  assay  of 
certain  vitamins.  Various  deficient  yeasts  and  filamentous  fungi  were 
used  as  test  organisms.  This  medium  is  suitable  for  testing  fungi  for 
vitamin  deficiencies. 

Preparing  for  an  assay.  In  general,  the  compound  being  assayed 
should  be  brought  into  aqueous  solution  before  assaying.  Many  vita- 
mins occur  in  a  ''bound"  condition  and  must  be  liberated  before  analysis. 
The  procedure  used  to  liberate  bound  vitamins  depends  upon  the  vita- 
min involved,  as  well  as  the  nature  of  the  substance  being  assayed. 
Snell  (1948)  has  listed  tentative  methods  for  the  liberation  of  the  various 
vitamins.  In  general,  acid  or  enzymatic  hydrolysis  is  used.  Proteins 
are  hydrolyzed  before  amino-acid  assay.     Acid  hydrolysis  is  destructive 


212  PHYSIOLOGY  OF  THE  FUNGI 

to  certain  amino  acids,  especially  tryptophane.  Alkaline  hydrolysis  of 
proteins  has  been  recommended  for  this  amino  acid  (Greene  and  Black, 
1944). 

The  concentrations  of  the  standard  compound  and  of  the  sample  for 
assay  should  be  so  chosen  that  the  response  of  the  test  organism  is  roughly 
linear.  Every  concentration  should  be  run  in  duplicate.  Control  flasks 
to  which  neither  the  standard  compound  nor  the  assay  sample  have  been 
added  should  form  a  part  of  every  assay.  This  provides  a  means  of 
evaluating  the  basal  medium  and  should  never  be  omitted. 

The  type  of  culture  vessel  and  the  volume  of  the  basal  medium  used 
will  depend  upon  the  test  organism.  Bacteria  are  frequently  cultured 
in  test  tubes.  These  are  also  useful  for  yeasts.  Uniform  test  tubes 
which  can  be  used  in  a  photoelectric  colorimeter  allow  measurement  of 
turbidity  without  transfer  (Lindegren  and  Raut,  1947).  The  filamentous 
fungi  are  usually  cultured  in  Erlenmeyer  flasks.  The  volume  of  medium 
should  be  so  chosen  that  the  liquid  is  less  than  1  cm.  deep.  All  glassware 
must  be  clean.  Accuracy  in  measuring  the  basal  medium  and  the  known 
and  unknown  solutions  is  essential. 

Inoculation  and  incubation.  The  medium  upon  which  the  inoculum 
is  grown  should  be  complete  and  contain  an  adequate  but  not  excessive 
amount  of  the  factor  under  investigation.  Certain  fungi,  especially  the 
yeasts,  cease  to  be  deficient  for  certain  vitamins  when  continuously  cul- 
tured upon  media  free  from  these  factors. 

Spore  inoculum  may  be  used  with  advantage  with  many  filamentous 
fungi.  Frequently  it  is  desirable  to  use  germinated  spores  for  inoculum. 
Phycomyces  blakesleeanus  spores  require  the  Z  factors  for  rapid  germina- 
tion (Robbins,  1940).  If  the  test  sample  contains  these  factors  and  the 
basal  medium  does  not,  early  growth  will  be  more  rapid  in  the  sample 
series.  It  is  convenient  to  germinate  the  spores  of  this  fungus  and  others 
by  preparing  a  spore  suspension  in  dilute  peptone  solution  a  few  hours 
before  inoculation.  These  germinated  spores  grow  essentially  without 
interruption  and  shorten  the  time  of  incubation.  Fragmented  mycelium 
may  also  be  used  to  advantage.  A  uniform  amount  of  inoculum  must  be 
used.  This  is  easy  to  achieve  when  a  suspension  of  spores  or  fragmented 
mycelium  is  used.  Inocula  of  these  types  provide  a  multitude  of  growing 
points,  which  results  in  uniform  growth.  Disks  of  mycelium  on  agar 
are,  in  general,  unsatisfactory. 

An  obvious  advantage  of  using  a  large  amount  of  inoculum  is  the 
shorter  time  required  for  an  assay.  However,  there  is  danger  of  intro- 
ducing with  a  large  inoculum  enough  of  the  substance  under  investigation 
to  give  abnormally  high  blanks.  Washing  the  inoculum  with  sterile 
distilled  water  reduces  this  hazard  but  increases  the  work  and  multiplies 
the  chances  of  contamination.  A  very  small  inoculum  results  in  a  longer 
lag  period,  and  the  time  required  for  analysis  may  be  prolonged. 


FUNGI  AS  TEST  ORGANISMS  213 

Test  organisms  during  an  assay  should  be  cultured  under  uniform 
conditions  with  respect  to  light  and  temperature.  In  general,  the  fila- 
mentous fungi  should  not  be  agitated  during  the  period  of  incubation. 
Yeasts  are  frequently  grown  with  continuous  or  intermittent  shaking. 

There  are  two  schools  of  thought  concerning  the  time  of  incubation 
for  assay.  The  first  recommends  a  uniform  short  period  of  growth  and 
determination  of  the  yield  before  the  organism  reaches  its  maximum 
development.  There  is  a  saving  in  time  in  this  method,  but  the  influence 
of  accessory  factors  in  the  sample  may  make  such  results  unreliable.  A 
comparison  should  always  be  made  between  the  analytical  data  for  a 
short  and  a  long  period  of  incubation  before  choosing  the  length  of  incu- 
bation period.  In  general,  we  feel  that  assays  tend  to  be  more  reliable 
when  the  period  of  incubation  is  long  enough  to  allow  maximum 
development  of  the  test  organism. 

Measuring  the  response.  The  methods  used  for  measuring  the 
response  of  test  organisms  vary.  The  growth  response  of  bacteria  may 
be  measured  either  by  titrating  the  acid  produced  or  by  determining  the 
turbidity  with  a  suitable  photoelectric  colorimeter.  The  growth  response 
of  yeasts  may  be  measured  as  turbidity,  or  the  cells  may  be  weighed. 
The  first  procedure  is  by  far  the  simpler.  The  growth  of  filamentous 
fungi  is  commonly  measured  by  collecting  the  mycelium  and  determining 
the  drj^  weight  (see  discussion  in  Chap.  3). 

Calculation  of  results.  A  growth  curve  (acidity,  turbidity,  or  weight) 
is  plotted  from  the  response  of  the  test  organism  to  the  different  concen- 
trations of  the  standard  substance.  The  concentration  of  the  substance 
in  the  sample  is  then  calculated  from  the  standard  curve.  It  is  necessary 
to  use  a  new  standard  curve  for  each  series  of  assays.  Unsuspected 
variations  in  the  basal  medium  and  in  technique  from  day  to  day  make 
this  precaution  necessary.  In  making  the  calculations,  it  is  assumed  that 
equal  amounts  of  the  substance,  whether  as  a  pure  compound  or  in  the 
sample,  will  cause  the  same  amount  of  response  by  the  test  organism. 
It  is  customary  to  report  the  concentrations  of  vitamins  and  micro  essen- 
tial elements  in  micrograms  per  gram  of  original  sample. 

As  an  example  of  the  type  of  calculation  involved  in  an  assay,  the 
standard  curve  (Fig.  43)  and  protocol  of  a  biotin  assay  are  given  below. 
The  substance  assayed  was  air-dry  yeast  cells.  Biotin  was  liberated 
from  the  sample  by  acid  hydrolysis,  and  the  cell  extract  was  neutralized 
and  made  up  to  such  volume  that  1  ml.  of  hydrolysate  was  equivalent 
to  50  mg.  of  original  yeast  cells.  The  test  organism,  Saccharomyces 
cerevisiae,  Gebriide  Mayer  strain,  was  incubated  for  72  hr  at  25°C. 
Twenty-five  milliliters  of  glucose-casein  hydrolysate  medium  was  used 
per  250-ml.  flask.  The  cultures  were  agitated  10  min.  every  hour.  The 
data  for  the  response  of  the  test  organism  to  varying  amounts  of  yeast 
hydrolysate  are  given  in  Table  35. 


214 


PHYSIOLOGY  OF  THE  FUNGI 


Table  35.     Yield  of  Saccharomyces  cerevisiae  Cells  Produced  when  Different 

Amounts  of  Yeast  Hydrolysate  Were  Added  to  25  Milliliters  of  a 

Biotin-free  Glucose-Casein  Hydrolysate  Medium 


Yeast  hydrolysate, 

Equivalent  weight 
of  sample,  mg. 

Yield,  mg. 

ml.  per  flask 

Flask  1 

Flask  2 

0.03125 
0.0625 
0.125 
0.25 

1.5625 
3.125 
6.25 
12.5 

9.2 
20.2 
32.8 
48.0 

8.9 
19.0 
32.8 
47.6 

The  amount  of  biotin  in  the  original  sample  may  then  be  calculated. 
The  amount  of  biotin  in  6.25  mg.  of  the  sample  produced  32.8  mg.  of 


60 


50 


«40 


S30 


©20 


10 


^'■^'"'^ 

1 

^( 

^ 

/ 

^ 

c 

/ 

i 

/ 

/ 

0  0.001      0002  O004  0.006  0.008 

Micrograms  of  biotin  per  flosk 

Fig.  43.  Standard  curve  for  a  biotin  assay  using  Saccharomyces  cerevisiae,  Gebriide 
Mayer  strain,  as  the  test  fungus.  Basal  medium  was  glucose-casein  hydrolysate, 
25  ml.  per  250-ml.  Erlenmeyer  flask.  Cultures  were  incubated  at  25°C.,  agitated  10 
min.  each  hour,  and  harvested  after  72  hr. 

dry  yeast  cells.  From  the  standard  curve  this  is  seen  to  be  equivalent 
to  0.0025  jug  of  biotin.  The  biotin  content  of  the  sample  is  therefore 
equal  to  0.0025  X  1,000/6.25,  or  0.4  /xg  of  biotin  per  gram  of  sample. 

VITAMIN  ASSAYS 

It  is  beyond  the  intent  of  this  chapter  to  include  detailed  information 
about  techniques  in  connection  with  individual  assays.  The  following 
references  are  useful  for  entry  into  the  hterature.  Schopfer  (1945)  has 
considered  the  philosophy  underlying  the  use  of  microorganisms  for 
assay.  Leonian  and  Lilly  (1945)  investigated  the  use  of  many  test 
organisms  to  assay  the  vitamin  content  of  a  single  substance.  This  work 
showed  that  widely  different  assay  values  for  some  vitamins  are  obtained 


FVNGI  AS  TEST  ORGANISMS  215 

when  different  test  organisms  are  used.     The  review  of  Snell   (1948) 
represents  the  critical  judgment  of  an  active  investigator  in  this  field. 

While  the  filamentous  fungi  are  frequently  passed  over  in  favor  of 
bacteria  and  yeasts,  they  offer  certain  advantages  when  only  simple 
apparatus  is  available,  or  where  occasional  assays  are  to  be  made.  The 
test  organisms  for  the  specific  vitamins  listed  below  are  in  part  those 
recommended  by  Snell  (1948). 

Thiamine.  Phycomyces  hlakesleeanus.  This  fungus  responds  to  the 
two  moities  of  thiamine.  Schopfer  (1935,  1945)  used  a  glucose-asparagine 
medium  and  used  dry  weight  of  mycelium  to  measure  growth.  This  is 
an  excellent  organism  to  use  in  gaining  experience  with  a  microbiological 
assay.  Schultz  et  al.  (1942)  used  Saccharomyces  cerevisiae  (Fleischmann's 
baker's  yeast)  and  measured  the  evolution  of  carbon  dioxide,  which  was 
proportional  to  the  thiamine  content  of  the  sample. 

Pyridoxine.  Saccharomyces  carlsbergensis.  Snell  (1945a)  found  that 
this  yeast  responds  about  equally  to  pyridoxine,  pyridoxal,  and  pyridoxa- 
mine.  Growth  may  be  measured  turbidimetrically  or  by  weighing  the 
cells.     Differential  assays  for  these  three  vitamers  have  been  devised. 

p-Aminobenzoic  acid.  Neurospora  crassa  mutant.  Various  labora- 
tories have  used  this  organism  (Tatum  et  al,  1946).  For  the  effect  of 
pH  on  utilization  of  this  vitamin  see  Wyss  et  al.  (1944). 

Pantothenic  acid.  Saccharomyces  carlsbergensis.  Most,  if  not  all, 
yeasts  respond  to  the  )3-alanine  moiety  of  pantothenic  acid.  Atkin  et  al. 
(1944)  noted  that  the  incorporation  of  Z-asparagine  in  the  basal  medium 
reduced  interference  due  to  /3-alanine. 

Nicotinic  acid.  Lactobacillus  arabinosus.  This  organism  responds 
equally  to  nicotinic  acid  and  nicotinamide.  Growth  may  be  measured 
either  by  titrating  the  acid  produced,  or  turbidimetrically  (Krehl  et  al., 
1943).  Zygosaccharomyces  marxianus  was  used  by  Leonian  and  Lilly 
(1945). 

Inositol.  Neurospora  crassa  mutant.  This  mutant  was  first  used  by 
Beadle  (1944)  to  assay  inositol.  It  is  an  easy  organism  to  handle,  and 
since  this  mutant  forms  few  conidia,  it  is  not  a  great  source  of  contamina- 
tion to  a  laboratoiy.  Snell  (1948)  recommends  the  use  of  Saccharomyces 
carlsbergensis  for  inositol  assay. 

Biotin.  Saccharomyces  cerevisiae.  Various  strains  have  been  used. 
Many,  if  not  all,  strains  respond  also  to  desthiobiotin  (Lilly  and  Leonian, 
1944).  The  existence  of  many  biotin  vitamers  makes  the  choice  of  a  test 
organism  difficult.  Neurospora  crassa  and  N.  sitophila  may  also  be  used. 
It  is  probable  that  some  of  the  divergence  of  assay  values  obtained  when 
different  test  organisms  are  used  is  due  to  biotin  complexes.  Such  a 
complex,  biocytin,  has  been  isolated  by  Wright  et  al.  (1950).  The 
analytical  results  were  unchanged  by  acid  hydrolysis  when  Lactobacillus 


216  PHYSIOLOGY  OF  THE  FUNGI 

casei  was  used  but  were  increased  when  L.   arahinosus  was  the  test 
organism. 

Riboflavin.  Lactohacillus  casei.  Fatty  acids  stimulate  growth. 
Growth  may  be  measured  by  titrating  the  acid  formed,  or  turbidimetri- 
cally  (Roberts  and  Snell,  1946).  It  is  probable  that  mutants  of  Neuro- 
spora  deficient  for  this  vitamin  may  also  be  used  in  assay. 

AMINO-ACID  ASSAYS 

The  importance  of  the  amino-acid  composition  of  proteins  used  in 
animal  nutrition  makes  any  advance  in  analytical  methods  of  great 
interest  and  value.  The  general  techniques  for  amino-acid  determina- 
tions by  microbiological  procedures  are  the  same  as  for  other  assays. 
The  first  requirement  of  this  type  of  microbiological  assay  is  a  suitable 
test  organism.  Few  fungi  isolated  from  nature  are  deficient  for  amino 
acids.  For  this  reason  bacteria  have  been  extensively  used.  The  follow- 
ing references  will  give  an  entry  into  the  literature  on  the  use  of  bacteria 
for  amino-acid  assay:  Hutchings  and  Peterson  (1943);  Shankman  (1943); 
Dunn  et  at.  (1944);  Snell  (1945);  and  Horn  et  at.  (1950). 

Some  mutants  of  N'eurospora  have  been  found  to  be  deficient  for  amino 
acids.  Mutants  having  the  following  amino-acid  deficiencies  have  been 
studied:  leucine,  isoleucine,  valine,  lysine,  methionine,  serine,  or  glycine. 
Only  the  mutant  deficient  for  leucine  appears  to  have  been  much  used  in 
microbiological  assay  (Ryan  and  Brand,  1944;  Brand  et  al.,  1945).  The 
growth  of  a  lysine-deficient  mutant  was  completely  inhibited  by  arginine 
when  the  molecular  ratio  of  arginine  to  lysine  was  2  to  1  (Doermann, 
1944). 

Ryan  (1948)  has  considered  the  possibility  of  microbiological  assay 
of  amino  acids  by  observing  the  percentage  of  germination  of  conidia 
from  deficient  mutants  in  the  presence  of  different  concentrations  of  the 
specific  amino  acid.  An  assay  can  be  completed  within  a  few  hours  by 
this  method.  Unfortunately  the  inhibiting  action  of  certain  amino  acids 
introduces  complications  into  the  proposed  method. 

Mutations  of  Neurospora  and  certain  other  fungi  have  been  induced 
by  chemicals,  such  as  nitrite  or  nitrous  acid,  colchicine,  nitrogen  mustard 
gas,  and  hydrogen  peroxide,  or  by  irradiation  with  ultraviolet  and  X  rays. 
These  mutants  are  frequently  characterized  by  inability  to  synthesize 
various  metabolites.  They  differ  from  the  parent  wild  type  in  that  one 
or  more  genes  have  been  inactivated.  It  is  thought  that  each  gene  con- 
trols a  single  biochemical  reaction.  Mutants  having  the  same  gross 
deficiency  may  differ  in  the  specific  gene  inactivated. 

Horowitz  (1947)  studied  four  mutants  of  Neurospora  which  were  unable 
to  synthesize  methionine  from  inorganic  sources  of  nitrogen  and  sulfur. 
One  of  these  mutants  was  able  to  grow  in  the  presence  of  cysteine,  cysta- 


FUNGI  AS  TEST  ORGANISMS  217 

thionine,  homocysteine,  and  methionine.  The  second  was  unable  to 
utihze  cysteine  but  was  able  to  utilize  the  other  three  compounds.  The 
third  isolate  utilized  either  homocysteine  or  methionine,  while  the  fourth 
isolate  utilized  only  methionine.  From  these  results  the  steps  in  the 
synthesis  of  methionine  and  the  genes  inactivated  may  be  summarized  as 

gene  4  gene  3  gene  2  gene  1 

follows : >  cysteine >  cystathionine >  homocysteine  — -^  methi- 
onine.    From  similar  studies  Srb  and  Horowitz  (1944)  concluded  that 

Neurospora  synthesizes  arginine  as  follows:  ornithine  — -^  citrulline > 

arginine. 

Fungus  mutants  have  proved  to  be  powerful  tools  for  investigating 
pathways  of  synthesis  and  utilization  of  vitamins,  amino  acids,  and  other 
compounds,  and  in  studies  of  biochemical  mutations.  From  these  studies 
also  comes  the  realization  that  each  step  in  the  synthesis  or  utilization  of 
a  compound  may  be  controlled  or  limited  by  a  specific  gene.  The  review 
papers  of  Bonner  (194G)  and  Beadle  (1945,  1945a)  should  be  consulted 
for  further  information  and  literature  citations. 

ASSAYS  FOR  ESSENTIAL  ELEMENTS 

Microorganisms  may  be  used  to  determine  the  presence  of  essential 
elements.  In  view  of  the  speed  and  accuracy  of  chemical  and  spectro- 
scopic methods,  it  might  be  assumed  that  microorganisms  would  be  of 
little  value  in  such  applications.  The  value  of  microbiological  tests 
would  appear  to  be  in  applications  where  availability  as  well  as  total 
amounts  are  of  importance.  Problems  of  this  sort  frequently  arise  in 
connection  with  mineral  deficiencies  in  soil.  It  is  recognized  that  the 
absolute  content  of  an  essential  element  in  a  soil  may  not  measure  the 
availability  of  that  element  for  green  plants.  Microbiological  and  chem- 
ical methods  of  analysis  must  be  correlated  with  plant  tests  before  they 
are  of  much  value. 

The  possible  number  of  test  organisms  is  unlimited  except  for  the 
important  considerations  of  sensitivity,  ease  of  handling,  and  time 
required  to  make  an  assay.  In  practice,  only  a  few  organisms  have  been 
used.  There  exists  a  wide  field  for  investigations  dealing  with  the  cor- 
relation between  availability  to  microorganisms  and  availability  to  green 
plants  of  certain  essential  elements  in  soil. 

Copper.  Mulder  (1939-1940)  used  Aspergillus  niger  to  determine 
copper  in  soil.  The  range  of  concentrations  in  the  standard  series  was 
0.0  to  2.5  Mg  Cu++  per  culture;  40  ml.  of  medium  was  used  in  liter  flasks. 
One  gram  of  sterile  soil  was  used  as  the  sample.  The  method  of  measur- 
ing the  response  of  A.  niger  to  copper  was  very  simple,  inasmuch  as  the 
number  and  color  of  the  spores  produced  were  functions  of  the  copper 
content  of  the  medium.     No  spores  developed  on  the  control  medium,  but 


218 


PHYSIOLOGY  OF  THE  FUNGI 


with  increasing  concentrations  of  copper  the  spores  were  yellow,  yellow- 
brown,  gray-brown,  brown,  and  black. 

The  color  of  the  spores  produced  on  copper-deficient  media  by  different 
isolates  of  A.  niger  varied.  Excellent  correlation  between  the  copper 
content  of  various  soils  as  determined  by  this  method  and  the  incidence 
of  copper  deficiency  in  grain  was  found.  Some  of  Mulder's  results  are 
presented  in  Table  36. 

Table  3G.     The  Correlation  of  Copper  Deficiency  in  White  Oats  and  the 
Copper  Content  of  the  Soil  as  Determined  by  Aspergillus  niger  Method 
All  soil  was  from  the  same  field.      (Mulder,  Antonie  van  Leeuwenhock  6,  1940.) 

Available  Copper, 
Condition  of  Oats  Mg  per  G.  of  Soil 

Severely  diseased 0 .  25 

Less  severely  diseased 0.8 

Healthy  (from  a  portion  of  the  field  not  showing  the  disease) 1.7 

Healthy  (copper  sulfate  added  to  the  soil) 2.5 

Magnesium.  Smit  and  Mulder  (1942)  postulated  that  a  microbio- 
logical method  would  show  better  correlation  with  magnesium  deficiency 
in  green  plants  than  chemical  methods.  This  was  confirmed  for  the 
Netherlands  soils  investigated.  Azotobacter  chroococcum  and  Aspergillus 
niger  were  used  as  test  organisms.  Preference  w^as  given  to  the  fungus 
inasmuch  as  only  4  to  5  days  w-ere  required  for  an  assay.  A  simple 
technique  was  used,  and  visual  comparison  w^as  sufficiently  accurate  to 
diagnose  magnesium  deficiency  in  soils. 

Potassium.  Aspergillus  niger  was  used  by  Niklas  and  Toursel  (1940) 
to  determine  available  potassium  and  other  elements  in  soils.  These 
authors  weighed  the  mycelium  produced.  Rogosa  (1944)  has  shown  that 
Lactobacillus  casei  may  be  used  to  determine  small  amounts  of  potassium. 

rABLE  37.     The  Effect  of  Molybdenum  Content  of  a  Glucose-Nitrate  Medium 

UPON  Yield  of  Mycelium  and  Sporulation  of  Aspergillus  niger 

(Mulder,  Plant  and  Soil  1,  1948.) 


Mg  Na2Mo04-2H.A 
in  50  ml.    medium 

Mg.  mycelium 
per  culture 

Sporulation 

Appearance  of 
mycelium 

0.0 
0.0025 
0.010 
0.050 

165 
294 

558 
868 

0 

0 

0 

Normal 

Entirely  mucous 
Partially  mucous 
Partially  mucous 
Normal 

Molybdenum.  The  amount  of  this  element  needed  by  fungi  and  green 
plants  is  greater  when  nitrogen  is  supplied  as  nitrate  than  when  ammo- 
nium nitrogen  is  furnished.  This  fact  introduces  a  complication  into  the 
microbiological  assaj^  of  molybdenum  in  that  the  sample  must  be  ashed 


FUNGI  AS  TEST  ORGANISMS  219 

before  analysis.  It  is  probable  that  amino  acids  and  other  nitrogen 
sources  containing  reduced  nitrogen  would  also  affect  the  amount  of 
molybdenum  needed.  Mulder  (1948)  investigated  the  use  of  Aspergillus 
niger  as  a  test  organism  (Table  37).  For  further  discussion  and  refer- 
ences to  the  use  of  microorganisms  in  essential-element  assay  see  Vande- 
caveye  (1948). 

SUGARS 

Yeasts  and  other  microorganisms  have  been  used  to  separate  optical 
isomers  and  complex  mixtures  of  sugars.  Pasteur  (18G0)  used  Pemcillium 
glaucum  to  obtain  the  "unnatural"  isomer  of  tartaric  acid  from  rfZ-tartaric 
acid.  Fischer  and  Hertz  (1892)  used  brewer's  yeast  to  ferment  D-galac- 
tose,  while  L-galactose  in  the  same  medium  was  not  utilized.  Auernheimer 
el  al.  (1948)  used  the  specific  fermentative  powers  of  Hansenula  suaveolens 
and  Candida  guilliermondi  in  the  separation  of  L-arabinose  and  D-xylose 
obtained  from  the  hydrolysis  of  straw  and  corn  cobs.  H.  suaveolens  does 
not  utilize  L-arabinose,  while  C.  guilliermondi  utilizes  both  pentoses. 
Saccharomyces  carlshergensis  was  used  to  demonstrate  the  absence  of 
D-glucose  in  the  hydrolysates.  These  yeasts  were  used  in  conjunction 
with  chemical  methods  of  analysis.  Appling  et  al.  (1947)  found  Sac- 
charomyces carlshergensis  var.  mandschuricus  to  ferment  D-galactose  but 
not  L-galactose.  Similarly,  H.  suaveolens  utilized  D-xylose  but  not 
L-xylose. 

These  citations  indicate  the  usefulness  of  yeasts  and  other  organisms 
in  the  solution  of  problems  difficult  to  solve  by  other  methods.  The 
value  of  microorganisms  in  such  applications  is  due  to  their  specificity. 

TESTS  FOR  CERTAIN  METABOLIC  PRODUCTS 

Fungi  excrete  into  the  media  in  w^hich  they  grow  various  physiologically 
active  substances.  In  the  older  literature  these  are  referred  to  as  staling 
'products.  Among  the  metabolic  products  are  those  which  may  stimulate 
or  inhibit  growth  and  reproduction.  The  kind  and  the  amount  of  com- 
pounds excreted  depend  upon  the  particular  fungus  involved  as  well  as 
the  composition  of  the  medium.  The  effect  of  the  metabolic  products  ot 
one  fungus  upon  another  is  simply  demonstrated  when  fungi  are  gro^vn 
in  association.  The  beneficial  effect  of  one  fungus  upon  another  was 
demonstrated  by  Kogl  and  Fries  (1937).  Neither  Nematospora  gossypii 
or  Polyporus  adustus  grew  when  inoculated  alone  into  a  synthetic  medium, 
but  when  both  fungi  were  inoculated  together  in  the  same  flask,  both 
began  to  grow  rapidly  after  about  a  week.  N.  gossypii  is  deficient  for 
biotin  but  synthesizes  thiamine,  while  P.  adustus  is  deficient  for  thiamine 
but  synthesizes  biotin.  Kogl  and  Fries  called  this  artificial  symbiosis. 
Schopfer  and  Guilloud  (1945)  cite  other  examples  in  connection  with 
work  on  strains  of  Candida  guilliermondi  involving  vitamin  deficiencies. 


220 


PHYSIOLOGY  OF  THE  FUNGI 


By  using  a  series  of  test  organisms  of  known  deficiencies,  it  is  easy  to 
demonstrate  that  fungi  excrete  vitamins.  It  is  a  common  experience  to 
find  deficient  fungi  growing  in  association  with  contaminants.  The 
method  is  simple  and  consists  of  inoculating  plates  of  vitamin-free  medium 
with  two  test  fungi  (Fig.  44).     Not  all  fungi  excrete  the  same  amount  of  a 


Fig.  44.  Test  demonstrating  the  excretion  of  biotin  by  Aspergillus  7'ugulosus  (right), 
whengrownwith Sordariafimicola  (biotin-deficient)  on  vitamin-free  glucose-asparagine 
medium.  Sordaria  (left)  made  only  slight  growth  until  it  approached  the  colony  of 
Aspergillus,  where  a  zone  of  stimulated  growth  is  evident. 


given  vitamin.  This  may  be  shown  by  choosing  test  fungi  such  as 
Sordaria  fimicola,  which  requires  more  biotin  for  fruiting  than  for  growth. 
Some  fungi  excrete  enough  biotin  to  allow  grow^th  of  S.  fimicola,  while 
others  excrete  enough  biotin  to  allow  reproduction  also.  Other  com- 
pounds besides  the  vitamins  may  be  excreted  and  favor  the  growth  of 
other  organisms.  Further  instances  of  the  favorable  effect  of  one  fungus 
on  the  sporulation  of  another  are  discussed  in  Chap.  14. 

The  metabolic  products  of  one  fungus  may  inhibit  the  growth  of 
another.  This  phenomenon  may  be  frequently  observed  on  contaminated 
plates  (Fig.  45).  Fleming  (1929)  discovered  the  action  of  penicillin  in 
this  way. 

Many  fungi  apparently  produce  substances  which  inhibit  the  germina- 
tion of  their  spores.  Schopfer  (1933)  found  that  spores  of  Phy corny ces 
blakesleeanus  would  not  germinate  on  agar  media  upon  which  this  fungus 
had  grown.  If  such  a  "staled"  plate  was  autoclaved,  the  medium  would 
then  allow  germination  and  growth  of  P.  blakesleeanus.     These  results 


FUNGI  AS  TEST  ORGANISMS  221 

indicate    that    the    spore-inhibiting    substance    was    either    volatile    or 
unstable.     This  inhibitory  substance  was  not  identified. 


Fig.  45.  Test  for  antibiotic  production  by  growing  two  organisms  in  association  on 
the  same  agar  plate.  Helminthosporium  sativum  on  the  left  and  an  unidentified 
actinomycete  on  the  right. 

TESTING  FABRIC  PROTECTANTS 

While  the  deterioration  of  cellulosic  materials  exposed  to  the  weather 
or  in  contact  with  the  soil  is  not  solely  due  to  the  action  of  bacteria  and 
fungi,  these  organisms  are  the  chief  agents  of  destruction.  The  problem 
of  deterioration  of  cellulosic  materials  has  received  a  vast  amount  of 
attention,  especially  in  connection  with  military  materiel  in  humid  tropic 
climates.  Work  on  this  problem  involves  the  identification  of  the  respon- 
sible microorganisms,  laboratory  tests,  and  use  of  test  fungi  in  evaluating 
protectants. 

The  basis  of  the  various  methods  for  determining  cellulolytic  activity 
consists  in  inoculating  cotton  duck  or  other  test  material  with  the  fungi 
under  test.  The  degree  of  cellulolytic  activity  is  determined  by  measur- 
ing the  decrease  in  tensile  strength  of  the  test  specimen.  The  test 
medium  used  is  usually  an  inorganic  salt  solution  having  pH  6.8.  It  is 
desirable  to  use  a  buffered  medium  inasmuch  as  cellulase  is  most  active 
around  pH  7.  White  et  at.  (1948)  note  that  many  fungi  which  are 
strongly  cellulolytic  under  laboratory  conditions  cause  but  httle  damage 
in  the  field.  They  believe  that,  under  a  given  set  of  natural  environ- 
mental conditions,  the  actual  decay  of  fibers  is  caused  by  a  relatively 
few  species  of  fungi.     Among  the  strongly  cellulolytic  fungi  are  Mem- 


222 


PHYSIOLOGY  OF  THE  FUNGI 


noniella  echinata  (the  variability  in  strains  in  laboratory  tests  is  possibly 
correlated  with  biotin  deficiency),  Chaetomium  spp.,  especially  C.  glo- 
bosian  (Greathouse  and  Ames,  1945),  Myrothecium  verrucaria  (as  strong 
a  cellulose  decomposer  as  yet  found  in  laboratory  tests),  Trichoderma 
viride,  and  Thielavia  sepedonium. 

The  reduction  in  tensile  strength  of  cotton  duck  maintained  under 
specified  conditions  is  used  as  a  measure  of  the  destructive  effects  of 
fungi  on  fabrics.     The  data  in  Table  38  are  taken  from  White  et  al.  (1948). 


Table  38. 


Assay  of  Fungi  for  Cellulolytic  Activity  Based  upon  Loss  of 
Tensile  Strength  of  Cotton  Duck 
(White  et  al,  Mrjcologia  40,  1948.) 


Species 


Aspergillus  niger  PQMD  25a 

A.  terreus  PQMD  72f 

Chaetomium  funicolum  PQMD  351. 

C.  globosum  PQMD  32b 

Fusarium  oxysponim  Fla  C-8 

Gliomastix  convoluta  PQMD  4c 

Myrothecium  verrucaria  PQMD  70h 
Thielavia  sepedonium  PQMD  47g.  . 

Trichoderma  viride  PQMD  6a 

T.  viride  PQMD  63d 


Strength  retained,  % 

6  days 

9  days 

12  days 

100 

103 

105 

67 

0 

18 

42 

32 

0 

— 

49 

36 

30 

0 

15 

51 

22 

0 

— 

18 

10 

8 

100 

99 

98 

Growth 

at  end  of 

experiment 

2 

4 
4 
4 
3 
4 
4 
4 
4 
0 


The  evaluation  of  protective  fungicides  for  fabrics,  paper,  and  other 
cellulosic  materials  consists  in  comparing  the  effects  of  known  cellulolytic 
fungi  upon  treated  and  untreated  specimens  of  material.  In  addition 
to  causing  loss  of  tensile  strength,  some  fungi  cause  great  damage  by 
surface  growth  (mildew).  Abrams  (1948)  has  reviewed  the  techniques 
used  at  the  Bureau  of  Standards  for  testing  mildew-  and  rotproofing 
agents.  Aspergillus  niger  was  used  to  determine  mildew  resistance,  and 
the  effectiveness  of  various  treatments  was  evaluated  by  visual  observa- 
tion. Chaetomium  globosum  and  a  species  of  Penicillium  (USDA  66) 
were  used  in  rot-resistance  tests.  Of  some  36  compounds  tested,  copper 
naphthenate  and  pyridyl  mercury  compounds  were  most  effective.  The 
effectiveness  of  the  fungicides  varies  with  the  test  organisms  used.  For 
data  on  fungicide  evaluation  the  reader  is  referred  to  Abrams  (1948). 

SUMMARY 

The  use  of  microorganisms  for  analytical  purposes  is  based  upon  specific 
biochemical  characteristics  of  selected  test  organisms.  Within  a  certain 
range  of  concentration,  the  response  is  proportional  to  the  amount  of 


FUNGI  AS  TEST  ORGANISMS  223 

test  substance  present  in  the  medium.  Among  the  substances  for  which 
quantitative  assay  procedures  have  been  developed  are  the  vitamins, 
amino  acids,  and  essential  elements.  Microorganisms  have  also  been 
used  to  discover  pathways  of  biochemical  synthesis  and  degradation,  to 
separate  isomers,  and  for  other  analytical  purposes. 

The  essential  features  of  a  microbiological  assay  are  (1)  a  suitable  test 
organism,  (2)  a  suitable  basal  medium  essentially  free  from  the  substance 
under  test,  (3)  preparation  of  the  sample,  (4)  a  reference  standard  (a 
pure  compound  where  possible),  (5)  two  series  of  cultures  to  which  a 
known  range  of  concentrations  of  the  standard  and  unknown  have  been 
added,  (6)  uniform  inoculation,  (7)  incubation  under  uniform  conditions, 
(8)  measuring  the  response  of  the  test  organisms,  (9)  construction  of  a 
standard  curve,  and  (10)  calculating  the  results. 

When  microbiological  assay  procedures  are  used,  it  is  unnecessary  to 
isolate  the  compound  being  assayed  from  the  other  constituents  present 
in  the  sample.  The  preparation  of  the  sample  for  assay  is  usually  simple 
and  ordinarily  involves  hydrolysis.  Microbiological  procedures  usually 
require  a  short  time  to  complete.  The  amount  of  sample  needed  is  small, 
which  is  an  important  consideration  in  some  problems.  Microbiological 
assays  are  invaluable,  provided  that  suitable  test  organisms  are  available, 
in  devising  chemical  procedures  for  the  isolation  of  new  vitamins  and 
other  physiologically  active  compounds. 

Biochemical  mutants  of  Neurospora  and  other  fungi  are  particularly 
useful  in  determining  the  pathways  of  synthesis  of  amino  acids  and  other 
compounds. 

REFERENCES 

Abrams,  E.:  Microbiological  deterioration  of  organic  materials,  its  prevention  and 
methods  of  test,  Natl.  Bur.  Standards  (U.S.),  Misc.  Pub.  188,  1948. 

Appling,  J.  W.,  E.  K.  Ratcliff,  and  L.  E.  Wise:  Chemical  and  microbiological 
differentiation  of  enantiomorphs  of  galactose  and  xylose,  Anal.  Chem.  19 :  496- 
497,  1947. 

Atkin,  L.,  W.  L.  Williams,  A.  S.  Schultz,  and  C.  N.  Frey:  Yeast  microbiological 
methods  for  determination  of  vitamins.  Pantothenic  acid,  Ind.  Eng.  Chem., 
Anal.  Ed.  16:  67-71,  1944. 

AuERNHEiMER,  A.  H.,  L.  J.  WicKERHAM,  and  L.  E.  Schniepp:  Quantitative  deter- 
mination of  hemicellulose  constituents  by  fermentation,  Anal.  Chem.  20: 
876-877,  1948. 

Beadle,  G.  W.  :  An  inositoUess  mutant  strain  of  Neurospora  and  its  use  in  bioassays, 
Joiir.  Biol.  Chem.  156 :  683-689,  1944. 

Beadle,  G.  W.  :  Biochemical  genetics,  Chem.  Revs.  37 :  15-96,  1945. 

Beadle,  G.  W.  :  Genetics  and  metabolism  in  Neurospora,  Physiol.  Revs.  25 :  643-663, 
1945a. 

Bonner,  D.:  Biochemical  mutations  in  Neurospora,  Cold  Spring  Harbor  Symposia 
Quant.  Biol.  11 :  14-24,  1946. 

Brand,  E.,  F.  J.  Ryan,  and  E.  M.  Diskant:  Leucine  content  of  proteins  and  food- 
stuffs, Jour.  Am.  Chem.  Soc.  67:  1532-1534,  1945. 


224  PHYSIOLOGY  OF  THE  FUNGI 

DoERMANN,  A.  II.:  A  lysineless  mutant  of  Neurospora  and  its  inhibition  by  arginine, 
Arch.  Biochem.  5 :  373-384,  1944. 

Dunn,  M.  S.,  S.  Shankman,  M.  N.  Camien,  W.  Frankland,  and  L.  B.  Rockland: 
Investigations  of  amino  acids,  peptides,  and  proteins.  XVIII.  The  amino  acid 
requirements  of  Leuconostoc  viesenteroides  P-60,  Jour.  Biol.  Chem.  156:  703-713, 
1944. 

FxscHER,  E.,  and  J.  Hertz:  Reduction  der  Schleimsaure,  Ber.  d.  deut.  chem.  Ges. 
25:  1247-12(51,  1892. 
■^Fleming,  A.:  On  the  antibacterial  action  of  cultures  of  a  Penicillium,  with  special 
reference  to  their  use  in  the  isolation  of  B.  influenzae,  Brit.  Jour.  Exptl.  Path. 
10:  226-236,  1929. 

Greathouse,  G.  a.,  and  L.  M.  Ames:  Fabric  deterioration  by  thirteen  described 
and  three  new  species  of  Chaetomium,  Mycologia  37  :  138-155,  1945. 

Greene,  R.  D.,  and  A.  Black:  The  microbiological  assay  of  trytophane  in  proteins 
and  foods.  Jour.  Biol.  Chem.  155 :  1-8,  1944. 

Horn,  M.  J.,  D.  B.  Jones,  and  A.  E.  Blum:  Methods  for  microbiological  and  chemi- 
cal determinations  of  essential  amino  acids  in  proteins  and  foods,  f7.»S.  De-pt. 
Agr.  Misc.  Pub.  696,  1950. 
*HoRowiTZ,  N.  H.:  Methionine  synthesis  in  Neurospora.     The  isolation  of  cysta- 
thionine. Jour.  Biol.  Chem.  171 :  255-264,  1947. 

Horowitz,  N.  H.,  and  G.  W.  Beadle  :  A  microbiological  method  for  the  determina- 
tion of  choline  by  use  of  a  mutant  of  Neurospora,  Jour.  Biol.  Chem.  150 :  325-333, 
1943. 

Horsfall,  J.  G.:  Fungicides  and  Their  Action,  Chronica  Botanica  Co.,  Waltham, 
1945. 

HuTCHiNGS,  B.  L.,  and  W.  H.  Peterson:  Amino  acid  requirements  of  Lactobacillus 
casei,  Proc.  Soc.  Exptl.  Biol.  Med.  52 :  36-38,  1943. 

Kogl,  F.,  and  N.  Fries:  Ueber  den  Einfluss  von  Biotin,  Aneurin  und  Meso-Inosit 
auf  das  Wachstum  verscheidener  Pilzarten,  Zeit.  physiol.  Chem.  249:  93-110, 
1937. 

KoGii,  F.,  and  B.  Tonnis:  Ueber  das  Bios-Problem.  Darstellung  von  krystallisier- 
tem  Biotin  aus  Eigelb,  Zeit.  physiol.  Chem.  242 :  43-73,  1936. 

Krehl,  W.  a.,  F.  M.  Strong,  and  C.  A.  Elvehjem:  Determination  of  nicotinic  acid. 
Modifications  in  the  microbiological  method,  Ind.  Eng.  Chem.,  Anal.  Ed.,  15: 
471-475,  1943. 

Leonian,  L.  H.,  and  V.  G.  Lilly:  The  comparative  value  of  different  test  organisms 
in  the  microbiological  assay  of  B  vitamins.  West  Va.  Agr.  Expt.  Sta.  Bull.  319, 
1945. 

Lilly,  V.  G.,  and  L.  H.  Leonian:  The  anti-biotin  effect  of  desthiobiotin,  Science  99 : 
205-206,  1944. 

LiNDEGREN,  C.  C,  and  C.  Raut:  The  effect  of  the  medium  on  apparent  vitamin- 
synthesizing  deficiencies  of  microorganisms,  Ann.  Missouri  Botan.  Garden  34: 
75-84,  1947. 

Mulder,  E.  G.:  On  the  use  of  micro-organisms  in  measuring  a  deficiency  of  copper, 
magnesium  and  molybdenum  in  soils.  Anionic  van  Leeuwenhoek  6:  99-109, 
1939-1940. 

Mulder,  E.  G.  :  Importance  of  molybdenum  in  the  nitrogen  metabolism  of  micro- 
organisms and  higher  plants,  Plant  and  Soil  1 :  94-1 19,  1948. 

NiKLAS,  H.,  and  O.  Toursel:  Die  Bodenuntersuchung  mittels  Aspergillus  niger, 
Bodenkunde  und  Pflxinzen  erndkr.  18 :  79-107,  1940. 

Pasteur,  L.  :  Note  relative  au  Penicillium,  glaucum  et  a  la  dissymetrie  moleculaire 
des  produits  organiques  naturels,  Compt.  rend.  acad.  set.  61 :  298-299,  1860. 


FUNGI  AS  TEST  ORGANISMS  225 

RoBBiNS,  W.  J.:  Effect  of  extracts  of  Phycomyces  upon  its  development,  Am.  Jour. 

Botany  27:  559-564,  1940. 
Roberts,  E.  C,  and  E.  E.  Snell:  An  improved  medium  for  microbiological  assays 

with  Lactobacillus  casei,  Jour.  Biol.  Chem.  163  :  499-509,  1946. 
RoGOSA,  M.:  Microbiological  method  for  the  quantitative  determination  of  small 

quantities  of  potassium,  Jour.  Biol.  Chem.  154:  307-308,  1944. 
*Ryan,  F.  J. :  The  germination  of  conidia  from  biochemical  mutants  of  Neurospora, 

Am.  Jour.  Botany  35:  497-503,  1948. 
Ryan,  F.  J.,  and  E.  Brand:  A  method  for  the  determination  of  leucine  in  protein 

hydrolysates  and  in  foodstuffs  by  the  use  of  a  Neurospora  mutant,  Jour.  Biol. 

Chem.  154:  161-175,  1944. 
ScHOPFER,  W.  H.:  Etude  d'un  cas  de  stimulation  unilaterale  et  d'un  cas  d'inhibition 

chez  un  micro-organisme,  Compt.  rend.  soc.  phys.  hist.  nat.  Genhve.  50:  152-154, 

1933. 
ScHOPFER,  W,  H.:  Recherches  sur  I'emploi  possible  d'un  test  vegetal  pour  la  vitamine 

Bi.     Essai  d'etalonnage.  Bull.  soc.  chim.  biol.  17:  1097-1109,  1935. 
ScHOPFER,  W.  H. :  Les  Tests  microbiologiques  pour  la  determination  des  vitamines, 

Experientia  1 :  1-68,  1945. 
ScHOPFER,    W.  H.,  and  M.  Guilloud:  Un  cas  de  stimulation  unilateral  chez  las 

microorganismes  explique  par  une  action  vitaminique,  Verhandl.  naturforsch. 

Ges.  Basel  61:  299-314,  1945. 
ScHULTZ,  A.  S.,  L.  Atkin,  and  C.  N.  Frey:  Determination  of  vitamin  Bi  by  yeast 

fermentation  method.     Improvements  related  to  the  use  of  sulfite  cleavage  and 

a  new  fermentometer,  Ind.  Eng.  Chem.,  Anal.  Ed.  14:  35-39,  1942. 
Shankman,  S.:  Amino  acid  nutrition  of  Lactobacillus  arabinosus,  Jour.  Biol.  Chem. 

150:305-310,  1943. 
Smit,  J.,  and  E.  G.  Mulder:  Magnesium  deficiency  as  the  cause  of  injury  in  cereals, 

Mededeelingen  van  de  Landbouwhoogeschool,  Deel  46,  Verhandeling  3,  1942. 
Snell,  E.  E.  :  The  microbiological  assay  of  amino  acids.  Advances  in  Protein  Chem.  2 : 

85-118,  1945. 
Snell,  E.  E.  :  The  vitamin  Be  group.     Evidence  for  the  occurrence  of  pyridoxamine 

and  pyridoxal  in  natural  products,  Jour.  Biol.  Chem.  157:  491-505,  1945a. 
*Snell,  E.  E.:  Use  of  microorganisms  for  assay  of  vitamins,  Physiol.  Revs.  28:  255- 

282,  1948. 
Srb,  a.  M.,  and  N.  H.  Horowitz:  The  ornithine  cycle  in  Neurospora  and  its  genetic 

control,  Jour.  Biol.  Chem.  154:  129-139,  1944. 
Tatum,  E.  L.,  M.  G.  Ritchey,  E.  V.  Cowdry,  and  L.  F.  Wicks:  Vitamin  content  of 

mouse  epidermis  during  methylcholanthrene  carcinogenesis.     I.  Biotin,  choline, 

inositol,  p-aminobenzoic  acid  and  pyridoxine,  Jour.  Biol.  Chem.  163 :  675-682, 

1946. 
Vandeca\teye,  S.  C.  :  Biological  methods  of  determining  nutrients  in  soil  in  Diagnos- 
tic Techniques  for  Soils  and  Crops  (edited  by  H.  B.  Kitchen),  The  American 

Potash  Institute,  Washington,  1948. 
*  White,  W.  L.,  R.  T.  Darby,  G.  M.  Stechert,  and  K.  Sanderson:  Assay  of  cellu- 

lolytic  activity  of  molds  isolated  from  fabrics  and  related  items  exposed  in  the 

tropics,  Mycologia  40 :  34-84,  1948. 
Wright,  L.  D.,  E.  L.  Cresson,  H.  R.  Skeggs,  T.  R.  Wood,  R.  L.  Peck,  D.  E.  Wolf, 

and  K.  Folkers:  Biocytin,  a  naturally-occuring  complex  of  biotin.  Jour.  Am. 

Chem.  Soc.  72:  1048,  1950. 
Wyss,  O.,  V.  G.  Lilly,  and  L.  H.  Leonian:  The  effect  of  pH  on  the  availability  of 

p-aminobenzoic  acid  to  Neurospora  crassa,  Science  99:  18-19,  1944. 


CHAPTER  11 
METABOLITE  ANTAGONISTS 

This  chapter  and  the  one  following  will  deal  with  chemical  compounds 
which  inhibit,  injure,  or  kill  fungi.  Much  can  be  learned  about  "normal" 
physiological  processes  by  studying  the  factors  which  interfere  with  them. 
The  ideas  to  be  discussed  here  are  applicable  to  the  entire  field  of  phys- 
iology, and  some  of  our  illustrative  material  will  deal  with  organisms 
other  than  fungi.  The  reviews  of  Woolley  (1944),  Welch  (1945),  Wright 
(1947),  Mcllwain  (1947),  and  Roblin  (1946,  1949)  are  extensive  and  well 
documented  and  should  be  consulted  for  additional  references. 

Metabolites  are  chemical  substances  which  are  essential  for  the  func- 
tioning and  maintenance  of  living  cells.  Metabolites  may  be  synthesized 
by  the  organism  or  obtained  from  the  medium,  e.g.,  vitamins,  amino 
acids,  etc. 

An  antimetabolite,  or  antagonist,  is  a  compound  which  interferes  with 
the  utilization  of  a  normal  metabolite.  Wright  (1947)  has  classified 
antagonists  (more  specifically  antivitamins)  on  the  basis  of  their  mode  of 
action:  (1)  those  which  act  by  virtue  of  destroying  or  inactivating  a 
metabolite;  (2)  those  which  combine  irreversibly  with  enzymes  (non- 
competitive inhibition) ;  and  (3)  those  which  combine  with  enzymes  but 
w^hich  may  be  displaced  by  increased  concentration  of  the  normal  metabo- 
lite (competitive  inhibition). 

Noncompetitive  enzyme  inhibition  is  so  called  because  an  increase  in 
the  concentration  of  the  normal  coenzyme  or  metabolite  molecules  does 
not  reverse  the  inhibition.  Inhibitors  of  this  type  act  by  combining 
with  some  atom  or  molecular  group  of  either  a  coenzyme  or  an  apoenzyme. 
Among  inhibitors  of  this  type  we  may  list  the  heavy  metals,  various 
organic  mercury  and  arsenic  compounds,  iodoacetate,  and  quinones, 
which  inactivate  enzymes  by  combining  with  free  sulfhydryl  groups  (see 
Singer,  1945,  and  McElroy,  1947,  for  references).  Among  the  inhibitors 
which  act  on  the  iron-porphyrin  enzymes  are  cyanide,  azide,  hydrogen 
sulfide,  and  carbon  monoxide.  Most  of  the  discussion  to  follow  will  deal 
with  competitive  antagonists. 

Metabolite  antagonists  are  analogues  of  normal  metabolites,  but  not 
all  analogues  of  a  metabolite  are  necessarily  antagonists.  These  "for- 
eign" molecules,  because  of  their  close  resemblance  to  normal  metabolites, 
combine  with  enzymes  in  the  same  manner  as  normal  metabolites.     How- 

226 


METABOLITE  ANTAGONISTS 


227 


ever,  these  foreign  molecules  are  not  transformed  by  the  enzyme  to  which 
they  are  bound.  If  the  antagonist  is  an  analogue  of  a  coenzyme,  it 
presumably  forms  a  pseudoholoenzyme  which  is  unable  to  function.  The 
close  structural  relation  between  a  metabolite  (p-aminobenzoic  acid)  and 
its  antagonist  (sulfanilamide)  is  shown  in  Fig.  46. 


6.7  A. 


H      H 

N 


J     A     ! 

— ►|2.3A.|-^ 
p-Aminobenzoate  Ion 


—►[2.4  A. 
Sulfanilamide 


Fig.  46.  Interatomic  distances  and  structural  relationships  of  p-aminobenzoate  ion 
and  sulfanilamide.  (Courtesy  of  Roblin,  Chem.  Eng.  News  27 :  3624,  1949.  Published 
by  permission  of  American  Chemical  Society.) 

In  spite  of  the  large  number  of  compounds  which  have  been  tested  for 
antagonism,  it  is  not  possible  to  specify  exactly  what  changes  in  metabolite 
molecules  are  required  to  produce  antagonists.  A  single  modification  of  a 
metabolite  molecule  is  more  hkely  to  produce  an  antagonist  than  two  or 
more  changes  in  structure.  This  is  to  be  expected,  for  an  antagonist 
must  closely  resemble  the  corresponding  metabolite.  Replacing  a 
carboxyl  group  with  a  sulfonic-acid  group  has  been  effective  in  many 
instances. 

The  specific  action  of  enzymes  has  been  likened  to  the  relation  of  a 
lock  and  its  key.  Unless  an  enzyme  and  a  substrate  molecule  are  related 
in  this  fashion,  no  reaction  will  take  place.  A  modern  diagrammatic 
representation  of  the  lock-and-key  simile  is  shown  in  Fig.  47.  The 
mechanism  of  competitive  inhibition  may  be  visuahzed  by  referring  to 
this  figure.  Metabolite  antagonists  may  be  thought  of  as  "wrong"  keys, 
which  jam  the  lock  mechanism.  As  long  as  a  false  key  is  in  the  lock, 
it  prevents  the  true  key  from  entering  and  opening  the  lock. 

Compounds  which  resemble  coenzymes  in  structure  compete  for  the 
active  surface  of  apoenzymes.  Because  of  similarity  in  structure,  an 
apoenzyme-foreign  molecule  complex,  or  pseudoholoenzyme  is  formed. 
Such  a  pseudoenzyme  is  unable  to  function.  The  reversal  of  enzyme 
inhibition  in  such  instances  is  caused  by  the  addition  of  more  coenzyme 
molecules.  The  argument  is  the  same  when  substrate  analogues  are 
involved.     For  example,  3-fluorophenylalanine  inhibits  the  utihzation 


228 


PHYSIOLOGY  OF  THE  FUNGI 


of  phenylalanine  (a  normal  metabolite)  by  Neurospora  crassa  (Mitchell 
and  Neimann,  1947). 

The  effect  of  an  antagonist  will  depend  upon  the  concentration  of  the 
normal  metabolite  present  in  the  medium  and  cells  and  upon  the  organ- 
ism. In  general,  enzymes  have  a  greater  affinity  for  metabolites  than 
for  antimetabolites.  Since  both  metabolite  and  antagonist  compete 
for  the  same  enzyme,  the  amount  of  inhibition  will  depend  upon  the 
relative  concentrations  rather  than  upon  the  absolute  amounts  of  these 
compounds  present.     The  amount  of  an  inhibitor  required  to  reduce  the 


nAP 

Substrate 

+ 


Products 
free  enzyme 


Enzyme -substrate 
complex 

Fig.  47.  A  diagrammatic  illustration  of  Fischer's  simile  that  an  enzyme  and  its 
substrate  are  related  as  are  a  lock  and  its  key.  (Courtesy  of  McElroy,  Quart.  Rev. 
Biol.  22 :  26,  1947.     Published  by  permission  of  The  Williams  &  Wilkins  Company.) 

amount  of  growth  to  one-half  will  depend  upon  the  ratio  of  inhibitor  and 
metabolite  present.  In  simple  instances,  at  least,  this  ratio  is  equal  to  a 
constant  and  is  called  the  inhibition  constant,  or  index.  The  amounts 
of  sulfadiazine  and  p-aminobenzoic  acid  required  to  reduce  the  amount 
of  growth  of  Streptococcus  faecalis  R  to  one-half  the  normal  value  gave  an 
inhibition  index  of  333  (Lampen  and  Jones,  1946). 

The  inhibition  index  is  valid  only  for  the  particular  conditions  used 
in  an  experiment  and  for  the  particular  strains  of  the  organism  used.  In 
the  case  of  self-sufficient  organisms  the  use  of  an  amount  of  inhibitor  less 
than  that  required  for  total  inhibition  will  only  decrease  the  rate  of 
growth,  and  thus  the  inhibition  index  will  change  with  the  time  of  incuba- 


METABOLITE  ANTAGONISTS  229 

tion.  This  is  due  to  the  synthesis  of  the  metabohte  by  the  organism. 
Sulfanilamide  inhibits  the  growth  of  Aspergillus  niger,  but  the  fun- 
gus overcomes  this  inhibition  as  the  time  of  incubation  is  prolonged 
(Hartelius,  194G).  The  concentration  of  a  metabolite  in  the  control 
cultures  should  be  less  than  the  amount  which  allows  maximum  growth, 
because  of  the  nonlinear  response  of  an  organism  to  the  metabolite  at 
high  concentrations. 

The  composition  of  the  medium  is  an  important  consideration  in  any 
investigation  of  metabolite  antagonism.  If  adequate  amounts  of  a 
natural  metabolite  are  present,  the  action  of  an  inhibitor  may  be  over- 
looked. Synthetic  media  should  be  used.  The  composition  of  the 
medium  used  may  also  affect  the  action  of  an  inhibitor  in  another  way. 
If  metabolite  A  is  transformed  into  metabolite  B  by  an  organism,  the 
presence  of  metabolite  B  in  sufficient  amount  for  optimum  growth  may  be 
expected  to  nullify  any  amount  of  an  antagonist  for  metabolite  A.  An 
antagonist  of  metabolite  B,  however,  would  exhibit  normal  competitive 
inhibition,  Shive  and  Macow  (1946)  have  pointed  out  that,  by  the  use 
of  a  suitable  series  of  inhibitors,  it  is  possible  to  follow  the  transformations 
of  a  given  metabolite  step  by  step.  These  authors  designate  this  use  of 
metabolite  antagonists  as  inhibition  analysis.  Rydon  (1948)  found 
Bacterium  typhosum  to  synthesize  tiyptophane  by  the  following  steps: 
anthranilic  acid  — >  indole  — >  tryptophane.  The  2-  and  4-methylanthran- 
ihc  acids  were  potent  inhibitors  against  anthranilic  acid  but  not  against 
indole  or  tryptophane.  Certain  analogues  of  indole  and  tryptophane 
were  inhibitors  of  these  metabolites. 

In  discussing  metabolite  antagonists  in  a  general  way,  it  should  be 
borne  in  mind  that  these  compounds  may  inhibit  only  certain  organisms, 
or  a  particular  organism  only  under  certain  conditions.  For  example, 
desthiobiotin  is  a  biotin  antagonist  for  Ceratostomella  pini  and  Lacto- 
hacillus  casei,  while  this  compound  replaces  biotin  for  many  strains  of 
Saccharomyces  cerevisiae  (Lilly  and  Leonian,  1944).  Woolley  (1944, 
1946)  is  of  the  opinion  that  the  established  facts  of  inhibition  and  reversal 
are  more  important  than  the  hypotheses  which  are  adopted  to  explain 
these  phenomena.  However,  the  concept  of  competitive  metabohte 
antagonism  has  been  very  useful  in  correlating  a  vast  amount  of  experi- 
mental work  in  apparently  unrelated  fields. 

ANTIVITAMINS 

Antivitamins  are  known  for  all  the  water-soluble  vitamins  which  have 
been  synthesized  and  for  at  least  one  of  the  fat-soluble  vitamins  (vitamin 
K). 

p-Aminobenzoic  acid  antagonists.  When  the  sulfonamides  were  intro- 
duced into  medicine,  it  was  quickly  found  that  serum  and  other  natural 


230 


PHYSIOLOGY  OF  THE  FUNGI 


products  antagonized  the  inhibitive  action  of  sulfanilamide  on  the  growth 
of  certain  bacteria.  Rubbo  and  Gillespie  (1940)  discovered  that  p-amino- 
benzoic  acid  was  a  growth  factor  for  certain  bacteria.  Woods  (1940) 
found  that  p-aminobenzoic  acid  in  low  concentration  overcame  sulfanila- 
mide inhibition.  A  general  theory  was  proposed  by  Fildes  (1940)  to 
explain  the  antagonism  between  metabolites  and  compounds  having 
closely  related  structures. 


1400 


8  10  12 

Doys  of  incubotion 


16 


18 


20 


Fig.  48.  The  effect  of  various  concentrations  of  sulfanilamide  (amounts  per  flask) 
upon  the  time  of  spore  germination  and  upon  the  rate  and  amovnt  of  growth  of 
Aspergillus  niger  in  flasks  containing  55  ml.  of  sucrose— ammonium  sulfate  medium  at 
32°C.  (Drawn  from  the  data  of  Hartelius,  Compt.  rend.  trav.  lab.  Carlsberg,  S^r. 
physiol.  24:  181,  1946.) 

Sulfanilamide  was  first  considered  to  be  antagonized  by  p-amino- 
benzoic  acid,  rather  than  the  reverse.  This  was  due  to  the  discovery 
of  the  therapeutic  value  of  sulfanilamide  before  it  was  known  that 
p-aminobenzoic  acid  was  a  vitamin.  The  structural  relation  between 
these  compounds  has  already  been  noted.  The  literature  dealing  with 
the  sulfonamides  is  abundant,  but  most  of  it  relates  to  bacteria  and 
medicine.  Relatively  few  papers  have  been  published  on  the  effects 
of  these  compounds  on  the  growth  of  fungi. 

Hartelius  (1946)  investigated  the  effect  of  sulfanilamide  upon  the 
growth  of  Aspergillus  niger  and  found  that  the  amount  of  inhibition  was 
dependent  upon  the  amount  of  inoculum  used,  the  concentration  of 
sulfanilamide  in  the  medium,  and  the  time  of  incubation.  The  curves 
in  Fig.  48  illustrate  the  effect  of  time  of  incubation  on  inhibition,  a  factor 
which  is  too  often  overlooked  in  experiments  of  this  kind.     The  curves 


METABOLITE  ANTAGONISTS  231 

in  Fig.  48  indicate  that  A.  niger  synthesizes  either  p-aminobenzoic  acid 
or  some  other  compound  which  reverses  the  inhibitory  action  of  sulfa- 
nilamide. When  p-aminobenzoic  acid  was  added  to  the  medium,  sulfa- 
nilamide no  longer  inhibited  the  growth  of  yl.  niger  (Hartelius  and  Roholt, 
1946).  Other  fungi  have  been  shown  to  react  like  A.  niger  when  cultured 
in  media  containing  sulfanilamide  (Fourneau  et  al.,  1936). 

It  has  been  assumed  that  self-sufficient  fungi  require  the  same  vitamins 
as  the  deficient  species.  The  synthesis  of  a  vitamin  may  suggest  its 
need  but  does  not  demonstrate  it.  Antivitamins  (or  other  antimeta- 
bolites) provide  a  way  of  demonstrating  the  need  of  self-sufficient  fungi 
for  the  vitamins  they  synthesize.  Thus  A.  niger  requires  p-aminobenzoic 
acid  just  as  Rhodotorula  aurantica  does,  but  this  need  can  be  demonstrated 
only  in  the  presence  of  a  specific  reversible  inhibitor  such  as  sulfanilamide. 
This  technique  offers  a  possible  way  of  discovering  new  vitamins  and 
other  metabolites.  If  a  compound  inhibits  growth,  it  is  worth  while  to 
search  for  compounds  which  overcome  this  inhibition  reversibly. 

For  most  purposes  sulfanilamide  has  been  replaced  by  other  sulfona- 
mides. However,  sulfanilamide  appears  to  be  the  most  active  sulfona- 
mide against  fungi.  For  a  review  of  the  clinical  aspects  of  the  sulfona- 
mides in  mycoses  and  for  literature  citations,  see  Wolf  (1947). 

Stoddard  (1947)  has  reported  the  sulfonamides  to  be  of  some  value  in 
controlling  the  X  disease  of  peach  (a  virus).  Addition  of  p-amino- 
benzoic  acid  lessened  the  effectiveness  of  the  treatment. 

It  is  recognized  that  the  simple  Woods-Fildes  theory  of  competitive 
inhibition  is  inadequate  to  explain  completely  the  mechanism  of  sulfona- 
mide therapy.  In  vivo  the  environment  is  much  more  complex  than  in 
simple  laboratory  media.  For  further  information  and  references  to  the 
literature,  see  Sevag  et  al.  (1945)  and  Mudd  (1945). 

Thiamine  antagonists.  Thiamine  may  be  inactivated  by  an  enzyme, 
thiaminase,  which  is  found  in  fish  viscera  (Sealock  et  al.,  1943)  and  prob- 
ably occurs  in  other  organisms.  Foxes  which  are  fed  raw  fish  may 
develop  a  thiamine-deficiency  disease  (Chastek  paralysis).  The  mode  of 
inactivation  was  further  investigated  by  Krampitz  and  Woolley  (1944), 
who  found  that  thiamine  was  destroyed  by  a  process  of  enzymatic  hydrol- 
ysis whereby  the  thiazole  and  pyrimidine  moieties  were  formed.  Mucor 
ramanniamis  (thiazole-deficient)  and  Endomyces  vernalis  (pyrimidine- 
deficient)  were  used  as  test  organisms  in  the  preliminary  work.  Another 
thiamine  antagonist  of  unknown  nature  has  been  reported  to  occur  in 
bracken  fern  (Weswig  et  al.,  1946). 

Pyrithiamine,  an  analogue  of  thiamine,  has  been  used  in  studies  of 
competitive  thiamine  inhibition.  Unfortunately,  the  exact  structure  of 
this  compound  is  not  known.  In  papers  published  before  1949  it  was 
assumed  that  pyrithiamine  had  the  structure  now  assigned  to  neopyri- 


232 


PHYHIOLOGY  OF  THE  FUNGI 


thiamine  (Wilson  and  Harris,  1949).  Pyrithiamine  appears  to  differ 
from  neopyrithiamine  in  the  amount  of  pyrimidine  moiety  it  contains. 
The  formula  for  neopyrithiamine  is  given  below. 

N=C— NH2-HBr  CH3  CH2— CH2— OH 


CHs — C     C — CHj 


N— CH  Br 

Neopyrithiamine 

Robbins  (1941)  found  low  concentrations  of  pyrithiamine  to  replace 
thiamine  for  Pythiomorpha  gonapodyoides  (pyrimidine-deficient),  while 
high  concentrations  inhibited  growth.  Pyrithiamine  did  not  replace 
thiamine  for  Phycomyces  blakesleeanus  (requires  both  moieties)  or  Phyto- 
phthora  cinnamomi  (requires  intact  thiamine).  The  inhibition  of  growth 
of  various  fungi  and  bacteria  caused  by  pyrithiamine  was  overcome  by 
increasing  the  thiamine  content  of  the  medium  (Woolley  and  White, 
1943).  The  inhibition  index  is  given  in  Table  39.  The  efficiency  of 
pyrithiamine  as  a  thiamine  antagonist  is  related  to  the  specific  vitamin 
requirements  of  the  organisms  tested.  The  inhibition  index  was  low  for 
those  species  which  require  intact  thiamine,  intermediate  for  those  which 
require  either  or  both  moieties,  and  high  for  self-sufficient  species. 


Table  39. 


The  Efficiency  of  Pyrithiamine  as  an  Inhibitor  of  Fungus  and 
Bacterial  Growth 
(Woolley  and  White,  Jour.  Exptl.  Med.  78,  1943.) 


Organism 

Inhibition  index 

pyrithiamine 

thiamine 

Thiamine 
requirement 

Ceratastomella  fimbriata 

C.  penicillata 

Phytophthora  cinnamomi 

7 

10 

12 

11 

130 

800 

800 

400,000 

40,000 

5,000,000 

Intact  thiamine 
Intact  thiamine 
Intact  thaimine 

Chalaropsis  thielavioides 

Intact  thaimine 

Fndomyces  vernalis 

Pyrimidine 

Mucor  ramannianus 

Thiazole 

Saccharomyces  cerevisiae 

Neurospora  crassa 

Lactohacillus  arabinosus 

L.  casei 

Both  moieties 
None 
None 
None 

Pyrithiamine  was  found  to  inhibit  sporulation  of  Ceratostomella  fimbri- 
ata, Choanephora  cucurhitarum,  and  Chaetomium  convolutum  (Lilly  and 
Barnett,  1948).  This  inhibition  was  overcome  by  thiamine.  Pyrithia- 
mine was  reported  to  be  a  more  efficient  antagonist  for  diphosphothia- 
mine  than  for  thiamine  when  Penicillium  digitatum  was  used  as  a  test 
organism  (Sarett  and  Cheldelin,  1944). 


METABOLITE  ANTAGONISTS  233 

Pyrithiamine  causes  a  thiamine  deficiency  disease  in  mice,  which  may 
be  cured  or  prevented  by  the  administration  of  sufficient  thiamine  (Wool- 
ley  and  White,  1943).  Neopyrithiamine  is  reported  to  be  four  times  as 
active  as  pyrithiamine  for  the  rat  (Wilson  and  Harris,  1949). 

Biotin  antagonists.  Many  biotin  vitamers  are  known  which  are  highly 
specific.  The  efficiency  of  an  antibiotin  in  some  instances  may  depend 
upon  whether  biotin  or  one  of  its  vitamers  is  the  competing  metabolite. 
The  formulas  of  two  of  the  compounds  are  given  below.  Compare  with 
the  formulas  of  biotin  and  desthiobiotin  given  in  Chap.  9. 

CO  CO 

/   \  /   \ 

HN  NH  HN  NH 

CHs— CH Cn(CH.2)5— SO3H  H2C CH(CH2)5— COOH 

Sulfonic-acid  analogue  of  desthiobiotin  Imidazolidonecaproic  acid 

Desthiobiotin  and  imidazolidonecaproic  acid  differ  only  by  a  methyl 
group.  Desthiobiotin  was  found  to  act  as  a  biotin  vitamer  for  Saccha, 
romyces  cerevisiae  and  other  yeasts  (Dittmer  et  al.,  1944;  Lilly  andLeonian- 
1944),  while  imidazolidonecaproic  acid  is  an  antibiotin  for  S.  cerevisiae 
(Dittmer  and  Du  Vigneaud,  1944).  Both  compounds  are  antibiotins 
for  Lactobacillus  casei.  The  sulfonic-acid  analogue  of  desthiobiotin  was 
shown  by  Duschinsky  and  Rubin  (1948)  to  be  more  active  against  desthio- 
biotin and  oxybiotin  than  against  biotin  for  S.  cerevisiae.  The  replace- 
ment of  a  carboxyl  group  by  a  sulfonic-acid  group  appears  to  be  a  rather 
general  method  of  changing  a  metabolite  into  an  antagonist.  Further 
examples  of  this  will  be  cited  in  connection  with  pantothenic  and  amino- 
acid  antagonists. 

Egg  white  contains  a  specific  protein  which  combines  with  biotin  and 
thus  renders  this  vitamin  inactive.  This  inactivity  is  due  to  the  molecu- 
lar size  of  the  avidin-biotin  complex,  which  prevents  its  absorption  by 
organisms.  Raw  egg  white  may  be  used  to  produce  experimental  biotin 
deficiency  in  animals.  Avidin  is  no  longer  active  after  heating,  and 
bound  biotin  is  released  by  this  treatment.  This  specific  protein  has 
been  used  to  separate  biotin  vitamers  into  two  groups,  for  avidin  com- 
bines only  with  those  compounds  w^hich  have  an  intact  urea  ring  structure. 
The  papers  of  Eakin  et  al.  (1941)  and  Burk  and  Winzler  (1943)  may  be 
consulted  for  further  details. 

Pantothenic  acid  antagonists.  Yeasts  are  the  only  fungi  which  have 
been  reported  to  be  deficient  for  pantothenic  acid,  and  in  most  instances 
/3-alanine  replaces  the  intact  vitamin  molecule.  One  of  the  commonly 
studied  pantothenic  acid  antagonists  is  the  compound  called  pantoyl- 
taurine.  The  formulas  for  pantothenic  acid  and  pantoyltaurine  are 
given  belo'rt 


234 


PHYSIOLOGY  OF  THE  FUNGI 


HO— CH2— C(CH3)2- 
HO— CH2— C(CH3)2- 


CHOH— CO— NH- 

Pantothenic  acid 

-CHOH— CO— NH- 

Pantoyltaurine 


CH2— CHo— COOH 


-CH2— CH2— SO3H 


Pantoyltaurine  is  the  sulfonic-acid  analogue  of  pantothenic  acid. 
Snell  (1941)  studied  the  competitive  inhibition  of  yeast  growth  by  pan- 
toyltaurine and  found  that  this  compound  was  effective  when  pantothenic 
acid  was  the  metabolite  supplied  in  the  medium  but  that  pantoyltaurine 
did  not  compete  with  /3-alanine.  The  data  in  Table  40  illustrate  this 
difference. 

Table  40.     The  Effect  of  Pantoyltaurine  on  the  Growth  of  Saccharomyces 
cerevisiae  in  the  Presence  of  Pantothenic  Acid  and  (3-Alanine 
Inoculum  used,  0.02  mg.,  time  of  incubation,  16  hr.      (Snell,  Jour.  Biol.  Chem.  141, 
1941.) 


Calcium 

pantothenate, 

yug/lO  ml. 

Sodium  salt 
of  pantoyl- 
taurine, 
Mg/10  ml. 

Moist 

cells, 

mg/10  ml. 

/3- Alanine, 
Mg/10  ml. 

Sodium  salt 
of  pantoyl- 
taurine, 
Mg/10  ml. 

Moist 

cells, 

mg./lO  ml. 

0.0 

0 

0.3 

0.0 

0 

0.03 

0.5 

0 

6.6 

0.3 

0 

2.8 

0.5 

1,000 

2.9 

0.5 

0 

5.5 

0.5 

5,000 

0.4 

0.3 

1,000 

3.0 

0.5 

10,000 

0.3 

0.3 

10,000 

3.0 

30.0 

10,000 

6.6 

0.5 

5,000 

5.5 

0.5 

10,000 

5.7 

The  synthesis  of  pantothenic  acid  via  /3-alanine  by  Escherichia  colt 
is  inhibited  by  cysteic  acid  (sulfonic-acid  analogue  of  aspartic  acid). 
This  inhibition  is  reversed  by  /3-alanine  or  pantothenic  acid  (Ravel  and 
Shive,  1946).  For  further  information  concerning  other  pantothenic 
acid  and  other  antagonists,  the  review  of  Roblin  (1946)  should  be 
consulted. 

Pyridoxine  antagonists.  Some  of  the  pyridoxine  analogues  studied 
by  Robbins  and  Ma  (1942)  inhibited  the  growth  of  Ceratostomella  ulmi. 
This  inhibition  was  reversed  by  additional  pyridoxine.  The  substitution 
of  an  ethyl  group  for  the  methyl  group  of  pyridoxine  produced  an  antag- 
onist for  C.  ulmi,  but  ethyl  pyridoxine  was  as  active  for  excised  tomato 
roots  as  pyridoxine  itself.  The  above  authors  suggest  that  ethyl  pyri- 
doxine might  be  a  chemotherapeutic  agent  for  the  Dutch  elm  disease. 
The  formulas  of  ethyl  pyridoxine  and  desoxypyridoxine  are  given  below : 


HO- 


C2H2- 


/^-CHoOH 


-CH2OH 


HO- 


/%— CH2OH 


\n^ 


CH3— 


Ethyl  pyridoxine 


Desoxypyridoxine 


METABOLITE  ANTAGONISTS 


235 


Martin  et  at.  (1948)  found  desoxypyridoxine  to  be  slightly  more  effec- 
tive against  pyridoxal  than  pyridoxine,  when  Saccharomyces  cerevisiae 
was  used. 

Vitamin  K  antagonists.  There  are  at  least  two  naturally  occurring 
compounds  which  have  vitamin  K  activity.  Certain  synthetic  analogues 
are  used  in  medicine  to  replace  the  natural  vitamins.  All  these  com- 
pounds are  substituted  1,4-naphthoquinones.  The  structural  formula 
for  vitamin  K2  is  given  below: 


O 


O 


-CHs  CH3  CH3 

I  I 

-CHo— (CH=C— CH2— CH2)6— CH=C— CH3 

Vitamin  K2 


Horsfall  (1945)  has  reported  2-methyl-l,4-naphthoquinone  to  be  a  weak 
fungicide,  although  this  compound  replaces  natural  vitamin  K  in  medi- 
cine. On  the  other  hand,  2,3-dichloro-l,4-naphthoquinone  (Phygon)  is 
a  potent  fungicide  (Ter  Horst  and  Felix,  1943). 


O 


0 


— CHa 


O 
2-Methy  1- 1 , 4-naphtho  quinone 


—CI 
—CI 


o 


2 , 3-Dichloro- 1 , 4-naphtho  quinone 


Phygon  may  act  as  a  fungicide  by  virtue  of  combination  of  the  quinone 
with  free  amine  or  sulfhydryl  groups.  This  mechanism  probably  inac- 
tivates certain  enzymes  noncompetitively.  On  the  other  hand,  Phygon 
is  structurally  related  to  vitamin  K,  and  a  competitive  type  of  inhibition 
should  also  be  possible.  Woolley  (1945)  investigated  the  inhibitory  effect 
of  2,3-dichloro-l,4-naphthoquinone  and  2-methyl-l,4-naphthoquinone  on 
the  growth  of  Saccharomyces  cerevisiae  and  Endomyces  vernalis.  The  first 
compound  was  more  toxic  than  the  second.  In  less  than  toxic  concen- 
trations, the  second  compound  partially  overcame  the  toxicity  of  the 
first.  The  amount  of  2,3-dichloro-l,4-naphthoquinone  required  to 
inhibit  yeast  (half  maximum  growth)  was  1.7  jug  per  liter,  while  230  ng 
of  2-methyl-l,4-naphthoquinone  were  required  to  produce  the  same 
amount  of  inhibition.  Some  of  Woolley 's  data  are  presented  in  Table  41. 
Many  potent  antimalarial  drugs  are  1,4-naphthoquinone  derivatives 
(Fieser  et  al.,  1948). 

It  has  been  assumed  in  our  discussion  of  the  effects  of  antagonists  on 


236  PHYSIOLOGY  OF  THE  FUNGI 

organisms  that  antimetabolites  are  active  by  virtue  of  interfering  with 
various  enzymatic  processes.  It  is  also  interesting  to  note  that  com- 
petitive inhibition  has  been  demonstrated  with  isolated  enzyme  systems, 
Schopfer  and  Grob  (1949)  found  the  action  of  urease  to  be  inhibited  by 
2-chloro-l,4-naphthoquinone.  Most  of  the  activity  was  restored  by  the 
addition  of  2-methyl-l,4-naphthoquinone  (vitamin  K3). 

Table  41.     The  Reversal  of  Inhibition  Caused  by  2,3-Dichloro-1,4-naphtho- 
QuiNONE  by  2-Methyl-1,4-naphthoquinone 
Test  fungus,  Saccharomyces  cerevisiae.     Concentration  of  2,3-dichloro-l,4-naphtho- 
quinone,  0.005  jug/ml.      (Woolley,  Proc.  Soc.  Exptl.  Biol.  Med.  60,  1945.     Published 
by  permission  of  the  Society  for  Experimental  Biology  and  Medicine.) 

2-methyl-l,4-  Turbidity 
naphthoquinone,  ^g/ml.        (100  =  no  growth) 

0.0  93 

0.04  60 

0.02  68 

0.01  77 

0.005  85 

Other  vitamin  antagonists.  The  sulfonic-acid  analogue  of  nicotinic 
acid  inhibits  the  growth  of  certain  bacteria  (Mcllwain,  1940).  Appar- 
ently this  analogue  has  not  been  tested  in  nicotinic  acid-deficient  fungi. 
Woolley  (194Ga)  has  reported  maize  to  contain  a  "pellagragenic"  agent 
which  may  tentatively  be  considered  as  a  naturally  occurring  anti- 
nicotinic-acid  factor. 

Among  the  recently  developed  insecticides,  7-hexachlorocyclohexane 
is  of  considerable  value.  Kirkwood  and  Phillips  (1946)  have  shown  that 
the  growth  of  Saccharomyces  cerevisiae  is  inhibited  by  this  compound,  and 
that  the  inhibition  is  overcome  by  meso-inositol.  The  other  isomers  of 
hexachlorocyclohexane  were  not  very  effective  inhibitors  of  yeast  growth ; 
neither  are  they  of  much  value  as  insecticides.  These  observations  point 
to  competitive  inhibition  as  a  possible  mechanism  of  insecticidal  action 
of  this  compound. 

AMINO-ACID  ANTAGONISTS 

Organisms  must  either  synthesize  or  obtain  from  exogenous  sources 
the  different  amino  acids  they  require  for  the  synthesis  of  protein.  Anti- 
metabolites which  antagonize  the  synthesis  or  utilization  of  essential 
amino  acids  would  have  a  profound  effect  upon  growth  or  other  functions 
of  organisms.  The  role  of  amino  acids  is  not  confined  to  the  synthesis 
of  proteins  but  extends  to  the  synthesis  of  other  essential  metabolites. 
An  amino-acid  antagonist  may  act  in  two  ways,  (1)  by  inhibiting  protein 
synthesis  and  (2)  by  inhibiting  the  synthesis  of  essential  metabolites 
which  are  derived  from  amino  acids,  either  directly  or  indirectly.  If 
an  amino  a^id  functions  in  more  than  one  way,  the  action  of  an  amino- 


METABOLITE  ANTAGONISTS 


237 


acid  antagonist  may  be  overcome,  at  least  in  part,  by  the  action  of  second- 
ary metabolites  as  well  as  the  primary  metabolite.  The  toxic  effect  of 
3-acetylpyridine  on  rats  is  reversed  by  either  nicotinic  acid  amide  or 
tryptophane  (Woolley,  1945a). 

Analogues.  Mitchell  and  Niemann  (1947)  found  that  the  halogenated 
derivatives  of  phenylalanine  and  tyrosine  competitively  inhibit  growth 
of  the  wild  strain  of  Neurospora  crassa  (Table  42).  The  most  effective 
of  these  inhibitors  was  3-fiuoro-DL-phenylalanine.  The  structural  form- 
ulas for  this  analogue  and  the  natural  metabolite  are  shown  below: 


^^-CHo— CHNH2— COOH 


/\ 


CHo— CHNH2— COOH 


V 


Phenylalanine 


3-Fluorophenylalanine 


Table  42.     Inhibition  of  Growth  of  Neurospora  crassa  by  Some  Halogenated 

Alpha-amino  Acids 
Basal  medium  contained  30  mg.  of  DL-phenylalanine  or  20  mg.  L-tyrosine  per  liter 
depending  upon  the  antagonist  tested.      (Mitchell  and  Niemann,  Jour.  Am.  Chem. 
Sac.  69,  1947.     Published  by  permission  of  the  American  Chemical  Society.) 


Compound 

Mg./ml.  for  50% 
inhibition 

Moles  inhibitor 
Moles  amino  acid 

3-Fluoro-DL-phenylalanine 

3-Fluoro-DL-tyrosine 

3-Fluoro-L-tyrosine 

3-Fluoro-D-tyrosine 

0.04 
0.23 
0.15 
0.41 

1.2 

10.5 

6.8 

18.5 

The  other  halogen  derivitives  (chloro,  bromo,  and  iodo)  were  less  effective 
inhibitors.  3-Fluorophenylalanine  was  shown  to  be  an  effective  inhibitor 
for  various  other  fungi  and  bacteria. 

The  effect  of  j8-2-thienylalanine  on  the  growth  of  a  strain  Saccharomyces 
cerevisiae  and  certain  bacteria  has  been  studied  by  Ferger  and  Du 
Vigneaud  (1948).  The  formula  for  this  thiophene  analogue  of  phenyl- 
alanine is  given  below: 

HC C— CHo— CHNH2— COOH 


HC 


CH 


i3-2-Thienylalanine 

Only  the  l  isomer  is  active  in  competing  with  phenylalanine.  The  replace- 
ment of  divalent  sulfur  ( — S — )  by  a  vinylene  group  ( — CH=CH — ),  or 
vice  versa,  often  leads  to  the  production  of  an  antimetabolite.  As 
another  example,  the  effect  of  replacing  sulfur  in  cysteine  by  radicals 


238 


PHYSIOLOGY  OF  THE  FUNGI 


containing  the  vinylene  group  may  be  cited.  Dittmer  et  al.  (1948)  found 
methallylglycine,  allylglycine,  and  crotylglycine  to  inhibit  the  growth  of 
Saccharomyces  cerevisiae  and  Escherichia  coli.  The  effects  of  these  three 
antimetabohtes  on  the  growth  of  yeast  are  shown  in  Fig.  49. 


50  100  150  200         250         300 

Micrograms  of  unsoturoted  amino  acids  per  7.5  ml. 

Fig.  49.  The  inhibition  of  growth  of  Saccharomyces  cerevisiae,  strain  139,  by  DL-allyl- 
glycine,  DL-methallylglycine,  and  DL-crotylglycine.  (Courtesy  of  Dittmer,  Goering, 
Goodman,  and  Cristol,  Jour.  Am.  Chem.  Soc.  70:  2501,  1948.  Published  by  permis- 
sion of  the  American  Chemical  Society.) 

Natural  amino  acids.  Antagonism  among  the  amino  acids  is  not 
limited  to  competitive  inhibition  between  naturally  occurring  amino 
acids  and  their  analogues.  Robbins  and  McVeigh  (1946)  found  hydroxy- 
proline  to  inhibit  the  growth  of  several  dermatophytes:  Tricho-phyton 
mentagrophytes,  T.  gypseum  (granular  form),  T.  purpureum,  Epidermo- 
phyton  fiocculosum,  and  Microsporum  canis.  This  inhibition  was  over- 
come by  proline.  The  relationship  of  these  two  naturally  occurring 
amino  acids  is  shown  below: 


H2C- 
H.C 


-CH2 
CH— COOH 


HOHC- 


H2C 


-CHo 

in- 


COOH 


N 


H 
Proline 


N 

H 
Hyd  ro  xy  proline 


Low  concentrations  of  hydroxyproline  stimulated  growth  of  Tricho- 
phyton purpureum,  while  higher  concentrations  inhibited  growth.  Addi- 
tion of  hydroxyproline  to  a  glucose-asparagine  medium  increased  the 
growth  of  Polyporus  squamosiis.  Hydroxyproline  was  without  effect 
on  the  growth  of  19  other  species  of  fungi.  Whether  amino-acid  antago- 
nisms may  limit  the  nitrogen  utilization  of  natural  mixtures  of  these 


METABOLITE  ANTAGONISTS  239 

compounds  is  unknown,  but  the  possibility  of  inhibition  should  be  kept 
in  mind  when  only  a  few  amino  acids  are  used  in  a  medium.  The  effect 
of  any  single  compound  upon  a  fungus  may  be  modified  by  the  other 
constituents  of  the  medium. 

Harteiius  (1946a)  found  glutamic  and  aspartic  acids,  glutamine,  and 
asparagine  to  inhibit  the  growth  of  a  strain  of  yeast  when  suboptimumal 
amounts  of  /^-alanine  were  used  in  the  medium.  These  amino  acids  did 
not  inhibit  growth  when  pantothenic  acid  was  used.  In  fact,  these  com- 
pounds stimulated  growth  under  these  conditions.  The  inhibitory  effect 
in  the  presence  of  /3-alanine  was  overcome  by  increasing  the  concentration 
of  this  provitamin.  To  obtain  maximum  growth  in  the  presence  of  50 
mg.  of  glutamic  acid  per  flask  (55  ml.),  twenty  times  as  much  j9-alanine 
was  required  as  when  glutamic  acid  was  omitted  from  the  medium. 
Harteiius  attributed  this  effect  to  the  combination  of  /3-alanine  and 
glutamic  acid  to  form  an  inactive  dipeptide. 

Among  the  naturally  occurring  amino  acids,  L-canavanine  is  found 
free  in  jack  beans.  Canavanine  is  an  analogue  of  arginine;  the  formulas 
are  shown  below. 

HjN— C(=NH)— NH— CH2— CH2— CH,— CH(NH2)— COOH 

Arginine 

H2N— C(=NH)— NH— O— CH2— CH2— CHCNH,)— COOH 

Canavanine 

Horowitz  and  Srb  (1948)  studied  the  effect  of  canavanine  on  three 
wild-type  strains  of  Neurospora  and  found  one  strain  to  be  inhibited 
completely  by  concentrations  greater  than  1.25  mg.  per  liter;  another 
strain  was  only  partially  inhibited,  while  the  third  strain  was  tolerant. 
Genetic  analysis  indicated  that  tolerance  and  susceptibility  segregated 
by  alternative  forms  of  a  single  gene.  L-Arginine  was  effective  in  over- 
coming canavanine  toxicity,  while  L-lysine  was  less  effective.  Three 
molecules  of  arginine  overcame  one  molecule  of  canavanine  in  the  strain 
of  Neurospora  most  sensitive  to  this  inhibitor.  A  similar  competitive 
inhibition  between  canavanine  and  arginine  in  various  bacteria  has  also 
been  observed  (Volcani  and  Snell,  1948). 

Other  metabolite  antagonists.  Woolley  (1944a)  found  benzimidazole 
to  inhibit  the  growth  of  Saccharomyces  cerevisiae  and  Endomyces  vernalis. 
This  inhibition  was  overcome  by  adenine  and  guanine.  The  structural 
relationship  between  benzimidazole  and  adenine  is  shown  below: 


N=C— NH2 

HC     C— NH 

\ 
CH 


— C— N 
Benzimidazole  Adenine 


240  PHYSIOLOGY  OF  THE  FUNGI 

DEVELOPMENT  OF  FASTNESS 

An  organism  which  has  become  tolerant,  or  resistant,  to  an  inhibitor 
(analogue,  drug,  antibiotic,  etc.)  after  exposure  is  said  to  be  fast,  or  more 
specifically  pyrithiamine-fast,  sulfanilamide-fast,  or  penicillin-fast,  as 
the  case  may  be.  Fastness  is  a  very  common  phenomenon,  although  it 
appears  to  have  been  but  little  studied  in  fungi.  It  is  an  important  factor 
which  limits  the  use  of  many  antibiotics  and  the  sulfonamides  in  medicine. 
This  phase  of  fungus  physiology  deserves  more  attention  than  it  has 
received.  It  is  conceivable  that  the  prolonged  use  of  a  single  fungicide 
to  control  a  fungus  pathogen  could  lead  to  the  development,  or  selection, 
of  a  strain  which  would  be  relatively  tolerant  to  the  effect  of  the  fungicide. 
Such  findings  do  not  appear  to  have  been  reported  from  field  studies,  but 
this  possibility  should  be  kept  in  mind. 

Fungi  do  become  fast  to  various  antagonists.  WooUey  (19446),  by 
repeatedly  subculturing  Endomyces  vernalis  in  a  medium  containing 
pyrithiamine,  developed  a  strain  which  withstood  twenty-five  times  the 
concentration  of  pyrithiamine  which  served  to  reduce  the  growth  of 
the  parent  strain  to  half  the  maximum.  In  this  instance,  fastness  was 
correlated  with  the  ability  of  the  pyrithiamine-fast  strain  to  cleave  the 
inhibitor  molecule  into  its  cyclic  moieties.  Thus,  the  development  of 
pyrithiamine  fastness  may  be  ascribed  to  the  formation  of  an  adaptive 
enzyme  which  destroyed  the  antagonist.  Escherichia  coU,  which  is  not 
inhibited  by  pyrithiamine,  also  hydrolyzed  this  compound.  These 
results  indicate  that  adaptive  enzymes  may  play  a  role  in  the  develop- 
ment of  fastness. 

In  addition  to  resistance  or  fastness  which  develops  in  organisms  cul- 
tured in  the  presence  of  an  inhibitor,  it  has  been  found  recently  that 
various  bacteria  not  only  develop  resistance  but  may  develop  strains 
which  are  actually  dependent  upon  the  presence  of  the  "inhibitor"  before 
they  can  grow.  Yegian  et  al.  (1949)  have  found  that  culturing  Myco- 
bacterium tuberculosis  in  the  presence  of  streptomycin  gave  rise  to  strains 
which  were  fast  to  this  antibiotic  and  also  produced  strains  which  cannot 
grow  unless  streptomycin  is  present  in  the  medium. 

SUMMARY 

The  normal  utilization  of  a  metabolite  may  be  prevented  or  inhibited 
in  three  ways:  (1)  destruction  or  removal  in  an  unavailable  combination 
of  a  metabolite;  the  enzymatic  hydrolysis  of  thiamine  and  the  combina- 
tion of  biotin  with  avidin  are  representative  examples  of  this  mode  of 
inactivation ;  (2)  the  noncompetitive  inhibition  of  various  enzymes  by 
such  compounds  as  iodoacetate,  cyanide,  and  azide;  (3)  competitive 
inhibition  due  to  metabolite  antagonists.     This  type  of  inhibition  is 


METABOLITE  ANTAGONISTS  241 

overcome  by  increasing  the  concentration  of  the  normal  metabohte. 
Antagonists  are  known  which  inhibit  the  functioning  of  vitamins,  amino 
acids,  and  other  metabohtes. 

It  is  postulated  that  a  metabolite  and  its  antagonists  compete  for  the 
active  surface  of  specific  enzymes.  The  ratio  of  inhibitor  to  metabolite 
required  to  reduce  growth  to  one-half  its  normal  value  is  called  the  inhibi- 
tion index.  Effective  inhibitors  have  small  inhibition  indexes.  The 
same  compound  may  act  as  an  antagonist  for  some  fungi  and  as  a  metabo- 
lite for  others;  e.g.,  desthiobiotin.  The  medium  used  for  investigating 
inhibition  is  important,  for  the  presence  of  a  normal  metabolite,  or  a 
secondary  metabolite  derived  from  it,  may  prevent  inhibition,  A  given 
compound  may  be  considered  as  an  antagonist,  but  it  is  only  an  antagonist 
for  certain  species,  and  then  only  under  certain  conditions.  Organisms 
may  acquire  a  tolerance  or  resistance  to  inhibitory  agents  and  become 
fast.  In  extreme  instances  they  become  dependent  upon  the  inhibitor, 
which  then  acts  as  a  kind  of  growth  factor. 

The  competitive  nature  of  many  inhibitions  is  firmly  established.  In 
most  instances  there  is  a  close  structural  relation  between  a  metabolite 
and  its  antagonists.  The  theories  which  have  been  advanced  to  explain 
these  phenomena  have  been  useful  in  correlating  the  results  of  research 
and  for  increasing  our  insight  into  metabolic  processes, 

REFERENCES 

BuRK,  D.,  and  R.  J.  Winzler:  Heat-labile,  avidin-uncombinable,  species-specific 

and  other  vitamers  of  biotin,  Science  97 :  57-60,  1943. 
DiTTMER,  K.,  and  V.  du  Vigneaud:  Antibiotins,  Science  100:  129-131,  1944. 
DiTTMER,  K.,  H.  L.  GoERiNG,  I.  GooDMAN,  and  S.  J.  Cristol:  The  inhibition  of 

microbiological  growth  by   allylglycine,    methallylglycine   and   crotylglycine, 

Jour.  Am.  Chem.  Soc.  70:  2499-2501,  1948. 
DiTTMER,  K.,  D.  B.  Melville,  and  V.  du  Vigneaud:  The  possible  synthesis  of 

biotin  from  desthiobiotin  by  yeast  and  the  antibiotin  effect  of  desthiobiotin  for 

L.  casei,  Science  99 :  203-204,  1944. 
DuscHiNSKY,  R.,  and  S.  H.  Rubin:  The  synthesis  and  biological  activity  of  4-methyl- 

5-(e-sulfoamyl)-2-imidazolidone,  a  sulfonic  acid  analog  of  desthiobiotin,  Jour. 

Am.  Chem.  Soc.  70 :  2546-2547,  1948. 
Eakin,  R.  E.,  E.  E.  Snell,  and  R.  J.  Williams:  The  concentration  and  isolation  of 

avidin,  the  injury-producing  protein  in  raw  egg  white,  Jour.  Biol.  Chem.  140: 

535-543,  1941. 
Ferger,  M.  F.,  and  V.  du  Vigneaud:  The  microbial  growth  inhibition  produced  by 

optical  isomers  of  /3,2-thienyl-alanine,  Jour.  Biol.  Chem.  174 :  241-246,  1948. 
FiESER,  L.   F.,  et  al.:   Naphthoquinone  antimalarials.     I.  General  survey,  Jour. 

Am.  Chem.  Soc.  70:  3151-3155,  1948. 
FiLDES,  P. :  A  rational  approach  to  research  in  chemotherapy,  Lancet  1 :  955-957, 

1940. 
FouRNEAU,  E.,  J.  Tr^ifouel,  J.  Tri^fouel.  F.  Nitti,  and  B  Bovet:  Action  du  para- 

aminophenylsulfamide  sur  les  moisisseures,  Compt.  rend,  s  jc.  biol.  122 :  652-654, 

1936. 


242  PHYSIOLOGY  OF  THE  FUNGI 

Hartelius,  v.:  Antivitamin  effect  of  sulfanilamide  on  Aspergillus  niqer,  Conipl. 

rend.  trav.  lab.  Carlsberg,  Sir.  physiol.  24:  178-184,  1946. 
Hartelius,  V.:  Glutamic  acid,  aspartic  acid,  asparagine  and  glutamine  as  anti- 
growth  substances  for  /3-alanine,  Conipt.  rend.  trav.  lab.  Carlsberg.  SSr.  physiol. 

24:  185-222,  1946a. 
Hartelius,  V.,  and  K.  Roholt:  Effect  of  marfanil  on  Streptobacterium  planatarum, 

Saccharomyces  cerevisiae  and  Aspergillus  niger,  Compt.  rend.  trav.  lab.  Carlsberg, 

Sir.  physiol.  24:  163-171,  1946. 
Horowitz,  N.  H.,  and  A.  M.  Srb:  Growth  inhibition  of  Neurospora  by  canavanine 

and  its  reversal.  Jour.  Biol.  Chern.  174:  371-378,  1948. 
HoRSFALL,  J.  G.:  Fungicides  and  Their  Action,  Chronica  Botanica  Co.,  Waltham, 

1945. 
KiRKWOOD,  S.,  and  P.  H.  Phillips:  The  anti-inositol  effect  of  7-hexachlorocyclo- 

hexane,  Jour.  Biol.  Chem.  163:  251-254,  1946. 
Krampitz,  L.  O.,  and  D.  W.  Woolley:  The  manner  of  inactivation  of  thiamine  by 

fish  tissue,  Jour.  Biol.  Chem.  152:  9-17,  1944. 
Lampen,  J.  O.,  and  M.  J.  Jones:  The  antagonism  of  sulfonamide  inhibition  of  certain 

lactobacilli  and  entrococci  by  pteroylglutamic  acid  and  related  compounds, 

Jour.  Biol.  Chem.  166 :  435-448,  1946. 
Lilly,  V.  G.,  and  H.  L.  Barnett:  The  effect  of  pyrithamin  upon  sporulation  of  three 

thiamin-deficient  fungi.  Am.  Jour.  Botany  35:  801,  1948. 
Lilly,  V.  G.,  and  H.  L.  Leonian:  The  anti-biotin  effect  of  desthiobiotin,  Science  99 : 

205-206,  1944. 
Martin,  G.  J.,  S.  Avakian,  and  J.   Moss:  Studies  of  pyridoxine  displacement, 

Jour.  Biol.  Chem.  174 :  495-500,  1948. 
*McElroy,  W.  D.:  The  mechanism  of  inhibition  of  cellular  activity  by  narcotics, 

Quart.  Rev.  Biol.  22:  25-58,  1947. 
McIlwain,  H.:  Pyridine-3-sulphonic  acid  and  its  amide  as  inhibitors  of  bacterial 

growth,  Brit.  Jour.  Exptl.  Path.  21 :  136-147,  1940. 
McIlwain,  H.:  Interrelations  in  microorganisms  between  growth  and  metabolism  of 

vitamin-like  substances.  Advances  in  Enzymol.  7 :  409-460,  1947. 
Mitchell,  H.  K.,  and  C.  Niemann:  The  competitive  inhibition  of  the  metabolism 

of  a-amino  acids  by  their  halogenated  analogs,  Jour.  Am.  Chem.  Soc.  69:  1232, 

1947. 
MuDD,  S.:  Can  chemotherapy  be  extended  to  include  intra-cellular  disease  agents?. 

Jour.  Bad.  49:  527-537,  1945. 
Ravel,  J.  M.,  and  W.  Shive:  Biochemical  transformations  as  determined  by  compe- 
titive analogue-metabolite  growth  inhibitions.     IV.   Prevention  of  pantothenic 

acid  synthesis  by  cysteic  acid.  Jour.  Biol.  Chem.  166:  407-415,  1946. 
Robbins,  W.  J.:  The  pyridine  analog  of  thiamin  and  the  growth  of  fungi,  Proc. 

Natl.  Acad.  Sci.  U.S.  27:  419-422,  1941. 
Robbins,  W.  J.,  and  R.  Ma:  Specificity  of  pyridoxine  for  Ceratostomella  ulmi,  Bull. 

Torrey  Botan.  Club  69:  342-352,  1942. 
Robbins,  W.  J.,  and  I.  McVeigh:  Effect  of  hydroxyproline  on  Trichophyton  menta- 

grophytes  and  other  fungi.  Am.  Jour.  Botany  33 :  638-647,  1946. 
RoBLiN,  R.  O.,  Jr.:  Metabolite  antagonists,  Chem.  Revs.  38:  255-377,  1946. 
■*RoBLiN,  R.  O.,  Jr.:  Metabolite  antagonists.     The  relation  between  structure  and 

biological  activity,  Chem.  Eng.  Neivs  27 :  3624-3629,  1949. 
RuBBo,  S.  D.,  and  J.  M.  Gillespie:  Para-aminobenzoic  acid  as  a  bacterial  growth 

factor.  Nature  146:  838-839,  1940. 
Rydon,  H.  N. :  Anthranilic  acid  as  an  intermediate  in  the  biosynthesis  of  tryptophane 

by  Bact.  typhosum,  Brit.  Jour.  Exptl.  Path.  29 :  48-57,  1948. 


METABOLITE  ANTAGONISTS  243 

Sarett,  H.  p.,  and  V.  H.  Cheldelin:  Inhibition  of  utilization  of  thiamine  and 
diphosphothiamine  for  the  growth  of  microorganisms,  Jour.  Biol.  Chem.  156 : 
91-100,  1944. 

ScHOPFER,  W.  H.,  and  E.  C.  Grob:  Ueber  den  Einfluss  von  1,4-Naphtochinon- 
derivaten  mit  Vitamin-K-  oder  Antivitamin-K-Character  auf  die  Urease, 
Helv.  Chim.  Acta  3^ :  829-838,  1949. 

Sealock,  R.  R.,  a.  H.  Ln'ERMORE,  and  C.  A.  Evans:  Thiamine  inactivation  by 
the  fresh-fish  or  Chastek-paralysis  factor.  Jour.  Am.  Chem.  Soc.  65:  935-940, 
1943. 

Sevag,  M.  G.,  R.  a.  Richardson,  and  J.  Henry:  Studies  on  the  action  of  sulfo- 
namides on  the  respiration  and  growth  of  bacteria.  IIB.  Effect  of  serum  on  the 
inhibition  of  respiration  and  growth  of  Pneumococcus,  Type  I,  and  Staphylo- 
coccus aureus  by  sulfonainides  and  p-aminobenzoic  acid.  Jour.  Bad.  49  :  139-147, 
1945. 
■^Shive,  W.,  and  J.  Macow:  Biochemical  transformations  as  determined  by  competi- 
tive analogue-metabolite  growth  inhibitions.  I.  Some  transformations  involv- 
ing aspartic  acid,  Jour.  Biol.  Chem.  162 :  451-462,  1946. 

Singer,  T.  P. :  Enzyme  inhibitors  and  the  active  groups  of  proteins.  Brewers  Digest 
20:  85-88     104-106,   1945. 

Snell,  E.  E.:  Growth  inhibition  by  N-(Q;,7-dihydroxy-/3,^-dimethylbutyryl)taurLne 
and  its  reversal  by  pantothenic  acid,  Jour.  Biol.  Chem.  141:  121-128,  1941. 

Stoddard,  E.  M.:  The  x  disease  of  peach  and  its  chemotherapy.  Conn.  Agr.  Expt. 
Sta.  Bull.  506,  1947. 

Ter  Horst,  W.  p.,  and  E.  L.  Felix:  2,3-Dichloro-l,4-naphthoquinone,  a  potent 
fungicide,  Ind.  Eng.  Chem.  35:  1255-1259,  1943. 

Volcani,  B.  E.,  and  E.  E.  Snell:  The  effect  of  canavanine,  arginine  and  related 
compounds  on  the  growth  of  bacteria,  Jour.  Biol.  Chem.  174 :  893-902,  1948. 

Welch,  A.  D.:  Interference  with  biological  processes  through  the  use  of  analogs  of 
essential  metabolites,  Physiol.  Revs.  25:  687-715,  1945. 

Weswig,  p.  H.,  a.  M.  Freed,  and  J.  R.  Haag:  Antithiamine  activity  of  plant  mate- 
rials. Jour.  Biol.  Chem.  165 :  737-738,  1946. 

Wilson,  A.  N.,  and  S.  A.  Harris:  Synthesis  and  properties  of  neopyrithiamine  salts, 
Jour.  Amer.  Chem.  Soc.  71:  2231-2333,  1949. 

Wolf,  F.  T.  :  The  action  of  sulfonamides  and  antibiotic  agents  on  the  pathogenic 
fungi  in  Biology  of  Pathogenic  Fungi  (edited  by  W.  J.  Nickerson),  Chronica 
Botanica  Co.,  Waltham,  1947. 

Woods,  D.  D.  :  The  relation  of  p-aminobenzoic  acid  to  the  mechanism  of  the  action 
of  sulphanilamide,  Brit.  Jour.  Exptl.  Path.  21 :  74-90,  1940. 

WooLLEY,  D.  W.:  Some  new  aspects  of  the  relationship  of  chemical  structure  to 
biological  activity,  Science  100 :  579-583,  1944. 

WooLLEY,  D.  W.:  Some  biological  effects  produced  by  benzimidazole  and  their 
reversal  by  purines.  Jour.  Biol.  Chem.  152  :  225-232,  1944a. 

WooLLEY,  D.  W. :  Development  of  resistance  to  pyrithiamine  in  yeast  and  some 
observations  on  its  nature,  Proc.  Soc.  Exptl.  Biol.  Med.  55:  179-180,  1944b. 
*WooLLEY,   D.   W. :  Observations  on  the  antimicrobial  action  of  2,3-dichloro-l,4- 
naphthoquinone  and  its  reversal  by  vitamins  K,  Proc.  Soc.  Exptl.  Biol.  Med.  60 : 
225-228,  1945. 

WooLLEY,  D.  W.:  Production  of  nicotinic  acid  deficiency  with  3-acetylpyridine,  the 
ketone  analogue  of  nicotinic  acid.  Jour.  Biol.  Chem.  157 :  455-459,  1945a. 

WooLLEY,  D.  W. :  Some  aspects  of  biochemical  antagonism  in  Currents  in  Bio- 
chemical Research  (edited  by  D.  E.  Green),  Interscience  Publishers,  Inc., 
New  York,  1946. 


244  PHYSIOLOGY  OF  THE  FUNGI 

Wool  LEY,  D.  W.:  The  occurrence  of  a  "pcllagragenic"  agent  in  corn,  Jour.  Biol. 

Chem.  163 :  773-774,  1946a. 
WooLLEY,  D.  W.,  and  A.  G.  C.  White:  Selective  reversible  inhibition  of  microbial 

growth  with  pyrithiamine,  Jour.  Exptl.  Med.  78 :  489-497,  1943. 
•  Wright,  L.  D.:  The  significance  of  anti-vitamins  in  nutrition.  Jour.  Am.  Dietet. 

Assoc.  23:  289-298,  1947. 
Yegian,  D.,  V.  BuDD,  and  R.  J.  Vanderlinde:  Streptomycin-dependent  bacilh: 

A  simple  method  for  isolation,  Jour.  Bad.  58 :  257-259,  1949. 


CHAPTER  12 
THE  ACTION  OF  FUNGICIDES 

The  never-ending  warfare  which  man  must  wage  against  parasitic  fungi 
in  order  to  protect  his  crops  has  been  ably  chronicled  by  Large  (1940). 
The  saprophytic  species  which  decay  wood  and  other  cellulosic  materials 
cause  great  economic  loss,  although  these  species  perform  a  necessary 
and  indispensable  role  in  maintaining  the  carbon  cycle  in  nature.  It  is  to 
man's  interest  and  profit  that  the  deterioration  of  textiles  and  lumber  be 
prevented  or  delayed  and  that  his  crops  be  protected  from  pathogenic 
fungi.  This  is  done  by  the  use  of  fungicides,  which  either  kill  or  inhibit 
the  action  of  fungi. 

By  definition,  an  agent  which  kills  fungi  is  a  fungicide.  A  fungistatic 
agent  merely  causes  inhibition.  The  same  agent  is  commonly  capable  of 
producing  both  actions.  A  discussion  of  the  terms  fungicidal  and  fungi- 
static is  given  by  McCallan  and  Wellman  (1942).  These  authors  point 
out  that  the  fungistatic  activity  of  an  agent  is  broader  than  its  fungicidal 
activity. 

Both  physical  and  chemical  agents  may  be  fungicidal  and  fungistatic. 
Of  the  physical  agents,  heat  and  ultraviolet  radiation  are  probably  most 
commonly  used,  while  many  chemical  compounds  are  ''toxic"  to  fungi. 
Whether  an  agent  is  fungicidal  or  fungistatic  is  primarily  a  matter  of 
degree  of  intensity  and  duration  of  exposure.  We  may  assume  that  these 
agents,  whether  chemical  or  physical,  act  directly  upon  certain  specific 
enzymes  or  enzyme  systems.  If  the  action  is  less  severe  and  may  be 
reversed,  the  result  is  fungistasis,  while  if  it  is  irreversible,  the  action  is 
fungicidal.  Since  most  of  the  agents  employed  by  man  are  chemical 
compounds,  much  of  the  following  discussion  will  be  limited  to  the 
mechanism  of  action  of  these  compounds. 

Chemical  fungicides  may  be  applied  as  eradicants  or  as  protectants.  A 
protectant  is  applied  to  the  plant  or  other  material  before  the  inoculum 
arrives  at  the  infection  court  and  often  functions  only  after  the  fungus 
spore  germinates.  An  eradieant  kills  the  fungus  already  present  on  or  in 
the  substrate  material. 

The  lethal  action  of  a  chemical  depends  upon  both  the  concentration 
of  the  active  compound  or  ion  and  the  time  of  exposure.  Species  of 
fungi  exhibit  great  variation  in  ability  to  resist  the  action  of  certain 
fungicides.     Many  fungi  are  killed  by  exposure  to  a  few  parts  per  million 

245 


246  PHYSIOLOGY  OF  THE  FUNGI 

of  cupric  ion,  while  a  few  species  have  been  reported  to  grow  in  a  saturated 
solution  of  copper  sulfate  (Starkey  and  Waksman,  1943).  There  is  no 
useful  universal  fungicide. 

The  intelligent  choice  of  a  fungicide  depends  upon  a  number  of  factors, 
the  major  ones  being  the  species  of  fungus  to  be  controlled  and  the  nature 
of  the  material  to  be  protected.  The  solubility  of  the  fungicide  is  of  great 
importance.  For  most  efficient  preservation  of  wood  or  protection  as  a 
spray,  a  fungicide  must  have  a  low  solubility  in  order  that  the  protection 
may  extend  over  a  long  period  of  time.  For  surface  sterilization  a  highly 
soluble  fungicide  is  used.  When  a  fungicide  is  to  be  used  on  a  living 
plant  (or  other  organism),  the  relative  sensitivity  of  the  host  and  of  the 
fungus  to  the  fungicide  must  be  considered.  Host  sensitivity  limits  the 
use  of  many  potent  fungicides.  A  useful  fungicide  must  be  more  toxic 
to  the  fungus  than  to  the  host.  For  example,  copper  fungicides  are  quite 
toxic  to  cabbage,  cucumber,  and  pea  seed,  while  beet,  eggplant,  pepper, 
and  spinach  seed  are  relatively  tolerant  to  copper. 

Although  there  is  an  enormous  accumulation  of  literature  on  fungicides, 
their  composition,  application,  limitations,  and  economic  value  (see 
Frear,  1948,  and  Horsfall,  1945),  relatively  little  has  been  published  on 
the  mechanism  of  fungicidal  action.  This  is  a  practical  as  well  as  an 
academic  question,  for  the  intelligent  use  of  known  fungicides  and  the 
search  for  new  and  better  ones  are  based  upon  a  knowledge  of  how  they 
act. 

In  the  past  the  most  important  inorganic  fungicides  have  contained 
compounds  of  copper,  mercury,  or  sulfur.  In  the  future,  however,  excel- 
lent fungicides  may  be  made  from  other  toxic  elements.  For  example, 
cadmium  is  of  potential  interest,  but  the  present  supply  is  limited.  In 
controlling  fungi  and  other  pests,  there  is  always  the  danger  that  they 
will  become  tolerant,  or  fast,  to  a  given  toxicant.  This  means  that  the 
more  susceptible  individuals  are  killed  and  that  a  greater  amount  of  a 
given  fungicide  is  required  to  control  the  more  tolerant  population  which 
is  then  built  up.  It  is  desirable  from  several  viewpoints  to  have  satis- 
factory reserve  fungicides  in  the  armory  of  the  plant  pathologist. 

COPPER 

The  first  copper  salts  to  be  used  as  fungicides  were  the  sulfate  and 
acetate  (Prevost,  1807).  These  salts  are  soluble,  and  even  in  low  con- 
centration they  are  too  toxic  for  many  uses.  Since  all  the  copper  is 
available  at  once,  these  salts  are  toxic  to  plants,  especially  to  young  parts. 
These  soluble  salts  have  a  further  disadvantage  when  used  as  a  spray,  for 
a  heavy  dew  or  rain  will  easily  wash  them  off.  However,  these  salts, 
especially  copper  sulfate,  were  successfully  used  for  treating  seed  grain  to 
destroy  surface  contaminants.  This  treatment  was  devised  by  Prevost 
to  control  bunt. 


ACTION  OF  FUNGIDICES 


247 


The  next  advance  in  copper  fungicides  was  not  until  1885,  whei 
Millardet  published  the  formula  for  making  the  famous  fungicide 
Bordeaux  mixture.  Millardet  recommended  that  8  kg.  of  copper  sulfate 
pentahydrate  (bluestone)  be  dissolved  in  100  liters  of  water.  This  solu- 
tion was  then  mixed  with  15  kg.  of  quicklime  slaked  in  30  liters  of  water. 
The  chemistry  of  Bordeaux  mixture  is  more  complicated  than  was 
assumed  at  first.  Instead  of  cupric  hydroxide,  a  series  of  basic  sulfates 
are  formed,  the  composition  being  dependent  upon  the  ratio  of  coppei 
sulfate  and  calcium  hydroxide  used  (Frear,  1948).  Bordeaux  mixture  is 
a  copper  compound  or  compounds  of  low  solubility.  According  to 
Goldsworthy  and  Green  (1936),  Bordeaux  mixture  in  equilibrium  with 
water  yields  a  solution  containing  about  4  p. p.m.  of  copper.  However, 
McCallan  and  Wilcoxon  (1936)  found  that  well-washed  4-4-50  Bordeaux 
mixture  was  soluble  only  to  the  extent  of  furnishing  1  p. p.m.  of  copper. 
After  this  material  was  thoroughly  dried,  as  in  a  spray  film,  the  solubihty 
in  terms  of  copper  decreased  to  0.2  to  0.3  p. p.m.  McCallan  and  Wilcoxon 
have  reported  a  comparison  between  the  amounts  of  Bordeaux  mixture 
and  copper  sulfate  required  to  inhibit  the  germination  of  90  per  cent  of 
the  spores  of  a  few  species.     These  data  are  given  in  Table  43. 

Table  43.     The  Relative  Efficiency  of  Bordeaux  Mixture  and  Copper  Sulfate 

IN  Inhibiting  Spore  Germination 
(McCallan  and  Wilcoxon,  Contribs.  Boyce  Thompsori  Inst.  6,  1936.) 


Cu,  mg. /liter,  for  LD  90 

Species 

Bordeaux 
mixture 

Copper 

sulfate 

Uromyces  caryophyllinus 

Sclerotinia  fructicola 

180 
120 
390 
500 
2,400 

1.74 
1  20 

Botridis  paeoniae 

2  23 

Glomerella  cingulata 

Alternaria  solani 

1.40 
6.72 

If  Bordeaux  mixture  or  other  copper  spray  or  dust  of  low  solubility 
furnishes  less  than  1  p. p.m.  of  copper  to  the  solutions  with  which  it  is  in 
equilibrium,  it  is  obvious  that  the  concentration  of  copper  is  too  low  for 
any  great  amount  of  toxicity.  We  must  also  take  into  account  the  rate 
^  of  solubility  of  the  ''insoluble"  copper  compounds,  for  if  the  rate  of 
solution  is  slow,  the  maximum  concentration  may  not  be  attained  in 
time  to  prevent  infection.  The  only  hypothesis  which  would  account 
for  the  lethal  action  of  copper  compounds  of  such  low  solubility  would  be 
that  of  cumulative  action.  A  germinating  spore  in  a  saturated  solution 
of  the  copper  compound  in  equilibrium  with  the  solid  copper  compound 
would  remove  copper  from  the  solution.     This  process  would  cause  more 


248  PHYSIOLOGY  OF  THE  FUNGI 

of  the  copper  compound  to  dissolve  until  the  spore  was  no  longer  able  to 
take  more  copper  from  the  solution.  This  theory  is  attractive  because 
of  its  simplicity,  but  there  seems  to  be  no  very  good  evidence  for  it 
(McCallan,  1929). 

In  practice,  Bordeaux  mixture  and  other  "insoluble"  copper  sprays 
act  as  if  they  were  more  soluble  than  is  indicated  by  chemical  tests. 
However,  in  practice  the  fungicide  is  exposed  to  the  action  of  the  atmos- 
phere, the  host  plant,  and  the  fungus  spores.  This  is  a  more  complicated 
situation  than  that  found  in  the  chemical  determination  of  solubility. 
Barker  and  Gimingham  (1911)  found  that  intact  leaves  increased  the 
soluble  copper  from  Bordeaux  mixture  to  some  extent  but  were  of  the 
opinion  that  the  host  plant  had  only  a  slight  influence  on  the  solubility 
of  such  sprays.  However,  if  the  leaves  were  injured,  they  were  quite 
effective  in  bringing  copper  into  solution.  The  possibility  that  the  spores 
exert  a  solvent  action  on  "insoluble"  copper  compounds  has  long  been 
considered  by  plant  pathologists.  The  spores  of  at  least  some  species 
do  exert  a  solvent  action  on  Bordeaux  mixture.  McCallan  and  Wilcoxon 
(1936)  showed  that  the  amount  of  copper  brought  into  solution  by  the 
soluble  materials  washed  from  or  excreted  by  100  million  spores  of  some 
species  varied  as  follows:  Uromyces  caryophyllimis,  1.01  mg.;  Sclerotinia 
fructicola,  0.76  mg.;  Neurospora  sitophila,  0.12  mg.;  Botrytis  paeoniae, 
0.10  mg.;  Glomerella  cingulata,  0.046  mg. ;  Aspergillus  niger,  0.023  mg.; 
and  Alternaria  solani,  0.013  mg.  Enough  spores  of  Neurospora  sitophila 
were  collected  so  that  the  nature  of  the  soluble  materials  from  the  spores 
could  be  identified  chemically.  Malic  acid  was  isolated  and  identified. 
The  presence  of  amino  acids  also  was  detected.  Both  malic  acid  (or 
malates)  and  various  amino  acids  dissolve  "insoluble"  copper  compounds 
under  nevitral  or  alkaline  conditions  with  the  formation  of  soluble  complex 
copper  compounds.  McCallan  and  Wilcoxon  showed  that  sodium 
cuprimalate  and  a  copper-glycine  compound  were  about  as  toxic  as  copper 
sulfate.  On  the  other  hand,  Goldsworthy  and  Green  (1936)  were  of  the 
opinion  that  spore  secretions  played  a  minor  role  in  increasing  the  solu- 
bility of  Bordeaux  mixture,  but  the  evidence  of  McCallan  and  Wilcoxon 
seems  quite  conclusive. 

Basic  copper  carbonate  (malachite,  Cu(OH)2-CuC03)  and  cuprous 
oxide  (CuiO)  are  used  in  treating  seeds.  Since  the  seed  covered  with 
these  materials  is  planted  in  soil  which  contains  a  variety  of  protein 
degradation  products,  it  is  easy  to  understand  how  these  substances  are 
made  sufficiently  soluble  to  be  fungicidal.  Marten  and  Leach  (1944) 
studied  the  effect  of  various  nitrogenous  compounds  upon  the  solubility 
of  cuprous  oxide.  Gelatin  and  peptone  were  less  efficient  in  dissolving 
cuprous  oxide  than  were  glycine,  aspartic  acid,  asparagine,  or  cystine. 
Ammonium  hydroxide  was  also  a  solvent  for  cuprous  oxide.     With  all 


ACTION  OF  FUNGICIDES  249 

these  "solvents"  the  solutions  were  blue  in  color,  which  indicates  that 
the  copper  was  oxidized  to  the  cupric  state.  Marten  and  Leach  investi- 
gated the  toxicity  of  the  copper-glycine  compound  to  Pythnim  debaryanum. 
It  was  noted  that  an  excess  of  glycine  protected  the  fungus  from  the 
action  of  the  copper.  Some  200  times  as  much  copper  was  required  to 
inhibit  growth  when  glycine  was  present  in  the  medium  as  when  it  was 
absent.  Thus,  it  seems  that  whether  a  given  amount  of  copper  is  toxic 
or  not  depends  upon  the  nature  and  amount  of  certain  constituents  in 
the  medium  or  substrate. 

One  may  ask,  By  what  mechanism  does  the  copper  ion  cause  fungistasis, 
or  how  does  the  copper  kill?  The  common  explanation  of  the  toxic 
action  of  the  heavj'^  metals  (copper,  mercury,  and  silver)  is  based  upon  the 
property  of  these  ions  of  precipitating  or  denaturing  proteins.  Enzymes 
are  proteins,  and  it  would  be  expected  that  the  heavy  metals  would 
inactivate  these  catalysts.  However,  not  all  enzymes  are  equally  inac- 
tivated by  low  concentrations  of  heavy-metal  ions.  The  enzymes  which 
require  free  sulfhydryl  groups  for  activity  appear  to  be  especially  suscepti- 
ble to  inactivation  by  ions  of  heavy  metals.  It  is  probable  that  copper 
causes  fungistasis  by  combining  with  the  sulfhydryl  groups  of  certain 
enzymes.  At  this  stage,  the  action  of  copper  is  reversible.  Goldsworthy 
and  Green  (1936)  found  that  spores  of  Sderotinia  fructicola  which  had 
been  treated  with  insufficient  copper  to  kill  made  normal  growth  when 
sown  on  copper-free  medium.  As  long  as  an  inhibition  is  reversible,  the 
process  is  one  of  fungistasis.  Death  of  the  spore  results  when  irreversible 
changes  occur. 

There  is  reason  to  believe  that  the  injurious  effect  of  copper  fungicides 
upon  the  host  plant  is  due  to  the  same  mechanism  that  operates  in  fungus 
spores.  Foster  (1947)  attributed  the  sensitivity  to  copper  of  certain 
seeds  to  their  content  of  sulfhydryl  enzymes. 

mercury' 

While  a  number  of  inorganic  salts  of  mercury  have  been  used  as  anti- 
septics, only  two  have  had  wide  application  as  fungicides.  Mercuric 
chloride  (corrosive  sublimate,  bichloride  of  mercuiy,  HgCl2)  is  a  soluble, 
highly  poisonous  compound.  It  is  commonly  used  for  surface  steriliza- 
tion in  a  concentration  of  1/1,000.  Mercuric  chloride  is  occasionally 
used  as  a  special-purpose  fungicide. 

Mercurous  chloride  (calomel,  HgCl  or  Hg2Cl2)  is  essentially  insoluble 
in  water,  sufficiently  so  to  be  used  in  medicine.  Calomel  slowly  decom- 
poses into  mercury  and  mercuric  chloride.  This  decomposition  is  accel- 
erated by  sunlight,  which  may  account  for  the  successful  use  of  calomel 
to  control  dollar  spot,  brown  patch,  and  other  turf  diseases. 

The  organic  mercury  compounds  have  won  wide  acceptance  in  the 


250  PHYSIOLOGY  OF  THE  FUNGI 

treating  of  seed  to  control  the  attack  of  fungi  which  cause  damping-off 
and  of  certain  seed-borne  pathogens.  The  organic  mercurials  are  free 
from  many  of  the  objections  inherent  in  the  inorganic  compounds  of 
mercury.  In  general,  they  combine  less  avidly  with  proteins,  are  more 
selective  in  their  action,  and  are  far  less  toxic  to  animal  life.  As  used  for 
seed  protection,  they  are  commonly  diluted  with  an  inert  carrier.  Most 
if  not  all  such  organic  mercury  compounds  are  sold  under  trade  names, 
but  the  active  components  are  required  by  law  to  be  stated  on  the  label. 
Among  these  organic  mercury  compounds  are  ethylmercury  phosphate 
(Semesan  Jr.  and  New  Improved  Ceresan),  ethylmercury  chloride 
(Ceresan)  and  hydroxymercurichlorophenol  (Semesan). 

The  organic  mercury  compounds  used  as  sprays  and  for  treating  seeds 
are  in  general  related  to  mercuric  chloride  in  the  following  way: 

C2H6— Hg— CI  CI— Hg~Cl 

Ethylmercury  chloride  Mercuric  chloride 

The  ethyl  group  has  replaced  a  chlorine  atom  in  mercuric  chloride. 
The  type  formula  for  compounds  like  this  may  be  written  as  R — Hg — X, 
where  R  may  be  any  alkyl  or  aryl  (or  other)  group  and  X  represents  any 
anion,  I~,  Cl~,  0H~,  NOs",  P04~.  The  anion  greatly  modifies  the  solu- 
bility of  the  compound  in  water.  In  general,  this  type  of  organic  mer- 
cury compound  is  volatile,  and  this  property  may  be  assumed  to  aid  in 
penetration.  Other  organic  mercury  fungicides  are  derivatives  of  alkjd 
and  aryl  mercuric  hydroxides.  These  compounds  can  react  with  organic 
acids  to  form  salts.  The  relation  of  these  compounds  to  mercuric  hydrox- 
ide is  shown  below: 

CeHs— Hg— OH  HO— Hg— OH 

Phenylmercury  hydroxide  Mercuric  hydroxide 

Compounds  of  this  type  are  used  to  protect  cellulose  and  leather  products. 

Parker-Rhodes  (1942)  investigated  the  toxicity  of  the  following  mer- 
cury compounds  to  Macrosporium  sarcinaeforme  and  Botrytis  allii:  mer- 
curic acetate,  mercuric  chloride,  methylmercury  nitrate,  and  tolylmer- 
curic  nitrate.  All  these  compounds  were  toxic  to  M.  sarcinaeforme,  and 
all  except  methylmercury  nitrate  were  toxic  to  B.  allii.  Perhaps  methyl- 
mercury  nitrate  is  not  soluble  enough  in  fat  for  the  spores  of  this  fungus 
to  absorb  a  toxic  amount  of  the  compound.  Dillon- Weston  and  Booer 
(1935)  found  that  vapor  of  ethylmercury  iodide  was  toxic  to  TiUetia  spores 
in  the  laboratory  but  afforded  no  control  in  the  field. 

The  soluble  inorganic  mercury  salts  are  protein  precipitants,  and  this 
property  may  explain  in  part  their  mode  of  action  when  used  in  high 
concentrations.  These  salts  are  frequently  fungistatic  or  bacteriostatic, 
since  the  very  firmness  of  the  union  between  the  mercuric  ion  and  the  cell 
membrane  may  form  a  barrier  to  further  penetration.     The  first  action 


ACTION  OF  FUNGICIDES  251 

of  mercuric  ion  is  to  cause  stasis,  which  may  be  reversed  by  treating  the 
cells  with  reagents  which  have  a  high  affinity  for  mercury.  McCalla 
(1940)  demonstrated  that  cells  of  Escherichia  coli  which  had  been  treated 
with  mercuric  chloride  could  be  revived  by  hydrogen  sulfide.  If  stasis 
due  to  mercury  is  not  overcome  within  a  certain  time,  irreversible  changes 
occur  and  death  of  the  cells  results. 

Organic  mercury  compounds  are  not  protein  precipitants,  and  this  is 
one  of  their  advantages  as  disinfectants  and  fungicides.  Fildes  (1940) 
ascribed  the  action  of  mercury  compounds  to  combination  of  mercuric 
ion  with  the  sulfhydryl  group  of  essential  metabolites  and  enzymes. 
Others  have  shown  that  organic  mercury  compounds  act  similarly. 
According  to  this  view,  enzyme  inhibition  is  the  basis  of  the  action  of 
mercury  compounds.  Fildes  found  that  the  action  of  mercury  was 
antagonized  by  compounds  which  contained  free  sulfhydryl  groups 
(thioacetate,  cysteine,  glutathione).  Neither  cystine  ( — S — S — )  nor 
methionine  ( — S — )  was  effective  in  overcoming  mercury  toxicity. 

Organic  mercury  compounds  appear  to  act  by  the  same  mechanism  as 
the  mercuric  ion.  p-Chloromercuribenzoate  was  found  to  inhibit  the 
action  of  various  sulfhydryl  enzymes  which  take  part  in  carbohydrate 
metabolism,  e.g.  succinic  acid  oxidase,  yeast  carboxylase,  malate  oxidase, 
and  ketoglutarate  oxidase.  This  organic  mercury  compound  also  inhib- 
ited the  action  of  c?-amino  acid  oxidase,  transaminase,  Z-glutamate  oxidase, 
and  other  enzymes  (Barron  and  Singer,  1945;  Singer  and  Barron,  1945). 
In  many  instances  the  inhibitory  action  of  p-chloromercuribenzoate  on 
these  enzymes  could  be  reversed  by  glutathione,  cysteine,  or  hydrogen 
sulfide. 

Cook  et  at.  (1946)  found  phenylmercuric  nitrate  to  depress  the  respira- 
tion of  Saccharomyces  cerevisiae.  This  depression  in  rate  of  respiration 
was  overcome  by  various  compounds  having  a  free  — SH  group;  e.g., 
cysteine  and  homocysteine,  while  cystine  and  methionine  were  without 
effect.  The  work  of  these  investigators  and  of  others  makes  it  highly 
probable  that  mercury  compounds  are  toxic  because  they  inactivate 
certain  essential  enzyme  systems.  The  enzyme  inhibitions  discussed 
above  are  examples  of  noncompetitive  inhibition.  These  inhibitions  are 
reversible,  as  in  the  case  of  competitive  inhibition,  but  the  reversing 
agents  are  nonspecific,  or  not  limited  to  a  single  metabolite. 

SULFUR 

Of  the  nonmetallic  elements,  sulfur  and  certain  of  its  compounds  are 
widely  used  as  protective  and  eradicant  fungicides.  The  toxicity  of  the 
nonmetallic  elements  is  dependent  upon  the  state  of  oxidation.  In  many 
instances,  compounds  in  the  higher  states  of  oxidation  are  the  least  toxic. 
For  example,  sulfur  in  the  form  of  the  free  element  (S)  and  of  sulfide 


252 


PHYSIOLOGY  OF  THE  FUNGI 


(S=)  forms  excellent  fungicides,  while  sulfites  (SOa^)  are  only  slightly- 
toxic  and  sulfates  (SO 4=)  are  nontoxic. 

According  to  Large  (1940),  elemental  sulfur  has  been  used  to  control 
powdery  mildew  for  slightly  over  100  years.  The  effectiveness  of  sulfur 
increases  as  the  particle  size  diminishes.  Finely  divided  sulfur  adheres 
to  plant  surfaces  much  better  than  larger  particles.  In  addition,  the 
distance  between  particles  tends  to  be  decreased  when  fine  particles  are 
used,  and  the  infection  court  is  thereby  better  protected.  The  odds  are 
increased  that  a  fungus  spore  falling  upon  a  treated  leaf  will  be  within  the 
range  of  action  of  a  particle  of  sulfur.  An  example  of  the  effect  of  particle 
size  of  sulfur  on  toxicity  is  given  in  Table  44.     The  greater  toxicity  of 

Table  44.     The  Relation  between  the  Particle  Size  and  Toxicity  of  a  Sulfur 
Dust  to  the  Conidia  of  Sclerotinia  americana 
(Wilcoxon  and  McCallan,  Phytopathology  20,  1930.) 


Treatment 

Mean  diameter  of 
sulfur  particles,  tx 

Germination,  % 

Control     

285 

142 

60 

33 

97.6 

Ground  roU  sulfur 

62.8 

Ground  roll  sulfur 

Ground  roll  sulfur 

Ground  roll  sulfur 

47.2 
29.1 
20.7 

the  finely  divided  sulfur  is  due  to  the  fact  that  sulfur  enters  the  spore  in 
the  form  of  vapor.  The  amount  of  vapor  formed  from  a  given  amount  of 
sulfur  in  a  given  time  depends  upon  the  area  of  the  exposed  surface,  as 
well  as  upon  temperature.  Therefore,  the  fineness  of  the  sulfur  particles 
governs  the  effective  concentration  of  sulfur  vapor  and  its  effectiveness 
as  a  fungicide. 

McCallan  (1946)  estimated  the  yearly  consumption  of  sulfur  in  the 
United  States  alone  to  be  142  million  pounds.  Of  this  amount  110 
million  pounds  is  used  as  sulfur  dust,  5  million  pounds  as  wettable  sulfur, 
and  27  million  pounds  as  lime-sulfur.  Approximately  62  per  cent  of 
this  is  used  primarily  for  the  control  of  apple  scab  alone. 

Since  elemental  sulfur  is  insoluble,  its  action  upon  fungi  cannot  be 
attributed  to  the  sulfur  in  this  form.  Two  general  theories  have  been 
proposed  to  explain  the  action  of  sulfur.  One  theory  holds  that  the 
action  is  due  to  oxidized  sulfur,  such  as  SO2  or  SO3  (which  form  sulfurous 
and  sulfuric  acid,  respectively,  with  water)  or  pentathionic  acid,  H2S5O6. 
According  to  the  second  theory,  the  reduced  form  of  sulfur,  H2S,  is  the 
active  toxic  agent.  Both  these  theories  are  supported  by  published 
experimental  evidence.  All  these  compounds  are  toxic  to  fungi  under 
certain    conditions,    if    in    high    enough    concentrations.     However,    to 


ACTION  OF  FUNGICIDES  253 

account  satisfactorily  for  the  toxic  properties  of  sulfur,  it  must  be  demon- 
strated that  the  toxic  agent  is  produced  under  the  conditions  which 
prevail  in  the  field  in  quantities  sufficient  to  account  for  the  observed 
effects.  Any  evaluation  of  these  hypotheses  must  take  into  account  all 
the  variables  involved.  The  caution  of  Wilcoxon  and  McCallan  (1930) 
is  pertinent: 

In  making  comparisons  of  the  toxicity  of  chemical  substances  to  fungus  spores, 
there  are  two  requisites  for  obtaining  accurate  results  which,  though  quite 
obvious,  have  not  always  received  the  consideration  they  deserve,  (a)  The 
substance  whose  toxicity  is  to  be  measured  must  be  available  in  a  pure  state 
and  of  known  concentration,  and  (6)  the  technique  employed  must  be  capable  of 
distinguishing  between  the  toxicity  of  the  substances  it  is  desired  to  compare. 

It  is  agreed  that  elemental  sulfur  is  not  the  toxic  agent  and  that  sulfur  is 
transformed  into  the  toxic  agent.  There  are  three  possible  agencies  for 
such  transformations:  the  atmosphere,  the  plant  on  which  the  sulfur  is 
dusted,  and  the  fungus  spores  or  mycelium.  Sulfur  acts  at  a  distance, 
and  since  sulfur  is  volatile  at  room  temperature,  this  property  offers  an 
explanation.  Sulfur  vapor  is  a  gas,  and  in  this  state  it  should  be  more 
easily  transformed  into  the  toxicant. 

Sulfur  is  slowly  oxidized  by  the  oxygen  of  the  atmosphere  to  form  sulfur 
dioxide,  but  the  rate  at  which  this  reaction  occurs  at  ordinary  tempera- 
tures makes  it  impossible  for  this  reaction  to  account  for  all  the  toxic 
properties  of  sulfur,  even  though  sulfur  dioxide  is  toxic  to  fungus  spores 
(McCallan  and  Weedon,  1940). 

Young  (1922)  set  forth  the  hypothesis  that  pentathionic  acid  is  the 
toxic  agent  formed  from  sulfur.  It  is  agreed,  even  by  those  who  do  not 
support  Young's  hypothesis,  that  this  acid  is  formed  on  the  surface  of 
sulfur  dust.  A  considerable  number  of  papers  were  published  during 
the  next  decade  which  gave  support  to  this  view  (Liming,  1932).  Wil- 
coxon and  McCallan  (1930)  investigated  this  theory  thoroughly  and 
concluded  that  pure  pentathionic  acid  had  no  toxic  properties  for  the 
spores  of  Sderotinia  americana,  Botrytis  sp.,  Macrosporium  sarcinaeforme, 
and  Uromijces  caryopJujUinus.  If  sufficient  pentathionic  acid  was  used 
to  reduce  the  pH  to  about  4,  spore  germination  was  inhibited.  Solutions 
of  sulfuric  acid  having  the  same  pH  were  equally  toxic.  Neutral  salts 
of  both  acids  were  nontoxic.  Roach  and  Glynne  (1928)  likewise  found 
pentathionic  and  sulfuric  acids  to  have  the  same  toxicity  when  tested 
against  the  winter  sporangia  of  Synchytrium  endohioticum.  Wilcoxon 
and  McCallan  (1930)  performed  a  decisive  experiment  when  they  washed 
one  lot  of  sulfur  dust  with  alkali  to  remove  pentathionic  acid  and  com- 
pared this  pentathionate-free  dust  with  the  original  sample,  which  con- 
tained a  trace  of  this  acid.  No  difference  in  toxicity  of  the  washed  and 
control  samples  of  this  sulfur  dust  was  found. 


254 


PHYSIOLOGY  OF  THE  FUNGI 


There  is  now  general  agreement  that  hydrogen  sulfide  is  the  common 
toxic  compound  produced  from  sulfur.  Not  only  is  hydrogen  sulfide 
toxic  to  fungus  spores,  but  the  mechanism  for  its  production  is  also  pres- 
ent. It  is  known  that  hydrogen  sulfide  is  produced  from  sulfur  both  by 
the  treated  plant  and  by  the  fungus  spores, 

McCallan  and  Wilcoxon  (1931)  made  qualitative  tests  for  the  ability 
of  the  spores  of  17  species  of  fungi  to  produce  hydrogen  sulfide  from  sulfur. 

All  produced  this  substance,  but  in 
varying  amounts  and  at  varying 
rates.  They  showed  that  the  spores 
need  not  be  in  direct  contact  with 
solid  sulfur  to  produce  hydrogen 
sulfide.  Figure  50  illustrates  the 
method  used  by  these  investigators 
to  demonstrate  this  phenomenon. 

These  authors  investigated  the 
toxicity  of  hydrogen  sulfide  to  the 
spores  of  eight  species  of  fungi. 
These  experiments  were  performed 
in  a  flowing  stream  of  air  which  con- 
tained known  amounts  of  hydrogen 
sulfide,  and  the  concentration  in  the 
water  droplet  in  which  the  spores 
were  suspended  was  calculated  from 
Henry's  law.  These  precautions  are 
necessary  because  hydrogen  sulfide  is 
unstable.     Neglect   of  this  fact  by 


LEAD    ACETATE 
PAPER       WHITE 

LEAD    ACETATE 
PAPER    BLACKENED 

COLLODION    SAC 
SPORE  SUSPENSION 
SULPHUR    PASTE 


Fig.  50.  The  production  of  hydrogen 
sulfide  by  Sclerotinia  spores  separated 
from  sulfur  by  a  collodion  membrane. 
Note  that  the  production  of  hydrogen 
sulfide  takes  place  on  the  spore  side  of 
the  membrane  and  not  on  the  sulfur 
side.  (Courtesy  of  McCallan  and 
WUcoxon,  Contribs.  Boyce  Thompson 
Inst.  3:  26,  1931.) 


earlier  investigators  led  to  an  underestimation  of  the  toxicity  of  hydrogen 
sulfide.  These  results  of  McCallan  and  Wilcoxon  are  presented  in  Fig. 
51.  From  these  curves  it  is  seen  that  spores  of  Venturia  inaequalis, 
Uromyces  caryophyllinus,  and  Puccinia  antirrhini  are  inhibited  by  very 
low  concentrations  of  hydrogen  sulfide,  while  the  spores  of  Botrytis  sp. 
and  Glomerella  cingulata  are  scarcely  affected  by  ten  times  as  much 
hydrogen  sulfide.  By  increasing  the  hydrogen  sulfide  concentration  to 
60  p. p.m.,  complete  inhibition  of  germination  of  the  spores  of  these  two 
species  was  obtained.  The  spores  of  these  eight  fungi  were  shown  to 
produce  varying  amounts  of  hydrogen  sulfide  per  unit  weight  of  spores. 
Whether  hydrogen  sulfide  produced  by  spores  would  prove  toxic  would 
therefore  depend  upon  the  ability  of  the  particular  spores  to  produce 
hydrogen  sulfide  and  the  sensitivity  of  the  spores  to  this  substance.  The 
correlation  is  shown  in  Table  45. 

The  actions  of  sulfur  and  hydrogen  sulfide  are  parallel,  and  it  may  be 
concluded  that  sulfur  is  toxic  to  the  spores  of  certain  species  by  virtue 


ACTION  OF  FUNGICIDES 


255 


of  absorption  of  sulfur  vapor  and  its  reduction  to  hydrogen  sulfide  within 
the  spore.  Thus,  the  spores  of  susceptible  species  destroy  themselves. 
It  is  not  thought  that  the  hydrogen  sulfide  evolved  from  leaves  or  other 
spores  is  absorbed  in  lethal  quantities  under  natural  conditions. 


0.6  0.8  1.0  1.2  1.4  1.6  1.8  2.0 

Concentration  -milligrams  per  liter 

Fig.  51.  Toxicity  of  hydrogen  sulfide  to  urediospores  of  Uromyces  caryophyllinus 
and  Puccinia  antirrhini  and  to  conidia  of  Venturia  inaequalis,  Sclerotinia  americana, 
Macrosporiuni  sarcinaeforme,  Pestalotia  stellata,  Glonierella  cingulata,  and  Botrytis  sp. 
(Courtesy  of  McCallan  and  Wilcoxon,  Contribs.  Boyce  Thompson  Inst.  3:  31,  1931.) 

Liquid  lime-sulfur  is  a  common  spray  material  and  is  prepared  by 
boiling  sulfur  and  calcium  hydroxide  together.  The  chief  active  ingre- 
dient is  calcium  polysulfide.  After  deposition  on  leaves  the  calcium 
polysulfide  is  quickly  decomposed,  yielding  sulfur  and  calcium  sulfide, 

Table  45.     Comparison  between  the  Toxicity  and  the  Production  of  Hydrogen 
Sulfide,  Expressed  in  Units  Equal  to  the  Amounts  of  Hydrogen  Sulfide 
Required  to  Reduce  Germination  50  Per  Cent 
(McCallan  and  Wilcoxon,  Contribs.  Boyce  Thompson  Inst.  3,  1931.) 


Species 

Mg.  H2S 
required  to 
reduce  ger- 
mination of 

1,000,000 
spores  50% 

Mg.  H2S 

produced  by 

1,000,000 

spores  in 

12  hr. 

Production  of  H2S 

expressed  in  units 

equal  to  the  amount 

of  H2S  required  to 

reduce  germination 

50% 

Venturia  inaeaualis 

0.001 
0.002 
0.006 
0.013 
0.043 
0.049 
0.532 
0.665 

0.002 

0.019 

0.13 

0.039 

0.013 

0.001 

0.027 

0.002 

2.0 

Uromyces  caryophyllinus 

Puccinia  antirrhini 

9.5 
2.2 

Sclerotinia  americana 

Macrosporiuni  sarcinaeforme 

Pestalotia  stellata 

Glonierella  cingulata 

3.0 
0.30 
0.02 
0.05 

Botrytis  sp 

0.003 

256  PHYSIOLOGY  OF  THE  FUNGI 

which  in  turn  may  decompose  by  hydrolysis  to  yield  hydrogen  sulfide 
and  calcium  hydroxide.  At  the  same  time  some  of  the  calcium  poly- 
sulfide  is  oxidized  to  calcium  thiosulfate  and  sulfur.  See  Frear  (1948) 
for  a  discussion  of  the  chemistry  involved. 

It  is  known  that  lime-sulfur  exerts  an  eradicant  action  on  some  fungi, 
including  Venturia  inaequalis,  when  first  applied.  After  a  few  days 
this  spray  exerts  only  a  protective  action  like  that  of  elemental  sulfur, 
which  probably  depends  on  the  elemental  sulfur  set  free  by  the  decom- 
position of  various  constituents  comprising  lime-sulfur.  The  eradicant 
action,  then,  depends  upon  either  the  calcium  polysulfide  or  calcium 
sulfide.  We  may  consider  that  sulfide  ion  (S=)  is  the  toxic  agent.  The 
alkalinity  of  the  spray  may  aid  in  penetration  into  the  mycelium  already 
present. 

Lime-sulfur  solution  may  be  treated  with  ferrous  sulfate  or  aluminum 
sulfate  in  the  spray  tank  to  produce  colloidal  sulfur  and  hydrogen  sulfide. 
Aluminum  sulfate,  AI2 (804)3,  hydrolyzes  to  form  aluminum  hydroxide 
and  sulfuric  acid.  A  lime-sulfur  spray  so  treated  has  only  a  protective 
action.  It  has  lost  its  eradicant  value.  We  may  assume,  therefore,  that 
the  decomposition  of  lime-sulfur  in  spray  tanks  when  treated  with  acid 
(aluminum  sulfate)  and  the  decomposition  on  the  leaf  follow  a  somewhat 
similar  pattern.  This  scheme  of  producing  colloidal  sulfur  has  the  draw- 
back that  the  added  iron,  when  ferrous  sulfate  is  used,  is  toxic  to  vegeta- 
tion, and  dangerous  amounts  of  hydrogen  sulfide  are  evolved. 

We  may  assume  that  hydrogen  sulfide  exerts  its  toxic  action  on  fungus 
spores  by  inactivating  certain  enzymes.  Hydrogen  sulfide  is  known  to 
inactivate  many  enzymes,  including  catalase,  cytochrome  oxidase,  dopa 
oxidase,  lactase,  and  others.  Generally,  hydrogen  sulfide  and  cyanide 
inhibit  the  same  enzymes.  It  is  thought  that  these  metalloenzymes  are 
inhibited  by  sulfide  or  cyanide  because  these  agents  react  with  iron  or 
copper  to  form  highly  insoluble  or  little-ionized  compounds  or  complexes. 

ORGANIC  FUNGICIDES 

The  newer  fungicides,  with  few  exceptions,  are  either  organic  or  organo- 
metallic  compounds.  The  organic  mercury  compounds  were  considered 
with  the  inorganic  compounds  of  mercury,  since  the  mechanism  of  action 
appears  to  be  the  same  in  both  types  of  compounds.  Many  of  the 
organic  fungicides  exhibit  greater  specificity  than  the  inorganic  fungicides. 
The  possibilities  of  modification  in  the  structure  of  organic  compounds 
are  almost  unlimited.  The  study  of  organic  fungicides,  therefore,  offers 
the  opportunity  of  correlating  structure  with  type  and  intensity  of  fungi- 
cidal action. 

Aldehydes.  The  first  organic  fungicide  to  attain  wide  acceptance  was 
formaldehyde.  At  one  time  this  compound  was  used  for  the  surface 
sterilization  of  grain  and  potato  tubers,  but  at  present  formaldehyde  is 


ACTION  OF  FUNGICIDES 


257 


little  used.  Formaldehyde  reacts  with  free  amino  groups,  and  it  is 
probable  that  its  fungicidal  action  depends  upon  this  property.  Some 
other  aldehydes  also  have  fungicidal  properties  (Uppal,  1926). 

Quinones.  While  there  are  two  series  of  quinones  (ortho,  or  1,2,  and 
para,  or  1,4),  we  shall  consider  only  the  1,4-quinones  as  fungicides.  The 
simplest  quinone  is  p-benzoquinone.  Quinones  are  cyclic  compounds 
which  possess  a  characteristic  pair  of  double  bonds.  Such  a  configuration 
of  double  bonds  is  called  quinoid  and  is  possessed  by  many  dyes,  some 
of  which  are  fungicides.  If  a  considerable  series  of  toxic  compounds 
possess  a  common  functional  group  or  groups,  it  may  be  assumed  that 
these  groups  are  involved  in  fungicidal  activity.  According  to  Horsfall 
(1945),  1,4-benzoquinone  has  a  slight  toxicity  to  fungi.  The  four  hydro- 
gens in  1,4-benzoquinone  can  be  replaced  by  chlorine  to  form  chloranil 
(Spergon),  which  greatly  increases  the  fungicidal  properties.  The  struc- 
tural formulas  for  these  compounds  are  given  below: 

O  O 


— H 


-H 


CI— 


Cl- 


— CI 


-CI 


o  o 

1,4-Benzoquinone  Chloranil  (Spergon) 

Spergon  has  been  used  as  a  seed  protectant. 

Substituted  naphthoquinones  are  more  important  fungicides  than, 
the  benzoquinones.  Among  these,  2,3-dichloro-l,4-naphthoquinone 
(Phygon)  is  reported  to  be  five  to  eight  times  as  effective  as  Spergon  (Ter 
Horst  and  Felix,  1943).  Some  of  the  naphthoquinones  synthesized  by 
plants  are  fungicides.  Juglone,  5-hydroxy-l,4-naphthoquinone,  is  found 
in  walnut  hulls  and  is  secreted  by  walnut  roots.  The  isomeric  2-hydroxy- 
1,4-naphthoquinone  (lawsone)  is  found  in  henna  leaves.  Juglone  is 
reported  to  be  as  toxic  to  fungus  spores  as  Bordeaux  mixture.  Juglone 
controls  black  spot  of  roses  as  well  as  sulfur  does  (Gries,  1943,  1943a). 
It  is  also  toxic  to  many  plants.  Little  et  al.  (1948)  isolated  2-methoxy- 
1,4-naphthoquinone  from  Impatiens  balsamina.  This  compound  was  an 
active  fungicide  which  exhibited  no  phytotoxicity  toward  tomato  and 
bean  plants.  The  formulas  of  two  naphthoquinone  fungicides  are  given 
below: 

O  O 


-OCH3 


o 

2-Methoxy-l,4-naphthoquinone 


OH   O 

Juglone 


258 


PHYSIOLOGY  OF  THE  FUNGI 


The  fungicidal  action  of  substituted  quinones  may  be  due  in  part  to 
their  property  of  reacting  with  free  amino  groups  of  proteins  (Theis, 
1945).  Substituted  naphthoquinones  as  antagonists  of  vitamin  K  were 
discussed  in  Chap.  11.  Most  of  the  available  evidence  indicates  that 
the  principal  mechanism  of  quinone  toxicity  lies  in  its  noncompetitive 
inhibition  of  sulfhydiyl  enzymes. 

It  has  been  suggested  that  the  mechanism  of  inhibition  is  dependent 
upon  the  structure  of  the  substituted  naphthoquinones.  Colwell  and 
McCall  (1946)  found  the  fungistatic  and  fungicidal  concentrations  of 
2-methyl-l,4-naphthoquinones  to  be  the  same  when  Aspergillus  niger 
and  an  unidentified  fungus  were  used  as  test  organisms.  Addition  of 
sodium  thioglycolate  or  cysteine  antagonized  the  toxic  action  of  this 
naphthoquinone.  These  authors  postulate  that  only  naphthoquinones 
unsubstituted  in  position  3  react  wdth  sulfhydryl  groups,  for  2-methyl-3- 
methoxy-l,4-naphthoquinone  was  not  antagonized  by  thioglycolate  or 
cysteine.  The  reaction  between  certain  naphthoquinones  and  sulfhydiyl- 
containing  compounds  can  be  demonstrated  in  vitro. 

The  amounts  of  various  substituted  1,4-naphthoquinones  required  to 
cause  a  50  per  cent  inhibition  of  isolated  yeast  carboxylase  and  similar 
reduction  in  the  germination  of  Monilinia  fructicola  spores  w^ere  roughly 
parallel  (Foote  et  al.,  1949).  Carboxylase  is  a  sulfhydryl  enzyme.  It  is 
probable  that  other  sulfhydryl  enzymes  are  also  inhibited  by  naphtho- 
quinones. For  further  information  on  the  mechanism  of  quinone  inhibi- 
tion, see  Geiger  (1946). 

Dyes.  Various  dyes  are  fungistatic  compounds.  Malachite  green 
and  crystal  violet  are  used  to  control  various  fungus  infections  of  the 
skin.     Both  these  dyes  have  a  benzoquinoid  structure,  as  is  shown  below: 


^        ^ 


-N(CH3)2 


(CH3)2N=' 


/ 


\/ 


^ 


CI 


Malachite  green 

Leonian  (1930)  made  a  study  of  the  toxicity  of  malachite  green  to  many 
species  and  strains  of  Phytophthora  and  found  only  three  species  {P. 
hydrophila,  P.  melongenae,  and  P.  sp.)  able  to  grow  in  the  presence  of 
1  p. p.m.  of  malachite  green.  Other  species  were  more  sensitive  to  this 
dye.  P.  colocasiae  and  P.  richardiae  failed  to  grow  in  nutrient  solutions 
containing  1  part  of  malachite  green  in  16  million  parts  of  medium. 
Leonian  (1932)  investigated  the  growth-inhibiting  properties  of  malachite 


ACTION  OF  FUNGICIDES 


259 


(CH3)2N 


=C 


N(CH3)2 


N(CH3)2 


Crystal  violet 


CI 


green  and  crystal  violet  upon  26  species  and  isolates  of  Trichophyton. 
Malachite  green  proved  greatly  superior  to  crystal  violet.  Over  half 
the  isolates  tested  failed  to  grow  in  the  presence  of  1  part  of  malachite 
green  to  50,000  parts  of  medium,  and  many  failed  to  grow  in  the  presence 
of  1  p.p.m.  of  this  dye.  Crystal  violet  allowed  some  growth  in  all  isolates 
tested  at  a  concentration  of  1  part  in  50,000  parts  of  medium.  Placing 
the  inoculum  in  direct  contact  with  the  medium  containing  the  dye  was 
more  lethal  than  placing  the  agar  inoculum  plug  with  the  mycelium  upon 
the  surface  of  the  test  medium.  Some  other  dyes  such  as  methylene 
blue  are  also  toxic  to  fungi.  Both  malachite  green  and  methylene  blue 
inhibit  carboxylase  (Horsfall,  1945). 

Dithiocarbamates  and  related  compounds.  Barratt  and  Horsfall 
(1947)  have  reported  extensive  investigations  on  the  homologues  and 
analogues  of  disodium  ethylenebisdithiocarbamate  (Nabam).  In  gen- 
eral, these  compounds  are  formed  when  primary  and  secondary  amines 
react  with  carbon  disulfide.     The  formula  for  Nabam  is  given  below: 


S     H 


H    S 


Na— S— C— N— CH2— CH2— N— C— S— Na 
Disodium  ethylenebisdithiocarbamate  (Nabam) 

The  zinc  (Ziram)  and  ferric  (Ferbam)  salts  of  dimethyldithiocarbamate 
are  effective  fungicides  for  the  control  of  certain  fungus  pathogens.  The 
formula  for  dimethyldithiocarbamate  is  given  below: 

S 

II 
(CH3)2— N— C— SH 
Dimethyldithiocarbamate 

The  oxidation  product  of  dimethyldithiocarbamate  is  tetramethylthluram 
disulfide  (Thiram),  which  has  some  value  as  a  seed  protectant. 

The  dithiocarbamate  fungicides,  such  as  Nabam,  yield  hydrogen  sulfide 
on  hydrolysis.  This  reaction  takes  place  spontaneously  in  the  presence 
of  moisture.  The  mechanism  of  hydrogen  sulfide  toxicity  has  already 
been  discussed.  The  second  mechanism  which  has  been  proposed 
involves  the  formation  of  insoluble  mercaptides  of  certain  essential  metals. 


2G0  PHYSIOLOGY  OF  THE  FUNGI 

In  addition,  Nabam  on  decomposition  yields  an  unidentified  toxic  gaseous 
compound,  which  is  neither  hydrogen  sulfide  nor  sulfur  dioxide  (Rich 
and  Horsfall,  1950). 

Specific  organic  reagents  for  metals.  The  essential  nature  of  certain 
micro  elements  for  fungus  growth  and  the  role  of  these  elements  in 
enzymes  were  discussed  in  Chaps.  4  and  5.  The  chemistry  of  these 
specific  organic  reagents  is  treated  by  Yoe  and  Sarver  (1941).  These 
reagents  form  insoluble  or  slightly  ionized  compounds  with  metals. 
Zentmeyer  (1944)  tested  various  organic  analytical  reagents  and  found 
8-hydroxyquinoline  (Oxine)  and  ammonium  nitrosophenylhydroxylamine 
(Cupferron)  to  be  fungistatic.  8-Hydroxyciuinoline  inhibited  the  growth 
of  Fusarium  oxysporum  var.  lycopersici,  Ceratostomella  ulmi,  and  a  species 
of  Penicillium.  The  effectiveness  of  8-hydroxyquinoline  in  forming 
chelate  salts  increases  as  the  pH  values  increase.  Below  pH  3.5  complex 
formation  does  not  take  place  with  zinc,  copper,  iron,  and  manganese. 
Zinc  ion  reacts  with  8-hydroxyquinoline  as  shown  below: 


+  H+ 


I  I     r 

OH  O— Zn 

8-Hydroxyquinoline         Zinc  complex  of  8-hydroxyquinoline 

The  fungistatic  effect  of  8-hydroxyquinoline  on  Fusarium  oxysporum 
var.  lycopersici  and  Ceratostomella  ulmi  was  overcome  by  increasing  the 
zinc  content  of  the  medium.  In  the  presence  of  8-hydroxyquinoline 
there  was  competition  between  this  compound  and  one  or  more  enzyme 
systems  for  the  zinc  present  in  the  medium.  Whether  or  not  an  organic 
compound  such  as  8-hydroxyquinoline  will  act  as  a  fungistatic  agent 
depends  upon  the  concentration  of  the  reagent,  the  amount  of  fungus 
mycelium,  and  the  concentration  of  the  metallic  ion  for  which  the  two 
sj^stems  compete.  One  would  expect  that  such  fungicides,  in  common 
with  all  others,  would  be  more  effective  when  the  mass  of  the  fungus  is 
small. 

Other  organic  fungicides.  Many  other  types  of  organic  compounds 
are  fungicides,  and  an  intensive  search  for  new  ones  is  in  progress.  Brief 
mention  of  some  of  these  developments  is  made  below.  Geiger  (1948) 
reports  various  unsaturated  ketones  to  be  active  against  Aspergillus 
niger,  Trichoderma  koningii,  Cryptococcus  neoformans,  and  Trichophyton 
mentagrophytes.  The  mode  of  action  resembles  that  of  the  naphtho- 
quinones in  that  sulfhydryl  enzymes,  including  succinic  acid  dehydro- 
genase, triose  phosphate  dehydrogenase,  and  urease,  are  inhibited.  The 
fungistatic   activity  of  ethylenic   and   acetylenic   compounds  has  been 


ACTION  OF  FUNGICIDES  261 

tested  on  Fusarium  graminearum,  Penicillium  digitatum,  and  Botrytis 
aim  (McGowan  et  al.,  1948).  The  fungicidal  action  of  substituted 
pyrazoles  was  tested  on  spores  of  Alter naria  oleracea  and  Sclerotinia 
americana  in  the  laboratory,  and  for  the  control  of  apple  scab,  cedar-apple 
rust,  and  late  blight  of  potato  and  tomato.  Some  of  these  compounds 
show  promise,  although  the  mechanism  of  action  is  not  known  (McNew 
and  Sandholm,  (1949).  For  a  survey  of  the  newer  fungicides  see  Well- 
man  (1948). 

EVALUATING  FUNGICIDES 

The  preliminary  tests  of  fungicidal  activity  are  made  in  the  laboratory 
in  order  to  eliminate  inactive  compounds  or  to  compare  the  activities 
of  different  compounds  under  identical  conditions.  Evaluation  in  the 
greenhouse  and  field  is  the  final  test  of  a  new  fungicide.  This  discus- 
sion will  be  limited  to  a  general  consideration  of  laboratory  testing  of 
fungicides. 

Fungus  spores  rather  than  mycelium  are  used  in  most  laboratory  tests 
because  it  is  the  function  of  a  protectant  fungicide  to  kill  or  inhibit  spore 
germination.  Three  basic  types  of  procedures  may  be  used  in  laboratory 
tests  (McCallan,  1947):  (1)  Spores  are  suspended  in  solutions  or  suspen- 
sions of  the  fungicide  under  test,  and  the  inhibition  of  germination  is 
noted  as  a  function  of  time  of  exposure  and  concentration  of  the  fungicide. 
This  is  a  modification  of  the  Rideal- Walker  method  of  evaluating  anti- 
septics. (2)  The  compound  to  be  tested  is  incorporated  in  a  suitable 
solid  or  liquid  medium,  which  is  then  inoculated  with  spores  of  the  test 
fungi.  The  amount  of  inhibition  of  germination  or  growth  is  determined. 
(3)  Glass  slides  are  covered  uniformly  with  the  fungicide,  and  after  dry- 
ing, the  spores  are  sown  on  the  treated  slides.  The  inoculated  plates 
are  then  placed  in  constant-humidity  chambers  and  the  percentage  of 
germination  determined  after  20  to  24  hr. ;  or  the  effectiveness  of  a  fungi- 
cide may  be  studied  as  a  function  of  time  of  exposure. 

The  second  and  third  methods  appear  to  be  the  most  useful.  Fleury 
(1948)  studied  the  fungistatic  action  of  thiourea  on  Aspergillus  niger 
by  adding  this  substance  to  a  liquid  basal  medium.  Thiourea  was  a 
much  more  potent  inhibitor  w^hen  nitrate  nitrogen  was  used  than  when 
ammonium  or  organic  nitrogen  was  present  in  the  medium.  Agar 
medium  has  been  used  by  Leben  and  Keitt  (1949)  to  assay  the  amount  of 
toxicant  on  leaf  surfaces.  A  suspension  of  spores  of  Glomerella  cingulata 
was  prepared  in  warm  (38  to  40°C.)  agar  medium.  Five  milliliters  of 
this  seeded  medium  was  added  to  Petri  dishes  which  contained  15  ml.  of 
solidified  agar  medium.  After  the  seeded  agar  had  solidified,  leaf  disks 
of  uniform  size  were  cut  from  sprayed  leaves  and  placed  on  the  agar.  The 
amount  of  toxicant  present  on  the  leaf  surface  was  determined  by  measur- 


262  PHYSIOLOGY  OF  THE  FUNGI 

ing  the  diameter  of  the  zone  of  inhibition.  Disks  of  blotting  paper  to 
which  fungicides  have  been  added  may  be  used  to  determine  their  potency. 
Thornberry  (1950)  has  suggested  the  use  of  filter-paper  disks  for  the 
evaluation  of  fungicides  and  bactericides.  Filter-paper  disks  appear 
to  be  more  suitable  than  blotting  paper.  In  this  method  seven  filter- 
paper  disks  are  uniformly  spaced  on  a  Petri  dish,  and  0.09  ml.  of  the 
toxicant  in  aqueous  solution  is  added  per  disk.  The  zone  of  inhibition 
is  a  measure  of  the  effectiveness  of  the  fungicide. 

The  glass-slide  method  appears  to  simulate  more  closely  the  conditions 
under  which  the  spores  of  plant  pathogenic  fungi  germinate  in  nature. 
The  Committee  on  the  Standardization  of  Fungicidal  Tests  of  the  Ameri- 
can Phytopathological  Society  has  considered  this  method  important 
enough  to  publish  a  detailed  and  documented  summary  (1943),  to  which 
the  student  is  referred  for  further  information  and  references.  This 
committee  recommended  the  use  of  spores  of  the  following  species  for 
this  test:  Alternaria  solani,  Glomerella  cingulata,  Macrosporium  sarcinae- 
forme,  Sclerotinia  fructicola,  Penicillium  expansum,  and  Rhizopus  nigri- 
cans. For  accurate  work,  at  least  two  of  these  test  fungi  should  be  used. 
The  effectiveness  of  a  fungicide  is  determined  by  calculating  the  percent- 
age of  inhibition  of  spore  germination.  The  methods  of  evaluating  data 
obtained  in  fungicide  tests  are  discussed  by  Horsfall  (1945). 

SUMMARY 

A  fungicide  is  an  agent  capable  of  killing  some  fungi.  Fungicides  may 
be  either  water-soluble  or  nearly  insoluble.  The  action  of  fungicides 
of  the  first  class  is  immediate;  that  of  the  second  class  is  delayed.  Eradi- 
cant  fungicides  are  of  the  first  class,  w^hile  protective  fungicides  are  of 
the  second.  Fungistasis  is  the  complete  or  partial  inhibition  of  one  or 
more  life  processes  of  a  fungus.  This  inhibition  is  reversible.  The  same 
chemical  compound  may  cause  fungistasis  or  may  be  a  fungicide,  depend- 
ing upon  the  concentration  and  time  of  exposure.  The  same  substance 
may  be  a  fungicide  for  one  species,  cause  fungistasis  of  a  second,  and  be 
without   effect  upon   a  third.     Fungistasis  precedes  fungicidal   action. 

Before  a  fungicide  can  act  upon  a  fungus,  the  toxicant  must  get  into 
the  fungus  cells,  or  at  least  reach  the  protoplasmic  membrane.  While 
other  factors  undoubtedly  enter  into  the  mechanism  of  fungicidal  action, 
the  principal  point  of  attack  appears  to  be  enzyme  systems.  The  heavy- 
metal  fungicides  appear  to  act  by  inhibiting  various  sulfhydryl  enzymes. 
Fungus  spores  transform  sulfur  into  hydrogen  sulfide,  which  inhibits  the 
metalloenzymes.  Organic  fungicides,  so  far  as  is  known,  are  also  enzyme 
inhibitors. 

In  the  past,  fungicides  containing  copper,  mercury,  and  sulfur  have 
been  the  most  useful.     Recently,  organic  fungicides  have  become  impor- 


ACTION  OF  FUNGICIDES  263 

tant  and  promise  to  be  used  even  more  extensively  in  the  future.  Organic 
fungicides  are  generally  more  specific  than  inorganic  fungicides.  Satis- 
factory fungicides  for  the  control  of  certain  diseases  are  still  undiscovered. 

REFERENCES 

Barker,  B.  T.  B.,  and  C.  T.  Gimingham:  The  fungicidal  action  of  bordeaux  mix- 
tures, Jour.  Agr.  Set.  4:  76-94,  1911. 
Barratt,  R.  W.,  and  J.  G.  Horsfall:  Fungicidal  action  of  metallic  alkyl  bisdi- 

thiocarbamates,  Conn.  Agr.  Expt.  Sta.  Bull.  508,  1947. 
Barron,  E.  S.  G.,  and  T.  P.  Singer:  Studies  on  biological  oxidations.     XIX.  SuLf- 

hydryl  enzymes  in  carbohydrate  metabolism,  Jour.  Biol.  Chem.  157 :  221-240, 

1945. 
CoLWELL,  C.  A.,  and  M.  McCall:  The  mechanism  of  bacterial  and  fungus  growth 

inhibition  by  2-methyl-l,4-naphthoquinone,  Jo^ir.  Bad.  51:  659-670,  1946. 
*CoMMiTTEE  ON  THE  STANDARDIZATION  OF  FUNGICIDAL  Tests:  The  slide-germination 

method  of  evaluating  protectant  fungicides,  Phytopathology  33 :  627-632,  1943. 
Cook,  E.  S.,  G.  Perisutti,  and  T.  M.  Walsh:  The  action  of  phenylmercuric  nitrate. 

II.  Sulfhydryl  antagonism  of  respiratory  depression  caused  by  phenylmercuric 

nitrate,  Jour.  Biol.  Chem.  162 :  51-54,  1946. 
Dillon-Weston,  W.  A.  R.,  and  J.  R.  Booer:  Seed  disinfection.     1.  An  outline  of 

an  investigation  on  disinfectant  dusts  containing  mercury.  Jour.  Agr.  Sci.  26 : 

628-649,  1935. 
FiLDES,  P.:  The  mechanism  of  the  anti-bacterial  action  of  mercury,  Brit.  Jour. 

Exptl.  Path.  21 :  67-73,  1940. 
Fleury,  C.:  Action  de  la  thio-uree  sur  V Aspergillus  niger.     Role  particulier  joue 

par  la  source  d'azote  nitrique,  Bull.  soc.  botan.  Suisse  58 :  462-476,  1948. 
FooTE,  M.  W.,  J.  E.  Little,  and  T.  J.  Sproston:  On  naphthoquinones  as  inhibitors 

of  spore  germination,  Jo2ir.  Biol.  Chem.  181:  481-487,  1949. 
Foster,  A.  A.:  Acceleration  and  retardation  of  germination  of  some  vegetable  seeds 

resulting  from  treatment  with  copper  fungicides,  Phytopathology  37 :  390-398, 

1947. 
Frear,  D.  E.  H.:  Chemistry  of  Insecticides,  Fungicides  and  Herbicides;  2d  ed., 

D.  Van  Nostrand  Company,  Inc.,  New  York,  1943. 
Geiger,  W.  B.  :  The  mechanism  of  the  antibacterial  action  of  quinones  and  hydro- 

quinones.  Arch.  Biochem.  11 :  23-32,  1946. 
Geiger,  W.  B.:  Antibacterial  unsaturated  ketones  and  their  mode  of  action.  Arch. 

Biochem.  16:  423-435,  1948. 
GoLDSwoRTHY,  M.  C,  and  E.  L.  Green:  Availability  of  the  copper  of  Bordeaux 

mixture  residues  and  its  adsorption  by  the  conidia  of  Sclerotinia  fructicola, 

Jour.  Agr.  Research  52:  517-533,  1936. 
Gries,   G.   A.:  Juglone    (5-hydroxy-l,4-naphthoquinone) — a  promising    fungicide, 

Phytopathology  33:  1112,  1943. 
Gries,  G.  A.:  Juglone — the  active  agent  in  walnut  toxicity.  Northern  Nut  Growers 

Assoc.  Ann.  Kept.  34:  52-55,  1943a. 
*HoRSFALL,  J.  G.:  Fungicides  and  Their  Action,  Chronica  Botanica  Co.,  Waltham, 

1945. 
Large,  E.  C:  The  Advance  of  the  Fungi,  Henry  Holt  and  Company,  Inc.,  New 

York,  1940. 
Leben,   C,  and  G.   W.   Keitt:  Laboratory  and  greenhouse  studies  of  antimycin 

preparations  as  protectant  fungicides.  Phytopathology  39 :  529-540,  1949. 


264  PHYSIOLOGY  OF  THE  FUNGI 

Leonian,  L.  H.  :  Differential  growth  of  Phytophthoras  under  the  action  of  malachite 

green,  Am.  Jour.  Botany  17:  C71-677,  1930. 
Leonian,  L.  H.:  Effects  of  position  of  inoculum  on  growth  of  some  Trichophytons 

in  the  presence  of  dyes,  Arch.  Dermatol,  and  Syphilol.  25:  lOlG-1020,  1932. 
Liming,  O.  N.:  The  relation  of  pentathionic  acid  and  its  component  constituents  to 

the  toxicity  of  sulphur  fungicides,  Phytopathology  22:  143-165,  1932. 
*LiTTLE,  J.  E.,  T.  J.  Sproston,  and  M.  W.  Foote:  Isolation  and  antifungal  action  of 

naturally  occurring  2-methoxy-l,4^naphthoquinone,   Jour.   Biol.   Chem.   174: 

335-342,  1948. 
McCalla,  T.  M.:  Cation  adsorption  by  bacteria,  Jour.  Bad.  40:  23-32,  1940. 
*McCallan,   S.  E.  a.:  Studies  on  fungicides.     III.  The  solvent  action  of  spore 

excretions  and  other  agencies  on  protective  copper  fungicides,  Cornell  Univ. 

Agr.  Expt.  Sta.  Mem.  128,  1929. 
McCallan,  S.  E.  a.  :  Outstanding  diseases  of  agricultural  crops  and  uses  of  fungi- 
cides in  the  United  States,  Contribs.  Boyce  Thompson  Inst.  14:  105-115,  1946. 
McCallan,  S.  E.  A.:  Bioassay  of  agricultural  fungicides,   Agr.   Chemicals  2(9): 

31-34,  67;  2(10):  45,  1947. 
McCallan,  S.  E.  A.,  and  F.  R.  Weedon:  Toxicity  of  ammonia,  chlorine,  hydrogen 

cyanide,  hydrogen  sulphide,  and  sulphur  dioxide  gases.     II.  Fungi  and  bacteria, 

Contribs.  Boyce  Thompson  Inst.  11 :  331-342,  1940. 
McCallan,  S.  E.  A.,  and  R.  H.  Wellman:  Fungicidal  versus  fungistatic,  Contribs. 

Boyce  Thompson  Inst.  12:  451-463,  1942. 
■*McCallan,  S.  E.  a.,  and  F.  Wilcoxon:  The  fungicidal  action  of  sulphur.     II.  The 

production  of  hydrogen  sulphide  by  sulphured  leaves  and  spores  and  its  toxicity 

to  spores,  Contribs.  Boyce  Thompson  Inst.  3:  13-38,  1931. 
McCallan,  S.  E.  A.,  and  F.  Wilcoxon:  The  action  of  fvmgous  spores  on  Bordeaux 

mixture,  Contribs.  Boyce  Thompson  Inst.  6:  151-165,  1936. 
McGowan,  J.  C,  P.  W.  Brian,  and  H.  G.  Hemming:  The  fungistatic  activity  of 

ethylenic  and  acetylenic  compounds.     I.  The  effect  of  the  affinity  of  the  sub- 

stituents  for  electrons  upon  the  biological  activity  of  ethylenic  compounds, 

Ann.  Applied  Biol.  35:  25-36,  1948. 
McNew,  G.  L.,  and  N.  K.  Sandholm:  The  fungicidal  activity  of  substituted  pyra- 

zoles  and  related  compounds.  Phytopathology  39  :  721-751,  1949. 
Marten,  E.  A.,  and  J.  G.  Leach:  Some  factors  influencing  the  solubility  of  cuprous 

oxide  in  relation  to  its  toxicity  as  a  fungicide,  Phytopathology  34 :  459-470,  1944. 
Parker-Rhodes,  A.  F. :  Studies  on  the  mechanism  of  fungicidal  action.     IV.  Mer- 
cury, Ann.  Applied  Biol.  29:  404-411,  1942. 
Provost,  B.  :  Memoire  sur  la  cause  immediate  de  la  carie  ou  charbon  des  bl^s,  et  de 

plusieurs  autres  maladies  des  plantes,  et  sur  preservatifs  de  la  carie,   1807. 

Trans,  by  G.  W.  Keitt,  Phytopathological  Classics  No.  6,  American  Phyto- 

pathological  Society,  Menasha,  1939. 
Rich,  S.,  and  J.  G.  Horsfall:  Gaseous  toxicants  from  organic  sulfur  compounds, 

Am.  Jour.  Botany  37:  643-650,  1950. 
Roach,  W.  A.,  and  M.  D.  Glynne:  The  toxicity  of  certain  sulphur  compounds  to 

Synchytrium  endobioticum,  the  fungus  causing  wart  disease  of  potatoes,  Ann. 

Applied  Biol.  15:  168-190,  1928. 
Singer,  T.  P.,  and  E.  S.  G.  Barron:  Studies  on  biological  oxidations.     XX.  Sulf- 

hydryl  enzymes  in  fat  and  protein  metabolism.  Jour.  Biol.  Chem.  157:  241-253, 

1945. 
Starkey,  R.  L.,  and  S.  A.  Waksman:  Fungi  tolerant  to  extreme  acidity  and  high 

concentrations  of  copper  sulfate,  Jour.  Bad.  45:  509-519,  1943. 


ACTION  OF  FUNGICIDES  265 

Ter  Horst,  W.  p.,  and  E.  L.  Felix:  2,3-Dichloro-l,4-naphthoquinone,  a  potent 

fungicide,  Ind.  Eng.  Chem.  35:  1255-1259,  1943. 
Theis,  E.  H.:  The  collagen-quinone  reaction.     1.  Fixation  and  thermolability  as  a 

function  of  pH  values,  Jour.  Biol.  Chem.  157:  23-33,  1945. 
Thornberry,  H.  H.  :  a  paper-disk  plate  method  for  the  quantitative  evaluation  of 

fungicides  and  bactericides.  Phytopathology  40:  419-429,  1950. 
Uppal,  B.  N.:  Toxicity  of  organic  compounds  to  the  spores  of  Phytophthora  colocasiae 

Rac,  Jour.  Agr.  Research  32 :  1069-1097,  1926. 
Wellman,  R.  H.:  Synthetic  chemicals  for  agriculture.     II.  Fungicides,  nematocides, 

rodenticides  and  weed  killers,  Chem.  hids.  63 :  223-229,  1948. 
WiLCOXON,  F.,  and  S.  A.  E.  McCallan:  The  fungicidal  action  of  sulphur.     I.  The 

alleged  role  of  pentathionic  acid,  Phytopathology  20:  391-417,  1930. 
YoE,  J.  H.,  and  L.  A.  Sarver:  Organic  Analytical  Reagents,  John  Wiley  &  Sons 

Inc.,  New  York,  1941. 
Young,  H.  C:  The  toxic  property  of  sulphur,  Ann.   Missouri  Botan.  Garden  9: 

403-435,  1922. 
*Zentmeyer,  G.  a.  :  Inhibition  of  metal  catalysis  as  a  fungistatic  mechanism,  Science 

100 :  294-295,  1944. 


CHAPTER  13 
METABOLIC  PRODUCTS 

The  most  important  product  of  fungus  metabolism  is  carbon  dioxide, 
and  the  most  important  function  of  the  fungi  in  the  economy  of  nature 
is  the  destruction  of  plant  and  animal  remains.  The  use  of  fungi  for 
food  antedates  written  history.  The  use  of  fvmgi  for  the  preparation 
of  bread  and  wine  developed  as  a  household  art.  From  the  time  of 
Pasteur,  the  study  of  fermentation  has  led  to  an  ever-increasing  knowl- 
edge and  understanding  of  the  activities  of  microorganisms.  The  pro- 
duction of  antibiotics  and  vitamins,  alcohol  and  organic  acids,  and  the 
potential  utilization  of  waste  agricultural  products  are  current  fields  of 
research  and  industrial  activity.  For  extensive  treatment  of  these 
subjects  the  reader  is  referred  to  Prescott  and  Dunn  (1949)  and  Foster 
(1949). 

DECOMPOSITION  OF  ORGANIC  MATERIALS 

Brefeld  (1908)  called  fungi  "Organismen  der  Verwesung"  and  con- 
sidered them  to  be  indispensable  agents  in  maintaining  the  essential- 
element  balance  of  nature.  Saprophytic  fungi  and  bacteria  prevent  the 
accumulation  of  plant  and  animal  debris  and  return  the  elements  that 
compose  these  materials  to  the  storehouse  of  nature,  where  they  are 
reused  by  new  generations  of  plants  and  animals.  In  this  role,  sapro- 
phytic fungi  are  designated  as  "vegetable  vultures"  by  Rolfe  and  Rolfe 
(1926),  for  they  act  as  scavengers  in  the  plant  world. 

Green  plants  assimilate  carbon  in  the  form  of  carbon  dioxide.  Waks- 
man  (1938)  has  assembled  the  data  with  regard  to  the  amount  of  carbon 
in  the  biosphere.  It  is  estimated  that  the  atmosphere  contains  600 
billion  tons  of  carbon  in  the  form  of  carbon  dioxide,  and  plants  are  esti- 
mated to  remove  16  billion  tons  yearly.  Thus,  the  carbon  content  of  the 
atmosphere  is  sufficient  for  about  40  years,  if  no  carbon  dioxide  were 
returned  to  the  air. 

The  complete  destruction  of  plant  and  animal  remains  by  fungi  and 
bacteria  requires  a  long  time,  although  some  plant  constituents,  such  as 
soluble  sugars  and  other  carbohydrates,  are  quickly  utilized.  Presum- 
ably the  fungi  are  the  most  important  organisms  in  this  process.  Other 
plant  constituents,  such  as  the  waxes  and  lignin,  are  attacked  more 
slowly.     The  more  resistant  constituents  are  slowly  modified  to  form 

266 


METABOLIC  PRODUCTS  267 

humus.  Some  of  the  carbon  and  other  essential  elements  is  converted 
into  bacterial  and  fungus  protoplasm,  which  after  death  is  subject  to 
decay.  In  the  end,  humus  is  converted  into  carbon  dioxide,  water,  and 
other  simple  compounds,  which  are  used  again.  The  importance  of 
humus  as  a  soil  constituent  is  ably  discussed  by  Waksman  (1938).  In 
addition  to  the  carbon  cycle,  the  fungi  also  play  an  important  part  in  the 
cycles  involving  the  release  and  utilization  of  the  other  essential  elements. 

FUNGI  AS  FOOD 

Many  curious  details  about  the  early  use  of  fungi  as  food  have  been 
collected  from  classical  and  other  sources  by  Buller  (1914)  and  by  Rolfe 
and  Rolfe  (192G).  The  mushrooms  were  no  doubt  among  the  first  fungi 
used  as  food  by  man.  Yeast  became  part  of  his  diet  when  the  arts  of 
brewing  and  baking  were  discovered.  The  widespread  use  of  fermented 
beverages,  under  certain  dietetic  circumstances,  has  an  important  bearing 
on  nutrition  and  health.  J.  S.  Wallerstein  (1939)  has  discussed  primitive 
brewing  practices  and  the  geographical  distribution  of  the  art.  The  beer 
of  the  Middle  Ages  Avas  turbid,  owing  to  its  content  of  suspended  yeast 
cells  (Thaysen,  1943). 

The  nutritive  value  of  any  food  depends  upon  its  composition  and 
digestibility  and  the  assimilability  of  its  hydrolytic  products.  The  early 
writers,  in  the  absence  of  precise  information,  were  of  the  opinion  that 
fungi  had  little  value  as  food.  The  nutritive  value  of  fungi,  of  yeast  in 
particular,  will  be  discussed  from  the  standpoint  of  protein  content  and 
value,  vitamins,  fats,  and  minerals. 

Assuming  good  digestibility,  the  value  of  fungus  protein  is  determined 
by  its  amino-acid  composition.  Rose  (1938),  in  a  long  series  of  careful 
experiments,  has  determined  which  amino  acids  are  essential  for  man  and 
animals.  Some  nine  or  ten  amino  acids  were  found  to  be  essential  (Table 
20).  If  the  protein  part  of  a  diet  is  deficient  in  a  single  essential  amino 
acid,  nitrogen  is  lost  from  the  body,  or  inefficient  utilization  of  protein 
results.  More  of  a  poor  protein  must  be  consumed  in  order  to  increase 
the  intake  of  essential  amino  acids  to  satisfactory  levels.  The  amino- 
acid  composition  of  yeast  and  some  other  proteins  is  given  in  Table  46. 
Yeast  protein  compares  favorably  w4th  casein  or  meat  with  respect  to 
essential  amino  acids. 

Less  complete  data  are  available  for  the  amino-acid  composition  of 
fleshy  fungi.  According  to  Lintzel  (1941),  the  proteins  of  Psalliota 
campestris,  Cantherella  cibarius.  Boletus  edulis,  and  Morchella  esculenta 
are  about  equal  to  animal  protein.  From  100  to  200  gr.  (dry  weight)  of 
these  mushrooms  was  required  to  maintain  the  nitrogen  balance  in  a  man 
weighing  70  kg.  Fitzpatrick  et  al.  (1946)  found  the  tryptophane  content 
of  P.  campestris  to  be  5  mg.  per  100  g. 


268 


PHYSIOLOGY  OF  THE  FUNGI 


Table  46.     Approximate  Amino-acid  Composition  (in  Per  Cent)  of  Some  Plant 
AND  Animal  Proteins  Calculated  to  16  Per  Cent  Nitrogen 
(Block  and  Boiling,  Arch.  Biochem.  7,  1945.     Published  by  permission  of  Academic 
Press,  Inc.) 


Amino  acid 


Arginine 

Histidine 

Lysine 

Tyrosine 

Tryptophane . 
Phenylalanine 

Cystine 

Methionine .  . . 
Threonine. .  .  . 

Leucine 

Isoleucine .... 
Valine 


Yeasts  * 

Meat 

Casein 

Corn 
gluten 

Max. 

Min. 

5.3 

3.1 

7.7 

4.1 

3.1 

3.1 

2.3 

2.9 

2.5 

1.6 

8.1 

6.7 

7.2 

7.5 

0.8 

3.7 

3.4 

3.4 

6.4 

6.7 

1.5 

1.2 

1.3 

1.2 

0.7 

4.6 

2.9 

4.9 

5.2 

6.4 

1.1 

0.9 

1.3 

0.4 

1.1 

2.8 

2.6 

3.3 

3.5 

4.0 

6.0 

5.1 

5.4 

3.9 

4.1 

8.5 

6.1 

7.7 

12.1 

24.0 

6.2 

5.5 

5.2 

6.5 

5.0 

5.9 

4.6 

5.7 

7.0 

5.0 

Polished 
rice 


7.2 
1.5 
3.2 
5.6 
1.3 
6.7 
1.4 
3.4 
4.1 
9.0 
5.3 
6.3 


*  Eight  strains  analyzed. 

The  value  of  fungus  protein  in  nutrition  can  be  assessed  only  in  relation 
to  the  amino-acid  composition  of  the  remainder  of  the  diet.  If  the 
dietary  proteins  are  low  in  certain  essential  amino  acids,  the  supplemen- 
tary value  of  yeast  (or  other)  protein  may  be  great.  The  cereal  grains, 
which  furnish  the  bulk  of  protein  for  the  population  of  the  world,  are 
generally  low  in  one  or  more  essential  amino  acids.  Usually  cereal 
protein  is  low  in  lysine  or  tryptophane  or  both.  Sure  (1946,  1947)  studied 
the  effect  on  the  growth  of  rats  of  adding  1,  3,  and  5  per  cent  of  dried 
yeast  to  diets  which  contained  cereals  as  the  sole  source  of  protein.  The 
most  marked  effect  of  yeast  occurred  on  a  maize  diet.  At  the  end  of  a 
10-week  experimental  period  the  rats  receiving  only  cereal  weighed 
27.3  g.,  while  the  rats  which  received  an  additional  1  per  cent  yeast 
weighed  50.5  g.  Rats  which  received  the  cereal  plus  3  and  5  per  cent 
yeast  weighed  91.8  and  109.9  g.,  respectively.  The  effect  of  yeast  was 
not  so  great  when  wheat  or  rice  supplied  the  protein  in  the  diet.  In 
general,  the  most  promising  use  of  yeast  protein  in  human  nutrition  is 
as  a  supplement  rather  than  as  a  sole  source  of  protein. 

Yeasts  are  efficient  in  absorbing  and  concentrating  the  vitamins  present 
in  the  media  in  which  they  grow  (Gorcica  and  Levine,  1942).  The 
relative  value  of  yeast  as  a  source  of  vitamins  depends  upon  the  vitamin 
content  of  the  other  constituents  of  the  diet.  The  prevalence  of  vitamin 
deficiency  diseases  (beriberi,  pellagra,  and  others)  is  evidence  that  the 
vitamin  content  of  many  diets  is  inadequate. 


METABOLIC  PRODUCTS 


269 


A  dramatic  demonstration  of  the  value  of  yeast  as  a  source  of  vitamins 
is  reported  by  Bray  (1928),  onetime  medical  officer,  Nauru,  Central 
Pacific.  The  mandating  government  prohibited  the  brewing  of  toddy 
(palm  wine)  and  allowed  the  sale  of  refined  sugar.  The  results  of  these 
dietary  changes  were  appalling.  Soon,  40  per  cent  of  the  infants  born 
in  1  year  perished  of  infantile  beriberi  (thiamine  deficiency)  before  reach- 
ing the  age  of  6  months.  The  restoration  of  toddy  and  enforced  con- 
sumption of  the  dregs,  i.e.,  the  yeast,  reduced  the  incidence  of  beriberi 
to  one  death  in  16  months.  Truly,  Bray  was  right  in  calling  toddy  the 
elixir  of  life  of  the  Nauruans.  Piatt  and  Webb  (1945)  have  noted  that  a 
simple  maize  diet  which  was  inadequate  with  respect  to  riboflavin  and 
nicotinic  acid  was  made  adequate  in  these  respects  by  converting  a  por- 
tion of  the  dietary  maize  into  maize  beer. 

The  vitamin  content  of  yeasts  depends  upon  the  species  or  strain  and 
the  conditions  of  cultivation.  Some  representative  data  are  presented 
in  Table  47. 


Table  47.     Vitamin  Content  of  Seven  Food  Yeasts 
Results  in  milligrams  per  100  g.  of  dry  yeast.      (Von  Loesecke,  Jour.  Am.  Dietet. 
Assoc.  22,  1946.     Published  by  permission  of  the  American  Dietetic  Association.) 


Species 


Torula  utilis 

Saccharomyces  cerevisiae* 

S.  cerevisiae 

S.  cerevisiae  f 

S.  cerevisiae^ 

S.  cerevisiae  f 

S.  cerevisiaeX 


Thiamine 


1.7 
17.0 
20.5 
17.5 
17.5 
16.0 

3.0 


Riboflavin 


4.7 
8.0 
7.6 
4.2 
4.5 
3.6 
7.5 


Nicotinic 
acid 


19.0 
25.0 
29.0 
48.0 
37,0 
32.0 
38.0 


Pantothenic 
acid 


86.0 

112.0 
122.0 
86.0 
72.0 
74.0 
13.5 


*  Six  per  cent  salt  added. 
t  Debittered  brewer's  yeast. 
t  Primary  yeast. 

The  production  of  fats  by  fungi  is  discussed  elsewhere  in  this  chapter. 
The  usual  fatty  acids,  including  palmitic  and  oleic  acids,  are  found  in  fat 
synthesized  by  fungi.  Apparently  few  studies  have  been  made  on  the 
value  of  fungi  as  sources  of  fat  and  essential  minerals  in  human  nutrition. 

CULTIVATION  OF  FUNGI  FOR  FOOD 

The  ants  were  perhaps  the  first  to  cultivate  fungi  as  a  source  of  food 
(see  Leaoii,  1940,  for  discussion  and  references).  Fungi  have  been  used 
for  centuries  in  the  Orient  as  food  for  man.  The  Chinese  grow  Hirneola 
polytricha  and  the  Japanese  grow  Armillaria  shii-take  on  oak  saplings. 
The  mushroom  cultivated  almost  exclusively  in  the  Occident  is  Agaricus 


270  PHYSIOLOGY  OF  THE  FUNGI 

(Psalliota)  campestris.  The  method  of  cultivating  this  species  on  com- 
posted horse  manure  was  developed  near  Paris  before  1700.  For  informa- 
tion on  mushroom  growing  the  reader  is  referred  to  Duggar  (1915). 

While  attempts  to  replace  composted  horse  manure  by  other  substrates 
have  been  made,  none  appears  to  be  entirely  satisfactory.  Humfield 
(1948)  has  suggested  that  Psalliota  campestris  be  grown  in  large  fermentors 
and  the  mycelium  rather  than  the  fruit  bodies  be  used  for  food.  Aspara- 
gus butt  juice,  a  waste  agricultural  product,  is  a  suitable  medium.  The 
chemical  composition  of  mycelium  and  that  of  the  fruit  bodies  is  similar 
and  the  flavor  comparable.  This  approach  perhaps  offers  a  way  to  culti- 
vate other  desirable  species,  including  the  morels  and  the  truffles.  Nord 
(1948)  has  suggested  that  the  mycelium  of  Fusarium  lini  be  used  for  food. 

The  use  of  yeasts  to  convert  low-grade  carbohydrates,  such  as  wood 
sugar  and  molasses,  into  food  has  interesting  possibilities.  It  is  necessary 
to  fortify  these  carbohydrates  with  other  nutrients  for  the  cultivation  of 
yeast.  Phosphates,  a  source  of  potassium,  and  nitrogen,  in  the  form  of 
urea,  ammonia,  or  ammonium  salts,  are  added.  The  function  of  yeast  is 
to  convert  inorganic  nitrogen  into  protein.  Animals  are  unable  to 
assimilate  ammonia  or  urea  directly  but  require  nitrogen  in  the  form  of 
protein  or  amino  acids.  Inorganic  nitrogen  may  be  converted  into 
proteins  by  green  plants  or  by  certain  microorganisms.  The  use  of  urea, 
a  derivative  of  ammonia,  as  cattle  fodder  is  an  example  of  the  synthesis 
of  protein  by  the  microflora  of  the  rumen. 

The  possibility  of  using  wood  waste  for  yeast  propagation  was  investi- 
gated in  Germany  during  the  First  World  War.  In  1944  it  is  reported 
that  9,000  tons  of  food  yeast  were  produced  in  Germany.  Fermentable 
carbohydrates  are  obtained  from  wood  as  a  by-product  of  sulfite  paper 
manufacture,  or  by  direct  hydrolysis.  Before  sulfite  liquor  or  wood 
hydrolysate  is  used  for  yeast  culture,  it  is  treated  with  calcium  carbonate 
to  adjust  the  pH  and  precipitate  impurities.  After  the  addition  of 
nutrients  the  solution  is  heavily  inoculated  with  the  desired  strain  of 
yeast.  Aeration  is  necessary  for  high  yields  of  yeast.  The  weight  of 
yeast  produced  amounts  to  about  half  the  weight  of  sugar  utilized.  Such 
yeast  is  approximately  50  per  cent  protein  (Harris  et  al.,  1948).  The 
economics  of  fodder-yeast  production  from  sulfite  liquor  have  been  studied 
by  Schleef  (1948).  The  use  of  by-product  molasses  for  the  production 
of  food  and  fodder  yeasts  should  offer  fewer  technical  difficulties  than  the 

use  of  wood  sugar. 

FAT  PRODUCTION 

Serious  efforts  to  utilize  fungi  for  the  synthesis  of  fats  were  made  in 
Germany  during  the  First  World  War  and  continued  thereafter.  The 
technical  problems  encountered  proved  difficult,  but  some  success  was 
achieved  by  1942  (Hesse,  1949).     The  controHing  factor  in  fat  production 


METABOLIC  PRODUCTS 


271 


appears  to  be  the  carbon-nitrogen  ratio.  As  long  as  an  adequate  supply 
of  nitrogen  is  present,  little  fat  is  synthesized.  If  the  carbohydrate  sup- 
ply is  high  when  the  nitrogen  is  exhausted,  assimilable  fat  is  synthesized. 
Linder  (1922)  termed  these  two  phases  'protein  generation  and  Jat  genera- 
tion. Fat-laden  cells  of  many  fungi  appear  to  be  incapable  of  cell  division. 
Fat  formation  takes  place  only  in  the  presence  of  an  abundant  supply 
of  oxygen.  The  relation  between  sugar  concentration  and  amount  of  fat 
synthesized  by  Penicillium  javanicum  is  illustrated  in  Fig.  52. 


C7> 


2500 

> 

^ 

1. 

2000 

Myce 

//um"'' 

N 

\ 

40 

> 

/ 

\\ 

\ 

1500 

Per 

cent  fat 

/ 

\ 

\ 

35 

> 

/ 

t 

\ 

innn 

/ 

/ 

i 

k 

30 

/ 

^ 

\ 

3» 

E 


200 


300  400 

Groms  of  glucose  per  liter 


500 


Fig.  52.  The  effect  of  the  concentration  of  glucose  on  the  amount  of  mycelium  and 
amount  of  fat  synthesized  by  PenicilUum  javanimm  cultured  in  75  ml.  of  medium  for 
12  days.  (Drawn  from  the  data  of  Lockwood,  Catholic  Univ.  of  America  Biol.  Ser. 
13,  p.  8,  1933.     Published  by  permission  of  the  Catholic  University  of  America.) 


Among  the  fungi  investigated  for  fat  synthesis  are  Endomyces  vernalis, 
Oidium  lactis,  Tonda  utilis,  Rhodotorula  ghdinis,  and  species  of  Aspergil- 
lus, Penicillium,  Mucor,  and  Fusarium.  From  a  practical  standpoint, 
only  fungi  which  are  capable  of  synthesizing  fat  in  submerged  culture 
are  of  potential  value.  E.  vernalis  and  0.  lactis  do  not  produce  fat  effici- 
ently in  submerged  culture.  The  fat  content  of  various  filamentous  fungi 
was  determined  by  Preuss  et  al.  (1934)  and  Ward  et  al.  (1935).  The  use 
of  E.  vernalis  for  fat  and  protein  synthesis  has  been  reviewed  by  Raaf 
(1941).  Starkey  (1946)  studied  fat  production  by  an  unidentified  soil 
yeast,  which  under  favorable  conditions  contained  from  50  to  63  per  cent 


272  PHYSIOLOGY  OF  THE  FUNGI 

crude  lipide.     A  list  of  species  of  Penicillium  and  Aspergillus  which  syn- 
thesize considerable  fat  is  given  in  Table  48. 

Table  48.     The  Crude  Fat  Content  op  Dried  Mycelium  of  Various  Species  op 
PeniciUiuin  and  Aspergillus  as  Determined  by  Extraction  with  Ether 
(Ward  et  al.,  Ind.  Eng.  Chem.  27,  1935.     Published  by  permission  of  the  American 
Chemical  Society.) 

Species  Crude  Fat,  % 

Penicillium flavo-cinereum 28.5 

P.  piscarum 26-28 

P.  oxalicum 24.4 

P.  roqueforti 22.9 

P.  javanicum 22 . 2 

Aspergillus  flavus 16.0 

Various  theories  of  the  mechanism  of  fat  synthesis  have  been  published 
and  are  reviewed  by  Foster  (1949)  and  Hesse  (1949).  Most  of  these 
consider  acetaldehyde  or  acetate  to  be  the  product  of  intermediary  metab- 
olism used  in  fat  synthesis.  This  emphasizes  the  importance  of  pyruvic 
acid  in  fungus  metabolism.  Various  investigators  have  shown  that 
acetaldehyde  may  be  converted  into  fat  by  yeasts.  The  glycerol  required 
for  fat  synthesis  is  thought  to  arise  from  the  reduction  and  hydrolysis  of 
dihydroxyacetone  phosphate  or  3-phosphogly eerie  aldehyde  (scheme  VI, 

Chap.  7). 

PRODUCTION  OF  VITAMINS 

Only  a  few  species  of  fungi  and  bacteria  produce  vitamins  in  large 
enough  amounts  to  be  of  interest  in  industry.  Biological  synthesis  must 
compete  with  chemical  synthesis  on  a  cost  basis.  The  recovery  of  vita- 
mins as  a  by-product  of  commercial  processes  or  the  use  of  waste  materials 
as  the  basis  of  a  cheap  medium  may  make  biological  synthesis  attractive. 

Riboflavin  is  produced  so  abundantly  by  Candida  guilliermondi  under 
certain  cultural  conditions  that  it  crystallizes  in  the  medium  (Burkholder, 
1943).  Among  the  factors  found  to  influence  the  amount  of  riboflavin 
synthesized,  the  sources  of  carbon  and  nitrogen  and  aeration  are  impor- 
tant. Various  investigators  have  found  the  concentration  of  iron  in  the 
medium  to  have  a  profound  influence  on  the  amount  of  riboflavin  syn- 
thesized by  various  organisms.  Iron  concentrations  in  excess  of  10  ^g 
per  liter  decreased  the  amount  of  riboflavin  synthesized  by  C.  guillier- 
mondi and  C.  fiareri  (Tanner  et  al.,  1945;  Tanner  and  Van  Lanen,  1947). 
The  optimum  iron  concentration  for  riboflavin  synthesis  by  Clostridium, 
acetobutylicum  is  said  to  be  1  mg.  per  liter.  Hickey  (1945)  has  suggested 
the  use  of  2,2'-bipyridine  to  inactivate  excessive  concentrations  of  iron 
in  industrial  fermentations.  By  maintaining  the  iron  concentration 
between  40  and  60  ng  per  liter,  Levine  et  al.  (1949)  found  the  maximum 
yields  of  riboflavin  produced  by  C.  guilliermondi  and  C.  fiareri  to  be  175 
and  567  ng  per  ml.,  respectively.     Pilot-plant  yields  were  somewhat  less. 


METABOLIC  PRODUCTS  273 

Eremothecium  ashbyi  was  shown  to  produce  as  much  as  157  mg.  per 
liter  of  riboflavin  when  cultivated  on  glucose-peptone  medium  (Renaud 
andLachaux,  1945).  Aeration  was  necessary.  Foster  (1947)  has  recom- 
mended a  molasses  medium  for  the  commercial  production  of  riboflavin 
by  E.  ashbyi.  The  closely  related  species,  Ashbya  gossypii,  also  synthe- 
sizes riboflavin  in  large  amounts  (Tanner  et  al.,  1949). 

Peltier  and  Borchers  (1947)  determined  the  amount  of  riboflavin 
produced  by  240  isolates  of  soil  fungi  when  grown  on  wheat  bran.  Forty- 
five  isolates  produced  2  mg.  or  more  of  riboflavin  per  100  g.  of  dry  mold 
bran.  An  unidentified  species  of  Aspergillus  produced  5.8  mg.  of  ribo- 
flavin per  100  g.  of  substrate.  Species  of  Fusarium  and  Aspergillus  were 
outstanding  producers  of  riboflavin. 

The  commercial  microbiological  synthesis  of  riboflavin  depends  upon 
the  use  of  either  E.  ashbyi  or  C.  acetobutyliciim  (Tanner  et  al.,  1949). 

Vitamin  B12  was  isolated  in  crystalline  form  from  liver  and  shown  to 
contain  cobalt  (Rickes  et  al.,  1948;  Smith,  1948).  It  is  the  only  vitamin 
so  far  discovered  which  contains  a  metal  as  an  integral  part  of  the  mole- 
cule. Streptomyces  griseus  and  other  microorganisms  synthesize  this 
vitamin.  Sheep  and  cattle  pastured  on  cobalt-deficient  soils  (Florida, 
Australia,  New  Zealand)  develop  a  deficiency  disease.  Ingested  cobalt 
is  more  effective  than  injected  cobalt  in  overcoming  this  condition.  It 
may  be  assumed  that  cobalt  is  used  in  the  synthesis  of  vitamin  B12  by  the 
action  of  the  microorganisms  of  the  rumen  and  intestine.  Vitamin  B12 
appears  to  be  the  anti-pernicious-anemia  factor  (West,  1948).  Whether 
it  is  the  animal  protein  factor  is  undecided.  The  cow-manure  factor  may 
be  vitamin  B12  (Lillie  et  al.,  1948). 

Until  the  structure  of  vitamin  B12  is  determined  and  methods  of  syn- 
thesis developed,  certain  natural  products  will  remain  the  only  source 
of  this  vitamin.  The  only  organic  moiety  of  vitamin  B12  so  far  disclosed 
is  l-a-D-ribofuranosido-5,6-dimethylbenzimidazole  (Brink  et  al.,  1950). 
Vitamin  B12  is  obtained  as  a  by-product  from  various  industrial  processes, 
especially  streptomycin  production.  It  is  evident  that  the  medium  must 
contain  cobalt;  within  limits,  the  amount  of  vitamin  B12  synthesized  by 
Streptomyces  griseus  is  a  function  of  the  cobalt  content  of  the  medium. 
Maximum  synthesis  was  observed  when  the  medium  contained  1  to  2  mg. 
of  cobalt  per  liter  (Hendlin  and  Ruger,  1950). 

None  of  the  other  vitamins  appears  to  be  synthesized  by  fungi  in 
amounts  which  would  make  the  latter  attractive  sources  for  the  isolation 
of  pure  vitamins.  The  value  of  these  vitamins  in  fungi  used  for  food  was 
discussed  previously.  Yeast  can  be  fortified  with  thiamine  so  that  it 
may  serve  as  a  therapeutic  agent.  By  adding  synthetic  thiamine  to  an 
aerated  yeast  culture,  yeast  was  produced  which  contained  6  mg.  of 
thiamine  per  g.  (Van  Lanen  et  al.,  1942). 


274  PHYSIOLOGY  OF  THE  FUNGI 


ENZYME  PRODUCTION 


The  industrial  production  and  use  of  enzymes  from  microorganisms 
in  the  Occident  is  fairly  recent,  although  the  use  of  fungi  as  amylolytic 
agents  by  the  peoples  of  the  Orient  for  the  preparation  of  koji  and  other 
foods  is  an  old  art.  For  this  purpose,  mixed  cultures  of  species  of  Asper- 
gillus and  Rhizopus  are  grown  upon  the  rice  or  soybean  substrates,  the 
enzymes  being  used  without  separation.  The  pioneering  work  of  Taka- 
mine  (1914)  on  the  amylases  of  A.  oryzae  was  especially  important. 

The  ability  of  fungi  to  produce  amylase  is  widely  distributed,  but  only 
a  few  species  are  used  commercially  for  this  purpose.  The  amount  of 
amylase  produced  varies  with  the  species  or  isolate  and  the  cultural  condi- 
tions. Le  Mense  et  al.  (1947)  screened  359  isolates  of  Penicillium  and 
Aspergillus  and  found  42  isolates  to  produce  amylase  in  submerged 
culture.  The  activity  of  the  species  of  Penicillium  ranged  from  0.1  to 
0.6  enzyme  unit  per  milliliter  of  culture  medium.  One  isolate  of  A. 
niger  (NRRL  337)  was  found  to  be  especially  adapted  for  the  production 
of  amylase  in  submerged  culture.  The  production  of  amylase  was  highly 
dependent  upon  the  composition  of  the  medium.  Corn  meal  was  espe- 
cially valuable  in  increasing  amylase  production  when  added  to  basal 
media  composed  of  corn  steep  liquor,  dried  tankage,  soybean  meal,  or 
thin  stillage.  Amylase  production  was  stimulated  by  the  addition  of  10 
to  20  mg.  of  sodium  chloride  per  liter  of  culture  medium.  Addition  of  a 
mixture  of  chlorinated  phenols  (Dowcide  G)  inhibited  sporulation  and 
increased  amylase  production  (Erb  et  al.,  1948). 

Others  have  found  different  isolates  of  the  same  species  to  produce 
varying  amounts  of  amylase.  Hao  et  al.  (1943)  studied  the  production 
of  amylase  by  27  isolates  of  various  species  of  fungi  when  grown  upon 
wheat  bran.  A.  oryzae,  Rhizopus  delemar,  and  R.  oryzae  produced  the 
largest  amounts  of  amylase.  A.  oryzae  was  the  fungus  of  choice  because 
of  ease  of  handling. 

In  practice,  fungus  amylases  are  produced  and  utilized  in  three  general 
ways.  (1)  In  the  amylo  process,  starch  is  solubilized  by  autoclaving  with 
a  trace  of  a  mineral  acid,  and  the  mash  is  inoculated  with  a  species  of 
Rhizopus,  which  produces  amylase  abundantly,  and  a  species  of  yeast. 
The  function  of  the  Rhizopus  species  is  to  convert  the  starch  into  ferment- 
able sugars,  from  which  the  yeast  produces  alcohol.  For  a  description 
of  this  process  see  Owen  (1933).  (2)  The  fungus  may  be  grown  upon  a 
solid  substrate  such  as  bran  and  the  resulting  moldy  mass  (mold  bran) 
dried  (Underkofler  et  al.,  1946).  The  fresh  material  may  be  used  without 
drying  (Roberts  et  al.,  1944).  (3)  Fungus  amylases  may  be  produced  in 
submerged  aerated  cultures  much  as  antibiotics  are  produced.  The 
culture  medium  may  be  used  directly  to  replace  malt  as  a  saccharifying 
agent. 


METABOLIC  PRODUCTS  275 

Fungus  amylases  are  used  to  replace  malt  amylase  for  the  saccharifica- 
tion  of  starch,  Myrback  (1948)  is  of  the  opinion  that  amylase  from  A. 
niger  is  an  a-amylase,  but  it  differs  from  a-amylase  of  malt  in  that  it  has  a 
higher  capacity  for  saccharification.  For  a  comparison  of  fungus  and 
malt  amylase  and  the  economic  considerations  involved,  see  Underkofler 
et  al.  (194G).  The  yield  of  alcohol  is  said  to  be  slightly  higher  when 
fungus  amylase  is  used  in  place  of  malt  for  saccharification. 

Fungi  are  the  source  of  other  enzymes  of  commercial  interest,  including 
pectinase  and  sucrase.  Pectinase  is  used  in  the  clarification  of  fruit 
juices.  For  a  survey  of  the  commercial  production  of  fungus  enzymes 
seeL.  Wallerstein  (1939). 

ALCOHOLIC  FERMENTATION 

Yeasts  are  used  almost  exclusively  for  the  commercial  production  of 
fermentation  alcohol,  but  alcoholic  fermentation  is  not  restricted  to  these 
fungi.  Pasteur  (1872)  observed  that  Penicillium  glaucum,  Aspergillus 
glaucus,  and  Mucor  raceniosus  produced  alcohol  under  anaerobic  condi- 
tions. Further  information  on  alcohol  production  by  filamentous  fungi 
may  be  found  in  the  monograph  of  Raistrick  et  al.  (1931),  who  determined 
complete  carbon  balances  for  96  species  of  Aspergillus,  75  species  of 
Penicillium,  8  species  of  Citromyces  (Penicillium) ,  23  species  of  Fusarium 
and  36  miscellaneous  species.  The  original  report  should  be  consulted 
for  details  and  the  quantitative  methods  used.  All  the  23  species  of 
Fusarium  studied  produced  alcohol.  From  this  and  other  reports  in  the 
literature,  it  must  be  concluded  that  this  property  is  common  among 
species  of  this  genus.  Many  species  of  Aspergillus  and  Penicillium  pro- 
duced alcohol,  as  did  species  of  other  genera.  Only  a  few  of  the  species 
studied  failed  to  produce  detectable  amounts  of  alcohol.  The  apparatus 
used  in  these  studies  is  shown  in  Fig.  53. 

The  concentration  of  alcohol  which  inhibits  the  growth  of  fungi  varies 
with  the  species  or  strain.  In  general,  yeasts  are  more  tolerant  of  alcohol 
than  the  filamentous  fungi.  The  upper  limit  for  most  yeasts  is  about  12 
per  cent  alcohol,  although  some  strains  are  more  tolerant.  The  suscepti- 
bility to  alcohol  limits  the  alcohol  concentration  of  naturally  fermentt>? 
beverages.  The  rate  of  fermentation  decreases  as  the  concentration  (k 
alcohol  increases. 

Not  all  isolates  of  a  species  are  equally  efficient  in  producing  alcohcu 
For  example,  eight  isolates  of  Fusarium  lini  produced  varying  amountfe 
of  alcohol  on  the  same  medium.  The  more  virulent  pathogens  on  flax 
produced  the  most  alcohol  (Letcher  and  Willaman,  1926).  A  correlation 
between  sporulation  and  alcohol  production  by  Aspergillus  flavus  was 
noted  by  Yuill  (1928).  In  general,  sporulating  cultures  produced  less 
alcohol  than  nonsporulating  cultures. 

The  most  important  condition  w^hich  governs  alcoholic  fermentation 


276 


PHYSIOLOGY  OF  THE  FUNGI 


^  ^ 


r^HI 


sterilization 


Cone. 


Cotton  wool 


mh: 


Fig.  53.  Apparatus  for  studying  the  metabolic  products  of  fungi  and  other  micro- 
organisms. The  apparatus  consists  of  five  units:  A  gasholder,  P;  a  train  for  the  puri- 
fication and  sterilization  of  air  or  other  gases,  A-E\  the  culture  flask,  F;  a  train  for  the 
quantitative  absorption  of  carbon  dioxide,  H-M;  an  aspirator,  Q,  for  the  collection 
of  gaseous  products  of  metabolism  other  than  carbon  dioxide.  (Redrawn  from  Birkin- 
shaw  and  Raistrick,  Trans.  Roy.  Soc.  (London),  Ser.  B,  220:  14,  1931.  Published  by 
permission  of  the  Royal  Society.) 

is  the  supply  of  oxygen.  The  relation  between  fermentation  and  anaero- 
bic conditions  was  recognized  by  Pasteur,  who  summarized  his  extensive 
investigations  on  fermentation  as  "la  vie  sans  air."  The  essential  feature 
of  fermentation  is  anaerobic  dissimilation  of  carbohydrates.  Growth  and 
fermentation  are  competitive  processes,  for  fungi  require  oxygen  for 
growth.     In  practice  it  is  advantageous  to  carry  out  fermentations  in  the 


METABOLIC  PRODUCTS  277 

presence  of  some  air,  especially  at  the  start.  This  allows  some  increase 
in  the  number  of  cells  and  reduces  the  amount  of  inoculum  required.  The 
amount  of  oxygen  available  to  submerged  mycelium  or  cells,  unless 
vigorous  aeration  is  used,  is  insufficient  to  inhibit  alcoholic  fermentation 
by  certain  species. 

Alcoholic  fermentation  has  been  studied  since  the  time  of  Lavoisier. 
Few  fields  of  study  have  been  so  valuable  in  increasing  our  understanding 
of  the  life  processes  of  microorganisms.  Harden  (1932)  has  concisely 
reviewed  the  early  work  and  theories  on  fermentation.  The  idea  that 
yeasts  as  living  fungi  were  the  proximate  cause  of  fermentation  did  not 
gain  acceptance  for  many  decades.  The  eminent  Wohler  (1839)  ridiculed 
this  idea  in  a  lively  skit,  in  which  he  declared  that  he  had  followed  the 
entire  process  microscopically.  Briefly,  he  states  that  the  responsible 
organism  developed  from  an  egg  and  had  the  shape  of  a  Beindorf  distilling 
flask;  "...  diese  Infusorien  fressen  Zucker,  entleeren  aus  dem  Darm- 
kanal  Weingeist,  und  aus  den  Harnorganen,  Kohlensaure." 

The  enzymatic  nature  of  alcohol  fermentation  was  established  by 
Buchner  (1897).  The  enzymatic  transformations  involved  in  fermenta- 
tion were  discussed  in  Chap.  7.  Further  information  and  references  may 
be  found  in  Summer  and  Somers  (1947),  Tauber  (1949),  Prescott  and 
Dunn  (1949),  Meyerhof  (1944,  1949),  Nord  and  Mull  (1945),  and  Foster 
(1949). 

The  larger  part  of  the  world-wide  fermentation  industry  is  devoted 
to  the  production  of  ethyl  alcohol.  During  the  war  year  of  1945  some 
600  million  gallons  of  95  per  cent  ethyl  alcohol  was  produced  in  the  United 
States  alone.  Less  than  one-third  this  amount  was  produced  in  1948. 
Of  this  amount  64  per  cent  was  produced  by  fermentation  (Lee,  1949). 
While  any  source  of  fermentable  sugars  may  be  used  for  the  production 
of  alcohol,  the  more  common  raw  materials  include  molasses,  starch  from 
various  sources,  hydrolyzed  cellulose  or  wood  sugar,  and  fruit  juices.  It 
is  beyond  the  scope  of  this  text  to  discuss  the  commercial  production  of 
industrial  and  beverage  alcohol.  For  information  on  these  subjects  see 
Prescott  and  Dunn  (1949). 

ORGANIC  ACIDS 

Many  fungi  synthesize  organic  acids,  which  accumulate  in  the  medium. 
These  acids  include  oxalic,  citric,  succinic,  fumaric,  malic,  lactic,  itaconic, 
kojic,  gluconic,  and  others.  Commonly,  a  species  may  synthesize  a 
variety  of  related  acids.  The  isolates  of  a  given  species  may  differ  widely 
in  synthetic  capacity.  To  obtain  maximum  yields,  it  is  necessary  to 
control  nutritional  and  environmental  factors  closely.  The  optimum 
conditions  for  one  isolate  may  differ  from  those  of  another  isolate  of  the 


278  PHYSIOLOGY  OF  THE  FUNGI 

same  species.     Acid  production  by  fungi  is  discussed  in  detail  by  Foster 
(1949),  Prescott  and  Dunn  (1949),  and  Wallvcr  (1949). 

The  meaning  of  the  term  fermentation  has  been  expanded  by  most 
authors  to  include  aerobic  as  well  as  anaerobic  processes.  The  produc- 
tion of  most  organic  acids  and  antibiotics  Ijy  fungi  takes  place  in  the 
presence  of  oxygen,  and  these  processes  are  not  fermentations  in  the 
restricted  (anaerobic)  sense  of  the  term.  Indeed,  adequate  aeration  is 
one  of  the  salient  features  of  such  processes.  Aeration  may  be  achieved 
by  cultivating  the  fungi  on  the  surface  of  shallow  layers  of  medium  in 
pans  or  trays ;  or  the  fungi  may  be  cultivated  in  closed  tanks,  which  may 
contain  as  much  as  15,000  gal.  of  medium.  Aeration  is  provided  by 
mechanical  stirring  and  blowing  in  sterile  air  under  pressure. 

The  organic  acids  discussed  in  this  chapter  are  derived  from  carbo- 
hydrates present  in  the  medium.  In  general,  media  highly  unbalanced 
with  respect  to  carbohydrates  are  used.  The  balanced  medium  devel- 
oped by  Steinberg  for  the  cultivation  of  Aspergillus  niger  (Chap.  2)  has 
a  carbon-to-nitrogen  ratio  of  29  to  1,  while  the  medium  recommended  by 
Currie  (1917)  for  the  production  of  citric  acid  by  ^.  niger  has  a  carbon-to- 
nitrogen  ratio  of  72  to  1.  A  fungus  first  utilizes  the  nutrients  in  the 
unbalanced  medium  for  the  production  of  mycelium  (growth  phase). 
The  excess  carbohydrate  which  remains  when  the  nitrogen  is  exhausted 
is  dissimilated  ("fermentation"  phase).  Advantage  is  taken  of  such 
preformed  mycelium,  for  if  the  original  medium  is  replaced  by  fresh 
medium,  the  mycelium  continues  to  dissimilate  carbohydrate.  The 
replacement  medium  is  frequently  more  unbalanced  than  the  growth 
medium.  For  example,  Karow  and  Waksman  (1947)  used  for  A.  wentii 
a  growth  medium  with  a  carbon-to-nitrogen  ratio  of  135  to  1,  while  the 
replacement  medium  had  a  carbon-to-nitrogen  ratio  of  270  to  1. 

Economic  amounts  of  organic  acids  may  accumulate  in  the  medium 
because  the  normal  use  of  these  compounds  for  the  synthesis  of  mycelium 
is  prevented  by  the  imposed  experimental  conditions.  If  the  nitrogen 
supply  is  exhausted,  no  more  protoplasm  can  be  formed.  The  mycelium 
then  dissimilates  sugars  enzymatically.  Enough  nutrients  are  supplied 
in  replacement  media  to  repair  and  maintain  the  enzyme  systems  of  the 
fungus  in  a  vigorously  functioning  state.  The  enzymes,  other  than  those 
concerned  with  certain  phases  of  carbohydrate  dissimilation,  are  largely 
idle  because  of  the  lack  of  suitable  substrates. 

A  fungus  commonly  produces  several  organic  acids  at  the  same  time. 
Citric  and  oxalic  acids  are  produced  by  many  isolates  of  A .  niger,  and  the 
relative  amounts  of  these  acids  may  be  varied  by  controlling  the  pH  of 
the  medium.  In  general,  a  highly  acid  medium  (pH  2.0  to  3.0)  favors  the 
synthesis  of  citric  acid,  while  less  acid  media  favor  the  production  o/ 
oxalic  acid. 


METABOLIC  PRODUCTS 


279 


Citric  acid.  Wehmer  was  the  first  to  recognize  the  commercial  possi- 
bilities of  citric  acid  synthesis  by  two  species  of  Citromyces  {Penicillium). 
Selected  isolates  of  Aspergillus  niger  appear  to  be  used  in  industry, 
although  the  propertj^  of  producing  citric  acid  is  common  to  many  fungi. 
The  following  fungi  have  been  suggested  for  commercial  citric  acid  pro- 
duction (Von  Loesecke,  1945):  Citromyces  pfejferianus,  C.  glaber,  C. 
citricus,  Aspergillus  carhonarius,  A.  glaucus,  A.  clavatus,  A.  cinnamomeus, 
A.fumaricus,  A.  awamori,  A.  aureus,  Penicillium  arenarium,  P.  olivaceum, 
P.  divaricatum,  P.  sanguifluus,  P.  glaucum,  Mucor  pyriformis. 

The  production  of  citric  acid  in  the  United  States  increased  from  about 
5  million  to  26  million  pounds  between  1935  and  1945  (Von  Loesecke. 
1945).  Presumably  most  of  this  was  "fermentation"  citric  acid.  At 
present  it  is  beheved  that  most  citric  acid  is  produced  by  surface  cultures. 

Citric  acid  is  formed  from  many  sources  of  carbon.  Sucrose  is  said 
to  be  the  best  carbon  source  for  the  production  of  citric  acid.  There  is 
less  agreement  upon  the  value  of  other  sugars.  Different  investigators 
have  found  glucose,  fructose,  and  maltose  to  vary  from  good  to  poor.  In 
part,  this  is  to  be  attributed  to  the  use  of  different  isolates  and  different 
experimental  conditions.  Beet  molasses  is  used  in  industry.  The 
suitability  of  this  substrate  is  said  to  vary  with  the  source  and  year  of 
production  (Bernhauer  and  Knobloch,  1941).  The  evaluation  of  carbon 
sources  is  complicated  by  the  metallic  elements  they  contain,  especially 
iron  and  manganese.  Methods  of  treating  beet  and  cane  molasses  to 
remove  inhibiting  impurities  are  described  by  Perlman  et  al.  (1946), 
Gerhardt  et  al.  (1946),  and  Karow  and  Waksman  (1947).  The  inhibiting 
effect  of  metallic  ions  on  the  production  of  citric  acid  from  sugars  is  illus- 
trated by  the  data  in  Table  49. 

Table   49.     The   Effect   of  Removing   Metallic   Contaminants  from  Three 
Sugars,  by  the  Process  of  Cationic  Exchange,  on  the  Prodlction  of 
Citric  Acid  by  Aspergillus  niger,  Wisconsin  Strain  62 
(Perlman  et  al.,  Arch.  Biochem.  11,  1943.     Published  by  permission  of  Academic 
Press,  Inc.) 


Sugar  used 

Treatment 

Yield* 

of  citric  acid,  % 

Sucrose  from  cane 

Not  treated 

21.4 

Treated 

64.0 

Sucrose  from  beet 

Not  treated 

11.3 

Treated 

66.8 

Glucose 

Not  treated 

20.5 

Treated 

60.0 

*  Theoretical  yield  123  per  cent. 


The  production  of  citric  acid  in  submerged  culture  was  tried  at  an  early 
date  and  abandoned  in  favor  of  surface  culture.     However,  recent  litera- 


280  PHYSIOLOGY  OF  THE  FUNGI 

ture  indicates  that  submerged  culture  may  be  the  preferred  process  in 
the  future.  Average  yields  of  72  g.  of  anhydrous  citric  acid  per  100  g. 
of  sucrose  in  the  medium  have  been  obtained  in  the  laboratory  (Shu 
and  Johnson,  1948). 

The  formula  for  citric  acid  is  given  below: 

CH2— COOH 

I 
HO— CH— COOH 

CH2— COOH 

Citric  acid 

Any  theory  of  citric  acid  formation  must  take  into  account  the  following 
facts:  Citric  acid,  a  branched-chain  compound,  is  synthesized  from  carbon 
sources  containing  from  two  to  seven  carbon  atoms.  Yields  of  citric 
acid  may  approach  90  per  cent  of  the  sugar  used  (Wells  et  al.,  1936). 
The  amount  of  carbon  dioxide  evolved  is  low,  which  suggests  either 
reutilization  or  a  mechanism  of  producing  the  necessary  intermediates 
without  the  production  of  carbon  dioxide.  Reutilization  of  carbon 
dioxide  seems  the  more  probable,  for  Foster  et  al.  (1941)  showed  Asper- 
giUus  niger  to  utilize  radioactive  carbon  dioxide  in  the  synthesis  of  citric 
acid.  The  more  probable  pathway  of  synthesis  is  via  the  Krebs  cycle 
(Chap.  7)  and  the  supplementary  formation  of  oxalacetic  from  pyruvic 
acid  and  carbon  dioxide  (Wood-Werkman  reaction). 

Gluconic  acid.  A  considerable  number  of  fungi  produce  gluconic  acid. 
These  include  Aspergillus  niger  (various  isolates),  A.  fuscus,  A.  cinna- 
momeus,  A.  oryzae,  Penicillium  glabrum,  P.  glaucum,  P.  purpurogenum 
var.  ruhrisclerotium,  P.  chrysogenum,  P.  crustaceum,  and  Fumago  vagans. 
Most  investigators  have  used  selected  isolates  of  A.  niger  for  the  produc- 
tion of  gluconic  acid.  Details  of  laboratory  and  semi-pilot-plant  investi- 
gations may  be  found  in  the  papers  of  Wells  et  al.  (1937),  Gastrock  et  al. 
(1938),  and  Forges  et  al.  (1941). 

Many  factors  influence  the  formation  of  gluconic  acid.  Isolates  of 
A.  niger  differ  in  ability  to  synthesize  this  acid.  Not  all  isolates  produce 
the  maximum  amount  of  acid  under  identical  conditions.  Adequate 
aeration  is  necessary  for  the  enzymatic  conversion  of  glucose  to  gluconic 
acid.  Gluconic  acid  is  produced  most  abundantly  when  the  pH  of  the 
medium  is  kept  near  5.  Calcium  carbonate  is  used  for  neutralizing  the 
gluconic  acid  formed.  This  is  advantageous,  for  calcium  gluconate  is 
used  in  medicine  as  a  source  of  readily  assimilable  calcium.  Frecipitation 
of  calcium  gluconate  during  formation  may  be  prevented  by  the  addition 
of  boric  acid  or  borax  to  the  culture  medium  in  amounts  vaiying  up  to 
2,000  p. p.m.  (Moyer  et  al.,  1940).  Boron  compounds  are  added  after 
the  growth  of  mycelium  is  essentially  complete.  The  mycelium  may  be 
used  as  many  as  thirteen  times  by  removing  the  spent  medium  and  adding 


METABOLIC  PRODUCTS  281 

fresh  medium  with  a  high  glucose  content  but  low  in  other  nutrients 
(Forges  et  al,  1940,  1941). 

The  production  of  gluconic  acid  appears  to  be  a  direct  oxidation  of 
glucose.  The  enzyme  responsible  for  this  transformation  is  called  glucose 
aerodehydrogenase.  This  enzyme,  when  free  from  catalase,  catalyzes  a 
reaction  between  glucose  and  oxygen.  Gluconic  acid  and  hydrogen 
peroxide  are  the  products  formed.  Glucose  aerodehydrogenase  was  first 
isolated  from  Penicillium  chrysogenum  and  was  called  notatin,  or  penicil- 
lin B,  at  first.  Its  antibiotic  activity  is  due  to  liberation  of  hydrogen 
peroxide.  For  recent  papers  on  this  enzyme  see  Keilin  and  Hartree 
(1948,  1948a). 

Lactic  acid.  Various  lactic  acid  bacteria  are  used  in  the  commercial 
production  of  lactic  acid.  These  bacteria  require  a  complex  natural 
medium,  which  makes  the  purification  of  lactic  acid  laborious.  Many 
species  of  Phycomycetes  produce  lactic  acid,  and  species  of  Rhizopus  are 
noteworthy  in  this  respect.  The  following  fungi  produce  lactic  acid: 
Rhizopus  arrhizus,  R.  chinensis,  R.  elegans,  R.  japonicus,  R.  nodosus,  R. 
oryzae,  R.  pseudodiinensis,  R.  salehrosus,  R.  shanghaiensis,  R.  stolonifer, 
R.  tritici,  Mucor  rouxii,  Monilia  tamari,  and  Blastocladia  pringsheimii. 
Most  of  these  fungi  appear  to  synthesize  c?-lactic  acid,  although  R. 
chinensis  synthesizes  Wactic  acid  (Saito,  1911). 

The  use  of  R.  oryzae  for  production  of  lactic  acid  has  been  intensively 
investigated  (Lockwood  et  al.,  1936;  Ward  et  at.,  1936,  1938).  Glucose 
appears  to  be  the  best  sugar.  Nitrate  nitrogen  is  not  used  by  this  fungus. 
Calcium  carbonate  is  used  in  the  medium  to  neutralize  lactic  acid  as  it  is 
formed.  Yields  increase  when  the  cultures  are  aerated.  As  much  as 
75  per  cent  of  the  glucose  utilized  is  converted  into  lactic  acid.  The 
presence  of  added  zinc  increases  mycelial  growth  but  depresses  the  yield 
of  lactic  acid. 

The  mechanism  of  lactic  acid  production  by  fungi  is  ably  discussed  by 
Foster  (1949).  Under  anaerobic  conditions,  ethyl  alcohol,  carbon 
dioxide,  and  lactic  acid  are  produced  in  equimolecular  amounts.  The 
amount  of  lactic  acid  produced  under  aerobic  conditions  increases,  while 
the  amount  of  alcohol  decreases  (Waksman  and  Foster,  1939).  The 
most  probable  intermediate  for  the  production  of  lactic  acid  is  pyruvic 
acid. 

Itaconic  acid.  Aspergillus  itaconicus  was  the  first  fungus  reported  to 
synthesize  itaconic  acid.  The  structural  formula  below  shows  that  this 
unsaturated  acid  is  related  to  succinic  acid. 

CH2=C— COOH 

HoC— COOH 
Itaconic  acid 


282  PHYSIOLOGY  OF  THE  FUNGI 

The  fungi  which  have  been  tested  for  itaconic  acid  production  are  mainly 
selected  isolates  of  A.  terreus.  Relatively  few  isolates  produce  sufficient 
itaconic  acid  to  have  commercial  possibilities  (Calam  et  al.,  1939;  Moyer 
and  Coghill,  1945). 

Various  attempts  have  been  made  to  produce  mutants  of  A.  terreus 
by  irradiating  conidia  with  ultraviolet  light  (Raper  et  al.,  1945).  Less 
success  attended  these  efforts  than  comparable  treatment  of  conidia  of 
Penicillium  chrysogenum  for  obtaining  mutants  with  enhanced  penicillin 
production. 

Among  the  factors  which  affect  the  production  of  itaconic  acid  by 
isolates  of  .4.  terreus  are  the  composition  of  the  medium,  hydrogen-ion 
concentration,  temperature,  and  aeration.  Glucose  and  ammonium 
nitrate  appear  to  be  the  best  sources  of  carbon  and  nitrogen.  The  pH 
range  in  which  itaconic  acid  accumulates  is  narrow  and  low,  1.9  to  2.3. 
The  aluminum  ion  is  toxic  to  A.  terreus,  but  aluminum  trays  may  be  used 
if  the  concentration  of  magnesium  ion  in  the  medium  is  high.  As  much 
as  4.75  g.  of  magnesium  sulfate  heptahydrate  per  liter  of  medium  may  be 
used.  It  is  probable  that  this  high  concentration  of  magnesium  ion  also 
enables  the  fungus  to  withstand  low  pH  values  (Lockwood  and  Ward, 
1945). 

Fumaric  acid.  This  unsaturated,  four-carbon,  dicarboxylic  acid  is 
produced  by  many  fungi,  although  only  a  relatively  few  species  synthe- 
size large  amounts.  With  few  exceptions,  the  fungi  which  synthesize 
fumaric  acid  in  significant  amounts  are  Phycomycetes.  The  formula  for 
fumaric  acid  is  given  below: 

HOOC— CH 

II 
HC— COOH 

Fumaric  acid 

The  factors  which  affect  the  production  of  fumaric  acid  by  Rhizopus 
nigricans  were  studied  by  Foster  and  Waksman  (1939).  The  concentra- 
tion of  zinc  was  found  to  be  especially  important.  Optimum  production 
of  fumaric  acid  occurred  in  cultures  receiving  less  zinc  than  that  required 
for  optimum  growth.  Not  all  isolates  of  R.  nigricans  synthesized  fumaric 
acid  in  equal  amounts  or  under  the  same  conditions.  One  isolate  studied 
by  Foster  and  Waksman  (1939a)  produced  fumaric  acid  anaerobically 
and  aerobically,  whereas  another  produced  fumaric  acid  aerobically  only. 

Various  proposals  have  been  made  to  explain  the  mechanism  of  fumaric 
acid  formation.  Anaerobic  synthesis  is  thought  to  involve  the  formation 
of  oxalacetic  acid  from  pyruvic  acid  and  carbon  dioxide  (Foster  and 
Davis,  1948).  The  follo^^•ing  steps  would  convert  oxalacetic  acid  to 
fumaric  acid:  oxalacetate  — ^  malate  — >  fumarate.  It  is  probable  that 
fumaric  acid  is  produced  aerobically  from  acetic  acid  as  follows:  2  (ace- 


METABOLIC  PRODUCTS  283 

cate) —>  succinate —>  fumarate  (Thunberg-Wieland  condensation).  R. 
nigricans  produces  high  yields  of  fumaric  acid  from  both  ethyl  alcohol 
and  acetic  acid,  which  is  evidence  in  favor  of  this  scheme  of  formation 
(Foster  and  Waksman,  1939). 

Other  organic  acids.  Apparently,  the  first  organic  acid  to  be  dis- 
covered as  a  product  of  fungus  metabolism  was  oxalic  acid.  Many  fungi 
in  nature  contain  calcium  oxalate  crystals.  This  was  noted  as  early  as 
1887  by  De  Bary.  Many  species  of  Aspergillus  and  Penicillium  produce 
large  amounts  of  oxalic  acid,  especially  if  enough  alkali  is  present  in  the 
medium  to  convert  the  acid  into  an  oxalate.  Many  species  of  Aspergillus 
which  produce  oxalate  in  the  presence  of  a  neutralizing  agent  also  produce 
citric  acid  in  acid  media  (Currie,  1917).  For  a  recent  discussion  of  oxalic 
acid  production  by  fungi  see  Foster  (1949). 

Various  species  of  Aspergillus,  including  .4.  oryzae,  A.  flavus,  A.  nidu- 
lans,  A.  giganteus,  and  some  other  fungi  produce  kojic  acid.  Kojic  acid 
is  a  cyclic  compound,  a  pyrone,  and  has  been  shown  to  have  antibiotic 
properties  (Morton  et  al.,  1945). 

ESTERS 

Among  the  esters  reported  to  be  formed  by  fungi  are  ethyl  acetate, 
methyl  cinnamate,  methyl  p-methoxycinnamate,  and  isobutyl  acetate. 
Various  reports  are  in  the  literature  concerning  a  "banana-oil"  odor 
being  produced  by  fungi,  but  apparently  amyl  acetate  has  not  been 
isolated  and  identified  as  a  product  of  fungus  metabolism.  Ethyl  acetate 
is  produced  by  Penicillium  digitatum  (Birkinshaw  et  al.,  1931)  and  by 
Endoconidiophora  moniliformis  (Gordon,  1950). 

ANTIBIOTICS  AND  DRUGS 

The  inhibition  of  one  organism  by  another  is  called  antagonism.  The 
phenomenon  has  been  known  since  the  time  of  Pasteur,  and  the  subject 
has  been  reviewed  by  Waksman  (1947)  in  a  book  containing  over  1,000 
references.  Antagonism  occurs  in  nature  as  well  as  in  the  laboratory 
and  is  of  such  common  occurence  that  it  is  frequently  overlooked.  Exam- 
ples are  easily  found  by  examining  contaminated  plates  for  clear  areas 
around  the  contaminants.  Antagonism  may  be  due  to  competition  for 
nutrients  or  to  toxic  substances.  This  discussion  will  deal  wdth  the  toxic 
substances  produced  by  fungi  which  inhibit  fungi  and  bacteria. 

General  discussion.  Fungi  and  other  organisms  produce  a  variety 
of  toxic  substances,  which  include  enzymes,  alkaloids,  toxins,  simple  and 
complex  organic  compounds,  and  inorganic  compounds.  Organic  com- 
pounds produced  by  fungi  and  other  organisms,  especially  bacteria  and 
actinomycetes,  which  inhibit  the  life  processes  of  microorganisms  are 
called  antibiotics.     Waksman  (1947)  would  restrict  the  term  antibiotic 


284  PHYSIOLOGY  OF  THE  FUNGI 

to  organic  compounds  produced  by  microorganisms  which  inhibit  the 
functioning  of  other  microorganisms.  General  usage  of  the  term  anti- 
biotic is,  however,  wider  than  this  and  appHes  the  term  to  those  organic 
compounds  of  fairly  simple  structure  produced  by  organisms  which  inhibit 
microorganisms.  These  substances  are  referred  to  more  specifically  as 
antibacterial,  antifungal,  or  antiviral  substances. 

There  are  no  universal  antibacterial  or  antifungal  substances.     Anti- 
biotics are  specific  in  action.     Penicillin,  for  example,  is  active  against 


A 

Fig.  54.  Method  of  assay  for  antibiotics.  A,  control  culture  of  Penicillium  notatum 
on  agar  medium;  radial  series  of  plugs  cut  at  6  days.  B,  agar-plug  assay  plate  show- 
ing zones  of  inhibition  of  Staphylococcus  developed  after  agar  blocks  removed  from  A 
have  been  incubated  for  16  hr.  at  37°C.  (Courtesy  of  Raper,  Alexander,  and  Coghill, 
Jour.  Bad.  48:  644,  1944.  Published  by  permission  of  The  Williams  &  Wilkins 
Company.) 

many  Gram-positive  bacteria  and  only  a  relatively  few  Gram-negative 
organisms. 

The  occurrence  of  antibiotics  is  probably  far  more  widespread  than 
suspected  at  present.  The  reason  for  this  lies  in  the  way  in  which  anti- 
biotics are  discovered.  Antibiotics  are  detected  by  their  inhibiting  action 
on  living  organisms.  A  susceptible  test  organism  is  essential  for  the 
detection  of  an  antibiotic.  For  obvious  reasons,  human  pathogenic 
bacteria  are  most  used  for  screening  tests.  If  one  desires  to  obtain  anti- 
fungal substances  active  against  pathogenic  fungi,  these  fungi  should  be 
used  as  test  organisms. 

The  same  principle  underlies  all  methods  for  detecting  antibiotic  action. 
The  test  organisms  are  brought  into  contact  with  the  products  elaborated 
by  the  organism  suspected  of  producing  an  antibiotic.  This  may  be  done 
by  growing  two  organisms  on  the  same  Petri  dish.     A  clear  zone  between 


METABOLIC  PRODUCTS  285 

the  colonies  indicates  inhibition  (Fig.  45).  A  second  method  consists 
in  growing  an  organism  on  agar  and  cutting  radially  a  series  of  agar  plugs 
and  placing  these  agar  disks,  which  contain  the  antibiotic,  on  agar  plates 
sown  uniformly  with  the  test  organism  (Raper  et  al.,  1944).  This  method 
is  illustrated  in  Fig.  54.  Other  methods  of  detecting  antibiotics  have 
been  summarized  by  Waksman  (1947). 


Fig.  55.     The  antibiotic  effect  of  Streptomyces  sp.  on  two  plant  pathogenic  fungi, 
Monilinia  fructicola,  on  the  left,  and  Helminthosporiuin  sativum,  on  the  right. 

The  production  of  antibiotic  substances  by  fungi  is  common.  In  a 
screening  test  of  over  400  species,  which  included  over  300  wood-inhabit- 
ing fungi  and  22  dermatophytes,  somewhat  over  200  species  produced 
substances  active  against  Staphylococcus  aureus  and  Escherichia  coli 
(Robbins  et  al.,  1945).  A  large  number  of  Basidiomycetes  and  other 
fungi  have  been  tested  for  the  presence  of  antibiotics  by  Wilkins  and 
Harris  (1944).  The  actinomycetes  are  the  source  of  many  useful  anti- 
biotics including  streptomycin,  Chloromycetin,  aureomycin,  terramycin, 
and  other  unidentified  compounds  (Waksman,  1947).  With  the  excep- 
tion of  Phytophthora  erythroseptica  none  of  the  Phycomycetes  appear  to 
have  been  reported  as  producing  antibacterial  substances.  For  a  survey 
of  Fungi  Imperfecti  in  the  role  of  producing  antibacterial  substances 
(against  Staphylococcus  aureus)  and  antifungal  substances  (against 
Botrytis  allii),  see  Brian  and  Hemming  (1947).  The  inhibiting  effect  of 
Streptomyces  sp.  on  two  plant  pathogenic  fungi  is  shown  in  Fig.  55. 

Many  soil  organisms  produce  antibiotics.  Whether  these  organisms 
produce  antibiotics  in  sufficient  amounts  to  inhibit  plant  pathogens  under 
natural  conditions  in  the  soil  is  not  certain.     It  is  known,  however,  that 


286 


PHYSIOLOGY  OF  THE  FUNGI 


the  incidence  of  certain  diseases  may  be  decreased  by  adding  certain 
bacteria,  actinomycetes,  and  fungi  to  soil.  For  references,  see  Grossbard 
(1948),  Henry  (1931),  Waksman  (1937),  and  Anwar  (1949). 

The  influence  of  various  soil-inhabiting  organisms  in  decreasing  infec- 
tion of  barley  by  Helminthos'porium  sativum  has  been  reported  by  Anwar 
(1949).  Figure  56  illustrates  some  of  these  results.  It  is  by  no  means 
certain  that  these  effects  were  due  to  the  antifungal  substances  produced 
by  the  antagonistic  organisms. 


Fig.  56.  The  effects  of  certain  soil  organisms  on  the  pathonogcnicit}^  of  Helmintho- 
s'porium sativum  on  barley.  Seedlings  grown  at  80°F.  in  steamed  soil  infested  with: 
A,  no  organisms;  B,  H.  sativum  and  Bacillus  subtilis:  C,  H.  sativum  and  Penicillium  sp. ; 
D,  H.  sativum  and  Trichoderma  lignorum:  E,  H.  sativum.  (Courtesy  of  Anwar, 
Phytopathology  39:  1011,  1949.) 

The  situation  in  soil  is  very  complicated.  Basic  antibiotics  such  as 
streptomycin  are  adsorbed  on  clay;  acidic  antibiotics  like  clavacin  are 
apparently  held  less  firmly.  Gottlieb  and  Siminoff  (1950)  are  of  the 
opinion  that  competition  is  more  of  a  factor  than  antibiotic  action  as  the 
cause  of  one  organism  inhibiting  another  in  the  soil.  Thus,  either  As-per- 
gillus  clavatus  or  Streptomyces  griseus  inhibits  the  growth  of  Bacillus 
suhtilis  in  soil.  No  difference  was  noted  between  a  strain  of  S.  griseus 
which  produced  streptomycin  and  one  which  did  not. 

Schatz  and  Hazen  (1948)  reported  that  124  of  the  243  soil  Actiyiomyces 


METABOLIC  PRODUCTS 


287 


tested  were  antagonistic  to  four  test  human  pathogens,  Candida  albicans, 
Cryptococciis  neoformans,  Trichophyton  gypseum,  and  T.  rubrum. 

Table  50.     Antibiotics  Produced  by  Soil-inhabiting  Actinomycetes  and  Fungi 
(Brian,  Cheni.   Industry  1949.     Published  by  permission  of  the  Society  of  Chemical 

Industry.) 


Organism 


Streptomyces  griseus 


Nocardia  gardneri 
Actinomyces  lavendulae 

Proactinomyces  cyaneus 

Streptomyces  venezuelae 

Aspergillus  jlavus 

A.  terreus 

Fusarium  orthoceras 

Penicillium  brevi-compactum 

P.  chrysogenum 

P.  griseofulvum 
P.  janczewiskii 
P.  patulum 

Trichoderma  viride 


Antibiotic 


Grisein 


Actidione 
Streptomycin 

Proactinomycin 
Streptothricin 

Litmocidin 

Chloromycetin 


Aspergillic  acid 

Citrinin 

Enniatin  B 

Mycophenolic  acid 

Penicillin 

Griseofulvin 
Griseofulvin 
Patulin 

Gliotoxin 
Viridin 


Properties 


Antibacterial  (Gram  positive  and 
negative) ;  antirickettsial ;  not 
antifungal 

Not  antibacterial;  antifungal 

Antibacterial  (Gram  positive  and 
negative  and  acid  fast) 

Antibacterial  (Gram  positive) 

Antibacterial  (Gram  positive) ; 
antifungal 

Antibacterial  (Gram  positive  and 
negative) 

Antibacterial  (Gram  positive  and 
negative);  not  antifungal;  anti- 
rickettsial 

Antibacterial  (Gram  positive  and 
negative);  antifungal 

Antibacterial  (Gram  positive  and 
negative);  antifungal 

Antibacterial  (Gram  positive  and 
acid  fast) ;  not  antifungal 

Antibacterial  (Gram  positive  and 
negative);  antifungal 

Antibacterial  (Gram  positive) ;  not 
antifungal 

Not  antibacterial;  antifungal 

Not  antibacterial;  antifungal 

Antibacterial  (Gram  positive  and 
negative) ;  antifungal 

Antibacterial  (Gram  positive  and 
negative  and  acid  fast) ;  anti- 
fungal 

Not  antibacterial;  antifungal 


A  list  of  antibiotics  produced  by  some  soil-inhabiting  actinomycetes 
and  fungi  is  given  in  Table  50.  Note  that  some  organisms  produce  more 
than  one  antibiotic  and  that  the  same  antibiotic  substance  may  be  pro- 
duced by  more  than  one  species.  Organisms  differ  in  susceptibility  to 
antibiotics.  This  range  of  effectiveness  is  frequently  called  the  antibioHc 
spectrum.  Thus,  Penicillum  luteum-purpurogenum  is  some  12  thousand 
times  as  sensitive  to  gliotoxin  as  to  streptomycin.  Not  all  fungi  are 
equally  inhibited  by  the  same  concentration  of  an  antibiotic;  some  11 


288  PHYSIOLOGY  OF  THE  FUNGI 

times  as  much  clavacin  is  required  to  inhibit  the  growth  of  Aspergillus 
clavatus  as  Trichophyton  mentagrophytes  (Reilly  et  al.,  1945). 

Fungi  produce  substances  which  are  capable  of  inactivating  certain 
plant  viruses.  The  Basidiomycetes  are  especially  noteworthy  in  this 
respect  (Utech  and  Johnson,  1950).  Extracts  of  Trichotheciiim  roseum 
reduce  infectivity  of  southern  bean  mosaic,  tobacco  mosaic,  and  tobacco 
necrosis  viruses  (Gupta  and  Price,  1950).  These  authors  believe  that 
this  reduced  infectivity  is  due  to  increased  resistance  of  the  host.  There 
is  no  evidence  which  indicates  that  any  of  the  known  antibiotics  are 
involved  in  the  destruction  of  plant  viruses.  However,  antibiotics  are 
known  which  are  effective  against  virus  diseases  in  man. 

Preliminary  studies  indicate  that  certain  antibiotics  may  be  used  to 
control  fungi  which  cause  plant  diseases.  Actidione  has  been  reported 
by  Vaughn  et  al.  (1949)  to  control  powdery  mildew  on  beans  and  roses. 
Actidione  was  toxic  to  young  rose  leaves  at  a  concentration  of  2.5  p.p.m. 
but  less  toxic  to  bean  plants.  Laboratory  tests  indicated  that  actidione 
is  a  fungistatic  substance  for  a  considerable  number  of  plant  pathogenic 
fungi,  including  Sclerotinia  fructicola,  Cladosporium  cucumerinum,  and 
Colletotrichum  lagenarium.  Further  data  on  the  effect  of  actidione  on 
plant  pathogenic  fungi  are  reported  by  Whiffin  (1950). 

The  protective  action  of  an  antibiotic  obtained  from  an  unidentified 
species  of  Streptomyces  against  Venturia  inaequalis  on  apple  has  been 
reported  by  Leben  and  Keitt  (1949).  This  antibiotic  has  been  named 
antimycin. 

Penicillin  has  been  used  successfully,  to  a  limited  extent,  in  controlling 
necrosis  of  giant  cactus,  caused  by  Erwinia  carnegieana  (Boyle,  1949). 
Injections  of  penicillin  into  the  necrotic  tissue  apparently  diffused  through 
the  plant  tissues  for  some  distance,  killing  the  bacteria.  This  is  one  of 
the  few  cases  in  which  an  antibiotic  has  been  used  successfully  in  thera- 
peutic treatment  of  plant  disease. 

The  principal  use  of  antibiotics  is  to  control  disease  in  man  and  animals. 
Only  a  relatively  few  antibiotics  are  useful  for  this  purpose.  In  addition 
to  killing  or  inhibiting  pathogenic  organisms,  an  antibiotic,  to  be  useful 
in  medicine,  must  be  relatively  nontoxic  to  the  host.  Some  of  the  older 
and  more  useful  antibiotics  used  in  medicine  will  be  discussed  in  greater 
detail  on  the  following  pages. 

Penicillin.  This  antibiotic  drug  is  produced  in  industry  by  selected 
isolates  or  mutants  of  Penicillium  chrysogcnum  and  P.  notatum.  The 
original  isolate  of  Fleming  produced  from  2  to  4  units  of  penicillin  per 
milliliter  of  culture  filtrate.  P.  chrysogenum  Q-17Q  has  produced  in 
excess  of  1,000  units  per  milliliter.  The  synthesis  of  penicillin  is  not 
limited  to  species  of  the  P.  chrysogenum-notatum  group  but  includes  cer- 
tain species  of  Aspergillus  belonging  to  the  A.  flavus  group.     A  few  fungi 


METABOLIC  PRODUCTS 


289 


belonging  to  other  genera  also  produce  penicillin.  Sterile  culture  condi- 
tions must  be  maintained  at  all  times,  as  penicillin  is  rapidly  destroyed 
by  the  enzyme,  penicillinase,  which  is  excreted  by  many  bacteria.  Peni- 
cillin is  extracted  from  the  "fermentation"  liquid,  or  penicillin  beer, 
either  by  extraction  at  pH  2.0  to  2.5  with  water-immiscible  solvents  such 

S 


R— 


— CO— NH— CH- 
0=0— 


-CH  C— (CH3)2 

-N CH— COOH 


Type  formula  of  the  penicillins 

as  amyl  acetate,  or  by  adsorption  on  activated  carbon.  The  extraction 
must  be  carried  out  quickly  from  acid  solutions,  owing  to  the  instability 
of  penicillin  under  these  conditions.  Penicillin  forms  crystalline  salts 
with  the  alkali  metals.     The  sodium  salt  is  usually  produced. 

Table   51.     Names   and   American   and   British-type   Designations   of   Four 

Naturally  Occurring  Penicillins 


R 

Name 

Type 

American 

British 

Benzylpenicillin 
A^-Pentenylpenicillin 
p-HydroxybenzylpenicillLn 
n-Heptylpenicillin 

G 
F 
X 
K 

II 

CHs- CHo— CH=CH— CHo— 

HOC6H4CH2— 

(j-H.3(0-H.'>)5 — C-H.'> — 

I 

III 

IV 

Penicillin,  as  produced  by  P.  chrysogenum,  is  a  mixture  of  related  com- 
pounds. The  ratios  among  the  various  penicillins  depends  upon  the 
isolate  and  conditions  used.  By  the  use  of  suitable  precursors  the  yield 
of  the  desired  compound,  penicillin  G,  is  greatly  increased.  This  is 
desirable  because  penicillin  G  salts  crystallize  well  and  are  most  useful 
in  medicine.  The  type  formula  for  the  penicillins  is  shown  below.  The 
precursors  used  for  the  production  of  penicillin  G  are  related  to  phenyl- 
acetic  acid.  Many  other  penicillins  are  known,  some  of  which  are  pro- 
duced only  in  the  presence  of  precursors  which  do  not  occur  in  nature 
(Behrens,  1949). 

The  growth  of  the  penicillin  industry  in  the  United  States  is  shown  by 
the  production  figures  in  Table  52. 

Penicillin  is  chiefly  active  against  Gram-positive  bacteria.  A  few 
important  Gram-negative  pathogens,  including  Neisseria  gonorrhoeae  and 
Treponema  pallidum,  are  controlled  by  penicillin.  Penicillin  is  not  active 
against  acid-fast  bacteria  or  fungi.  Many  bacteria  may  become  resist- 
ant, or  fast,  to  penicillin.  Whether  natural  selection  or  mutation  or  both 
are  involved  in  this  phenomenon  is  uncertain.     The  morphology  and 


290  PHYSIOLOGY  OF  THE  FUNGI 

physiology  of  penicillin-fast  bacteria  vao^y  be  abnormal  (Bellamy  and 
Klimek  1948).  Penicillin  is  most  active  against  young  cells,  in  that  it 
inhibits  the  process  of  cell  division.  For  papers  on  the  mechanism  of 
penicillin  action  see  Cavallito  et  al.  (1945),  Chain  and  Diithie  (1945); 
Bailey  and  Cavallito  (1948) ;  and  a  series  of  papers  by  Pratt  and  Dufrenoy 
(1949). 

Table  52.     The  Production  of  Penicillin  in  the  United  States  for  the  Years 

1943  TO  1948 
A  unit  of  penicillin  is  0.6  ^g.      (Coghill  and  Koch,  Chem.  Eng.  Neivs  23,  1045;  Lee, 
Ind.   Eng.   Chem.   41,    1949.     Published  by  permission   of  the   American   Chemical 
Society.) 

Year     Billions  of  Units 


1943 

21 

1944 

1,633 

1945 

6,852 

1946 

25,809 

1947 

41,426 

1948 

95,855 

Further  details  may  be  found  in  the  following  selected  references.  For 
a  concise  authoritative  account  of  all  phases  of  penicillin,  see  Foster 
(1949).  The  medical  aspects  of  penicillin  therapy  are  discussed  by 
Fleming  (1949).  The  chemistry  of  penicillin  is  covered  in  the  monograph 
edited  by  Clarke  et  al.  (1949).  The  early  history  of  penicillin  is  pre- 
sented by  Chain  and  Florey  (1944)  and  Waksman  (1947). 

Streptomycin.  This  antibiotic  was  discovered  in  Waksman's  labora- 
tory in  1943,  and  three  years  later,  commercial  production  of  this  drug 
began.  Streptomycin  is  synthesized  by  some  isolates  of  Streptomyces 
griseus.  The  techniques  used  in  industry  resemble  those  used  for  the 
production  of  penicillin  in  submerged  aerated  culture.  Streptomycin 
is  adsorbed  on  activated  carbon  as  the  first  step  in  isolation  and  purifica- 
tion. In  contrast  to  penicillin,  streptomycin  is  a  basic  compound.  The 
production  of  streptomycin  in  the  United  States  increased  from  1,175 
billion  units  in  1946  to  37,710  billion  units  in  1948  (Lee,  1949).  The 
chemistry  of  streptomycin  is  reviewed  by  Lemieux  and  Wolfrom  (1948). 

Streptomycin  is  mainly  active  against  Gram-negative  bacteria  and 
certain  acid-fast  organisms,  including  Mycohacterium  tuberculosis.  This 
drug  controls  many  pathogens  which  are  unaffected  by  penicillin.  Organ- 
isms exposed  to  streptomycin  frequently  become  fast.  Indeed,  some 
bacteria  have  been  reported  to  become  dependent  upon  the  drug. 

The  composition  of  the  medium  influences  streptomycin  production. 
Soybean  meal  appears  to  be  a  suitable  source  of  nitrogen  for  commercial 
production  (Rake  and  Donovick,  1946).  The  influence  of  carbon  and 
nitrogen  sources  in  synthetic  media  has  been  studied  by  Dulaney  (1948, 
1949).     These  results  may  be  summarized  as  follows:  Glucose  and  man- 


METABOLIC  PRODUCTS  291 

nose  are  the  best  hexoses;  maltose  is  the  best  disaccharide ;  starch  and 
its  degradation  product  dextrin  are  good  carbon  sources.  Streptomyces 
griseus  does  not  utihze  nitrate  nitrogen.  L-ProHne  is  the  most  favorable 
single  amino  acid,  but  addition  of  this  amino  acid  to  other  sources  of 
nitrogen  does  not  increase  yields  (see  also  Thornberry  and  Anderson, 
1948).  A  popular  account  of  the  development  of  streptomycin  is  to  be 
found  in  Epstein  and  Williams  (1946). 

Aureomycin.  This  antibiotic  is  produced  by  Streptomyces  aureofaciens. 
Aureomycin  has  been  prepared  in  pure  crystalline  form  (Duggar,  1948). 
In  addition  to  being  effective  against  many  Gram-positive  and  Gram- 
negative  bacteria,  aureomycin  is  also  active  against  certain  rickettsial 
and  viral  agents.  Preliminary  reports  on  the  use  of  aureomycin  for 
treating  Rocky  Mountain  spotted  fever  have  been  favorable  (Schoenbach 
et  al.,  1948).  Aureomycin  appears  to  be  of  great  clinical  value  in  treating 
lymphogranuloma  venereum  and  granuloma  inguinale,  two  virus  diseases 
of  man  (Wright  et  al,  1948). 

Chloromycetin.  This  antibiotic  was  discovered  independently  in  two 
laboratories  (Ehrlich  et  al.,  1947,  and  Carter  et  al.,  1948).  It  is  produced 
by  Strepto77iyces  venezuelae  and  is  the  first  useful  antibiotic  to  be  produced 
synthetically  on  a  commercial  scale.  Chloromycetin  is  active  against 
certain  Gram-positive  and  Gram-negative  bacteria,  acid-fast  bacteria, 
and  Rickettsia  (Smith  et  al.,  1948).  It  has  been  reported  that  this  anti- 
biotic has  been  used  successfully  to  treat  epidemic  typhus  (Payne  and 
Knaudt,  1948). 

Chloromycetin  contains  non-ionic  chlorine  and  a  nitro  group,  two 
unusual  features  in  compounds  of  biological  origin.  The  formula  is  given 
below : 

O  H    NHCOCHClo 


N— <f  >^C— C— CH2OH 

O  HO     H 

Chloromycetin  (Chloramphenicol) 

Ergot.  Among  the  alkaloids  produced  by  fungi,  only  those  obtained 
from  the  sclerotia  of  Claviceps  purpurea  appear  to  be  used  in  medicine. 
Seven  related  alkaloids  have  been  isolated  from  ergot.  These  alkaloids 
have  different  pharmacological  properties.  The  most  useful  alkaloid, 
ergonovine,  is  isolated  from  the  others  before  use.  Certain  undesirable 
effects  of  the  other  alkaloids  are  avoided  by  this  procedure.  Ergonovine, 
and  formerly  a  mixture  of  ergot  alkaloids,  is  used  to  stimulate  uterine 
contraction.  At  present  there  is  no  satisfactory  synthetic  substitute  for 
ergonovine  (Mass,  1950). 

In  1941,  571,000  lb.  and  in  1944,  85,000  lb.  of  ergot  sclerotia  were 
imported  into  the  United  States.     If  the  sclerotia  are  consumed  in  large 


292  PHYSIOLOGY  OF  THE  FUNGI 

enough  quantities  by  man  or  animals,  they  cause  ergotism,  a  disease  also 

known  as  St.  Anthony's  fire. 

C.  purpurea  has  been  cultured  under  laboratory  conditions  but  forms 

neither  sclerotia  nor  alkaloids  under  these  conditions   (Michener  and 

Snell,   1950).     Apparently  alkaloids  are  formed  only  in  the  sclerotia. 

Ergotamine,  when  added  to  mycelial  cultures  of  C.  purpurea,  was  largely 

destroyed. 

TOXINS 

Numerous  toxic  substances  are  produced  by  fungi  in  nature,  and  their 
effects  on  man  and  animals  are  varied.  The  most  severe  toxins  are 
produced  by  some  of  the  Agaricaceae,  particularly  by  species  of  Amanita. 
It  is  not  known  whether  the  toxins  are  present  in  the  mycelium  of  these 
species  as  well  as  in  the  fruit  bodies.  A  few  of  the  outstanding  examples 
of  fungus  toxins  will  be  discussed  briefly. 

Amanita  toxin  (phalloidin)  is  stable  to  heat  and  drying  and  to  the  action 
of  digestive  juices.  The  great  majority  of  the  deaths  due  to  mushroom 
poisoning  are  caused  by  Amanita  phalloides,  A.  virosa,  and  A.  verna, 
which  contain  amanita  toxin.  The  action  of  this  toxin  is  slow,  the  symp- 
toms being  delayed  for  6  to  15  hr.  after  the  mushrooms  are  eaten.  By 
this  time  the  toxin  has  been  absorbed,  and  the  patient  seldom  responds 
to  treatment.  No  antidote  for  this  toxin  is  known.  The  mortality  rate 
is  high,  varying  from  60  to  100  per  cent  (Fischer,  1918).  In  addition  to 
the  three  species  of  Amanita  mentioned  above,  the  same  or  a  similar  toxin 
is  present  in  A.  spreta,  A.  porphyria,  A.  strohiliformis,  A.  radicata,  and 
A.  chlorinosma.  Hygrophorus  conicus  and  Pholiota  autumnalis  produce 
similar  symptoms  and  may  contain  this  toxin  (Krieger,  1936). 

About  1  g.  of  pure  crystalline  toxin  can  be  extracted  from  40  kg.  of 
A.  phalloides  fruit  bodies.  The  toxic  dose  for  white  mice  is  50  jug;  death 
results  in  from  1  to  2  days.  Chemically,  phalloidin  is  a  polypeptide 
containing  six  amino-acid  residues.  Wieland  and  Witkop  (1940)  report 
that  phalloidin,  on  hydrolysis  wdth  sulfuric  acid,  yields  1  mole  each  of 
Z-Q!-oxytryptophane  and  cysteine,  and  2  moles  each  of  ^-hydroxy proline  b 
(not  found  in  protein  digests)  and  /-alanine.  Kuhn  et  al.  (1939)  found 
methionine  in  addition  to  cysteine  (ratio  1  to  5)  in  phalloidin.  Among 
the  antibiotics,  gramicidin  and  tyrocidine  are  polypeptides  which  contain 
"unnatural"  amino  acids  and  are  toxic  when  injected  into  experimental 
animals. 

Muscarin  is  the  principal  toxic  agent  present  in  A.  muscaria.  It  is  a 
quick-acting  toxin,  producing  symptoms  within  1  to  6  hr.  after  being 
consumed.  The  patient  usually  responds  well  to  treatment  and  recovery 
is  rapid,  although  death  may  occur.  Atropin  is  an  antidote  for  muscarin 
which  is  closely  related  to  choline.  Muscarin  has  also  been  demonstrated 
in  A.  pantherina,  Russula  emetica.  Boletus  luridus,  and  B.  satanas.     A 


METABOLIC  PRODUCTS  293 

similar  or  the  same  toxin  is  present  in  Clitocyhe  illudens,  Inocyhe  infelix, 
I.  infida,  Lactarius  torminosus,  and  B.  miniato-olivaceus  var.  sensibilis 
(Krieger,  1936;  Wolf  and  Wolf,  1947). 

HelveUic  acid  is  known  to  be  present  only  in  Gyromitra  esculenta.  Ap- 
parently there  is  considerable  variability  in  the  reaction  of  individuals 
who  eat  this  fungus.  Many  people  have  eaten  it  with  no  ill  effects, 
although  a  number  of  cases  of  poisoning  and  even  a  few  deaths  have  been 
reported.  Helvellic  acid  is  soluble  in  hot  water.  Its  toxic  action  is  due 
to  its  blood-dissolving  power. 

While  a  number  of  other  toxic  substances  have  been  detected  in  the 
fleshy  fungi,  the  exact  identity  of  most  of  them  is  not  known.  For 
example,  species  of  Paneolus  may  cause  temporary  paralysis  or  intoxica- 
tion similar  to  alcoholic  intoxication.  Some  species  of  Amanita  have 
been  reported  to  contain  other  toxins  in  addition  to  those  discussed  above 
(Fischer,  1918). 

The  alkaloids  produced  by  Claviceps  purpurea  and  C.  paspali  are  toxic 
to  man  and  animals  if  consumed  in  large  enough  quantities.  The  specific 
alkaloids  produced  were  mentioned  previously.  The  production  of 
toxins  is  also  common  in  the  genus  Fusarium.  Gihberella  zeae,  the  cause 
of  scab  of  small  grains,  produces  toxic  substances  which  are  poisonous 
to  livestock  fed  on  scabby  grain  (Christensen  and  Kernkamp,  1936). 
Numerous  other  fungi  which  cause  plant  diseases  are  known  to  produce 
toxic  substances  which  kill  the  host  or  modify  its  activity  (see  Chap.  17). 

PIGMENTS 

Colored  compounds  produced  by  fungi  and  other  organisms  are  called 
pigments.  In  the  fungi,  some  pigments  accumulate  in  the  mycelium  and 
spores,  while  others  diffuse  into  the  culture  medium.  The  pigments 
produced  by  a  fungus  are  in  part  determined  by  genetic  factors  and  in 
part  by  the  environment.  Mycelium,  fruit  bodies,  and  spores  may  be 
pigmented,  or  in  some  species  the  pigment  is  confined  to  the  spores. 

Among  the  fleshy  fungi,  brown  is  one  of  the  most  common  colors  of 
fruit  bodies,  with  yellow,  orange,  and  red  being  somewhat  less  common. 
Often  a  number  of  pigments  are  obviously  present.  Few  fungi  are  green. 
Yet,  Chlorosplenium  aeruginosum  produces  a  green  pigment,  sylindein 
(Wolf  and  Wolf,  1947),  which  stains  the  wood  in  which  it  grows.  Blue- 
stain  fungi  (CeratostomeUa  spp.)  excrete  blue  pigments  into  wood.  Some 
species  of  Boletus  produce  a  blue  or  bluish-green  pigment  when  bruised  or 
wounded.  Tricholoma  personatum  and  Laccaria  amythestina  are  among 
the  mushrooms  producing  purple  or  violet  pigments.  It  is  said  that  the 
red-orange  pigment  of  the  fruit  bodies  of  Echinodontium  tinctorium,  the 
Indian  paint  fungus,  was  used  by  the  Indians  as  make-up.     Few  of  the 


294  PHYSIOLOGY  OF  THE  FUNGI 

larger  fruit  bodies  of  the  fungi  are  entirely  black,  although  this  is  a  com- 
mon color  for  perithecia,  pycnidia,  and  spores. 

Among  the  nutritional  factors  which  modify  the  production  of  pigments 
by  fungi  in  culture,  the  micro  essential  elements,  the  carbon  and  nitrogen 
sources,  the  initial  pH  of  the  medium,  and  the  temperature  are  important. 
Perhaps  the  first  of  these  factors  to  be  studied  was  the  effect  of  iron,  cop- 
per, zinc,  and  other  micro  elements  upon  the  spore  color  of  Aspergillus 
niger.  Copper  seems  to  play  an  outstanding  role  in  the  production  of 
dark  spores  by  this  fungus  (Mulder,  1939),  but  low  concentrations  of 
other  micro  essential  elements  also  affect  spore  color  of  this  fungus.  The 
influence  of  iron,  copper,  and  zinc  on  the  pigmentation  of  mycelium  and 
spores,  and  the  production  of  soluble  pigments  by  certain  species  was 
studied  by  Metz  (1930). 

The  investigation  of  the  chemical  structure  of  fungus  pigments  has 
formed  an  essential  part  of  a  comprehensive  study  of  the  products  of 
fungus  metabolism  at  the  University  of  London.  The  citations  in  this 
paragraph  will  give  the  reader  entry  into  this  excellent  work.  Many 
fungi  produce  anthraquinone  pigments.  Helminthosporium  gramineum 
stores  in  its  mycelium  two  pigments  (helminthosporin  and  hydroxyisohel- 
minthosporin),  which  may  account  for  30  per  cent  of  the  dry  weight  of 
the  mycelium.  Helminthosporin  is  2-methyl-4,5,8-trihydroxyanthra- 
quinone  (Charles  et  at.,  1933).  H.  cynodontis  and  H.  euchlaenae  form 
cynodontin,  l,4,5,8-tetrahydroxy-2-methylanthraquinone,  which  is  closely 
related  to  helminthosporin  (Raistrick  et  at.,  1933).  Some  12  anthra- 
quinone pigments  are  produced  by  fungi  (Howard  and  Raistrick,  1949). 
Xanthone  pigments  are  produced  by  H.  ravenelUi  and  H.  turcicum  (Rai- 
strick et  at.,  1936).  The  production  of  anthraquinone  pigments  is  not 
restricted  to  species  of  Helminthosporium,  for  Penicillium  islandicum 
synthesizes  chrysophanic  acid,  4,5-dihydroxy-2-methylanthraquinone 
(Howard  and  Raistrick,  1950).  In  general,  the  production  of  these  and 
other  pigments  is  modified  by  the  cultural  conditions  used.  The  produc- 
tion of  helminthosporin  by  H.  gramineum  was  increased  when  nitrate  or 
organic  sources  of  nitrogen  were  used.  Ammonium  nitrogen  was  not 
favorable  for  pigment  production.  More  pigment  was  produced  when 
the  initial  pH  was  8  than  in  more  acid  media. 

Many  of  the  water-soluble  pigments  produced  by  fungi  are  indicators. 
P.  phoeniceum  and  P.  ruhrum  produce  such  an  indicator  pigment,  phoe- 
nicine  (2,2'-dihydroxy-4,4'-ditoluquinone).  The  color  changes  of  this 
indicator  are  from  yellow  to  red  in  the  pH  range  of  1.8  to  3.4  and  from 
red  to  violet  in  the  range  5.4  to  6.4.  As  much  as  2  g.  of  this  pigment  is 
produced  by  P.  ruhrum  per  liter  of  medium  (Curtin  et  al.,  1940). 

The  functions  of  fungus  pigments  are  not  well  understood.  It  is 
known  that  certain  of  these  pigments  are  enzyme  inhibitors.     Others,  like 


METABOLIC  PRODUCTS  295 

citrinin,  are  antibiotics.  The  physiological  activity  of  solanione,  a  purple 
pigment  produced  by  Fusarium  solani,  decreases  growth  and  the  efficiency 
of  fat  formation  by  F.  lini  (Weiss  and  Nord,  1949).  Solanione  is  a 
1,4-naphthoquinone,  and  this  activity  is  in  accord  with  the  effects  of  other 
compounds  of  this  series.  F.  graminearum  synthesizes  an  orange-red 
pigment,  rubofusarin,  Avhich  is  a  xanthone  (Ashley  et  at.,  1937).  Rubo- 
fusarin  was  found  to  stimulate  growth  and  to  inhibit  the  enzymatic 
dehydrogenation  of  isopropyl  alcohol  by  F.  lini  (Sciarini  et  al.,  1943).  It 
is  suggestive  that  pigment  production  frequently  occurs  near  the  time  of 
maximum  development  of  mycelium.  Perhaps  pigments  influence 
sporulation  in  some  way. 

Carotene  is  produced  by  Mucor  hiemalis  and  Phycomyces  blakesleeanus, 
and  probably  by  Mucor  mucedo,  Pilairia  anomala,  and  Dicranophora  fulva 
(Schopfer,  1935).  The  amount  of  carotene  produced  by  Phycomyces 
blakesleeanus  was  increased  by  increasing  the  concentration  of  asparagine 
in  the  medium.  The  plus  strain  of  this  fungus  synthesized  more  carotene 
than  the  minus  strain.  Carotene  occurs  in  nature  as  three  isomeric 
compounds,  all  of  w'hich  may  be  converted  into  vitamin  A.  The  carotene 
found  in  M.  hiemalis  and  P.  blakesleeanus  is  ;S-carotene.  Emerson  and 
Fox  (1940)  found  7-carotene  to  be  associated  with  the  male  gametangia 
of  a  certain  species  of  Allomyces.  Apparently  carotene  is  also  common 
among  the  yellow  or  orange  Ascomycetes  and  Basidiomycetes. 

SUMMARY 

The  saprophytic  fungi  play  an  important,  if  not  indispensable,  part 
in  the  degradation  and  decay  of  plant  and  animal  residues.  The  most 
important  product  of  fungus  metabolism  in  nature  is  carbon  dioxide. 
Humus  is  in  part  a  result  of  the  activities  of  soil-inhabiting  fungi. 

Various  species  of  fungi  have  been  used  for  the  preparation  of  food  and 
beverages.  Fungi  may  be  used  to  increase  the  world's  supply  of  food. 
Yeasts  and  other  fungi  are  able  to  convert  w-aste  carbohydrate  and 
inorganic  nitrogen  compounds  into  protein,  fats,  and  vitamins.  Yeast 
protein,  because  of  its  content  of  essential  amino  acids,  has  value  as  a 
protein  supplement. 

Alcoholic  fermentation  is  not  restricted  to  the  yeasts,  although  these 
fungi  are  used  almost  exclusively  in  industrj^  for  this  purpose.  The 
production  of  alcohol  requires  anaerobic  or  partially  anaerobic  conditions. 

Fungi  may  be  used  to  produce  organic  acids,  of  which  citric  acid  is  the 
most  important  commercially.  In  general,  a  medium  high  in  carbo- 
hydrate and  low  in  nitrogen  favors  the  production  of  organic  acids,  which 
are  synthesized  in  quantity  only  after  growth  is  essentially  complete. 

In  nature  one  organism  may  be  antagonistic  to  another  because  of  the 
competition  for  nutrients  or  because  of  the  production  of  antibiotics. 


296  PHYSIOLOGY  OF  THE  FUNGI 

Relatively  feAV  of  the  antibiotics  known  to  be  produced  meet  the  pre- 
requisite of  being  nontoxic  to  man,  but  some  of  those  which  do  are  enor- 
mously important. 

Ergonovine,  one  of  the  ergot  alkaloids  obtained  from  the  sclerotia  of 
Claviceps  purpurea,  is  a  useful  drug  for  which  no  synthetic  substitute  is 
available. 

A  number  of  fleshy  fungi  produce  toxins,  some  of  which  are  deadly  to 
man  if  consumed  in  sufficient  quantity.  The  toxins  vary  in  chemical 
nature,  in  severity,  and  in  the  symptoms  they  produce. 

Pigments  are  assumed  to  serve  a  definite  function  in  fungi,  at  least  in 
some  instances.  Some  pigments  are  known  to  be  antibiotics;  others  such 
as  carotene  are  provitamins;  but  in  general  the  functions  of  the  fungus 
pigments  remain  unknown. 

REFERENCES 

Anwar,  A.  A.:  Factors  affecting  the  survival  of  Helrninthosporium  satirum  and 
Fusarium  lini  in  soil,  Phytopathologij  39:  1005-1019,  1949. 

Ashley,  J.  N.,  B.  C.  Hobbs,  and  H.  Raistrick:  Studies  in  the  biochemistry  of 
micro-organisms.  LIII.  The  crystalHne  colouring  matters  of  Fusarium  cul- 
moruyn  (W.  G.  Smith)  Sacc.  and  related  forms,  Biochem.  Jour.  31 :  385-397,  1937. 

Bailey,  J.  H.,  and  C.  J.  Cavallito:  The  reversal  of  antibiotic  action,  Joiir.  Bad. 
55:  175-182,  1948. 

Behrens,  O.  K.  :  Biosynthesis  of  penicillins  in  The  Chemistry  of  Penicillin,  Prince- 
ton University  Press,  Princeton,  N.J.,  1949. 

Bellamy,  W.  D.,  and  J.  W.  Klimek:  Some  properties  of  penicillin-resistant  Staphy- 
lococci, Jour.  Bad.  55:  153-160,  1948. 

Bernhauer,  K.,  and  H.  Knob  loch:  Ueber  die  Saiirebildung  aus  Zucker  durch 
Aspergillus  niger.  XI.  ]\Iitteilung.  Factoren  der  Citronensaureanhaufung  2, 
Biochem.  Zeit.  309:  151-178,  1941. 

BiRKiNSHAW,  H.,  J.  H.  V.  Charles,  and  H.  Raistrick:  Studies  in  the  biochemistry 
of  micro-organisms.  XVIII.  Biochemical  characteristics  of  species  of  Peni- 
cillium  responsible  for  the  rot  of  citrus  fruits,  Trans.  Roy.  Soc.  (London),  Ser.  B, 
220:355-362,  1931. 

Birkinshaw,  J.  H.,  and  H.  Raistrick:  Studies  in  the  biochemistry  of  micro- 
organisms. II.  Quantitative  methods  and  technique  of  investigation  of  the 
products  of  metabolism  of  micro-organisms,  Trans.  Roy.  Soc.  (London),  Ser.  B, 
220:  11-26,  1931. 

Block,  R.  J.,  and  D.  Rolling:  The  amino  acids  yielded  by  various  yeasts  after 
hydrolysis    of    the    fat-free    material.     A    comparative    investigation.    Arch. 
Biochem.  7:  313-321,  1945. 
*Boyle,  a.  M.:  Further  studies  of  the  bacterial  necrosis  of  the  giant  cactus,  Phyto- 
pathology 39 :  1029-1052,  1949. 

Bray,  C.  W.  :  Vitamin  B  deficiency  in  infants:  its  possibility,  prevalence  and  prophyl- 
axis, Trans.  Roy.  Soc.  Trap.  Med.  Hyg.  22 :  9-42,  1928. 

Brefeld:  O.:  Untersuchungen  aus  dem  Gesamtgebeit  der  Mykologie,  Vol.  XIV, 
Kommissions-Verlag  von  Heinrich  Schoningh,  Muenster,  1908. 

Brian,  P.  W.:  The  production  of  antiobiotics  by  soil  microorganisms,  Chem.  & 
Industry,  1949,  391-393. 

Brian,  P.  W.,  and  H.  G.  Hemmimg:  Production  of  antifungal  and  antibacterial 


METABOLIC  PRODUCTS  297 

substances  by  fungi :  Preliminary  examination  of  1G6  strains  of  Fungi  Imperfecti, 

Jour.  Gen.  Microbiol.  1:  158-167,  1947. 
Brink,  N.  G.,  F.   W.  Holly,  C.  H.  Shunk,  E.  W.    Peel,    J.    J.    Cahill,    and 

K.   FoLKfiRs:   Vitamin   B12.     IX.   l-a/p/ia-D-ribofuranosido-.5,6-dimethylimida- 

zole,  a  degradation  product  of  vitamin  B12,  Jour.  Am.  Chem.  Soc.  72,  1866,  1950. 
BucHNER,  E. :  Alcoholische  Garung  ohne  Hefezellen,  Ber.  d.  deut.  chem.  Ges.  30: 

117-124,  1897. 
BuLLER,  A.  H.  R. :  The  fungus  lore  of  the  Greeks  and  Romans,  Trans.  Brit.  Mycol. 

Soc.  5:21-66,  1914. 
BuRKHOLDER,  P.  R.  I  Influence  of  some  environmental  factors  upon  the  production 

of  riboflavin  by  a  yeast,  Arch.  Biochem.  3 :121-129,  1943. 
Calam,  C.  T.,  a.  E.  Oxford,  and  H.  Raistrick:  Studies  in  the  biochemistry  of 

micro-organisms.     LXIII.     Itaconic  acid,  a  metabolic  product  of  a  strain  of 

Aspergillus  terreus  Thom.,  Biochem.  Jour.  33:  1488-1495,  1939. 
Carter,  H.  E,  D.  Gottlieb,    and  H.  W.  Anderson:  Chloromycetin  and  strepto- 

thricin,  Science  107:  113,  1948. 
*Cavallito,   C.  J.,  J.  H.  Bailey,  T.  H.  Haskell,  T.  H.  McCormick,  and  W.  F. 

Warner:  The  inactivation  of  antibacterial  agents  and  their  mechanism  of 

action.  Jour.  Bad.  50:  61-69,  1945. 
Chain,  E.,  and  E.  S.  Duthie:  Bacteriocidal  and  bacteriolytic  action  of  penicillin 

on  the  Staphylococcus,  Lancet  248 :  652-657,  1945. 
Chain,  E.,  and  H.  W.  Florey:  Penicillin,  Endeavour  3:  3-14,  1944. 
Charles   J.  H.  V.,  H.  Raistrick,  R.  Robinson,  and  A.  R.  Todd:  Studies  in  the  bio- 
chemistry   of  micro-organisms.     XXVIII.  Helminthosporin   and   hydroxy/so- 

helminthosporin,  metabolic  products  of  the  plant  pathogen  Helminthosporium 

gramineum  Rabenhorst.,  Biochem.  Jour.  27:  499-511,  1933. 
Christensen,  J.  J.,  and  H.  C.  H.  Kernkamp:  Studies  on  the  toxicity  of  blighted 

barley  to  swine,  Minn.  Agr.  Expt.  Sta.  Tech.  Bull.  113,  1936. 
Clarke,  H.  T.,  J.  R.  Johnson,  and  R.  Robinson  (Editors) :  The  Chemistry  of 

Penicillin,  Princeton  University  Press,  Princeton,  N.J.,  1949. 
CoGHiLL,  R.  D.,  and  R.  S.  Koch:  Penicillin,  a  wartime  accomplishment,  Chem. 

Eng.  Neivs  23:  2310-2316,  1945. 
CuRRiE,  J.  N.:  The  citric  acid  fermentation  of  Aspergillus  niger,  Jour.  Biol.  Chem. 

31:  15-37,  1917. 
CuRTiN,  T.,  G.  Fitzgerald,  and  J.  Reilly:  Production  of  phoenicine  on  synthetic 

media.      1.  Penicillium     phoeniceum     van      Beyma.     2.  Penicillium     ruhrum 

Grassbergcr-Stoll.,  Biochem.  Jour.  34:  1605-1610,  1940. 
DuGGAR,  B.  M.:  Mushroom  Growing,  Orange  Judd  Publishing  Co.,  Inc.,  New  York, 

1915. 
DuGGAR,  B.  M.:  Aureomycin:  a  product  of  the  continuing  search  for  new  antibiotics, 

Ann.  N.Y.  Acad.  Sci.  51:  177-181,  1948. 
Dulaney,  E.  L.  :  Observations  on  Streptomyces  griseus.     III.  Nitrogen  sources  for 

growth  and  streptomycin  production.  Jour.  Bact.  56:  305-313,  1948. 
Dulaney,  E.  L.  :  Observations  on  Streptomyces  griseus.     II.  Carbon  sources  for 

growth  and  streptomycin  production,  Mycologia  41 :  1-10,  1949. 
Ehrlich,  J.,  Q.  R.  Bantz,  R.  M.  Smith,  D.  A.  Joslyn,  and  P.  Bunkholder:  Chlo- 
romycetin, a  new  antibiotic  from  a  soil  Actinomycete,  Science  106:  417,  1947. 
Emerson,  R.,  and  D.  L.  Fox:  Gamma-carotene  in  the  sexual  phase  of  the  aquatic 

fungus  Allomyces,  Proc.  Roy.  Soc.  (London),  Ser.  B,  128:  275-293,  1940. 
♦Epstein,  S.,  and  B.  Williams:  Miracles  from  Microbes,  the  Road  to  Streptomycin, 

Rutgers  University  Press,  New  Brunswick,  1946. 
Erb,  N.  M.,  R.  T.  Wisthoff,  and  W.  L.  Jacobs:  Factors  affecting  the  production  oi 


298  PHYSIOLOGY  OF  THE  FUNGI 

amylase  by  Aspergillus  niger,  strain  NRRL  337,  when  grown  in  submerged 
culture,  Jour.  Bad.  55:  813-821,  1948. 

Fischer,  O.  E.:  Mushroom  poisoning  in  C.  H.  Kauffman,  The  Agaricaceae  of  Michi- 
gan, Vol.  I,  Publication  26,  Biological  Series  of  the  Michigan  Geological  and 
Biological  Survey,  Lansing,  1918. 

FiTZPATRicK,  W.  H.,  W.  B.  EssELEN,  and  E.  Weir:  Composition  and  nutritive 
value  of  mushroom  protein,  Jour.  Atn.  Dietet.  Assoc.  22  :  318-322,  1946. 

Fleming,  A.:  Penicillin,  Its  Practical  Application,  The  Blakiston  Company,  Phila- 
delphia, 1949. 

Foster,  J.  W.:  Riboflavin,  British  patent  593,027,  Oct.  7,  1947.  Chem.  Abs.  42: 
1710,  1948. 

Foster,  J.  W.:  Chemical  Activities  of  Fungi,  Academic  Press,  Inc.,  New  York,  1949. 

Foster,  J.  W.,  S.  F.  Carson,  S.  Ruben,  and  M.  D.  Kamen:  Radioactive  carbon  as 
an  indicator  of  carbon  dioxide  utilization.  VII.  The  assimilation  of  carbon 
dioxide  by  molds,  Proc.  Natl.  Acad.  Sci.  U.S.  27:  590-596,  1941. 

Foster,  J.  W.,  and  J.  B.  Davis:  Anaerobic  formation  of  fumaric  acid  by  the  mold 
Rhizopus  nigricans,  Jour.  Bad.  56:  329-338,  19-8. 

Foster,  J.  W.,  and  S.  A.  Waksman:  The  specific  effect  of  zinc  and  other  heavy  metals 
on  growth  and  fumaric  acid  production  by  Rhizopus,  Jour.  Bad.  37:  599-617, 
1939. 

Foster,  J.  W.,  and  S.  A.  Waksman:  The  production  of  fumaric  acid  by  molds 
belonging  to  the  genus  Rhizopus,  Jour.  Am.  Chem.  Sac.  61 :  127-135,  1939a. 

Gastrock,  E.  a.,  N.  Porges,  P.  A.  Wells,  and  A.  J.  Moyer:  Gluconic  acid  produc- 
tion on  pilot-plant  scale.  Effect  of  variables  on  production  by  submerged 
mold  growths,  Ind.  Eng.  Chem.  30:  782-789,  1938. 

Gerhardt,  p.,  W.  W.  Dorrell,  and  I.  L.  Baldwin:  Citric  acid  fermentation  of 
beet  molasses,  Jour.  Bad.  52 :  555-564,  1946. 

GoRciCA,  H.  J.,  and  H.  Levine:  Vitamin  Bi  assimilation  by  yeast,  U.S.  patent 
2,295,036,  Sept.  8,  1942. 

Gordon,  M.  A. :  The  physiology  of  a  blue  stain  mold  with  special  reference  to  pro- 
duction of  ethyl  acetate,  Mycyologia  42  :  167-185,  1950. 

Gottlieb,  D.,  and  P.  Siminoff:  The  role  of  antibiotics  in  soil.  Phytopathology  40: 
11,  1950. 

Grossbard,  E.:  Production  of  an  antibiotic  substance  on  wheat  straw  and  other 
organic  materials  in  the  soil,  Nature  161:  614-615,  1948. 

Gupta,  B.  M.,  and  W.  C.  Price:  Production  of  plant  virus  inhibitors  by  fungi. 
Phytopathology  40 :  642-652,  1950. 

Hao,  L.  C,  E.  I.  FuLMER,  and  L.  A.  Underkofler:  Fungal  amylases  as  saccharify- 
ing agents  in  the  alcoholic  fermentation  of  corn,  Ind.  Eng.  Chem.  35:  814-818, 
1943. 

Harden,  A.:  Alcoholic  Fermentation,  4th  ed.,  Longmans,  Roberts  and  Green, 
London,  1932. 

Harris,  E.  E..  M.  L.  Hannan,  and  R.  R.  Marquardt:  Production  of  food  yeast 
from  wood  hydrolysates,  Ind.  Eng.  Chem.  40 :  2068-2072,  1948. 

Hendlin,  D.,  and  M.  L.  Ruger:  The  effect  of  cobalt  in  the  microbial  synthesis  of 
LLD-active  substances.  Science  111 :  541-542,  1950. 

Henry,  A.  W.:  The  natural  microflora  of  the  soil  in  relation  to  the  foot-rot  problem 
of  wheat.  Can.  Jour.  Research  4 :  69-77,  1931. 

Hesse,  A.:  Industrial  biosyntheses.  I.  Fats,  Advances  in  Enzymol.  9:  653-704, 
1949. 

HiCKEY,  R.  J.:  The  inactivation  of  iron  by  2,2'-bipyridine  and  its  effect  on  ribo- 
flavin synthesis  by  Clostridium  acetobutylicum.  Arch.  Biochem.  8:  439-447,  1945. 


METABOLIC  PRODUCTS  299 

♦Howard,  B.  H.,  and  H.  Raistrick:  Studies  in  the  biochemistry  of  micro-organisms. 

80.  The  colouring  matters  of  PeniciUium  islandicum  Sopp.  Part  1.  1,4,5-tri- 
hydroxy-2-methylanthraquinone,  Biochem.  Jour.  44:  227-233,  1949. 

Howard,  B.  H.,  and  H.  Raistrick:  Studies  in  the  biochemistry  of  micro-organisms. 

81.  The  colouring  matters  of  PeniciUium  islandicum  Sopp.  Part  2.  Chryso- 
phanic  acid,  4,5-dihydroxy-2-methylanthraquinone,  Biochem.  Jour.  46:49-53, 
1950. 

HuMFiELD,  H.:  The  production  of  mushroom  mycelium  {Agaricus  campestris)  in 

submerged  culture.  Science  107 :  373,  1948. 
Karow,  E.  O.,  and  S.  A.  Waksman:  The  production  of  citric  acid  in  submerged 

culture,  Ind.  Eng.  Chem.  39 :  821-825,  1947. 
Keilin,  D.,  and  E.  F.  Hartree:  Properties  of  glucose  oxidase  (Notatin),  Biochem. 

Jour.  42  :  221-229,  1948. 
Keilin,  D.,  and  E.  F.  Hartree:  The  use  of  glucose  oxidase  (Notatin)  for  the  deter- 
mination of  glucose  in  biological  material,  and  for  the  study  of  glucose-producing 

systems  by  manometric  methods,  Biochem.  Jour.  42  :  230-238,  1948a. 
Krieger,  L.  C.  C:  The  Mushroom  Handbook,  The  Macmillan  Company,  New  York 

193G. 
KuHN,  R.,  L.  BiRKHOFER,  and  F.  W.  Quackenbush:  Jodometrische  Titration  von 

SH-gruppen:  Mikromethode  zur  Bestimmung  von  Cystein  und  Methionin  in 

Proteinen,  Ber.  d.  deut.  chem.  Ges.  72 :  407-416,  1939. 
Leach,  J.  G.:  Insect  Transmission  of  Plant  Diseases,  McGraw-Hill  Book  Company, 

Inc.,  New  York,  1940. 
*Leben,  C,  and  G.  W.  Keitt:  Laboratory  and  greenhouse  studies  of  antimycin 

preparations  as  protectant  fungicides.  Phytopathology  39:  529-540,  1949. 
Lee,  S.  B.:  Fermentation,  Ind.  Eng.  Chem.  41 :  1838-1879,  1949. 
Le  Mense,  E.  H.,  J.  CoRMAN,  J.  M.  vanLanen,  and  A.  F.  Langlykke:  Production 

of  mold  amylase  in  submerged  culture.  Jour.  Bact.  54:  149-159,  1947. 
Lemiei-x,  R.  U.,  and  M.  L.  Wolfrom:  The  chemistry  of  streptomycin,  Advances  in 

Carbohydrate  Chem.  3 :  337-384,  1948. 
Letcher,  H.,  and  J.  J.  Willaman:  Biochemistry  of  plant  diseases.     VIII.  Alcoholic 

fermentation  of  Fusarium  lini,  Phytopathology  16:  941-949,  1926. 
Levine,  H.,  J.  E.  Oyaas,  L.  Wanerman,  J.  C.  Hoogerheide,  and  R.  M.  Stern: 

Riboflavin  production  by  Candida  yeasts,  Ind.  Eng.  Chem.  41 :  1665-1668,  1949. 
LiLLiE,   R.  J.,   C.  A.  Denton,  and   H.   R.   Bird:  Relation  of  vitamin  B12  to  the 

growth  factor  present  in  cow  manure,  Jour.  Biol.  Chem.  176:  1477-1478,  1948. 
Linder,  p.:  Das  Problem  der  biologischen  Fettbildung  und  Fettgewinnung,  Zeit. 

angeiv.  Chem.  35:  110-114,  1922. 
Lintzel,  W.:  Ueber  den  Nahrwert  des  Eiweisses  der  Speisepilze,  Biochem.  Zeit. 

308:  413-419,  1941. 
LocKWOOD,  L.  B.:  A  study  of  the  physiology  of  PeniciUium  javanicum  van  Beijma 

with  special  reference  to  the  production  of  fat.  Catholic  Univ.  of  America  Biol. 

Ser.  13,  1933. 
LocKwooD,  L.  B.,  G.  E.  Ward,  and  O.  E.  May:  The  physiology  of  Rhizopus  oryzae, 

Jour.  Agr.  Research  53:  849-857,  1936. 
LocKwooD,  L.  B.,  and  G.  E.  Ward:  Fermentation  process  for  itaconic  acid,  Ind. 

Eng.  Chem.  37:  405-406,  1945. 
Mass,  J.  M.:  Personal  communication,  1950. 
Metz,  O.  :  Ueber  Wachstum  und  Farbstoffbildung  einiger  Pilze  unter  dem  Einfluss 

von  Eisen,  Zink  und  Kupfer,  Arch.  Mikrobiol.  1:  197-251,  1930. 
Meyerhof,  O.:  Energy  relationships  in  glycolysis  and  phosphorylation,  Ann.  N.Y. 

Acad.  Sci.  45 :  377-393,  1944. 


300  PHYSIOLOGY  OF  THE  FUNGI 

Meyerhof,  O.  :  Glycolysis  of  animal  tissue  extracts  compared  with  the  cell-free 

fermentation  of  yeast,  Wallerstein  Labs.  Comniuns.  12  :  255-264,  1949. 
MiCHENER,  H.  D.,  and  N.  Snell:  Studies  on  cultural  requirements  of  Claviceps 

purpurea  and  inactivation  of  ergotamine,  Am.  Jour.  Botany  37:  52-59,  1950. 
Morton,  H.  E.,  W.  Kocholaty,  R.  Junowicz-Kocholaty,  and  A.  Kelner:  Toxic- 
ity and  antibiotic  activity  of  kojic  acid  produced  by  Aspergillus  luteo-virescens, 

Jour.  Bad.  50 :  579-584,  1945. 
MoYER,  A.  J.,  and  R.  D.  Coghill:  The  laboratory-scale  production  of  itaconic  acid 

by  Aspergillus  terreus,  Arch.  Biochem.  7:  1G7-183,  1945. 
♦MoYER,  A.  J.,  E.  J.  Umberger,  and  J.  J.  Stubbs:  Fermentation  of  concentrated 

solutions  of  glucose  to  gluconic  acid.     Improved  process,  Ind.  Eng.  Chem.  32 : 

1379-1383,  1940. 
Mulder,  E.  G.:  Ueber  die  Bedeutung  des  Kupfers  fiir  das  Wachstum  von  Mikro- 

organismen   und   iiber   eine   mikrobiologische    Methode   zur   Bestimmung   des 

pflanzenverfiigbaren  Bodenkupfers,  Arch.  Mikrobiol.  10 :  72-80,  1939. 
Myrback,   K.:  Products  of  the  enzymatic   degradation  of  starch  and  glycogen. 

Advances  in  Carbohydrate  Chem.  3,  1948. 
NoRD,  F.  F. :  Food  composition  containing  Fusaria,  U.S.  patent  2,450,055,  Sept.  28, 

1948. 
NoRD,  F.  F.,  and  R.  P.  Mull:  Recent  progress  in  the  biochemistry  of  Fusaria, 

Advances  in  Enzymol.  5:  165-205,  1945. 
Owen,  W.  L.  :  Production  of  industrial  alcohol  from  grain  by  the  amylo  process, 

Ind.  Eng.  Chem.  25:  87-89,  1933. 
Pasteur,  L.:  Etudes  sur  la  biere,  1872.     Studies  on  Fermentation.     The  Diseases 

of  Beer  (trans.  F.  Faulkner  and  D.  C.  Robb),  Macmillan  &  Co.,  Ltd.,  London, 

1879. 
Payne,  E.  H.,  and  J.  H.  Knaudt:  Treatment  of  epidemic  typhus  with  Chloromycetin, 

Proc.  Intern.  Cong.  Trop.  Med.  and  Malaria,  4th  Congr.,  4:  426-428,  1948. 
Peltier,  G.  L.,  and  R.  Borchers:  Riboflavin  production  by  molds.  Jour.  Bad. 

64 :  519-520,  1947. 
Perlman,  D.,  W.  W.  Dorrell,  and  M.  J.  Johnson:  Effect  of  metallic  ions  on  the 

production  of  citric  acid  by  Aspergillus  niger,   Arch.   Biochem.   11:  131-143, 

1946. 
Platt,  B.  S.,  and  R.  A.  Webb:  Fermentation  and  human  nutrition,  Biochem.  Jour. 

39:31,  1945. 
PoRGEs,    N.,    T.    F.    Clark,    and    E.    A.    Gastrock:    Gluconic    acid   production. 

Repeated  use  of  submerged  Aspergillus  niger  for  semi-continuous  production, 

Ind.  Eng.  Chem.  32:  107-111,  1940. 
PoRGEs,    N.,    T.    F.    Clark,    and    S.    I.    Aronovsky:    Gluconic   acid   production. 

Repeated  recovery  and  re-use  of  submerged   Aspergillus  niger  by  filtration, 

Ind.  Eng.  Chem.  33:  1065-1067.     1941. 
■^Pratt,  R.,  and  J.  Dufrenoy:  Cytochemical  mechanisms  of  penicillin  action.     VIII. 

Involvement  of  ribonucleic  acid  derivatives,  Jour.  Bad.  57:  9-13,  1949. 
Prescott,  S.  C,  and  G.  G.  Dunn:  Industrial  Microbiology,  2d  ed.,  McGraw-Hill 

Book  Company,  Inc.,  New  York,  1949. 
Preuss,  L.  M.,  E.  C.  Eichinger,  and  W.  H.  Peterson:  The  chemistry  of  mold 

tissue.     III.  Composition  of  certain  molds  with  special  reference  to  the  lipid 

content.  Cent.  Bakt.,  Abt.  II,  89:  370-377,  1934. 
Raaf,  H.:  Beitrage  zur  Kenntnis  der  Fett-  und  Eiweissyntheses  bei  Endomyces 

vernalis  und  einigen  anderen  Mikroorganismen,  Arch.  Mikrobiol.  12 :  131-182, 

1941. 
Raistrick,  H.,  J.  H.  Birkinshaw,  J.  H.  V.  Charles,  P.  W.  Clutterbuck,  F.  P. 

Coyne,  A,  C.  Hetherington,  C.  H.  Lilly,  M.  L,  Rintoul,  W.  Rintoul, 


METABOLIC  PRODUCTS  301 

R.  Robinson,  J.  A.  R.  Stoyle,  C.  Thom,  and  W.  Young:  Studies  in  the  bio- 
chemistry of  micro-organisms,  Trans.  Roy.  Soc.  (London),  Ser.  B,220:  1-367, 
1931. 
Raistrick,   H.,   R.  Robinson,  and  A.  R.  Todd:  Studies  in  the  biochemistry  of 
micro-organisms.     XXXII.   Cynodontin         (l,4,5,8-tetrahydroxy-2-methylan- 
thraquinone),  a  metabolic  product  of  H elminthosporium  cynodontis  Marigroni 
and  H elminthosporium  euchlaenae  Zimmermann,  Biochem.  Jour.  27:  1170-1175, 
1933. 
Raistrick,  H.,  R.  Robinson,  and  D.  E.  White:  Studies  in  the  biochemistry  of 
micro-organisms.     L.   Ravenelin(3-methyl-l,4,8-trihydroxyxanthone),     a    new 
metaboUc    product   of   H elminthosporium   ravenellii   Curtis   and    H.    turcicum 
Passerini,  Biochem.  Jour.  30:  1303-1314,  1936. 
Rake,  G.,  and  R.  Donovick:  Studies  on  the  nutritional  requirements  of  Strepto- 
myces  griseus  for  the  formation  of  streptomycin.  Jour.  Bad.  62 :  223-228,  1946. 
*Raper,  K.  B.,  D.  F.  Alexander,  and  R.  D.  Coghill:  Penicillin.     II.  Natural 
variation  and  penicillin  production  in  Penicillium  notatum  and  allied  species, 
Jour.  Bad.  48:  639-659,  1944. 

Raper,  K.  B.,  R.  D.  Coghill,  and  A.  Hollaender:  The  production  and  charac- 
terization of  ultraviolet-induced  mutants  in  Aspergillus  terreus.  II.  Cultural 
and  morphological  characteristics  of  the  mutations.  Am.  Jour.  Botany  32:  165- 
176,  1945. 

Reilly,  H.  C,  a.  Schatz,  and  S.  A.  Waksman:  Antifungal  properties  of  antibiotic 
substances.  Jour.  Bad.  49 :  585-594,  1945. 

Renaud,  J.,  and  M.  Lachatltx:  Recherches  sur  la  formation  de  lactoflavine  a  partir 
de  r  Eremothecium  ashbyii.  Influence  des  constituants  du  milieu  sur  la  produc- 
tion du  pigment,  Compt.  rend.  acad.  sci.  221 :  187-188,  1945. 

Rickes,  E.  L.,  N.  G.  Brink,  F.  R.  Koniuszy,  T.  R.  Wood,  and  K.  Folkers:  Com- 
parative data  on  vitamin  B12  from  liver  and  a  new  source,  Streptomyces  griseus, 
Science  108 :  634-635,  1948. 

RoBBiNS,  W.  J.,  A.  Hervey,  R.  W.  Davidson,  R.  Ma,  and  W.  C.  Robbins:  A  survey 
of  some  wood-destroying  and  other  fungi  for  antibacterial  activity,  Bidl.  Torrey 
Botan.  Club  72 :  165-190,  1945. 

Roberts,  M.,  S.  Laufer,  E.  D.  Stew^art,  and  L.  T.  Saletan:  Saccharification  of 
wheat  by  fungal  amylases  for  alcohol  production,  Ind.  Eng.  Chem.  36 :  811-812, 
1944. 

RoLFE,  R.  T.,  and  F.  W.  Rolfe:  The  Romance  of  the  Fungus  World,  J.  B.  Lippin- 
cott  Company,  Philadelphia,  1926. 

Rose,  W.  C.  The  nutritive  significance  of  the  amino  acids,  Physiol.  Revs.  18: 
109-136,  1938. 

Saito,  K.:  Ein  Beispiel  von  Milchsaurebildung  durch  Schimmelpilze,  Cent.  Bakt., 
Abt.  II,  29:  289-290,  1911. 

Schatz,  A.,  and  E.  L.  Hazen:  The  distribution  of  soil  microorganisms  antagonistic 
to  fungi  pathogenic  for  man,  Mycologia  40 :  461-477,  1948. 

ScHLEEF,  M.  L.:  The  economics  of  fodder  yeast  from  sulfite  waste  liquor,  Wash. 
State  Coll.  Economic  and  Business  Studies  Bull.  7,  1948. 

Schoenbach,  E.  B.,  M.  S.  Bryer,  and  P.  H.  Long:  The  pharmacology  and  clinical 
trial  of  aureomycin:  a  preliminary  report,  Ann.  N.Y.  Acad.  Sci.  51:  267-279, 
1948. 

Schopfer,  W.  H.:  Etude  et  identification  d'xm  carotinoide  de  champignon,  Compt. 
rend.  soc.  biol.  118 :  3-5,  1935. 

SciARiNi,  L.  J.,  R.  p.  Mall,  J.  C.  Wirth,  and  F.  F.  Nord:  Concerning  the  relation 
between  structure  and  action  of  xanthones  on  dehydrogenations  by  Fusaria, 
Proc.  Natl.  Acad.  Sci.  U.S.  29:  121-126,  1943. 


302  PHYSIOLOGY  OF  THE  FUNGI 

*Shu,  p.,  and  M.  J.  Johnson:  Critic  acid  production  by  submerged  fermentation 
with  Aspergillus  niger,  Ind.  Eng.  Chem.  40:  1202-1205,  1948. 

Smith,  E.  L.  :  Purification  of  antiperinicious  anemia  factors  from  liver,  Nature  161 : 
638-639,  1948. 

Smith,  R.  M.,  D.  A.  Joslyn,  O.  M.  Gruhzit,  I.  W.  McLean,  M.  A.  PENNER,and 
J.  Ehrlich:  Chloromycetin:  biological  studies.  Jour.  Bad.  55:  425-448,  1948. 

Starkey,  R.  L.  :  Lipid  production  by  a  soil  yeast,  Jour.  Bad.  51 :  33-50,  1946. 

Sumner,  J.  B.,  and  G.  F.  Somers:  Chemistry  and  Methods  of  Enzymes,  2d  ed.. 
Academic  Press,  Inc.,  New  York,  1947. 
*Sure,  B.  :  Nutritional  improvement  of  cereal  flours  and  cereal  grains.  Jour.  Am.  Dietet. 
Assoc.  22:  494-502,  1946. 

Sure,  B.  :  Further  studies  on  nutritional  improvement  of  cereal  flours  and  cereal 
grains  with  yeast,  Jour.  Am.  Dietd.  Assoc.  23:  113-119,  1947. 

Takamine,  J.:  Enzymes  of  Aspergillus  oryzae  and  the  application  of  its  amyloclastic 
enzyme  to  the  fermentation  industry.  Jour.  Ind.  Eng.  Chem.  6:  824-828,  1914. 

Tanner,  F.  W.,  Jr.,  and  M.  J.  van  Lanen:  Production  of  riboflavin  by  Candida 
flareri,  U.S.  patent  2,424,003,  July  15,  1947.     Chem.  Abs.  41 :  6368,  1947. 

Tanner,  F.  W.,  Jr.,  C.  Vojnovich,  and  J.  M.  van  Lanen:  Riboflavin  production  by 
Candida  species,  Science  101:  180-181,  1945. 
.^Tanner,  F.  W.,  Jr.,  C.  Vojnovich,  and  J.  M.  van  Lanen:  Factors  affecting  ribo- 
flavin production  by  Ashbya  gossypii,  Jour.  Bad.  58:  737-745,  1949. 

Tauber,  H.:  Chemistry  and  Technology  of  Enzymes,  John  Wiley  &  Sons,  Inc., 
New  York,  1949. 

Thaysen,  a.  C:  Value  of  microorganisms  in  nutrition  (food  yeast).  Nature  151: 
406-408,  1943. 
*Thornberry,  H.  H.,  and  H.  W.  Anderson:  Synthetic  medium  for  Streptomyces 
griseus  and  the  production  of  streptomycin,  Arch.  Biochem.  16 :  389-397,  1948. 

Underkofler,  L.  a.,  G.  M.  Severson,  and  K.  J.  Goering:  Saccharification  of 
grain  mashes  for  alcoholic  fermentation,  Ind.  Eng.  Chem.  38 :  980-985,  1946. 

Utech,  N.  M.,  and  J.  Johnson:  The  inactivation  of  plant  viruses  by  substances 
obtained  from  bacteria  and  fungi.  Phytopathology  40:  247-265,  1950. 

Van  Lanen,  J.  M.,  H.  P.  Broquist,  M.  J.  Johnson,  I.  L.  Baldwin,  and  W.  H. 
Peterson:  Synthesis  of  vitamin  Bi  by  yeast,  Ind.  Eng.  Chem.  34:  1244-1247, 
1942. 

Vaughn,  J.  R.,  J.  L.  Lockwood,  G.  S.  Rajstdwa,  and  G.  Hamner:  Action  of  acti- 
dione  on  plant  tissue  and  upon  certain  fungi,  Mich.  Agr.  Expt.  Sta.  Quart.  Bull. 
31 :  456-464,  1949. 

Von  Loesecke,  H.  W.  :  A  review  of  information  on  mycological  citric  acid  produc- 
tion, Chem.  Eng.  News  23 :  1952-1959,  1945. 

Von  Loesecke,  H.  W.:  Controversial  aspects:  yeast  in  human  nutrition.  Jour.  Arji. 
Dietet.  Assoc.  22 :  485-493,  1946. 

Waksman,  S.  a.:  Associative  and  antagonistic  effects  of  microorganisms.  I.  His- 
torical review  of  antagonistic  relationship.  Soil  Sci.  43:  51-68,  1937. 

Waksman,  S.  A.:  Humus,  Origin,  Chemical  Composition  and  Importance  in  Nature, 
2d  ed..  The  Williams  &  Wilkins  Company,  Baltimore,  1938. 

Waksman,  S.  A.:  Microbial  Antagonism  and  Antibacterial  Substances,  Common- 
wealth Fund,  Division  of  Publication,  New  York,  1947. 

Waksman,  S.  A.,  and  J.  W.  Foster:  Respiration  and  lactic  acid  production  by  a 
fungus  of  the  genus  Rhizopus,  Jour.  Agr.  Research  57 :  873-899,  1939. 

Walker,  T.  K.  :  Pathways  of  acid  formation  in  Aspergillus  niger  and  in  related 
molds,  Advances  in  Enzymol.  9 :  537-584,  1949. 


METABOLIC  PRODUCTS  303 

Wallerstein,  J.  S. :  On  the  genealogy  of  beer,  Wallerstein  Labs.  Communs.  2 :  35-45, 
1939. 
*Wallerstein,  L.  :  Enzyme  preparations  from  microorganisms.     Commercial  pro- 
duction and  industrial  application,  Ind.  Eng.  Chevi.  31 :  1218-1224,  1939. 

Ward,  G.  E.,  L.  B.  Lockwood,  O.  E.  May,  and  H.  T.  Herrick:  Production  of  fat 
from  glucose  by  molds.  Cultivation  of  Penicillium  javanicum  van  Beyma  in 
large  scale  laboratory  apparatus,  Ind.  Eng.  Chem.  27 :  318-322,  1935. 

Ward,  G.  E.,  L.  B.  Lockwood,  0.  E.  May,  and  H.  T.  Herrick:  Studies  in  the  genus 
Rhizopus.  I.  The  production  of  dextro-lactic  acid,  Jour.  Am.  Chem.  Soc. 
58:  1286-1288,  1936. 

Ward,  G.  E.,  L.  B.  Lockwood,  B.  Tabenkin,  and  P.  A.  Wells:  Rapid  fermentation 
process  for  dextrolactic  acid,  Ind.  Eng.  Chem.  30:  1233-1235,  1938. 

Weiss,  S.,  and  F.  F.  Nord:  On  the  mechanism  of  enzyme  action.  XXXVIL 
Solanione,  a  pigment  from  Fusarium  solani  D2  purple,  Arch.  Biochem.  22 : 
288-313,  1949. 

Wells,  P.  A.,  A.  J.  Moyer,  and  O.  E.  May:  The  chemistry  of  citric  acid  fermenta- 
tion.    I.  The  carbon  balance.  Jour.  Am.  Chem.  Soc.  58:  555-558,  1936. 

Wells,  P.  A.,  A.  J.  Moyer,  J.  J.  Sti^tbbs,  H.  T.  Herrick,  and  O.  E.  May:  Gluconic 
acid  production.  Effect  of  pressure,  air  flow,  and  agitation  on  gluconic  acid 
production  by  submerged  mold  growths,  Ind.  Eng.  Chem.  29:  653-656,  1937. 

West,  R.  :  Activity  of  vitamin  B12  in  Addisonian  pernicious  anemia,  Science  107: 
398,  1948. 

Whiffin,  a.  J.:  The  activity  in  vitro  of  cycloheximide  (acti-dione)  against  fungi 
pathogenic  to  plants,  Mycologia  42 :  253-258,  1950. 

WiELAND,  H.,  and  B.  Witkop:  Giftstoffe  des  Knollenblatterpilzes.  V.  Zur  Kon- 
stitution  des  Phalloidins,  /.  Liebigs  Ann.  d.  Chem.  543:  171-183,  1940. 

WiLKiNS,  W.  H.,  and  G.  C.  M.  Harris:  Investigations  into  the  production  of  bac- 
teriostatic substances  by  fungi.  VI.  Examination  of  the  larger  Basidiomycetes, 
Ann.  Applied  Biol.  31 :  261-270,  1944. 

WoHLER,  F.:  Das  Entrathselte  Geheimniss  der  geistigen  Garung  (Vorljiufige  brief- 
Hche  Mittheilung),  J.  Leibigs  Ann.  d.  Chem.  29  :  100-104,  1839. 

Wolf,  F.  A.,  and  F.  T.  Wolf:  The  Fungi,  Vol.  II,  John  Wiley  &  Sons,  Inc.,  New 
York,  1947. 

Wright,  L.  T.,  M.  Sanders,  M.  A.  Logan,  A.  Prigot,  and  L.  M.  Hill:  The  treat- 
ment of  lymphogranuloma  venereum  and  granuloma  inguinale  in  humans  with 
aureomycin,  Ann.  N.Y.  Acad.  Sci.  151:  318-330,  1948. 

Ytjill,  J.  L. :  Alcoholic  fermentation  by  Aspergillus  flavus  Brefeld,  Biochem.  Jour. 
22:  1504-1507,  1928. 


CHAPTER  14 
FACTORS  INFLUENCING  SPORULATION  OF  FUNGI 

The  life  of  the  individual  fungus  is  usually  short  and  of  uncertain  dura- 
tion. The  continuance  of  the  species  (in  most  instances)  depends  upon 
the  production  and  dissemination  of  sexual  or  asexual  spores.  The 
importance  of  spore  production  in  the  spread  of  epiphytotics  is  sufficient 
reason  to  study  the  factors  which  control,  modify,  or  inhibit  this  stage  of 
development  in  the  life  of  the  fungi.  Some  of  the  most  difficult  problems 
which  arise  in  the  study  of  the  life  processes  of  the  fungi  are  to  be  found 
in  the  events  and  conditions  which  control  the  production  of  spores. 

In  nature,  we  find  many  examples  of  the  influence  of  certain  environ- 
mental and  nutritional  factors  upon  reproduction  of  the  fungi.  A  num- 
ber of  parasitic  fungi  produce  the  perfect  stage  only  in  the  spring,  on  or  in 
dead  host  tissue.  This  is  true  of  Venturia  inaequalis,  Clavtceps  purpurea, 
Gnomonia  ulmi,  Monilinia  jructicola,  Coccomyces  hiemalis,  Guignardia 
bidwelln,  and  others.  Is  the  production  of  the  sexual  stage  dependent 
upon  the  pretreatment  of  cold,  or  freezing  and  thawing?  Is  it  dependent 
upon  a  favorable  temperature  and  perhaps  favorable  intensity  and  dura- 
tion of  light?  Or  is  it  a  matter  of  the  proper  nutrients  which  are  made 
available  only  after  decay  of  the  host  tissues?  These  are  difficult  ques- 
tions to  answer,  for  it  is  likely  that  the  production  of  the  sexual  stage 
depends  upon  the  proper  balance  of  a  number  of  factors.  Similarly,  we 
may  speculate  about  the  stimuli  involved  in  the  formation  of  the  peri- 
thecia  by  the  Erysiphales.  Most  of  these  obligate  parasites  form  fruit 
bodies  late  in  the  growing  season.  Perhaps,  at  least  in  some  cases,  this 
is  a  reaction  to  cooler  weather ;  or  perhaps  the  formation  of  perithecia  is  a 
result  of  a  decreasing  or  changing  food  supply  as  the  host  nears  maturity. 
Other  physical  factors  are  probably  involved,  since  we  know  that  the 
abundance  of  perithecia  varies  from  year  to  year.  It  is  also  of  physio- 
logical interest  that  many  parasitic  fungi  produce  conidia  only  while  the 
mycelium  is  actively  attacking  the  living  host. 

A  critical  investigation  of  the  factors  influencing  reproduction  requires 
that  the  fungi  be  brought  into  the  laboratory  or  greenhouse  where  external 
factors  can  be  controlled.  Only  one  variable  should  be  studied  at  a  time, 
and  all  other  influencing  factors  must  be  controlled.  It  is,  therefore,  of 
great  advantage  in  physiological  studies  to  be  able  to  grow  a  fungus  in 
pure  culture  on  synthetic  or  semisynthetic  media.     However,  it  n\ust  be 

304 


SPORULATION  305 

pointed  out  that  the  responses  of  a  fungus  in  nature  cannot  always  be 
duplicated  in  the  laboratory. 

Snyder  and  Hansen  (1947)  have  given  a  brief  and  clear  statement 
regarding  the  advantages  of  culturing  fungi  on  natural  media  and  under 
natural  environmental  conditions.  These  conditions  are  important,  if 
one  desires  to  obtain  reproduction  of  a  fungvis  which  does  not  sporulate 
readily  in  culture.  However,  if  one  desires  to  study  critically  the  indi- 
vidual nutritional  and  environmental  requirements  and  their  effects  upon 
reproduction  of  a  fungus  which  sporulates  abundantly  on  the  usual 
cultural  media,  it  is  often  necessary  to  subject  the  fungus  to  unfavorable 
conditions.  Thus,  only  by  preventing  sporulation,  by  varying  but  one 
factor  at  a  time,  may  we  discover  the  need  for  that  factor. 

Riker  and  Riker  (193G)  have  listed  11  methods  which  have  been  suc- 
cessfully employed  to  induce  sporulation  of  different  fungi  in  culture. 
Since  the  writing  of  their  manual  much  has  been  learned  about  this  phase 
of  fungus  physiology.  A  revised  list  of  the  conditions  known  to  influence 
sporulation  of  fungi  is  presented  in  the  summary  of  this  chapter. 

Kauffman  (1929)  called  attention  to  the  views  of  Klebs,  who  held  that 
living  cells  are  influenced  during  their  lifetime  in  three  ways:  (1)  by  the 
specific  structure;  (2)  by  the  internal  conditions;  and  (3)  by  the  external 
conditions.  Kauffman  equated  the  first  of  these  to  heredity  and  the  last 
two  to  environment.  The  external  environment  comprises  the  various 
physical  and  chemical  factors,  such  as  temperature,  light,  composition 
of  the  medium,  and  the  like.  Kauffman  used  the  term  internal  environ- 
ment to  designate  the  complicated  influences  and  reactions  between  cells 
wdthin  the  organism.  The  physical  and  chemical  effects  of  the  external 
environment  may  be  transmitted  through  the  cells  and  become  evident  at 
some  distance  from  the  point  of  the  stimulus. 

The  meaning  of  these  statements  may  be  clearer  if  we  consider  the 
effect  of  various  external  environmental  factors  upon  fruiting.  It  is  well 
known  that  various  external  stimuli  may  initiate  the  reactions  which  lead 
to  reproduction.  These  stimuli  must  act  through  the  internal  environ- 
ment. Most  of  the  discussion  that  follows  will  be  concerned  with  the 
external  environment  and  the  resulting  development  of  the  fvmgus. 
Some  external  factors  may  so  modify  the  internal  milieu  as  to  favor 
sporulation,  w^hile  others  may  inhibit  or  prevent  sporulation. 

Not  all  fungi  respond  in  the  same  way  to  the  external  factors  such  as 
light,  temperature,  or  nutrition.  Each  species  produces  spores  when  the 
internal  environment  is  suitable,  but  the  external  factors  do  not  operate 
upon  the  internal  environment  of  all  fungi  alike.  Thus,  there  is  no  univer- 
sal set  of  external  conditions  which  lead  to  fructification  in  all  fungi.  The 
external  conditions  favorable  for  sporulation  must  be  studied  for  each 
species.     This  does  not  imply,  however,  that  no  two  fungi  react  alike  or 


306  PHYSIOLOGY  OF  THE  FUNGI 

that  certain  helpful  generalizations  concerning  sporulation  cannot  be 
drawn.  It  does,  however,  imply  that  the  only  sure  way  of  understanding 
the  conditions  governing  reproduction  in  a  specific  fungus  lies  in  the 
experimental  approach. 

Again  it  must  be  emphasized  that  all  the  physical  and  chemical  condi- 
tions may  be  at  the  optima,  but  no  reproduction  can  occur  without  the 
presence  of  favorable  genetic  factors.  Too  often  we  may  fail  to  realize 
the  genetic  requirements.  The  appropriate  steps  should  be  taken  to 
determine  whether  the  fungi  under  study  are  homothallic  or  heterothallic. 
It  may  be  difficult,  indeed,  to  determine  whether  failure  of  a  fungus  to 
reproduce  in  culture  is  due  to  unfavorable  environmental  conditions  or 
to  unfavorable  genetic  factors.  There  is  much  yet  to  be  learned  regard- 
ing the  physiology  of  reproduction,  but  each  new  investigation  is  certain 
to  add  to  our  knowledge  of  this  interesting  and  important  phase  of  fungus 
physiology. 

Vegetative  growth  must  precede  reproduction.  The  length  of  the 
vegetative  phase  varies  from  organism  to  organism,  and  the  same  organ- 
ism may  remain  in  the  vegetative  phase  for  a  longer  or  shorter  period  of 
time  depending  upon  the  external  environment.  One  of  the  functions 
of  the  vegetative  phase  is  concerned  with  the  building  up  of  protoplasm 
and  the  storage  of  energy  reserves.  Reproduction  is  a  process  that  draws 
heavily  on  the  reserve  food.  The  spore  is  usually  well  stocked  with  these 
materials.  Asexual  reproduction  differs  less  from  vegetative  growth  than 
does  sexual  reproduction.  We  shall  find  that  the  conditions  limiting- 
sexual  reproduction  are  usually  more  narrow  than  conditions  which  allow 
asexual  reproduction  and  growth. 

Klebs  (1900)  summarized  his  views  on  reproduction  in  the  fungi  in  the 
form  of  four  laws  or  principles  as  follows:  (1)  Growth  and  reproduction 
are  life  processes,  which,  in  all  organisms,  depend  upon  different  condi- 
tions. In  the  lower  organisms  the  external  conditions  mainly  determine 
whether  growth  or  reproduction  takes  place.  (2)  Reproduction  in  the 
lower  organisms  does  not  occur  as  long  as  characteristic  external  condi- 
tions are  favorable  for  growth.  The  conditions  which  are  favorable  for 
reproduction  are  always  more  or  less  unfavorable  for  growth.  (3)  The 
processes  of  growth  and  reproduction  differ,  in  that  growth  may  take 
place  under  a  wider  range  of  environmental  conditions  than  reproduction. 
Growth  may  take  place,  therefore,  under  conditions  which  inhibit  repro- 
duction. (4)  Vegetative  growth  appears  to  be  mostly  a  preliminary  step 
for  reproduction  in  that  it  creates  a  suitable  internal  environment  for  it. 
To  a  certain  degree  it  is  not  growth  in  itself  but  the  prolonged  period  of 
assimilation  accompanying  growth  that  is  decisive  for  reproduction. 

These  generalizations  were  published  in  1900  and  were  based  upon 
Klebs's  own  work,  as  well  as  that  of  others.     Many  more  fungi  have  been 


SPORULATION 


307 


studied  during  the  past  50  years,  and  some  new  factors  have  been  brought 
to  light.  It  would  not  be  surprising  if  some  modifications  in  these  con- 
clusions would  be  necessaiy  in  the  light  of  50  years  of  research.  We  shall 
find,  however,  that,  in  the  main,  many  of  these  "laws"  are  still  valid. 

ENVIRONMENTAL  FACTORS 

Temperature.  Temperature  was  recognized  by  Bisby  (1943)  as  an 
important  natural  factor  governing  the  geographical  distribution  of  the 
fungi.  The  temperature  must  be  favorable  not  only  for  growth  but  also 
for  the  production  and  germination  of  the  spores,  if  the  fungus  is  to 
survive.  Certain  fungi  are  limited  by  high  temperatures.  Among  these 
are  Plasmodiophora  brassicae,  Colletotrichum  lindemuthianum,  Urocystis 
cepulae,  and  certain  Phycomycetes.  On  the  other  hand,  certain  genera 
of  the  Gasteromycetes,  such  as  Podaxis,  Battarrea,  Chlamydopus,  and 
Phellorina,  are  confined  to  the  hot  arid  regions  of  southwestern  United 
States,  northern  Africa,  central  Australia,  and  western  India.  Between 
these  extremes  we  may  observe  many  examples  where  seasonal  tempera- 
ture limits  or  favors  reproduction. 

Klebs  (1900)  pointed  out  that  the  temperature  range  which  allowed 
sporulation  was  more  narrow  than  the  range  for  growth.  In  general,  the 
temperature  limits  for  sexual  reproduction  are  narrower  than  the  limits 
for  asexual  reproduction.  Some  of  Klebs's  data  are  presented  in 
Table  53. 


Table  53.     Minimum  and  Maximum  Temperatures  (in  Degrees  Centigrade)  for 

Growth  axd  Sporulation  of  Various  Fungi 

(Klebs,  Jahrb.  iriss.  Botan.  35,  1900.) 


Fungus 


Aspergillus  repens.  .  . 
Sporidinia  grandis.  . . 
Piloholus  rnicrosporus 
Saprolegnia  mixta.  . .  . 


Growth 


Min. 


7-8 
1-2 
2-4 
0-1 


Max. 


37-38 
31-32 
33-34 
36-37 


Asexual  spores 


Min. 


8-9 

5-6? 

10-12 

1-2 


Max. 


35-36 
29-30 
28-30 
32-33 


Sexual  spores 


Min. 


5-6 


1-2 


Max. 


33-34 

27-28 

26-27 


It  will  be  noted  that  the  upper  temperature  which  allowed  the  produc- 
tion of  oospores  by  Saprolegnia  mixta  is  a  full  10°C.  less  than  the  upper 
temperature  limit  at  which  growth  took  place.  Coons  (1916)  found  the 
temperature  limits  for  the  growth  of  Plenodomus  fuscomacidans  to  be 
0  to  33°C.,  while  pycnidia  formed  between  6  and  30°C.  Perithecia  failed 
to  form  in  cultures  of  Ceratostomella  fimbriata  kept  at  18°C.  for  60  days 
(Barnett  and  Lilly,  1947a).     Cultures  of  this  fungus  on  the  same  medium 


308 


PHYSIOLOGY  OF  THE  FUNGI 


produced  abundant  perithecia  and  ascospores  at  25°C.  within  11  days. 
Conidia  were  formed  at  18°C. 

The  most  noteworthy  effect  of  culturing  a  fungus  at  temperatures 
below  the  optimum  is  the  decrease  in  the  rate  of  growth.  It  has  been 
found  by  various  investigators  that  there  is  an  optimum  temperature  for 
sporulation  as  well  as  for  growth.  The  two  optima  may  be  different. 
Figure  57  shows  the  effect  of  temperature  on  the  time  required  to  produce 
conidia  by  Aspergillus  repens. 


14 


12 


10 


2   6 


Q,      4 


o 
o 


• 

\ 

\ 

L 

\ 

\ 

•\ 

V 

\ 

• 

i 

1 

X. 

— .- 

y 

10  15  20  25  30 

Temperature  in  degrees  centigrode 


35 


40 


Fig.  57.     The  influence  of  temperature  on  the  time  required  to  produce  conidia  by 
Aspergillus  repens.      (Drawn  from  data  of  Klebs,  Jarhh.  wiss.  Botan.  35 :  137,  1900.) 


A  temperature  of  28°C.  was  optimum  for  sporulation  of  Piricularia 
oryzae  (Henry  and  Andersen,  1948).  Higher  and  lower  temperatures  of 
incubation  decreased  the  numbers  of  spores  produced.  At  32°C.  the 
number  of  spores  was  only  10  to  15  per  cent  of  that  produced  at  the 
optimum  temperature.  Reducing  the  temperature  of  incubation  to 
24°C.  reduced  the  numbers  of  spores  to  about  80  per  cent  of  the  maxi- 
mum. Thus,  a  small  temperature  increase  above  the  optimum  has  a 
much  greater  effect  upon  the  number  of  spores  produced  than  a  small 
decrease  in  temperature  below  the  optimum  (Fig.  58). 

In  nature,  fungi  are  exposed  to  fluctuating  temperatures.  Whether  a 
fluctuating  temperature  is  more  favorable  in  inducing  sporulation  than  a 
constant  temperature  appears  to  have  been  studied  but  little.  Jones 
(1946)  concluded  that  temperature  was  the  important  controlling  factor 
in  the  production  of  resistant  sporangia  of  Allomyces  arhiiscula  in  culture, 
and  he  beheved  that  "the  total  amount  of  temperature"  to  which  the 
cultures  were  subjected  was  more  important  than  the  maximum,  mini- 


SPORULATWN 


309 


mum,  or  degrees  of  fluctuation,  IMathur  et  al.  (1950)  reported  that  15 
to  20°C.  favors  conidium  formation  by  Colletotrichum  lindemuthianum  in 
culture.  Sporulation  was  less  at  25°C.  and  ceased  at  30°C.  Mrak  and 
Bonar  (1938)  found  that  temperature  influenced  the  relative  size  of  asci 
and  spores  of  Debaryomyces.  The  ascus  was  much  larger  than  the  spore 
cluster  at  4°C.,  but  the  spores  nearly  filled  the  ascus  at  25°C. 


13 


15 


7  9  11 

Incubation  period  (days) 

Fig.  58.  The  effects  of  temperature  and  time  of  incubation  on  sporulation  of  Piri- 
rularia  oryzae  on  rice-polish  agar.  (Courtesy  of  Henry  and  Andersen,  Phytopathology 
38 :  272,  1948.) 

An  interesting  selective  effect  of  temperature  upon  type  of  asexual 
sporulation  is  found  in  Choanephora  cucurbitarum  (Barnett  and  Lilly, 
1950).  This  fungus  produces  two  types  of  asexual  spores,  those  produced 
in  typical  sporangia  and  conidia  borne  in  heads.  Only  the  conidia  are 
found  commonly  in  nature,  while  both  types  are  abundant  in  culture. 
When  the  fungus  was  grown  in  Petri  dishes  at  25°C.,  87  per  cent  of  the 
reproductive  structures  were  conidial  heads,  while  13  per  cent  were 
sporangia  (Table  54).  When  the  temperature  was  increased  to  30°C., 
this  proportion  was  nearly  reversed.  At  31°C.  many  sporangia  but  no 
conidia  were  formed.  No  sporulation  occurred  at  34°C.,  but  mycelial 
growth  was  abundant.  Temperature  also  affected,  either  directly  or 
indirectly,  the  size  of  the  sporangia.  Those  produced  at  25°C.  averaged 
60  to  90  n  in  diameter,  while  those  formed  at  30  or  31°C.  were  much  larger, 
averaging  approximately  145  fi.  It  seems  likely  that  this  effect  is  indirect, 
being  a  reflection  of  the  relative  number  of  conidia,  which  are  formed  first 
under  favorable  conditions.  We  may  assume  that  the  production  of 
abundant  conidia  uses  much  of  the  food  materials  which  might  also  go 


310 


PHYSIOLOGY  OF  THE  FUNGI 


into  the  formation  of  sporangia.  Under  conditions  unfavorable  to 
conidium  production,  yet  favorable  to  sporangium  formation,  both  the 
size  and  abundance  of  sporangia  are  increased.  The  effect  of  temperature 
was  also  evident  when  pumpkin  flowers  artificially  inoculated  with  C. 
cucurbitarum  were  brought  into  the  laboratory  and  placed  at  30°C. 
Under  these  conditions  both  conidia  and  sporangia  were  produced. 


Table 


54.     The    Effect    of    Temperature    upon    Asexual    Reproduction    op 

Choanephora  cucurbitarum 
(Barnett  and  Lilly,  Phytopathology  40:  83,  1950.) 


Temperature  during 
sporulation,  °C. 

Conidial  heads 
per  culture 

Sporangia 
per  culture 

Average  size  of 
sporangia,  fx 

25 
30 
31 
34 

2,000 

150 

0 

0 

300 

1,300 

1,200 

0 

60-90 
148 
145 

Other  critical  temperature  studies  are  needed,  particularly  those 
designed  to  show  the  interrelated  effects  of  temperature  with  other 
environmental  or  nutritional  factors  and  to  determine  the  effects  of  tem- 
perature upon  the  "internal  environment"  of  the  fungi.  The  tempera- 
ture of  incubation  affects  zygospore  formation  by  Phycomyces  blakes- 
leeanus  indirectly  through  the  amount  of  acid  formed  in  the  medium 
(Robbins  and  Schmitt,  1945). 

Light.  Light  has  been  a  neglected  and  often  ignored  factor  in  many 
studies  of  sporulation.  Too  often  we  place  fungi  in  the  laboratory  or 
refrigerator  according  to  our  own  convenience,  not  to  their  needs,  and 
expect  them  to  reproduce  as  they  would  in  nature.  Under  natural  con- 
ditions many  fungi  fruit  only  when  exposed  to  light,  often  to  the  direct 
rays  of  the  sun,  for  a  part  of  the  time.  Numerous  observations  have  been 
reported  regarding  the  need  for  light,  but  too  few  of  these  reports  give 
data  as  to  the  intensity,  duration,  or  quality  of  the  light  required  to 
initiate  sporulation.  We  should  not  conclude  that  intensity  and  duration 
are  without  effect. 

A  review  of  the  early  work  on  the  influence  of  light  on  the  growth  and 
fruiting  of  the  fungi  is  presented  by  Coons  (1916).  Brefeld  (1877)  found 
that  some  species  of  Coprinus  failed  to  fruit  in  the  dark.  A  culture  of 
Coprimis  exposed  to  light  for  2  or  3  hr.  was  then  able  to  fruit  in  the  normal 
manner  when  removed  from  the  light.  He  also  found  that  higher  tem- 
peratures replaced,  in  part,  the  beneficial  effect  of  light  for  some  species. 

Sphaerographium  fraxini  produced  a  few  pycnidia  in  the  dark  at  30°C., 
whereas  none  were  produced  at  room  temperature  in  the  dark  (Leonian, 
1924).     Pycnidia   were    produced    at    room    temperature    in    the    light. 


SPORULATION  311 

Ascochyta  nymphaeae,  Cytosporella  mendax,  Endothia  parasitica,  Keller- 
mania  yuccagena,  Naemosphaera  sp.,  Plenodomus  destruens,  and  Phoma 
urens  formed  more  pycnidia  at  30°C.  in  the  dark  than  in  the  light  at  room 
temperature.  The  following  fungi  failed  to  fruit  in  the  dark  at  8°C.  but 
fruited  at  the  same  temperature  in  the  presence  of  light:  Hendersonia 
sp.,  Melanconium  hetulinum,  Naemosphaera  sp.,  Pestalotia  guepinia, 
Phoma  urens,  Phyllosticta  opuntiae,  Sphaerographium  fraxini ,  and  Sphaero- 
nema  pruinosum.  Light  favored  pycnidial  formation  by  Plenodomus 
fuscomacidans  (Coons,  191G).  The  above  examples  make  it  clear  that 
light  and  temperature  may  serve  as  interchangeable  stimuli  to  sporulation 
in  some,  but  not  all,  instances.  Since  the  response  (sporulation)  is  the 
same  whether  light  or  temperature  is  the  stimulus,  this  means  that  these 
stimuli  in  some  way  brought  about  the  same  or  equivalent  changes  in  the 
internal  environment  of  the  fungus. 

Drayton  (1937)  was  able  to  produce  the  perfect  stage  of  Botryotinia 
convoluta  by  controlling  light,  temperature,  and  nutrition.  The  technique 
is  somewhat  involved,  but  it  should  be  remembered  that  in  nature  the 
external  environment  varies  a  great  deal  during  the  course  of  a  year. 
Fluctuations  in  temperature,  moisture,  light,  and  food  supply  are  the 
normal  result  of  the  procession  of  the  seasons.  Drayton  found  autoclaved 
W'hole  wheat  to  be  an  excellent  substratum  for  this  fungus.  The  most 
favorable  results  w^ere  obtained  by  allowing  the  culture  to  develop  at 
14°C.  in  the  dark  for  45  days.  At  the  end  of  this  time  the  sclerotia  w^ere 
placed  in  moist  quartz  sand  at  0°C.  for  3  to  4  months,  then  stored  at 
5°C.  When  the  apothecial  fundaments  were  2  to  3  mm.  long,  the  cultures 
were  moved  to  a  greenhouse  and  placed  under  cheesecloth  and  the  tem- 
perature held  at  7°C.  at  night  and  below  15°C.  during  the  day.  The 
apothecia  matured  within  4  wrecks. 

Yarwood  (1936,  1941)  observed  parasitic  fungi  under  natural  condi- 
tions and  found  that  the  production  and  liberation  of  the  conidia  of 
Erysiphe  polygoni  and  the  ascospores  of  Taphrina  deformans  followed  a 
definite  diurnal  pattern  in  nature. 

The  combined  effects  of  temperature  and  light  upon  sporulation  of 
Helminthosporium  gramineum  are  clearly  shown  by  Houston  and  Oswald 
(1946).  Best  sporulation  was  obtained  under  outdoor  conditions,  with 
14  to  15  hr.  of  daylight  and  the  average  maximum  and  minimum  tem- 
peratures 26.8  and  8.2°C.,  respectively.  No  conidia  were  produced  on 
potato-glucose  agar  in  the  absence  of  light,  either  outdoors  or  inside. 
Artificial  light  apparently  was  less  effective  than  daylight.  However, 
continuous  light  at  13°C.  allowed  the  formation  of  a  few  conidia.  On 
pieces  of  infected  barley  leaves,  conidia  were  formed  without  exposure  to 
light,  over  a  considerable  range  in  temperature.  As  an  explanation  of 
these  differences,  the  authors  believe  that  the  mycelium  in  the  leaf  in 


312  PHYSIOLOGY  OF  THE  FUNGI 

nature  stored  up  the  "necessary  potentialities,"  which  then  permitted 
conidium  production  in  darkness.  MyceHum  growing  from  the  pieces 
of  leaf  into  the  agar  did  not  produce  spores  in  darkness.  This  is  an 
interesting  theory  regarding  a  possible  delayed  action  of  light  upon  sporu- 
lation.  It  also  seems  possible  that  the  leaf  tissue  of  the  host  may  furnish 
some  nutrient  necessary  for  sporulation  which  is  not  contained  in  potato- 
glucose  agar.  Perhaps  light  is  essential  to  the  synthesis  of  this  material 
by  the  fungus. 

It  was  demonstrated  recently  (Barnett  and  Lilly,  1950)  that  an  isolate 
of  Choanephora  cucurhitarum  requires  both  light  and  darkness  for  the  forma- 
tion of  conidia,  but  these  factors  have  little  or  no  apparent  influence  upon 
the  formation  of  sporangia.  This  fungus  was  grown  under  a  number  of 
conditions,  but  none  was  found  which  overcame  the  need  for  either  light 
or  darkness.  Cultures  incubated  in  the  laboratory  under  natural  alter- 
nating light  and  darkness  produced  abundant  conidial  heads  during  the 
second  and  third  nights  after  inoculation.  Exposure  to  artificial  light 
for  2  days  after  inoculation  followed  by  darkness  gave  similar  results,  but 
an  exposure  in  the  reverse  order  resulted  in  no  conidia.  Cultures  under 
continuous  artificial  light  (65  foot-candles)  and  those  in  total  continuous 
darkness  failed  to  form  conidial  heads.  Continuous  light  of  low  intensity 
(less  than  1  foot-candle) ,  however,  did  permit  the  formation  of  numerous 
conidial  heads  in  the  usual  period.  A  summary  of  the  important  results 
is  presented  in  Fig.  59,  together  with  an  outline  of  a  proposed  hypothesis 
to  explain  the  results.  We  may  assume  that  light,  or  its  absence,  affects 
two  metabolic  reactions,  or  groups  of  reactions,  which  are  essential  to 
conidium  formation  by  C.  cucurhitarum.  Light,  which  is  essential  to 
reaction  A,  apparently  inhibits  reaction  B,  which  must  occur  in  darkness 
or  weak  light.  The  reaction  in  light  must  be'^followed  by  the  reaction  in 
darkness,  if  conidia  are  to  be  formed.  Continuous  bright  light  favors 
only  reaction  A,  while  continuous  darkness  permits  only  reaction  B. 
Both  reactions  occur  simultaneously  in  continuous  light  of  low  intensity. 

A  different  isolate  of  C.  cucurhitarum  was  studied  by  Christenberry 
(1938),  who  found  that  alternating  periods  of  light  and  dark,  12  hr.  each, 
gave  the  best  sporulation.  Red-yellow  light  was  more  favorable  to 
conidium  formation  than  the  shorter  rays.  This  isolate  formed  conidia 
in  total  darkness. 

The  beneficial  effect  of  alternating  light,  or  a  period  of  light  followed 
by  darkness,  was  demonstrated  (Timnick  et  al.,  1951)  for  the  formation  of 
ascospores  by  Diaporthe  phaseolorum  var.  hatatafis.  Cultures  grown  in 
continuous  darkness  formed  only  a  few  perithecia,  which  contained 
abundant  ascospores.  In  continuous  bright  light  numerous  perithecia 
were  formed,  but  relatively  fewer  ascospores  were  produced.  A  long 
period  of  light  followed  by  darkness  gave  many  perithecia  with  abundant 
ascospores. 


SPORULATION 


313 


Marked  morphologic  differences  were  found  in  strains  of  Fusarium 
subjected  to  different  exposures  of  light  and  darkness  (Snyder  and  Han- 
sen, 1941).  Some  of  the  characters  affected  were  color,  zonation,  type  of 
colony,  presence  or  absence  of  sporodochia,  occurrence  of  the  perithecial 
stage,  and  size,  shape,  and  septation  of  macroconidia.  Light  was  usually 
found  necessary  for  the  formation  of  macroconidia.  Exposures  were 
made  to  continuous  total  darkness  but  not  to  continuous  light.  Evidence 
in  these  experiments  indicated  that  the  effect  of  light  is  only  upon  the 
actively  growing  portion  of  the  mycelium. 


Bright  light 


Reaction  A 


Darkness 


Reaction  B 


=  Conidia 


Continuous  bright  light 


Reaction  A 


Continuous  darkness 


Reaction  B 


=  No  conidia 


=  No  conidia 


Bright  light 


Reaction  A 


=  No  conidia 


Continuous  light  low  intensity 


Conidia 


Reactions  A  +  B  simultaneously 
Fig.  59.  Conidium  formation  by  Choanephora  cucurbitarum  under  different  light 
conditions,  shownig  the  possible  metabolic  reactions  controlled  by  light.  Under 
variable  conditions,  the  cultures  were  exposed  to  the  first  condition  (on  the  left)  for  2 
days  and  to  the  second  condition  for  24  hr.  (After  Barnett  and  Lilly,  Phytopathology 
40:  88,  1950.) 

The  length  of  exposure  necessary  to  stimulate  spore  formation  may  be 
very  short,  as  demonstrated  by  Bisby  (1925)  for  Fusarium  discolor  sul- 
phureum.  He  observed  that  brief  exposure  to  light,  while  Petri  dish 
cultures  were  being  examined,  resulted  in  the  formation  of  rings  of  conidia. 
Using  a  photographic  shutter,  he  further  demonstrated  that  an  exposure 
as  brief  as  }i  sec.  to  outdoor  light  on  a  bright  day  was  sufficient  to  stimu- 
late the  formation  of  a  ring  of  conidia. 

Coons  (1916),  in  his  work  with  Plenodomus  fuscomacidans,  reasoned 
that  the  effect  of  light  might  be  replaced  by  various  oxidizing  agents,  since 
light  is  known  to  promote  various  oxidations.  Cultures  treated  with 
hydrogen  peroxide  and  other  oxidizing  agents  produced  a  few  pycnidia. 
The  age  of  the  culture  when  these  chemicals  were  added  was  important. 


314 


PHYSIOLOGY  OF  THE  FUNGI 


For  these  chemicals  to  stimulate  pycnidium  formation,  the  culture  had 
to  be  in  such  a  physiological  condition  that  1-hr.  exposure  to  light  would 
induce  sporulation. 

The  sporulation  of  a  number  of  other  species  in  our  laboratory  has  been 
observed  to  be  influenced  by  the  presence  or  absence  of  light  (Figs.  60, 
61).     Among  these  are  Dendrophoma  obscurans,  Trichoderma  lignorum, 


Fig.  60.  The  effects  of  light  on  the 
production  of  conidia  by  Trichoderma 
lignorum  after  3  days  at  25°C.  A, 
exposed  to  continuous  artificial  light. 
Note  the  more  or  less  even  distribution 
of  conidia.  B,  exposed  to  alternate 
ight  and  darkness,  12  hr.  each.  Note 
the  rings  of  conidia.  C,  grown  in 
continuous  darkness.  Note  the  ab- 
sence of  conidia. 


Sphaeropsis  malorum,  Ceratostomella  ulmi,  Botrytis  sp.,  Endothia  para- 
sitica, Septoria  nodorum.  The  reaction  of  some  fungi  to  light  is  appar- 
ently dependent,  to  a  certain  extent,  upon  the  composition  of  the  medium. 

Still  another  effect  of  light  should  be  emphasized,  i.e.,  the  inhibitory 
effect.  The  depressing  effect  of  strong  light  upon  growth  and  length  of 
sporangiophores  of  Phy corny ces  hlakesleeanus  is  easily  demonstrated. 
Elfving  (1890)  noted  that  the  amount  of  inhibition  of  growth  by  light 
varied  with  the  composition  of  the  medium. 

Ultraviolet  light.  The  destructive  action  of  sunlight  upon  micro- 
organisms, especially  bacteria,  was  recognized  about  the  time  that  pure- 


SPORULATION 


315 


culture  methods  came  into  wide  use.  The  lethal  action  of  ultraviolet 
light  is  conditioned  by  the  wave  length  of  the  irradiation,  by  the  time  of 
exposure,  and  by  the  particular  nature  of  the  microorganism.  A  con- 
siderable number  of  investigators  have  studied  the  effect  of  ultraviolet 
radiation  upon  sporulation.  Both  favorable  and  unfavorable  results 
have  been  obtained.     It  should  be  recognized  that  length  of  exposure 


Fig.  61.  The  effect  of  light  on  the  production  of  pycnidia  by  an  isolate  of  Dendro- 
phoma  obscurans  when  grown  on  malt  extract-agar  plates  at  25°C.  A,  grown  under 
continuous  artificial  light,  .\lternate  light  and  darkness  gave  similar  results.  B, 
grown  in  continuous  darkness. 

is  a  very  important  factor  in  these  experiments.  In  addition,  the  medium 
used,  the  age  of  the  culture,  and  the  temperature  rise  during  irradiation 
also  modify  the  results. 

Stevens  (1928)  found  that  ultraviolet  radiation  induced  the  formation 
of  perithecia  by  various  isolates  of  Glomerella  cingulata  a  few  days  after 
irradiation.  While  old  cultures  produced  a  few  perithecia  without 
irradiation,  many  more  were  produced  by  young  cultures  within  a  short 
time  following  irradiation.  One  effect  of  such  irradiation  is  the  killing 
of  the  aerial  mycelium.  Short  exposures  allowed  the  formation  of  super- 
ficial perithecia,  while  long  exposures  prevented  their  formation.  The 
majority  of  the  perithecia  formed  following  intermediate  dosages  were 
embedded  in  the  medium.     It  was  noted  that  the  age  of  the  mycelium 


316  PHYSIOLOGY  OF  THE  FUNGI 

at  the  time  of  irradiation  had  an  effect  on  the  number  of  perithecia  formed. 
Colonies  4  days  old  when  irradiated  produced  perithecia,  which  were  most 
abundant  on  mycelium  1  day  old  at  the  time  of  irradiation.  Irradiation 
of  colonies  12  days  old  led  to  the  formation  of  but  few  perithecia.  No 
evidence  was  obtained  that  irradiation  of  the  medium  alone  had  any 
effect  on  perithecium  formation.  A  species  of  Coniothyrium  which 
formed  pj^cnidia  only  after  the  cultures  were  very  old  was  stimulated  to 
produce  pycnidia  within  3  days  after  irradiation.  This  work  of  Stevens 
is  apparently  the  first  which  demonstrated  that  ultraviolet  radiation 
stimulated  sporulation  by  fungi. 

Spore  production  by  Macrosporium  tomato  and  Fusarium  cepae  was 
greatly  increased  by  the  proper  exposure  to  ultraviolet  radiation  (Ramsey 
and  Bailey,  1930).  A  12-  to  15-fold  increase  in  the  numbers  of  spores 
produced  by  these  two  species  was  obtained  by  the  optimum  exposure. 
These  investigators  also  showed  that  irradiation  of  the  medium  before 
inoculation  had  no  subsequent  effect  on  sporulation  by  these  two  fungi. 
The  range  of  wave  lengths  which  stimulated  the  most  abundant  sporula- 
tion  was  found  to  be  2,300  to  2,800  A.  Smith  (1935)  points  out  that 
many  workers  have  neglected  the  precaution  of  controlling  the  tempera- 
ture of  cultures  during  irradiation.  She  found  it  necessary  to  control 
the  temperature  of  the  cultures  of  Fusarium  eumartii  in  order  to  separate 
the  effects  of  increased  temperature  and  ultraviolet  radiation. 

Ultraviolet  radiation  stimulated  or  depressed  sporulation  of  Diaporthe 
phaseolorum  var.  hatatatis  depending  on  the  medium  used  (Timnick 
et  al.,  1951).  Neither  stromata  nor  perithecia  were  formed  on  casein 
hydrolysate-glucose  medium,  unless  the  cultures  were  irradiated.  Cul- 
tures grown  on  potato-glucose  agar  produced  stromata  and  long-beaked 
perithecia  without  irradiation.  Irradiation  of  cultures  on  potato-glucose 
medium  resulted  in  the  formation  of  fewer  and  smaller  short-beaked 
perithecia.  Although  the  mode  of  action  of  ultraviolet  radiation  in 
stimulating  sporulation  is  unknown,  long  exposures  are  known  to  be 
lethal.  We  may  assume  that  even  short  exposures  injure  or  kill  some  of 
the  exposed  cells.  Perhaps  some  substance  is  thereby  released  which 
stimulates  sporulation.  The  presence  of  such  a  substance  in  the  potato- 
glucose  medium  might  explain  why  irradiation  was  not  necessary  for  the 
production  of  perithecia  by  D.  phaseolorum  var.  hatatatis  on  this  medium 

Aeration.  Although  the  fungi  are  aerobic  organisms,  the  amount  of 
free  oxygen  that  they  need  to  carry  out  their  life  processes  varies  from 
fungus  to  fungus.  The  amount  of  oxygen  required  is  less  for  growth  than 
for  reproduction.  The  aquatic  fungi  would  be  expected  to  grow  and 
reproduce  in  a  more  limited  supply  of  oxygen  than  terrestial  forms. 
While  many  aquatic  Phycomycetes  produce  their  spores  under  water,  a 
large  number  of  fungi  fail  to  fruit  until  some  aerial  mycelium  has  been 


SPORULATION  317 

formed.  Examples  of  the  inhibiting  effect  of  insufficient  aeration  on 
sporulation  are  numerous.  Coons  (1916)  found  that  lowered  oxygen  ten- 
sion inhibited  pycnidium  formation  by  Plenodomusfuscomaculans,  though 
there  was  still  sufficient  oxygen  supply  to  allow  some  growth.  Leonian 
(1924)  tested  the  effect  of  reduced  oxygen  on  pj^cnidium  formation  by 
various  Sphaeropsidales.  This  experiment  was  carried  out  by  culturing 
these  fungi  in  Petri  dishes,  some  of  which  were  placed  in  desiccators,  while 
the  controls  were  placed  on  a  table.  The  following  fungi  produced  fewer 
pycnidia  in  sealed  desiccators  than  in  the  control  cultures:  Ascochyta 
tiym'phaeae,  Phoma  urens,  Plenodomus  destruens,  Phyllosticta  opuntiae,  and 
Septosporium  acerinum.  It  is  possible  that  this  effect  may  have  been 
due  to  the  increased  concentration  of  carbon  dioxide  in  the  closed  vessels. 
Denny  (1933)  made  an  accurate  study  of  the  effect  of  oxygen  supply  on 
growth  and  formation  of  perithecia  by  Neurospora  sitophila.  Only  a 
trace  of  oxygen  was  required  for  limited  growth,  for  it  was  necessary  to 
keep  cultures  in  the  presence  of  alkaline  pyrogallol  to  inhibit  growth 
entirely.  Oxygen  concentrations  of  less  than  0.5  per  cent  inhibited 
perithecium  formation  for  30  days,  while  perithecia  formed  in  air  within 
4  days.  This  paper  should  be  consulted  for  the  details  of  conducting 
experiments  of  this  nature  under  closely  controlled  conditions.  Some  of 
Denny's  data  are  given  in  Table  55. 

Table  55.     The  Effect  of  Oxygen  Concentration  on  the  Formation  of  Peri- 
thecia BY  Neurospora  sitophila 
(Prepared  from  the  data  of  Denny,  1933.     Contribs.  Boyce  Thompson  Inst.  5,  1933.) 
Oxygen  Cbncentration,  %        Days  Required  to 

Form  Perithecia 
20.8  4 

9.4  7 
3.75  9 

1.5  12 

0.24  None  at  30  days 

Conidium  production  by  Choanephora  cucurhitarum  was  poor  in  tight- 
fitting  Petri  dishes  (Barnett  and  Lilly,  1950).  Sealing  the  dishes  pre- 
vented conidium  formation,  while  well-aerated  dishes  allowed  abundant 
conidial  heads  to  form.  Failure  to  form  conidia  under  these  conditions 
may  be  due  to  (1)  insufficient  oxygen  supply,  (2)  the  accumulation  of 
toxic,  volatile,  metabolic  by-products,  (3)  increased  carbon  dioxide  con- 
tent, or  (4)  unfavorable  humidity. 

Adequate  aeration  was  one  of  the  most  important  environmental  fac- 
tors necessary  for  conidium  formation  by  Piricularia  oryzae  (Heniy  and 
Andersen,  1948).  The  cultures  emitted  a  strong  odor  of  ammonia  after 
a  few  days'  incubation.  It  was  believed  that  aeration  removed  the 
ammonia   and   other  volatile  metabolic   by-products  which  prevented 


318 


PHYSIOLOGY  OF  THE  FUNGI 


abundant  sporulation.  Forced  aeration  of  the  culture  flasks  at  the  rate 
of  4  ml.  of  air  per  minute  per  milligram  of  oats-sorghum  medium  was 
found  to  be  optimum  for  sporulation.  Mader  (1943)  discussed  the 
factors  inhibiting  fruiting  of  Agnricus  campestris  and  concluded  that 
volatile  substances  are  important,  and  that  they  must  be  removed  by 
aeration  of  mushroom  cellars. 

Hydrogen-ion  concentration.  The  early  workers  recognized  that  the 
acidity  of  the  medium  influenced  sporulation.  Lock  wood  (1937)  studied 
the  formation  of  perithecia  and  asci  by  Penicillium  javanicum,  Aspergillus 
herhariorum,   and   Chaetomium.  globosum   in   buffered   media  of  various 


Fig.  62.  The  effect  of  glutamic  acid  on  gametic  reproduction  of  Phycomyces  hlakes- 
leeanus  at  26°C.  Left,  basal  medium;  right,  basal  medium  plus  10  mg.  d-glutamic 
acid,  neutraHzed  with  CaCOs.  Note  the  line  of  progametes  in  the  plate  on  the  right. 
Age,  6  days.     (Courtesy  of  Robbins  and  Schmitt,  A7n.  Jour.  Botany  32  :  321,  1945.) 

hydrogen-ion  concentrations  and  found  that  the  perithecia  produced  in 
the  more  acid  solutions  contained  few  if  any  asci  with  ascospores.  The 
percentage  of  fertile  perithecia  increased  as  the  pH  was  increased  to  7  or 
8.  Similarly,  in  our  laboratory,  we  have  noted  that  A .  rugulosus  produces 
many  perithecia  and  few  conidia  at  an  initial  pH  value  of  6  to  8,  while 
conidia  but  no  perithecia  form  at  pH  3  to  4. 

Robbins  and  Schmitt  (1945)  studied  the  sexual  reproduction  of  Phyco- 
myces hlakesleeanus  on  glucose-asparagine  medium  and  found  that  mature 
zygospores  did  not  form  at  26°C.  Zygospores  formed  when  various 
protein  hydrolysates,  amino  acids  (especially  glutamic  acid),  or  various 
organic  acids  were  added  to  the  medium.  These  buffers  prevented  the 
pH  from  falling  low  enough  to  inhibit  zygospore  formation  (Fig.  62). 
These  authors  also  noted  that  P.  hlakesleeanus  on  glucose-asparagine 
medium  produced  zygospores  at  20°C.  This  is  evidence  that  the  com- 
position of  the  medium  has  a  profound  effect  on  reproduction.     In  this 


SPORULATION  319 

instance,  it  was  possible  to  trace  the  connection  between  temperature  and 
the  composition  of  the  medium  to  a  specific  factor,  i.e.,  acidity 

Perithecia  were  not  formed  by  Sordaria  fimicola  until  the  pH  of  the 
culture  medium  was  6.5  or  greater  (Lilly  and  Barnett,  1947).  While 
acidity  of  the  medium  was  not  the  only  controlling  factor  affecting  the 
formation  of  perithecia  by  S.  fimicola,  perithecia  never  formed  when  the 
pH  was  less  than  6.5,  however  favorable  the  other  external  conditions 
were. 

OTHER  PHYSICAL  FACTORS 

It  has  frequently  been  observed  that  many  species  of  fungi  fruit  more 
readily  when  grown  upon  a  solid  or  semisolid  substratum  than  they  do  in 
liquid  media.  Leonian  (1924)  reported  that  only  6  out  of  20  species 
studied  formed  pycnidia  as  readily  in  liquid  medium  as  on  solid  medium. 
He  concluded  that  the  beneficial  effect  of  solid  media  was  due  to  better 
aeration  and  free  transpiration. 

The  favorable  effect  of  ozone  upon  the  formation  of  pycnidia  and  spores 
of  a  limited  number  of  fungi  was  recently  reported  by  Richards  (1949). 
The  production  of  viable  conidia  of  three  species  of  Alternaria  was  greatly 
increased  on  exposure  to  ozone.  Although  conidium  formation  of 
Mycosphaerella  citrullina  was  increased  by  exposure  to  ozone,  the  spores 
formed  did  not  germinate. 

The  transformation  and  elongation  of  basidia  of  certain  Polyporaceae 
in  nature  and  under  controlled  conditions  has  been  correlated  with  high 
humidity  by  Bose  (1943).  It  seems  likely  that  the  humidity  of  the 
atmosphere  may  have  a  greater  influence  upon  conidium  formation  in  the 
aerial  fungi  than  is  generally  supposed.  In  Rhizopus,  for  instance,  much 
more  liquid  moves  upward  through  the  sporangiophore  than  can  be  con- 
tained within  the  sporangium.  A  high  percentage  of  this  water  must  be 
transpired  in  order  to  condense  the  protoplasm  and  food  materials  stored 
in  the  spores.  A  change  in  relative  humidity  must  affect  the  rate  of 
transpiration.  On  the  other  hand,  Ternetz  (1900)  found  that  a  humidity 
of  98  per  cent  or  higher  was  necessary  for  fruit-body  production  by 
Ascophanus  carneus.  Actually,  we  know  little  about  the  influence  of 
humidity,  and  much  more  information  is  needed  on  this  subject. 

Emerson  and  Cantino  (1948)  showed  that  the  presence  of  high  concen- 
trations of  carbon  dioxide  favored  the  production  of  resistant  sporangia 
by  Blastocladia  pringsheimii. 

Mutilation  of  the  mycelium,  which  would  cause  the  death  and  release 
of  cellular  constituents,  has  been  used  to  stimulate  sporulation  (see  Rands, 
1917;  Kunkel,  1918;  and  McCallan  and  Chan,  1944).  Scraping  of  the 
mycelium  of  Alternaria  solani  followed  by  a  brief  exposure  to  ultraviolet 
rays  was  used  successfully  by  McCallan  and  Chan  (Fig.  63). 


320 


PHYSIOLOGY  OF  THE  FUNGI 


10,000 


<n 
o 

z 
< 

CO 

o 

I 


o 

LJ 


cr 
o 
a. 


1,000    r 


100    - 


640 


1280        2560 


20  40  80  160  320 

EXPOSURE    TIME  IN    SECONDS 
Fig.  63.     Effect  of  time  of  exposure  to  ultraviolet  radiation  on  the  production    of 
spores   from   scraped   and   unscraped   cultures   of   AUernaria   solarii.      (Courtesy   of 
McCallan  and  Chan,  Contribs.  Boyce  Thompson  Inst.  13 :  327,  1944.) 


NUTRITIONAL  FACTORS 

The  nutritional  conditions  under  which  a  fungus  produces  reproductive 
bodies  and  spores  are  often  quite  different  from  those  which  are  optimum 
for  vegetative  growth.  Not  all  media  are  equally  suitable  for  sporula- 
tion.  The  frequent  failure  to  obtain  sporulation  of  many  common  fungi 
in  culture,  even  though  they  grow  profusely,  testifies  to  the  extent  of  our 
ignorance  regarding  the  necessary  nutritional  factors.  However,  the 
following  factors  have  been  shown  to  be  important:  concentration  of 
medium,  carbon  and  nitrogen  sources,  carbon-nitrogen  ratio,  micro 
essential  elements,  specific  reproductive  factors,  and  vitamins. 

Concentration  of  nutrients.  Among  the  early  workers,  Klebs  (1900) 
gave  a  great  deal  of  attention  to  the  effect  of  nutrient  concentration  upon 
reproduction.  For  most  of  the  fungi  with  which  he  worked,  exhaustion 
of  the  food  supply  favored  sporulation.  Klebs  (1899)  kept  a  culture  of 
Saprolegnia  mixta  in  the  vegetative  condition  for  2^^  years  by  constant 
renewal  of  the  nutrient  solution.  Yet,  this  fungus  produced  spores  within 
a  few  days  when  the  food  supply  became  exhausted.  The  same  principle 
holds  true  for  the  Myxomycetes  as  well  as  the  filamentous  fungi.  Camp 
(1937)  grew  Physarum  polycephalum  and  studied  the  effect  of  the  number 


SPORULATION  321 

of  feedings  upon  the  time  of  fruiting.  The  protoplasm  continued  to 
grow  as  long  as  there  was  abundant  food,  but  when  the  food  was 
exhausted,  the  sHme  mold  passed  into  the  fruiting  stage. 

Leonian  (1923,  1924)  used  a  technique  in  studying  sporulation  which 
consisted  in  growing  a  fungus  in  a  medium  suitable  for  vigorous  vegetative 
growth,  and  then  transferring  it  to  solutions  of  different  concentrations 
to  stimulate  sporulation.  When  sterile  mycelium  of  Valsa  leucostoma 
was  transferred  from  a  medium  containing  1.5  per  cent  nutrients  to  a 
medium  containing  0.37  per  cent  nutrients  the  ratio  of  perithecia  to 
pycnidia  increased.  Transferring  sterile  mycelium  to  more  concentrated 
nutrient  solutions  favored  the  production  of  pycnidia  and  decreased  the 
number  of  perithecia  formed.  Endothia  'parasitica  showed  a  decrease  in 
the  number  of  pycnidia  when  the  mycelium  was  transferred  from  a  weak 
to  a  concentrated  medium.  When  the  sterile  mycelium  was  grown  in  a 
concentrated  medium  and  transferred  to  distilled  water,  the  pycnidia 
did  not  mature  but  an  enormous  number  of  pycnidium  initials  were 
formed.  If  such  a  culture  were  then  transferred  back  to  a  concentrated 
medium,  maximum  sporulation  was  obtained.  A  review  of  the  literature 
on  the  effect  of  concentration  on  fruiting  is  given  by  Leonian  (1924). 

From  the  above  examples  we  may  conclude  that  the  concentration  of 
nutrients  in  a  medium  may  have  a  profound  influence  upon  fruiting,  and 
that  the  different  types  of  fruiting  (sexual  and  asexual)  may  have  different 
requirements.  Not  only  the  amounts  of  the  different  nutrients  but  the 
proper  balance  between  the  components  of  the  medium  may  be  essential 
for  maximum  sporulation. 

Nitrogen  source.  The  source  of  nitrogen  influenced  the  formation  of 
pycnidia  and  spores  by  Phyllosticta  solitaria  (Mix,  1933).  The  specificity 
of  the  nitrogen  source  was  greater  for  the  production  of  spores  than  for  the 
formation  of  pycnidia.  The  different  isolates  of  this  fungus  responded 
differently  to  the  various  nitrogen  sources.  Nitrate  nitrogen  was  the 
most  favorable.  This  may  have  been  due  to  an  indirect  effect  on  the  pH 
of  the  medium,  for  this  fungus  sporulates  only  between  pH  4.2  and  5.8. 

In  our  laboratory  we  have  observed  that  sporulation  of  some  fungi  is 
favored  by  certain  sources  of  nitrogen,  which  are  not  necessarily  the  same 
as  those  which  are  favorable  for  growth  (Fig.  64) .  A  few  of  these  species 
with  the  more  favorable  nitrogen  sources  for  sporulation  are  Monilinia 
fructicola,  ammonium  tartrate,  glycine;  Phoma  betae,  glycine;  Neo- 
cosmopara  vasinfeda,  glutamic  acid,  glycine;  Septoria  nodorum,  glycine; 
Diaporthe  phaseolorum  var.  batatatis,  asparagine;  Choanephora  cucur- 
bitarum,  organic  nitrogen. 

Carbon  source.  Not  all  carbon  sources  are  equally  suitable  for  fruit- 
ing of  fungi.  Some  which  are  favorable  for  mycelial  growth  do  not  favor 
sporulation.     Hawker  (1939)  found  the  number  of  perithecia  produced 


322 


PHYSIOLOGY  OF  THE  FUNGI 


A  B  C 

Fig.  64.  The  effects  of  different  nitrogen  sources  on  sporulation  of  three  fungi  after 
19  days  on  a  ghicose-sucrose  medium  at  25°C.  The  nitrogen  sources  are:  A,  apsaragine; 
B,  casein  hydrolysate;  C,  potassium  nitrate;  D,  ammonium  sulfate;  E,  ammonium 
tartrate;  F,   glycine.     The  fungi  are:  top,  Glornerella  cingulata;   middle,  Pleurage 


SPORULATION 


323 


D  E  F 

^^.Jcolla;  bottom,  Melanospora  sp.  Note  that  asparagine  and  casein  hydrolysate 
are  good  sources  of  nitrogen  for  spore  production  of  all  three  fungi.  Discharged  spores 
of  P.  curvicolla  are  evident  only  on  these  two  media. 


curvic 


324 


PHYSIOLOGY  OF  THE  FUNGI 


by  Melanospora  destruens  to  be  influenced  by  the  concentration  and  kind 
of  sugar  used.  Glucose,  fructose,  or  an  equimolar  mixture  of  these 
sugars,  when  used  at  the  rate  of  5  g.  per  hter,  allowed  the  production  of 
perithecia,  but  no  perithecia  were  formed  when  50  g.  was  used.  Many 
perithecia  were  produced  when  50  g.  per  liter  of  sucrose  was  used. 

The  favorable  effect  of  sucrose  on  perithecial  formation  was  replaced 
by  various  hexose  phosphate  esters.  Glucose-1-phosphate  and  fructose- 
1,6-diphosphate  were  equally  active.  In  view  of  their  ready  enzymatic 
interconvertibility  in  organisms,  this  would  be  expected.     These  results 


V  ^^'.  Warn-       ^^MF  *  ^M 

ABC  D  E 

Fig.  65.  The  effects  of  different  carbon  sources  on  the  production  of  conidia  by 
Glomerella  cingidata  after  22  days  on  asparagine  medium  at  25°C.  The  carbon 
sources  are:  A,  glucose;  B,  sucrose;  C,  maltose;  D,  sorbose;  E,  starch.  Note  that 
sporulation  is  greatest  on  sucrose  and  least  on  starch. 

suggest  that  M.  destruens  phosphorylates  sucrose  with  greater  ease  than 
either  glucose  or  fructose.  This  is  in  line  with  the  experiments  of  Dou- 
doroff  (1945)  with  growth  of  Pseudomonas  saccharophila.  Since  M. 
destruens  makes  better  growth  upon  glucose  than  upon  sucrose,  it  may  be 
suggested  that  the  pathway  of  carbohydrate  utilization  is  different  in 
growth  and  reproduction.  These  findings  emphasize  again  that  the 
requirements  for  growth  and  reproduction  may  be  different. 

Glucose,  mannose,  fructose,  lactose,  and  sucrose  are  reported  (Mix, 
1933)  as  favorable  for  pycnidium  formation  by  Phyllosticta  solitaria. 
Lactose  was  the  most  favorable  sugar  for  the  production  of  perithecia  by 
Diaporthe  phaseolorum  var.  hatatatis  (Timnick  et  al.,  1951).  Brodie  (1948) 
induced  Cyathus  stercoreus  to  produce  normal,  fertile  fruit  bodies  on 
semisynthetic  media  containing  filter  paper. 

In  our  laboratory  we  have  observed  that  the  carbon  source  affects 
reproduction  of  a  number  of  other  fungi,  and  that  the  best  source  for 
sporulation  is  not  always  the  same  which  yields  maximum  vegetative 
growth  (Figs.  65  and  66).     Some  of  these  fungi  with  some  more  favorable 


SPORULATION 


325 


carbon  sources  for  reproduction  are  Aspergillus  niger,  glucose,  sorbose, 
sucrose;  Glomerella  cingidata,  sucrose;  Phoma  betae,  sucrose;  Monilinia 
fructicola,  sorbose,  sucrose;  Neocosmopara  vasinfecta,  maltose,  starch, 
glucose;  Pleurage  curvicolla,  maltose,  starch. 


ABC 
Fig.  66.  The  effect  of  three  carbon  sources  and  time  on  the  production  of  perithecia 
by  Melanospora  sp.  on  asparagine  medium  at  25°C.  The  carbon  sources  were:  A, 
glucose;  B,  sucrose;  C,  maltose.  Above,  cultures  11  days  old;  below,  the  same  cul- 
tures 22  days  old.  Note  the  poor  vegetative  growth  but  presence  of  perithecia  on 
sucrose  and  maltose,  and  the  abundant  early  vegetative  growth  but  delayed  produc- 
tion of  perithecia  on  glucose. 

For  further  information  on  the  effects  of  nutritional  factors  on  sporula- 
tion,  see  Hawker  (1950). 

Carbon-nitrogen  ratio.  It  seems  to  be  generally  held  that  a  proper 
balance  among  the  constituents  of  the  medium  is  quite  important  in 
growth  and  sporulation.  Westergaard  and  Mitchell  (1947)  investigated, 
among  other  factors,  the  influence  of  the  carbon-nitrogen  ratio  of  the 
medium  on  formation  of  perithecia  by  Neurospora  crassa.  Some  of  their 
data  are  given  in  Table  56.     It  is  evident  that  high  concentrations  of 


326 


PHYSIOLOGY  OF  THE  FUNGI 


glucose  and  potassium  nitrate  are  unfavorable  for  the  production  of 
perithecia  by  N.  crassa. 

Table  56.     The  Effect  of  the  Carbon-Nitrogen  Ratio  of  the  Medium  on  the 
Production  of  Perithecia  by  Neurospora  crassa 
Production  rated  on  scale  of   10.     Age,   11  days.     (Westergaard  and   Mitchell, 
Am.  Jour.  Botany  34,  1947.) 


Glucose  con- 

KNO3 

concentration,  % 

centration,  % 

0.001 

0.01 

0.05 

0.1 

0.5 

1.0 

0.2 

1 

3 

3 

3 

2 

1 

0.6 

2 

4 

6 

5 

3 

1 

1.0 

2 

5 

7 

7 

3 

1 

1.4 

1 

4 

7 

9 

3 

1 

1.8 

3 

5 

9 

10 

3 

2 

2.2 

1 

4 

7 

9 

2 

0 

Micro  essential  elements.  Steinberg  (see  references  in  Chap.  5)  found 
the  sporulation  of  Aspergillus  niger  to  be  depressed  by  the  omission  of 
various  of  the  essential  elements.  Lockwood  and  Ward  (1936)  found 
that  Rhizopus  oryzae  sporulated  on  the  thirteenth  day  of  incubation  when 
zinc  was  not  added  to  the  medium.  When  zinc  was  added,  sporulation 
occurred  on  the  third  day  of  incubation.  In  general,  when  any  essential 
element  is  low,  sporulation  tends  to  be  depressed  before  growth  is 
inhibited. 

Specific  reproductive  factors.  While  many  factors  may  influence 
reproduction,  there  is  little  evidence  that  the  fungi  need  specific  chemical 
substances  to  induce  reproduction.  Such  factors,  however,  do  exist 
among  certain  of  the  Phycomycetes. 

Four  specific  regulatory  substances,  called  hormones  (Raper,  1942,  and 
Raper  and  Haagen-Smit,  1942),  were  shown  to  initiate  and  control  sexual 
reproduction  of  Achlya  hisexualis.  A  specific  substratum  is  required  for 
the  production  of  hormone  A  in  large  quantities.  Hempseed  allows  a 
2-  to  10-fold  production  of  hormone  A  over  that  produced  by  similar  cul- 
tures grown  upon  corn,  rice,  lentils,  or  other  substances.  Hormone  A 
has  been  concentrated  70,000-fold  but  has  not  yet  been  obtained  in  pure 
form.     This  concentrate  of  hormone  A  is  active  in  dilutions  of  1  X  10^^^. 

Sexual  reproduction  in  Phytophthora  cactorum  was  greatly  stimulated  by 
an  extract  of  garden  peas  (Leonian,  1936).  This  substance  had  no 
growth-promoting  properties.  It  was  concluded  (Leonian  and  Lilly, 
1937)  that  the  sexuality  factor  was  none  of  the  known  vitamins,  and  that 
it  probably  was  not  carotene  or  xanthophyll,  although  it  was  concentrated 
by  methods  which  would  concentrate  these  substances.     When  vigorous 


SPORULATION  327 

sterile  mycelium  of  P.  cactorum  was  washed  in  distilled  water  and  trans- 
ferred to  the  optimum  concentration  of  the  sexuality  factor  in  0.1  per 
cent  agar,  oogonia  began  to  appear  within  15  hr.,  reaching  the  maximum 
development  in  3  days.  This  factor  was  also  effective  in  inducing  sexual 
reproduction  by  P.  erythroseptica,  P.  boehmeriae,  and  P.  megasperma.  In 
addition,  the  presence  of  this  sexuality  factor  induced  the  formation  of 
abundant  oogonia  within  a  week  by  15  of  the  20  species  of  Phythium 
tested.  However,  it  failed  to  induce  sporulation  of  various  Zygomycetes, 
Ascomycetes,  and  Basidiomycetes. 

7-Carotene  is  associated  with  the  male  cells  of  the  sexual  phase  and 
not  with  the  female  cells  of  Allomyces  (Emerson  and  Fox,  1940).  This 
specificity  of  association  with  the  male  cells  indicates  that  7-carotene  may 
be  associated  with  sexual  reproduction  in  some  species  of  this  genus. 
No  7-carotene  was  found  in  the  cells  of  these  fungi  in  the  asexual  phase. 

Association  with  other  organisms.  That  one  fungus  may  influence  the 
sporulation  of  another  has  been  known  for  a  long  time.  Sporulation  of 
Alternaria  and  Helminthosporium  was  increased  when  they  were  grown 
in  association  with  certain  other  organisms  (Porter,  1924).  This  paper 
has  a  valuable  bibliography  on  associative  effects. 

The  metabolic  products  of  Aspergillus  niger  are  known  to  promote 
conjugation  in  three  species  of  yeasts  belonging  to  the  genus  Zygosac- 
charomyces  (Nickerson  and  Thimann,  1943).  As  a  result  of  extensive 
investigations,  these  authors  found  that  part  of  the  activity  of  Aspergillus 
filtrate  could  be  replaced  by  glutaric  acid  and  riboflavin.  While  these 
substances  had  some  activity  w^hen  tested  separately,  the  combination  of 
glutaric  acid  and  riboflavin  greatly  exceeded  the  activity  of  either  alone. 
Riboflavin  was  shown  definitely  to  be  a  component  of  the  Aspergillus 
filtrate.  An  autolysate  from  Zygosaccharomyces  cells  had  a  favorable 
influence  on  sporulation  by  the  same  organism.  From  this  experiment 
and  from  the  work  of  Lindegren  and  Hamilton  (1944),  who  found  that 
ascus  formation  in  yeast  would  take  place  only  in  portions  of  the  yeast 
colony  where  autolysis  had  taken  place,  it  may  be  concluded  that  auto- 
lytic  products  favor  sporulation  in  some  instances.  Lindegren  and 
Lindegren  (1944)  found  that  addition  of  2  per  cent  dried  brewer's  yeast 
to  a  presporulation  medium  very  favorably  influenced  the  sporulation  of 
Saccharomyces  cerevisiae. 

The  presence  of  Bacillus  weidmaniensis  greatly  stimulated  growth  and 
production  of  macroconidia  by  Microsporum  audouini  (Benedek,  1943, 
and  Hazen,  1947).  The  addition  of  yeast  extract  to  the  medium  had  a 
similar  effect.  A  part  of  this  stimulating  effect  was  attributed  to  pyri- 
doxine.  The  addition  of  yeast  extract  to  a  basal  medium  of  honey  agar 
resulted  in  a  marked  increase  in  vegetative  growth  and  macroconidium 
production  (Hazen,  1947).     This  stimulation  was  attributed  to  the  pres- 


328 


PHYSIOLOGY  OF  THE  FUNGI 


ence  of  growth  factors  in  the  yeast  extract.  The  addition  of  pyridoxine 
to  the  basal  medium  caused  little  change  in  mycelial  growth,  but  a  great 
increase  in  abundance  of  macroconidia  resulted.  On  the  other  hand,  the 
addition  of  thiamine  or  of  a  mixture  of  thiamine  and  pyridoxine  caused 
no  increase  in  growth  or  production  of  macroconidia. 

An  interesting  observation  of  the  constant  natural  association  between 
Nectria  coccinea  and  Gonatorrhodiella  highlei  is  reported  by  Ayres  (1941). 
Because  of  this  constant  association  it  was  believed  that  G.  highlei  was 
either  parasitic  upon  A^.  coccinea  or  dependent  upon  it  for  some  nutri- 
tional substance.     On  potato-glucose,  malt  extract,  and  other  common 


Fig.  67.  The  stimulating  effect  of  Aspergillus  rugulosus  (small  colony  at  the  bottom) 
on  the  production  of  perithecia  by  Sordaria  fimicola  grown  on  glucose-asparagine 
medium  low  in  biotin.  The  zone  of  black  perithecia  of  Sordaria  around  the  colony  of 
Aspergillus  is  attributed  to  the  biotin  excreted  by  the  latter  fungus. 

media  G.  highlei  made  only  slight  growth  and  formed  no  conidiophores  or 
conidia.  However,  the  fungus  grew  well  and  produced  numerous  conidio- 
phores and  conidia  on  the  same  media  in  the  presence  of  A^.  coccinea,  N. 
galligena,  or  N.  cucurhitula.  Neither  N.  cinnabarina  nor  N .  coryli  caused 
stimulation.  G.  highlei  was  cultivated  successfully  with  production  of 
abundant  conidia  on  oatmeal  mush,  without  the  presence  of  other  fungi. 
These  results  strongly  suggest  a  nutritional  relation  between  G.  highlei 
and  N .  coccinea,  other  than  that  of  parasitism. 

The  beneficial  effect  of  one  fungus  upon  reproduction  of  another  can 
easily  be  demonstrated  by  placing  Aspergillus  rugulosus  and  Sordaria 
fimicola  on  a  plate  of  agar  containing  little  or  no  biotin  (Fig.  67).  S. 
fimicola,  being  a  more  rapid  grower,  produces  a  sparse  mycelium,  which 
surrounds  the  slow-growing  colony  of  Aspergillus.     The  Sordaria  myce- 


SPORULATION  329 

lium  next  to  the  Aspergillus  colony  soon  shows  stimulated  growth,  which 
is  followed  by  the  formation  of  abundant  perithecia  in  this  area.  Since 
we  know  that  Sordaria  is  deficient  for  biotin  and  requires  an  exogenous 
supply  of  this  vitamin  for  reproduction,  we  may  assume  that  the  stimulat- 
ing effect  is  due  to  the  extra  biotin  produced  by  A.  rugulosus  diffusing  into 
the  medium. 

Vitamins.  Many  fungi  cannot  synthesize  suflricient  amounts  of  certain 
vitamins  and  must  depend  upon  an  outside  source  of  these  vitamins  for 
optimum  growth.  Since  many  nutritional  factors  may  influence  both 
growth  and  reproduction,  it  would  be  logical  to  expect  that  the  vitamin 
supply  would  affect  the  reproduction  of  vitamin-deficient  fungi.  Robbins 
and  Ma  (1942)  ventured  the  opinion  that,  although  the  sex  organs  of 
certain  deficient  fungi  were  formed  only  in  the  presence  of  the  growth 
factor  for  which  these  fungi  were  deficient,  they  were  doubtful  if  there 
was  any  direct  relation  between  vitamin  supply  and  the  formation  of 
sex  organs.  They  regarded  the  failure  of  Ceratostomella  pluriannulata  to 
produce  perithecia  in  the  absence  of  thiamine  to  be  a  disturbance  of  the 
physiology  of  the  fungus  and  ventured  the  prediction  that  the  formation 
of  sex  organs  in  other  deficient  fungi  would  be  found  associated  with  the 
vitamins  for  which  the  fungi  was  deficient. 

Melanospora  destruens  is  able  to  grow  in  the  presence  of  biotin  as  the 
only  growth  factor,  but  it  produces  perithecia  only  when  thiamine,  too,  is 
added  to  the  medium  (Hawker,  1939).  The  relationship  between  the 
amount  of  sugar  and  the  amount  of  thiamine  necessary  for  maximum 
fruiting  is  clearly  brought  out  by  the  thiamine-deficient  fungus,  C. 
fimbriata  (Barnett  and  Lilly,  1947a).  Whether  perithecia  are  formed  on  a 
given  medium  is  determined  by  the  amount  of  thiamine  relative  to  the 
amount  of  food  in  the  medium  (Table  57).  In  a  reduced  supply  of  sugar 
the  concentration  of  thiamine  necessary  to  induce  fruiting  is  also  reduced. 
The  abundance  of  perithecia  is  conditioned  both  by  the  amount  of  thia- 
mine and  by  the  amount  of  nutrients.  Less  thiamine  is  required  for 
vegetative  growth  than  for  the  production  of  perithecia.  A  similar  rela- 
tion between  amounts  of  sugar  and  biotin  and  sporulation  of  Memnoniella 
echinata  was  described  by  Buston  and  Basu  (1948).  It  should  be  noted 
from  Table  57  that,  in  a  medium  high  in  sugar  but  low  in  thiamine,  no 
perithecia  were  formed  even  though  as  much  as  30  mg.  of  mycelium  was 
present.  On  the  other  hand,  perithecia  were  produced  on  as  little  as  two 
mg.  of  mycelium  when  the  medium  contained  the  same  amount  of  thia- 
mine but  was  very  low  in  sugar.  These  results  are  not  in  accord  with  the 
idea  that  vigorous  or  abundant  mycelium  is  essential  to  the  formation  of 
perithecia;  they  indicate  a  more  direct  relation  between  vitamin  supply 
and  sexual  reproduction. 

This  direct  relationship  may  be  further  illustrated  by  a  simple  experi- 


330 


PHYSIOLOGY  OF  THE  FUNGI 


ment.  Sterile  thiamine-starved  mycelium  may  be  obtained  by  growing 
C.  fimhriata  on  a  synthetic  medium  containing  25  g.  glucose  and  less  than 
2  ng  of  thiamine  per  liter.  When  a  small  portion  of  this  sterile  mycelium 
is  transferred  to  the  surface  of  distilled  water,  no  perithecia  are  formed. 
How^ever,  under  the  same  treatment,  but  with  the  addition  of  thiamine 
to  the  water,  fertile  perithecia  are  formed  within  a  few  days.  We  may 
assume  that,  when  thiamine  is  added,  the  fungus  uses  this  vitamin  as  a 
coenzyme  in  transforming  the  protoplasmic  reserves  into  perithecia  and 

Table  57.     The  Effect  of  Various  Concentrations  of  Thiamine  upon  Growth 
AND  Estimated  Abundance  of  Perithecia  Formed  by  Ceratostomella  fimhriata 

in  the  Presence  of  Various  Concentrations  of  Nutrients 
Growth  in  milligrams.     Abundance  of  perithecia  indicated  by:  0  =  none;   +  = 

less  than  20;  ++  =  20  to  200;  +  +  +  =  200  to  1,000;  +  +  +  +  =  more  than  1,000. 

(Barnett  and  Lilly,  Mycologia  39,  1947.) 


Thiamine,  /xg  per 

Dilution  of  i 

Tiedium 

culture  (25  ml.) 

Undiluted 

Vi 

He 

^4 

2.5 

110 

33 

10 

2 

+  +  +  + 

+  +  +  + 

+  +  +  + 

_1__L- 
1      1 

0.156 

59 

36 

9 

2 

+  + 

+  +  +  + 

+  +  +  + 

+  + 

0.04 

30 

17 

10 

2 

0 

+  + 

+  +  +  + 

+ 

0.02 

18 

14 

9 

3 

0 

0 

+  +  + 

+ 

ascospores.  This  experiment  has  been  successfully  conducted  using  other 
fungi  deficient  for  thiamine,  and  similar  results  also  were  obtained  with 
biotin  and  biotin-deficient  fungi.  The  amount  of  biotin  added  to  the 
medium  affected  not  only  the  number  of  perithecia  formed  by  Sordaria 
fimicola,  but  also  the  time  required  for  this  fungus  to  form  mature  peri- 
thecia (Lilly  and  Barnett,  1947).  This  period  ranged  from  13  to  41  days, 
depending  on  the  concentration  of  biotin  used. 

There  is  a  pronounced  effect  of  biotin  deficiency  upon  the  formation 
and  development  of  the  ascospores  of  S.  fimicola  (Barnett  and  Lilly, 
1947).  This  fungus  is  well  suited  for  such  a  study,  for  normally  nearly 
all  the  ascospores  mature  at  the  same  time.  Figure  68  shows  the  effects 
of  biotin  starvation  upon  the  formation  of  ascospores.  Severe  effects 
are  evident  by  the  failure  of  the  protoplasm  of  the  asci  to  be  delimited 
into  ascospores  or  by  the  failure  of  many  of  the  ascospores  to  mature. 
Other  conditions  being  equal,  the  amount  of  biotin  required  for  the  pro- 


SPORULATION 


331 


duction  of  mature  ascospores  is  greater  than  the  amount  required  for 
the  formation  of  perithecia. 


.4  B 

Fig.  68.  The  effect  of  biotin  starvation  on  the  formation  of  ascospores  by  Sordaria 
fimicola.  A,  asci  from  a  perithecium  developed  in  a  suboptimum  concentration  of 
biotin.  Note  the  few  mature  ascospores  and  the  majority  of  asci  in  which  no  spores, 
or  only  aborted  ascospores,  have  formed.  B,  normal  asci  with  mature  ascospores 
developed  on  medium  with  optimum  biotin  concentration. 

Since  the  vitamins  are  not  considered  as  specific  reproductive  sub- 
stances, it  may  be  expected  that  an  adequate  vitamin  supply  may  be 
necessary  for  the  formation  of  asexual  spores  as  well  as  the  sexual  fruiting 
structures.  However,  it  should  be  kept  in  mind  that  asexual  reproduc- 
tion is  more  nearly  like  vegetative  growth  in  its  requirements  than  is 
sexual  reproduction.  A  reduction  in  the  supply  of  a  necessary  vitamin 
to  the  point  where  asexual  reproduction  is  inhibited  may  also  allow  but 
little  vegetative  growth.  Piricularia  orijzae  is  deficient  for  both  thiamine 
and  biotin  (Leaver  et  al.,  1947).  Conidia  did  not  form  unless  both 
vitamins  were  present  in  the  medium.  The  concentration  of  biotin 
could  be  reduced  to  such  a  level  that  conidial  production  was  inhibited 
but  some  mycelial  growth  was  still  allowed.  In  an  adequate  supply  of 
biotin,   growth   and   sporulation   were   apparently   parallel.     It  is   also 

Table  58.     The  Effects  of  Concentration  of  Glucose  and  Thiamine  upon  the 

Number  of  Conidial  Heads  Formed  by  Choanephora  cucurbitarum 

(Barnett  and  Lilly,  Phytopathology  40,  1950.) 


Glucose, 
g.  per  liter 


25 

25 

25 

2 


Thiamine, 
Mg  per  liter 


Conidial  heads 
per  plate 


8 

210 

450 

2,000 


332  PHYSIOLOGY  OF  THE  FUNGI 

evident  that  a  proper  balance  between  the  amounts  of  vitamins  and 
supply  of  nutrients  is  necessary  for  maximum  production  of  asexual 
spores.  The  effects  of  the  concentrations  of  thiamine  and  glucose  upon 
the  production  of  conidial  heads  by  Choanephora  cucurhitarum  are  illus- 
trated in  Table  58.  On  the  other  hand,  we  have  observed  that  an  excess 
of  certain  vitamins  may  cause  a  decrease  in  sporulation  of  some  fungi. 

OTHER  FACTORS 

Method  of  inoculation.  The  method  of  inoculation  and  the  type  of 
inoculum  used  are  often  important  factors  affecting  sporulation  in  cul- 
ture. Some  fungi  sporulate  more  quickly,  and  often  more  abundantly, 
on  agar  when  the  medium  is  flooded  with  a  spore  suspension  than  when 
the  inoculum  is  placed  at  only  one  point.  This  is  particularly  applicable 
to  certain  pycnidium-  or  acervulus-producing  fungi,  which  usually  pro- 
duce abundant  mycelium  before  fruiting. 

It  is  possible  that  the  spores  of  certain  fungi  carry  over  sporulation- 
inducing  substances  to  the  next  generation.  The  production  of  conidia 
of  CoUctotrichum  lindeviuthianum  was  greater  on  agar  media  inoculated 
at  one  point  with  spores  than  when  bits  of  mycelium  were  used  as  inocu- 
lum (Mathur  et  at.,  1950). 

Method  of  sterilizing  media.  The  most  common  method  of  sterilizing 
culture  media  is  the  use  of  steam  pressure  in  an  autoclave.  The  standard 
time  is  usually  15  to  20  min.  at  15  lb.  pressure.  This  temperature  is 
known  to  cause  a  breakdown  of  certain  sugars,  with  an  accompanying 
change  in  acidity  of  the  medium.  When  natural  media  are  used,  other 
chemical  changes  occur,  which  may  or  may  not  be  beneficial  to  reproduc- 
tion of  the  fungi.  Shanor  (1936)  reported  that  fruiting  structures  of 
Cordyceps  militaris  were  not  formed  when  autoclaved  insects  were  inocu- 
lated, whereas  the  fungus  produced  stromata  and  perithecia  when  living 
pupae  were  inoculated.  A  new  approach  to  the  problem  of  sterilization  is 
suggested  by  Hansen  and  Snyder  (1947),  who  propose  the  use  of  propylene 
oxide.  This  method  has  great  possibilities  but  has  not  been  used  exten- 
sively enough  to  determine  all  its  virtues  and  limitations. 

Influence  of  the  host.  Numerous  parasitic  fungi  which  have  been 
grown  on  artificial  media  have  not  been  induced  to  form  the  sexual  repro- 
ductive stages  under  any  conditions  in  culture.  Some  of  these  will 
produce  the  reproductive  stages  when  grown  on  their  respective  hosts. 
For  instance,  most  smuts  will  produce  mycelium  in  culture,  but  few  have 
produced  teliospores  (''chlamydospores")  under  these  conditions.  Claviceps 
has  never  been  induced  to  form  sclerotia  (which  necessarily  precede  the 
perfect  stage)  in  artificial  culture.  Other  fungi  seem  to  lose  their  ability 
to  sporulate  profusely  in  culture  but  may  regain  that  ability  when  grown 
on  the  appropriate  host.     It  may  be  pointed  out  that  many  of  the  fungi 


SPORULATION  333 

pathogenic  on  man  do  not  produce  the  same  reproductive  stages  in  culture 
that  they  do  in  their  host  (Conant  et  al.,  1944).  Much  more  intensive 
study  needs  to  be  made  of  these  pathogens. 

SUMMARY 

Reproduction  in  the  fungi,  particularly  sexual  reproduction,  with  all  the 
necessary  preliminary  metabolic  activities,  is  a  complex  phenomenon. 
Some  fungi  appear  to  be  relatively  indifferent  to  and  independent  of  their 
environment  with  respect  to  sporulation,  while  others  appear  to  require 
a  special  combination  of  environmental  conditions.  If  we  assume  that 
all  fungi  require  similar  internal  conditions  for  sexual  reproduction — 
and  we  do  not  believe  this  to  be  an  unreasonable  assumption — we  must 
recognize  the  existence  of  numerous  fundamental  differences  in  the 
metabolic  activities  of  the  many  different  fungi.  Some  are  capable  of 
creating  the  necessary  internal  conditions  in  spite  of  external  conditions 
which  may  be  limiting  factors  for  other  fungi.  Other  species  may  not 
have  the  ability  to  create  the  necessary  internal  conditions  without 
specific  action  of  certain  external  factors,  such  as  light,  proper  nutrients, 
vitamins,  and  others.  As  has  been  previously  emphasized,  these  proc- 
esses are  often  dependent  upon  the  enzyme  systems  of  the  fungus,  and 
these  systems  and  their  activity  vary  widely  among  the  different  species. 

Many  fungi  reproduce  sexually  only  as  the  vegetative  growth  is  near  or 
past  the  maximum.  At  this  time  many  of  the  cells  of  the  mycelium  are 
dead  or  dying.  Autolysis  follows,  and  the  cells  that  remain  alive  absorb 
certain  of  these  products  of  autolysis  and  are  thus  enabled  to  increase 
their  concentration  of  certain  essential  substances  to  such  a  degree  that 
reproduction  is  possible.  If  these  speculations  are  valid,  it  might  be 
expected  that  nearly  any  type  of  injury  which  causes  death  of  some  of  the 
cells  would  have  a  favorable  effect  on  reproduction  of  some  fungi.  Ultra- 
violet radiation,  which  inhibits  growth  and  often  favors  sporulation,  may 
act  by  killing  some  of  the  cells.  Heat  may  also  act  by  speeding  up  the 
life  processes  of  a  fungus  so  that  maturity  is  reached  more  quickly,  which, 
followed  by  death  of  cells  and  autolysis,  would  furnish  the  necessary 
stimulus  for  reproduction.  Treatment  with  hydrogen  peroxide  and 
certain  other  chemicals  may  also  result  in  death  and  autolysis  of  some 
cells.  No  claim  is  advanced  that  these  speculations  are  the  true  explana- 
tion in  all  instances,  but  they  do  provide  a  hypothesis  for  interpreting 
certain  puzzling  problems  connected  with  sexual  reproduction  in  the 
fungi. 

Factors  which  may  initiate  or  stimulate  sporulation  of  fungi  which 
grow  well  but  fruit  only  sparingly  or  not  at  all  under  the  usual  conditions 
of  artificial  culture  (assuming  that  genetic  factors  are  favorable),  are  as 
follows:  (1)  A  change  in  the  concentration  of  one  or  more  of  the  nutrients. 


334  PHYSIOLOGY  OF  THE  FUNGI 

A  reduction  in  sugar  alone  may  be  effective.  (2)  A  change  in  the  source 
of  carbon.  Replacing  glucose  by  the  same  amount  of  sucrose,  lactose, 
starch,  or  other  carbon  source  may  favor  reproduction.  (3)  A  change 
in  the  source  of  nitrogen.  (4)  A  change  in  the  carbon-nitrogen  ratio. 
(5)  The  addition  of  an  adequate  supply  of  vitamins  for  vitamin-deficient 
fungi.  (6)  The  addition  of  certain  micro  elements  to  the  medium,  if  it 
is  made  up  of  highly  purified  chemicals.  (7)  The  addition  to  the  medium 
of  certain  natural  products,  such  as  pieces  of  stems  or  leaves.  (8)  The 
addition  of  culture  filtrate  containing  metabolic  products  from  the  same 
or  other  fungus.  (9)  The  addition  of  special  compounds,  such  as  glutaric 
acid,  which  has  been  shown  to  be  effective  in  a  few  cases.  (10)  The 
addition  of  certain  specific  sexual  factors,  or  hormones,  known  to  be 
effective  for  certain  Phycomycetes.  (11)  The  addition  of  certain  chem- 
icals, such  as  hydrogen  peroxide,  to  the  mycelium.  (12)  Exposure  of 
cultures  to  ozone.  (13)  The  use  of  spores  instead  of  mycelium  as  inocu- 
lum. Flooding  of  agar  plates  with  spore  suspension  has  given  excellent 
results  with  some  fungi.  (14)  The  sterilization  of  media  by  means  other 
than  heat.  (15)  Transfer  of  certain  parasitic  fungi  to  their  living  hosts. 
(16)  Growing  the  fungus  in  the  presence  of  certain  bacteria  or  other 
fvmgi.  (17)  A  change  in  pH  of  the  medium,  (18)  A  change  in  the 
temperature  of  incubation.  (19)  Adequate  aeration.  (20)  Exposure  to 
light  or  alternate  light  and  darkness.  (21)  Short  exposures  to  ultraviolet 
radiation.  (22)  Variation  in  the  intensity  and  wave  length  of  light.  (23) 
Mechanical  injury  to  the  mycelium.  (24)  Gradual  desiccation  of  the 
cultures.  (25)  Allowing  the  cultures  to  age.  (26)  The  proper  combina- 
tion of  any  two  or  more  of  the  above  factors.  The  secret  of  the  sexual 
reproduction  of  many  fungi  no  doubt  lies  in  the  proper  combination  of 
factors  which  singly  are  known  to  favor  reproduction  in  other  fungi. 

REFERENCES 

Ayres,  T.  T.:  The  distribution  and  association  of  Gonatorrhodiella  highlei  with 

Nectria  coccinea  in  the  United  States,  Mycologia  33:  178-187,  1941. 
*Barnett,  H.  L.,  and  V.  G.  Lilly:  The  effects  of  biotin  upon  the  formation  and 

development  of  perithecia,  asci  and  ascospores  by  Sordaria  fimicola  Ces.  and 

de  Not.,  Am.  Jour.  Botamj  34:  196-204,  1947. 
Barnett,  H.  L.,  and  V.  G.  Lilly:  The  relation  of  thiamin  to  the  production  of 

perithecia  by  Ceratostomella  fimbriata,  Mycologia  39:  699-708,  1947a. 
*Barnett,  H.  L.,  and  V.  G.  Lilly:  Nutritional  and  environmental  factors  influencing 

asexual  sporulation  of  Choanephora  cucurbitarurn  in  culture.  Phytopathology  40 : 

80-89,  1950. 
Benedek,  T.  :  Unilateral  stimulation  of  Microsporum  audouini  by  a  new  species 

of  Bacillus,  Mycologia  35 :  222-242,  1943. 
BisBY,  G.  R. :  Zonation  in  cultures  of  Fusarium  discolor  sulphureum,  Mycologia  17 : 

89-97,  1925. 
BisBY.  G,  R.:  Geographical  distribution  of  Fungi,  Botan.  Rev.  9:  466-482.  1943. 


SPORULATION  335 

BosE,  S.  R. :  Moisture-relation  as  a  determinant  factor  in  the  transformation  of  the 

basidia  of  certain  Polyporaceae,  Mycologia  35:  33-46,  1943. 
Bhefeld,   O. :  Botanische  Untersuchungen  liber  Schimmelpilze,  Heft  III;  1-266, 

Verlag  von  Arthur  Felix,  Leipzig,  1877. 
Brodie,  H.  J.:  Variation  in  fruit  bodies  of  Cyathus  stercoreus  produced  in  culture, 

Mycologia  40:  614-626,  1948. 
Buston,  H.  W.,  and  S.  N.  Basu:  Some  factors  affecting  the  growth  and  sporulation 

of  Chaetomium  globosum  and  Memnoniella  echinata,  Jour.  Gen.  Microbiol.  2 : 

162-172,  1948. 
Camp,  W.  G.  :  The  fruiting  of  Physarum  polycephalutn  in  relation  to  nutrition,  Am. 

Jour.  Botany  24 :  300-303,  1937. 
Christenberry,  G.  A.:  a  study  of  the  effect  of  light  of  various  periods  and  wave 

lengths  on  the  growth  and  asexual  reproduction  of  Choanephora  ciicurbitarum 

(Berk,  and  Rav.)  Thaxter,  Jour.  Elisha  Mitchell  Sci.  Soc.  54:  297-310,  1938. 
CoNANT,  N.  F.,  D.  S.  Martin,  D.  T.  Smith,  R.  D.  Baker,  and  J.  L.  Callaway: 

Manual  of  Clinical  Mycology,  W.  B.  Saunders  Company,  Philadelphia,  1944. 
Coons,  G.  H.:  Factors  involved  in  the  growth  and  the  pycnidium  formation  of 

Plenodomus  fuscomaculans,  Jour.  Agr.  Research  5:  713-769,  1916. 
Denny,  F.  E.:  Oxygen  requirements  of  Neurospora  sitophila  for  the  formation  of 

perithecia  and  growth  of  mycelium,  Contribs.  Boyce  Thompson  Inst.  5 :  95-102, 

1933. 
DouDOROFF,  M.:  The  utilization  and  synthesis  of  sucrose  and  related  compounds 

by  some  microorganisms.  Fed.  Proc.  4 :  241-247,  1945. 
*Drayton,  F.  L.:  The  perfect  stage  of  Botrytis  convoluta,  Mycologia  29  :  305-318,  1937. 
Elfving,  F.:  Studien  iiber  die  Einwirking  des  Lichtes  auf  die  Pilze,  Helsingfors 

Central-Druckerei,  Helsingfors,  1890. 
Emerson,  R.,  and  E.  C.  Cantino.  The  isolation,  growth,  and  metabolism  of  Blasto- 

cladia  in  pure  culture.  Am.  Jour.  Botany  35:  157-171,  1948. 
Emerson,  R.,  and  D.  L.  Fox:  Gam/na-carotene  in  the  sexual  phase  of  the  aquatic 

fungus  Allomyces,  Proc.  Roy.  Soc.  (London),  Ser.  B,  128:  275-293,  1940. 
Hansen,  H.  N.,  and  W.  C.  Snyder:  Gaseous  sterilization  of  biological  materials 

for  use  as  culture  media,  Phytopathology  37:  369-371,  1947. 
*Hawker,  L.  E.:  The  influence  of  various  sources  of  carbon  on  the  formation  of 

perithecia  by  Melanospora  destruens  Shear  in  the  presence  of  accessory  growth 

factors,  Ann.  Botany  (N.S.)  3:  455-468,  1939. 
Hawker,  L.  E.:  Stimulation  of  the  formation  of  perithecia  of  Melanospora  destruens 

Shear  by  small  quantities  of  certain  phosphoric  esters  of  glucose  and  fructose, 

Ann.  Botany  (N.S.)  12:  77-79,  1948. 
Hawker,  L.  E.:  Physiology  of  Fungi,  University  of  London  Press,  Ltd.,  Bickley, 

Kent,  England,  1950. 
Hazen,  E.  L. :  Microsporum  audouini:  The  effect  of  yeast  extract,  thiamine,  pyrido- 

xine,  and  Bacillus  weidmaniensis  on  the  colony  characteristics  and  macroconidial 

formation,  Mycologia  39 :  200-209,  1947. 
*Herny,  W.  B.,  and  A.  L.  Andersen:  Sporulation  by  Piricularia  oryzae,  Phytopa- 
thology 38 :  265-278,  1948. 
Houston,  B.  R.,  and  J.  W.  Oswald:  The  effect  of  light  and  temperature  on  conidium 

production  by  Hebtiinthosporium  gramineum  in   culture,  Phytopathology  36: 

1049-1055,  1946. 
Jones,  R.  C:  Factors  affecting  the  production  of  resistant  sporangia  of  Allomyces 

arbuscula,  Mycologia  38:  91-102,  1946. 
■*Kauffman,   C.   H.:  Klebs'  theory  of  the  control  of  developmental  processes  in 


336  PHYSIOLOGY  OF  THE  FUNGI 

organisms  and  its  application  to  fungi,  Proc.  Intern.  Congr.  Plant  Sci.,  1st  Congr., 

2:  1G03-1611,  1929. 
Klebs,  G.:  Zur  Physiologic  dcr  Fortflanzung  einiger  Pilze.      11.  Saprolegnia  mixta 

de  Bary,  Jahrb.  wiss.  Botan.  33:  519-593,  1899. 
Klebs,    G.:    Zur    Physiologic    der    Fortflanzung    einiger    Pilze.     III.   Allegemeine 

Betrachtungen,  Jahrh.  iriss.  Botan.  35:  80-203,  1900. 
KuNKEL,  L.  O. :  A  method  of  obtaining  abundant  sporulation  in  cultures  of  Macro- 

sporium  solani,  Brooklyn  Botan.  Garden  Mem.  1 :  306-312,  1918. 
Leaver,  F.  W.,  J.  Leal,  and  C.  R.  Brewer:  Nutritional  studies  on  Piricularia 

oryzae,  Jour.  Bad.  54 :  401-408,  1947. 
Leonian,  L.  H.:  The  physiology  and  perithecial  and  pycnidial  formation  in  Valsa 

leucostoma,  Phytopathology  13:  257-272,  1923. 
*Leonian,  L.  H.:  a  study  of  factors  promoting  pycnidium  formation  in  some  Sphaer- 

opsidales,  Am.  Jour.  Botany  11 :  19-50,  1924. 
Leonian,  L.   H.:  Control  of  sexual  reproduction  of  Phytophthora  cactorum,  Am. 

Jour.  Botany  23:  188-190,  1936. 
Leonian,  L.  H.,  and  V.  G.  Lilly:  Partial  purification  of  a  vitamin-like  substance 

which  stimulates  sexual  reproduction  in  certain  fungi,  Am.  Jour.  Botany  24: 

700-702,  1937. 
Lilly,  V.  G.,  and  H.  L.  Barnett:  The  influence  of  pH  and  certain  growth  factors  on 

mycelial   growth  and   perithecial  formation  by  Sordaria  fimtcola,  Am.  Jour. 

Botany  34:  131-138,  1947. 
*Lindegren,  C.  C.,  and  E.  Hamilton:  Autolysis  and  sporulation  in  the  yeast  colony, 

Botan.  Gaz.  105:  316-321,  1944. 
Lindegren,   C.   C.,  and  G.  Lindegren:  Sporulation  in  Saccharomyces  cerevisiae, 

Botan.  Gaz.  105:  304-316,  1944. 
LocKWOOD,  L.  B. :  Hydrogen  ion  concentration  and  ascus  formation,  Mycologia  29 : 

289-290,  1937. 
LocKWOOD,  L.  B.,  and  G.  E.  Ward:  The  physiology  of  Ehizopus  oryzae,  Jour.  Agr. 

Research  53 :  849-857,  1936. 
McCallan,  S.  E.  A.,  and  S.  Y.  Chan:  Inducing  sporulation  of  Alternaria  solani  in 

culture,  Contribs.  Boyce  Thompson  Inst.  13 :  323-336,  1944. 
Mader,   E.   O.:   Some  factors  inhibiting  the  fructification  and  production  of  the 

cultivated  mushroom,  Agaricus  campestris  L.,  Phytopathology  33:  1134-1145, 

1943. 
Mathur,  R.  S.,  H.  L.  Barnett,  and  V.  G.  Lilly:  Factors  influencing  growth  and 

sporulation  of   CoUetotrichum  lindemuthianum  in   culture,  Phytopathology  40: 

104-114,  1950. 
Mex,  A.  J.:  Factors  affecting  the  sporulation  of  Phyllosticta  solitaria  in  artificial 

culture,  Phytopathology  23:  503-524,  1933. 
Mrak,  E.  M.,  and  L.  Bonar:  The  effect  of  temperature  on  asci  and  ascospores  in  the 

genus  Debaryomyces,  Mycologia  30:  182-186,  1938. 
Nickerson,  W.  J.,  and  K.  V.  Thimann:  The  chemical  control  of  conjugation  in 

Zygosaccharomyces.     I.  Am.  Jour.  Botany  28:  617-621,  1941. 
Nickerson,  W.  J.,  and  K.  V.  Thimann:  The  chemical  control  of  conjugation  in 

Zygosaccharomyces.     II.  Am.  Jour.  Botany  30:  94-101,  1943. 
Porter,  C.  L.:  Concerning  the  character  of  certain  fungi  as  exhibited  by  their 

growth  in  the  presence  of  other  fungi.  Am.  Jour.  Botany  11:  168-188,  1924. 
Ramsey,  G.  B.,  and  A.  Bailey:  Effects  of  ultra-violet  radiation  upon  sporulation  in 

Macrosporium  and  Fusaritim,  Botan.  Gaz.  89:  113-136,  1930. 
Rands,  R.  D.:  The  production  of  spores  by  Alternaria  solani  in  pure  culture,  Phyto- 
pathology 7:  316-317,  1917. 


SPORULATION  337 

Raper,  J.  R.:  Sexual  hormones  in  Achlya.     III.   Hormone  A  and  the  initial  male 

reaction,  Ajn.  Jour.  Botany  29:  159-16G,  1942. 
Raper,  J.  R.,  and  A.  J.  Haagen-Smit:  Sexual  hormones  in  Achlya.     IV.  Prop- 
erties of  Hormone  A  of  Achlya  hisexnalis,  Jour.  Biol.  Chem.  143:  311-320,  1942. 
Richards,  M.  C:  Ozone  as  a  stimulant  for  fungus  sporulation,  Phytopathology  39: 

20,  1949. 
Riker,  a.  J.,  and  R.  S.  Riker:  Introduction  to  Research  on  Plant  Diseases,  John  S. 

Swift  Co.,  St.  Louis,  1936. 
RoBBiNS,  W.  J.,  and  R.  Ma:  Vitamin  deficiencies  of  Ceratostomella  and  related  fungi. 

Am.  Jour.  Botany  29:  835-843,  1942. 
RoBBiNS,  W.  J.,  and  M.  B.  Schmitt:  Factor  Z2  and  gametic  reproduction  by  Phyco- 

myces,  Am.  Jour.  Botany  32:  320-326,  1945. 
Shanor,  L.:  The  production  of  mature  perithecia  of  Cordyceps  militaris  (L.)  Link  in 

laboratory  culture.  Jour.  Elisha  Mitchell  Sci.  Soc.  52 :  99-104,  1936. 
Smith,  E.  C:  Effects  of  ultraviolet  radiation  and  temperature  on  Fusarium.      IL 

Stimulation,  Bull.  Torrey  Botan.  Club  62:  151-164,  1935. 
Snyder,  W.  C,  and  H.  X.  Haxsex:  The  effect  of  light  on  taxonomic  characters  in 

Fusarium,  Mycologia  33:  580-591,  1941. 
Sntder,  W.  C,  and  H.  X.  Hansen:  Advantages  of  natural  media  and  environments 

in  the  cuHure  of  fungi.  Phytopathology  37:  420-421,  1947. 
*Stevens,  F.  L.  :  Effects  of  ultra-violet  radiation  on  various  fungi,  Botan.  Gaz.  86: 

210-225,  1928. 
Ternetz,    C.  :    Protoplasmabewegung   und    Fruchtkorperbildung   bei    Ascophanus 

carneus  Pers.,  Jahrb.  wiss.  Botan,.  35:  273-312,  1900. 
TiMNiCK,  M.  B.,  V.  G.  Lilly,  and  H.  L.  Barnett:  The  influence  of  hght  and  other 

factors  upon  sporulation  of  Diaporthe  phaseolorum  var.  batatatis  from  soybean. 

Phytopathology  41:  327-336,  1951. 
Westergaard,  M.,  and  H.  K.  Mitchell:  N'eurospora.     V.  A  synthetic  medium 

favoring  sexual  reproduction.  Am.  Jour.  Botany  34:  573-577,  1947. 
Yakwood,   C.  E.:  The  diurnal  cycle  of  the  powdery  mildew^,   Erysiphe  polygoni. 

Jour.  Agr.  Research  52 :  645-657,  1936. 
*Yarwood,  C.  E.  :  Diurnal  cycle  of  ascus  maturation  of  Taphrina  deformans,  Am. 

Jour.  Botany  28 :  355-357,  1941. 


CHAPTER  15 
SPORE  DISCHARGE  AND  DISSEMINATION 

While  the  production  of  mature  viable  spores  may  be  considered  the 
climax  in  the  life  of  a  fungus,  the  wide  dissemination  of  these  spores  is 
often  a  requisite  to  the  perpetuation  of  the  species.  Frequently,  this  is 
assured  by  the  production  of  enormous  numbers  of  spores,  which  increases 
their  chances  of  falling  into  favorable  environment.  The  chief  agent  of 
dissemination  among  the  fungi  is  air  currents.  Water,  insects,  and  other 
animals  play  lesser  roles  in  the  natural  dispersal  of  spores.  Light,  dry 
spores  are  usually  disseminated  by  air  currents,  which  may  be  strong 
enough  to  loosen  them  from  the  fruiting  structures  on  which  they  are 
produced.  On  the  other  hand,  spores  borne  in  a  gelatinous  matrix  are 
better  adapted  to  transmission  by  rain,  by  insects,  or  by  other  animals 
which  come  in  contact  with  them.  Of  particular  interest  are  those  fungi 
which  possess  certain  special  mechanisms  for  discharging  their  spores 
away  from  the  fruiting  structures.  A  study  of  the  functions  of  these 
special  adaptations  must  of  necessity  be  based  upon  a  knowledge  of  the 
structure  of  the  fruit  bodies  which  produce  the  spores. 

METHODS  OF  SPORE  DISCHARGE 

The  discharge  or  liberation  of  spores  from  the  reproductive  structures 
which  produce  them  may  take  place  by  (1)  violent  expulsion  of  the  spores 
or  sporangia,  due  to  internal  pressure,  (2)  motility,  as  in  the  zoospores  of 
the  aquatic  Phycomycetes,  and  (3)  external  forces  of  the  environment. 

The  violent  discharge  of  spores,  sporangia,  or  other  reproductive  bodies 
depends  upon  the  development  of  considerable  pressure  within  the  fungus. 
The  structure  of  the  fungus  cell  is  very  similar  to  that  of  algae  or  the 
parenchyma  of  the  higher  plants.  The  vacuole  is  filled  with  water  and 
its  dissolved  compounds,  such  as  sugars,  salts,  and  amino  acids.  Foods 
in  the  cell  m^ay  be  in  the  form  of  sugar,  glycogen,  or  oil.  If  the  amount  of 
soluble  materials,  such  as  sugar,  is  increased,  the  cell  has  a  tendency  to 
absorb  water.  As  a  result,  the  cell  becomes  more  distended  and  may 
continue  to  swell  until  the  elasticity  of  the  cell  wall  is  exceeded.  The 
increase  in  sugar  concentration  may  be  the  result  of  the  hydrolysis  of 
glycogen. 

Discharge  of  sporangia.  The  genus  Piloholus  illustrates  a  remarkable 
combination  of  adaptations  for  the  production,  discharge,  and  subsequent 

338 


SPORE  DISCHARGE  AND  DISSEMINATION  339 

dissemination  of  its  spores.  These  include  (1)  the  dependence  upon  Hght 
for  the  production  of  sporangia,  (2)  the  positively  phototropic  response 
of  the  sporangiophores  (see  frontispiece),  (3)  the  violent  discharge  of  the 
sporangium  into  the  air  toward  the  source  of  light,  (4)  the  sticky  nature 
and  the  heavy  black  wall  of  the  sporangium,  and  (5)  the  dissemination 
of  the  spores  by  the  passage  through  the  digestive  tract  of  animals  which 
ingest  them. 

We  owe  much  of  our  knowledge  regarding  the  structure  of  Piloholus,  its 
physiology,  and  its  life  history  to  the  careful  study  and  comprehensive 
descriptions  of  Buller  (1934).  Much  of  his  work  was  done  with  P.  kleinii 
and  P.  longipes.  Brefeld  (1881)  showed  that,  in  the  absence  of  light,  no 
sporangia  were  formed,  but  that  the  sporangiophores  continued  to  grow 
for  10  to  14  days  and  reached  the  length  of  8  to  10  in.  A  2-hr.  exposure  to 
light  was  sufficient  for  partially  formed  sporangiophores  to  complete  their 
development  in  the  dark.  Sporangiophores  and  sporangia  developed 
normally  in  blue  light  but  did  not  develop  in  red-yellow  light. 

Under  natural  conditions,  Piloholus  produces  successive  daily  crops  of 
sporangiophores  and  sporangia.  Each  crop  requires  approximately  24  hr. 
for  its  development.  The  sporangiophores  begin  to  form  near  midday 
or  early  afternoon.  By  evening  they  have  received  enough  light  to  allow 
the  further  development  and  production  of  the  sporangia  during  the 
night.  By  the  following  morning,  the  sporangia  are  completely  formed. 
During  the  morning  the  sporangiophores  react  phototropically,  directing 
the  sporangia  toward  the  source  of  light.  From  midmorning  to  early 
afternoon  the  sporangia  are  discharged  violently  into  the  air  for  a  con- 
siderable distance.  The  horizontal  distance,  according  to  Buller,  may 
be  as  great  as  8  ft.  7  in. 

To  understand  the  mechanism  of  sporangium  discharge  in  Piloholus,  it 
is  first  necessary  to  know  the  structure  of  the  sporangium  and  the  spo- 
rangiophore  (Fig.  69).  The  entire  sporangiophore  consists  of  a  single 
large  cell,  with  a  rather  slender  lower  portion,  a  subsporangial  swelling, 
and  a  conical  columella,  which  projects  upward  into  the  sporangium.  A 
rather  thin  layer  of  cytoplasm  lies  next  to  the  cell  wall  and  surrounds  a 
large  central  vacuole.  At  the  base  of  the  subsporangial  swelling  there  is 
a  thick  perforated  ring  of  protoplasm,  which  is  reddish  in  color,  containing 
carotene. 

When  the  sporangiophore  is  pointed  directly  toward  the  source  of  light, 
the  parallel  rays  of  light  which  strike  the  black  hemispherical  sporangium 
are  screened  out.  The  subsporangial  swelling  acts  as  a  lens,  and  the  rays 
falling  upon  it  are  bent  so  that  they  converge  on,  or  uniformly  near,  the 
red  mass  of  protoplasm  at  the  base  of  the  swelling;  this  results  in  an 
equilibrium,  i.e.,  no  bending  occurs.  When  the  sporangium  is  directed 
at  an  acute  angle  away  from  the  source  of  light,  the  side  of  the  sub- 


340 


PHYSIOLOGY  OF  THE  FUNGI 


Fig.  69.  A,  a  median  longitudinal  section  of  Pilobolus  kleinii  just  before  discharge  oi 
sporangium.  The  gun  is  pointed  at  an  acute  angle  away  from  the  light  source.  The 
light  rays  that  strike  the  sporangium  are  screened  out.  The  rays  that  strike  the  side  of 
the  subsporangial  swelling  are  bent  and  concentrated  on  the  oposite  side.  This  presum- 
ably gives  a  photochemical  stimulus  which  is  conducted  to  the  motor  region  below  the 


SPORE  DISCHARGE  AND  DISSEMINATION 


341 


N) 


A.H.RB, 


B 


subsporangial  swelling.  Bending  toward  the  source  of  light  then  occurs  until  an 
equilibrium  is  reached  and  the  light  rays  are  concentrated  at  the  base  of  the  sub- 
sporangial swelling.  B,  at  this  time  the  gun  is  pointed  directly  toward  the  light 
source.  (After  Buller,  Researches  on  Fungi,  Vol.  VI,  pp.  91,  92,  1934.  Reproduced 
by  permission  of  Longmans,  Roberts  and  Green,) 


342  PHYSIOLOGY  OF  THE  FUNGI 

sporangial  swelling,  acting  as  a  lens,  causes  the  light  rays  to  converge  on 
the  side  of  the  swelling  away  from  the  source  of  light  (Fig.  69A).  Pre- 
sumably, this  causes  a  photochemical  reaction  in  the  protoplasm,  and  the 
stimulus  is  transmitted  downward  to  the  motor  region,  the  portion  of  the 
sporangiophore  just  below  the  subsporangial  swelling.  The  growth  of 
this  region  is  more  rapid  on  the  side  away  from  the  light  source,  which 
results  in  a  bending  of  the  sporangiophore  until  an  equilibrium  of  light  is 
again  reached;  i.e.,  when  the  sporangium  is  pointing  directly  toward  the 
source  of  light  (Fig.  695). 

There  is  a  thicker  layer  of  protoplasm  near  the  upper  portion  of  the 
subsporangial  swelling.  This  layer  also  contains  some  carotene.  Buller 
believes  that  it  is  photochemical ly  reactive  and  may  serve  to  bring  about 
chemical  changes  which  result  in  the  increase  in  the  osmotic  pressure  of 
the  cell.  When  a  culture  with  nearly  mature  sporangia  is  placed  in  the 
dark,  a  much  greater  time  is  required  for  the  discharge  of  the  sporangia 
than  when  it  is  left  in  the  light. 

The  weakest  place  in  the  wall  of  the  Pilobolus  structure  is  located  just 
below  the  sporangium,  and  it  is  here  that  the  wall  of  the  subsporangial 
swelling  breaks  circularly  as  the  sporangium  is  discharged.  The  increased 
osmotic  pressure  becomes  too  great  for  the  resisting  elastic  wall,  and  the 
system  is  ruptured.  A  drop  of  cell  sap  is  squirted  out  of  the  tip  of  the 
subsporangial  swelling  as  discharge  occurs.  The  conical  columella  is  also 
carried  away  with  the  sporangium. 

In  nature  the  sporangia  adhere  to  the  surface  of  vegetation,  where  they 
may  be  ingested  by  herbivorous  animals.  The  spores  are  released  in  the 
digestive  tract  and  pass  out  in  the  feces  unharmed.  It  seems  probable 
that  exposure  to  gastric  juices  helps  to  break  dormancy  of  the  spores  and 
favors  immediate  germination.  The  sporangium  adheres  to  the  vegeta- 
tion by  the  lower  gelatinous  part,  with  the  black,  hemispherical,  non- 
wettable  portion  outward.  Thus,  the  injurious  ultraviolet  rays  are 
screened  out,  and  the  spores  remain  viable,  although  they  may  not  be 
eaten  for  weeks  or  even  months. 

Basidioholus  ranarurti  shows  a  great  many  characters  similar  to  those  of 
Pilobolus,  namely,  the  general  structure  of  the  sporangiophore,  the 
mechanism  of  discharge  of  the  sporangia,  the  coprophilous  habit,  and  the 
general  method  of  dissemination.  The  sporangiophore  consists  of  a 
slender  lower  portion  and  cylindrical  enlarged  upper  portion,  which 
supports  a  spherical  sporangium.  The  osmotic  pressure  in  the  spo- 
rangiophore increases  to  the  point  that  it  exceeds  the  tensile  strength  of 
the  resisting  wall,  which  is  suddenly  ruptured  circularly  near  the  base 
of  the  enlarged  portion  (Ingold,  1934).  At  this  instant,  the  upper  portion 
contracts  and  causes  the  cell  sap  to  be  squirted  backward,  giving  a  rocket- 
like effect.     The  dissemination  of  spores  is  accomplished  after  the  spo- 


SPORE  DISCHARGE  AND  DISSEMINATION  343 

rangia  are  eaten  by  beetles,  which  in  turn  are  eaten  by  frogs  or  hzards. 
After  the  beetles  are  digested  and  the  sporangia  are  released,  the  spores 
are  formed.  These  spores  are  then  capable  of  germination  and  produc- 
tion of  mycelium  on  frog  or  lizard  excreta. 

A  somewhat  different  method  of  discharge  is  described  for  the  genus 
Entomophthora  by  Fitzpatrick  (1930)  and  more  specifically  for  E.  sphaero- 
sperma  by  Sawyer  (1931).  Instead  of  being  due  to  a  squirting  action  of 
the  cell  contents,  as  in  the  case  of  Piloholus  and  Bastdiobolus,  the  discharge 
of  the  sporangium  in  Entomophthora  involves  the  opposing  forces  of 
osmotic  pressure  and  the  adhesive  power  between  the  two  walls  separating 
the  sporangium  and  sporangiophore.  As  the  sporangium  matures,  the 
pressure  on  both  sides  of  the  separating  walls  becomes  so  great  that  the 
outer  wall  is  suddenly  broken,  and  the  sporangium  is  thrown  into  the 
air.  The  sporangia,  being  sticky,  readily  adhere  to  the  objects  which 
they  strike. 

The  theory  of  violent  discharge  of  sporangia  in  certain  downy  mildews 
was  advanced  by  De  Bary  (1887),  using  Peronospora  parasitica  and 
Phytophthora  infestans  as  examples.  Later,  Pinckard  (1942)  found  the 
forcible  adiection  of  sporangia  in  Peronospora  tabacina  to  be  the  same  as 
that  described  by  De  Bary.  As  the  mature  sporangiophore  dries  out, 
the  entire  crown,  with  its  branches  and  sporangia,  begins  a  counterclock- 
wise rotation.  Each  portion  of  the  sporangiophore,  including  the 
sterigma,  rotates  independently.  The  sporangiophore  is  hygroscopic, 
and  as  the  air  becomes  more  moist,  the  movement  is  reversed.  The  effect 
is  a  sudden  release  of  the  mature  sporangia.  During  the  course  of  rota- 
tion many  of  the  branches  become  entangled  with  others,  and  the  spo- 
rangia are  dislodged  by  the  spring-like  action  as  the  branches  are  dis- 
engaged. The  discharge  of  sporangia  was  verified  by  observations  on 
single  isolated  sporangiophores,  showing  that  it  is  not  dependent  upon 
the  intermingling  of  the  sporangiophores.  No  sporangia  were  released 
in  a  saturated  atmosphere,  since  no  hygroscopic  movement  took  place. 
Other  species  which  were  observed  to  react  similarly  were  Peronospora 
parasitica,  P.  geranii,  P.  halstedii,  and  P.  effusa.  Similar  rotation  of 
conidiophores  upon  desiccation  is  apparently  not  uncommon  among  fungi 
of  other  groups,  particularly  those  with  long  conidiophores. 

Discharge  of  ascospores.  In  the  majority  of  fungi  (except  those  whose 
asci  deliquesce),  ascospore  discharge  is  accomplished  by  the  building  up 
of  osmotic  pressure  of  the  ascus  to  a  point  where  it  exceeds  the  resistance 
of  the  elastic  ascus  wall.  In  one  type  of  expulsion,  the  ascus  wall  is 
suddenly  ruptured,  usually  throwing  the  ascospores  outward  into  the  air 
simultaneously.  In  other  species,  the  ascospores  are  discharged  succes- 
sively through  an  apical  pore  in  the  ascus.  In  the  latter  case,  the  ellipsoid 
or  fusoid  shape  of  the  spore  is  apparently  important.     The  spore  pushes 


344  PHYSIOLOGY  OF  THE  FUXGI 

pai't  way  through  the  pore  to  its  broadest  point  and  is  then  suddenly 
squeezed  out  by  the  contraction  of  the  ascus  tip  (Ingold,  1933). 

The  increase  in  the  osmotic  pressure  within  the  maturing  ascus  must 
be  preceded  by  an  increase  in  the  soluble  materials  in  the  cell  sap.  This 
is  believed  to  be  accomplished  by  the  digestion  of  glycogen,  which  is 
known  to  occur  in  the  young  ascus.  In  the  majority  of  t  he  Pyrenomycetes 
the  asci  are  produced  within  a  spherical  or  flask-shaped  perithecium. 

There  are  three  general  ways  by  which  the  ascospores  are  released 
through  the  ostiole  of  the  perithecium,  two  of  which  depend  upon  the 
explosi^•e  rupture  of  the  ascus  wall.  In  the  first  type,  which  is  the  most 
common  and  believed  to  be  the  most  primitive,  the  ascus  wall  remains 
attached  at  its  base,  while  the  spores  are  discharged.  This  is  accom- 
plished by  the  elongation  of  the  elastic  ascus.  until  the  tip  reaches  or 
protrudes  through  the  ostiole.  The  ascus  then  explodes,  throwing  the 
ascospores  into  the  air.  The  wall  of  the  empty  ascus  contracts  to  the 
base  of  the  perithecium.  and  another  ascus  elongates.  The  process  is 
repeated  successively  as  the  asci  mature.  This  type  is  illustrated  by 
Sordaria,  Pleurage,  and  many  other  common  fungi.  An  interesting 
parallelism  exists  between  the  method  of  dissemination  of  Pilobolus  and 
that  of  Pleuragc,  Sordaria,  and  other  coprophilous  Pyrenomycetes.  The 
short  beaks  of  the  perithecia  are  positi\ely  phototropic  and,  as  they 
develop,  are  directed  toward  the  source  of  light.  In  nature  the  ascospores 
fall  upon  vegetation  and  are  subsequently  eaten  and  disseminated  bj' 
herbivorous  animals.  The  vertical  distances  to  which  ascospores  may 
be  shot  have  been  reported  as  6  cm.  for  S.  fimicola  and  45  cm.  for  P. 
curvicolla  (Weimer.  1920). 

A  second  general  type  of  ascospore  discharge  occurs  more  commonly  in 
species  with  long  perithecial  beaks.  The  asci  become  detached  from  the 
base  of  the  perithecium  and  are  pushed  up  through  the  beak  to  the  ostiole, 
where  the  spores  are  released  simultaneously  or  successively.  This  is  a 
rapid  method  of  spore  discharge.  Examples  of  this  type  are  Etidothia 
parasitica.  Gnomonia  rubi,  Guignardia  bidivcllii,  and  CcraiosfomclJa  anipid- 
lax:ea.  !Most  of  these  species  are  adapted  to  AA"ind  dissemination  of 
ascospores.  Some  idea  of  the  tremendous  numbers  of  ascospores  dis- 
charged is  given  by  Heald  and  Walton  (1914),  who  reported  that  some 
specimens  of  E.  parasitica  expelled  ascospores  every  day  for  168  days. 
The  rate  of  spore  discharge  from  one  perithecium  was  found  to  be  as  high 
as  one  ascus  explosion  about  every  2  sec.  At  this  rate  approximately 
14,000  ascospores  may  be  discharged  per  perithecium  per  hour.  On  the 
basis  of  these  figures,  it  is  little  wonder  that  the  fungus  spread  so  rapidly 
among  the  American  chestnuts. 

A  third  group  includes  the  nonexplosive  type  of  ascus,  in  which  the 
ascospores  are  released  by  the  deliquescence  of  the  ascus  wall.     Thej'  are 


SPORE  DISCHARGE  AND  DISSEMINATION  345 

embedded  in  mucilage,  and  as  they  accumulate  in  the  body  of  the  peri- 
theciu.m,  some  spores  ooze  out  through  the  ostiole,  much  like  tooth  paste 
from  the  tube.  Examples  of  this  type  are  Chaetomium  spp.,  Cerato- 
stomeUa  fimhriata,  and  C.  ulmi.  These  spores  are  not  adapted  to  wind 
dissemination  but  may  be  carried  in  moist  weather  by  insects  (C.  ulmi), 
by  other  contacts  (such  as  C.  fimhriata  on  stored  sweet  potatoes),  or  by 
rain. 

The  Discomycetes,  in  general,  show  a  marked  response  to  the  stimulus 
of  light  in  orienting  the  asci  so  that  the  ascospores  may  be  discharged 
into  the  air  away  from  the  apothecium.  The  apothecium  of  Ascobolus 
is  small,  and  only  a  few  asci  mature  at  one  time.  As  an  ascus  matures, 
it  enlarges  greatl}^  and  extends  well  beyond  the  surface  of  the  hymenium. 
It  then  reacts  phototropically  so  that  the  tip  is  pointed  directly  toward 
the  source  of  light.  When  the  ascus  bursts,  the  operculum  at  the  tip  is 
forced  open,  and  the  spores  are  expelled  simultaneously.  Most  species 
of  Ascobohis  are  coprophilous  and  are  disseminated  in  much  the  same 
manner  as  Piloholus  and  Sordaria. 

The  phenomenon  of  ''puffing"  in  manj^  of  the  larger  Discomycetes  is 
described  in  most  textbooks  of  mj^cology  and  plant  pathology.  It  is  due 
to  the  simultaneous  violent  spore  discharge  from  many  asci,  so  that  a 
cloud  of  spores  may  be  seen  to  rise  a  few  inches  from  the  apothecium. 
This  may  be  so  violent  that  a  faint  hissing  or  fizzing  sound  can  be  heard. 
If  the  asci  were  to  explode  singly  as  they  mature,  the  ascospores  would 
be  shot  up  into  the  air  only  bj^  the  initial  force  of  the  explosion.  For  most 
species,  this  distance  would  probably  not  exceed  1  or  2  in.  However, 
when  a  great  many  asci  explode  simultaneously,  an  air  blast  is  created 
which  carries  the  ascospores  vertically  to  a  much  greater  height,  as  great 
as  5  to  7  in.  (Buller,  1934).  This  additional  distance  above  the  fruit 
body,  which  is  commonly  located  on  or  near  the  ground,  increases  the 
chances  of  dissemination  by  air  currents. 

In  nature,  the  puffing  of  ascospores  may  be  initiated  by  a  sudden 
change  from  shade  to  open  sun,  by  the  passing  of  a  cloud,  or  by  swaying 
of  a  branch.  Strong  sunlight  is  not  the  only  stimulus,  for  the  phenomenon 
has  been  observed  in  the  laboratory  under  uniform  light  conditions.  A 
sudden  jar  of  the  fruit  body,  when  it  is  tapped  or  picked  up,  may  cause 
spore  discharge  in  some  species.  Likewise,  an  instant's  exposure  to 
alcohol  fumes  may  serve  as  the  stimulus. 

Buller  (1934)  has  shown  that  in  the  cupulate  or  V-shaped  apothecia, 
such  as  those  of  Sarcoscypha  protracta,  many  of  the  asci  are  pointing 
directly  toward  the  opposite  side  of  the  cup;  yet  the  ascospores  are  dis- 
charged upward,  free  from  the  fruit  body.  The  operculum  of  this  species, 
instead  of  being  centrally  located  at  the  tip  of  the  ascus,  is  obliqueh^ 
placed  toward  the  upper  side  of  the  ascus.     As  the  discharged  ascospores 


346 


PHYSIOLOGY  OF  THE  FUNGI 


leave  the  ascus,  they  are  directed  vertically.  Biiller  })elieves  that  the 
oblique  position  of  the  operculum  is  a  physiological  character  formed  as  a 
response  of  the  ascus  end  to  the  stimulus  of  light.  Seaver  (1928),  how- 
ever, believes  that  the  position  of  the  operculum  is  not  determined  by 


light. 


Fig.  70.  Sections  through  the  hymeniuni  of  Ascobolus  magnificus,  showing  the  photo- 
tropic  response  of  the  ascus  tips  to  Hght.  Discharge  of  the  ascospores  is  then  directly 
toward  the  source  of  hght.  (After  Buller,  Researches  on  Fungi,  Vol.  VI,  p.  272,  1U34. 
Reproduced  by  permission  of  Longmans,  Roberts  and  Green.) 

In  the  development  of  the  apothecium  the  paraphyses  are  formed 
before  the  asci  mature,  and  the  developing  asci  push  their  way  upward 
among  the  paraphyses.  In  some  species  (Ascobolus  spp.,  Lachnea  scuiel- 
lata)  the  paraphyses  are  straight,  and  only  the  portion  of  the  ascus  extend- 
ing beyond  the  paraphyses  tips  responds  phototropically  (Fig.  70).  The 
paraphyses  of  others  {Peziza  hadia,  Aleuria  vesiculosa)  bend  toward  the 
light,  and  the  developing  asci  are  likewise  bent  as  they  elongate.  Aleuria 
repanda  sometimes  shows  a  coarse  adjustment  toward  light,  by  the  turn- 
ing of  the  entire  apothecium,  and  a  fine  adjustment,  by  the  bending  of  the 
ascus  tips  in  the  same  direction. 


SPORE  DISCHARGE  AND  DISSEMINATION 


347 


Discharge  of  peridioles.  The  discharge  of  the  peridiole  (gleba-con- 
taining  basidiospores)  of  Sphaerobolus  depends  largely  upon  the  unique 
structure  of  the  fruit  body  (Fig.  71).  The  spherical  fruit  body  measures 
but  2  to  3  mm.  in  diameter.  The  peridium  is  made  up  of  six  distinct 
layers.  At  maturity,  the  peridium  breaks  open  at  the  top,  in  a  stellate 
manner,  through  all  but  the  sixth,  or  innermost,  layer,  which  surrounds 


Fig.  71.  Structure  of  fruit  body  and  mechanism  of  discharge  of  peridiole  of  Sphaero- 
bolus stellatus.  A,  section  of  mature  sporocarp,  with  six  layers  (1-6)  that  invest  the 
central  peridiole  (7).  B,  dehiscence  of  sporocarp  at  apex.  The  inner  membrane  is 
liquefied.  C,  eversion  of  the  remaining  two  layers  by  which  the  peridiole,  D,  is  sud- 
denly discharged.  (Reproduced  by  permission  from  Wolf  and  Wolf,  The  Fungi,  Vol. 
II,  p.  203,  John  Wiley  &  Sons,  Inc.,  New  York,  1947.) 

the  peridiole.  This  layer  deliquesces,  and  the  peridiole  then  rests  in  the 
watery  substance  produced.  A  split  then  occurs  between  layers  3  and  4, 
beginning  at  the  base  but  not  progressing  to  the  very  top.  Laj^er  4,  the 
fibrous  layer,  is  composed  of  small,  rather  closely  packed  cells,  while  layer 
5,  the  palisade  layer,  is  made  up  of  comparatively  large,  somewhat 
elongated  cells.  The  cells  of  the  palisade,  which  is  on  the  concave  sur- 
face, increase  in  turgor  and  in  size  and  are  held  under  great  tension  by 
the  relatively  inelastic  fibrous  layer.  These  inner  layers  are  suddenly 
everted,  acting  as  a  catapult,  throwing  the  peridiole  violently  upward. 
Walker  (1927)  has  reported  that  the  peridiole  of  S.  stellatus  may  be 
thrown  to  a  vertical  distance  of  14  ft.,  while  Buller  (1934)  reports  a  maxi- 
mum horizontal  distance  of  18  ft.  7  in. 


348  PHYSIOLOGY  OF  THE  FUNGI 

The  force  which  causes  the  discharge  is  apparently  located  in  the 
palisade  layer.  It  has  been  demonstrated  by  microchemical  tests  that 
the  palisade  cells  of  the  unopened  fruit  body  are  densely  filled  with 
glycogen,  which  disappears  before  the  discharge  of  the  peridiole  (Walker 
and  Andersen,  1925).  The  glycogen  is  converted  to  reducing  sugars,  one 
of  which  is  maltose,  and  this  leads  to  the  increase  in  osmotic  pressure. 
Light  hastens  the  opening  of  the  fruit  body  and  the  discharge  of  the 
peridiole  and  is  believed  to  speed  up  the  conversion  of  glycogen  into 
sugars.  S.  stellatus  is  a  coprophilous  or  lignicolous  species,  and  the 
peridioles  may  be  eaten  and  disseminated  by  herbivorous  animals. 

Dodge  (1941)  reports  his  own  observations  as  well  as  those  of  others 
upon  the  presence  of  peridioles  of  the  bird's-nest  fungi  attached  to  leaves 
and  branches  as  high  as  10  to  15  ft.  above  the  ground.  Dodge  describes 
the  attachment  of  the  peridioles  of  Cyathus  striatus  to  the  peridium  but 
offers  no  theory  to  explain  the  mechanism  of  the  peridiole  discharge  or 
the  force  which  is  responsible.  The  slender  mucilaginous  threads  which 
attach  the  peridioles  in  the  fruit  bodies  also  serve  to  attach  the  discharged 
peridioles  to  certain  objects. 

Discharge  of  basidiospores.  The  mechanism  and  the  force  involved 
in  the  discharge  of  basidiospores  in  the  Hymenomycetes  and  of  the 
sporidia  of  the  smuts  and  rusts  have  not  been  satisfactorily  explained. 
There  is  no  evidence  that  the  explanation  used  for  any  of  the  types 
described  above  can  be  applied  to  the  discharge  of  basidiospores.  How- 
ever, certain  structural  features  are  present  which  may  be  adaptations 
for  this  special  method  of  spore  discharge. 

In  all  Basidiomycetes  in  which  the  spores  are  shot  off  forcibly,  the 
sterigma  is  attached  slightly  to  one  side  of  the  tip  of  the  spore  (Fig.  72). 
Just  before  a  spore  is  discharged,  a  small  drop  of  liquid  appears  at  the 
tip  of  the  sterigma.  Its  invariable  presence  is  believed  to  be  an  important 
feature  in  the  process  of  spore  discharge.  After  discharge,  there  appears 
to  be  no  pore  present,  either  in  the  spore  or  in  the  tip  of  the  sterigma. 
Buller  (1922,  1924)  suggests  that  the  spore  may  be  shot  from  the  sterigma 
by  hydrostatic  pressure,  but  that  the  amount  of  liquid  ejected  as  the 
explosion  occurs  at  the  sterigma  may  be  so  minute  that  it  may  not  be 
detected  by  the  microscope.  He  further  suggests  that  the  surface  ten- 
sion of  the  drop  of  water  may  in  some  way  bring  about  the  abjection  of  the 
spore.  While  neither  theory  is  completely  satisfactory,  they  are  the  best 
yet  offered. 

Discharge  of  aeciospores.  Experimental  work  showing  that  the 
aeciospores  of  the  rust  fungi  are  violently  discharged  was  reported  by 
Buller  (1924)  and  Dodge  (1924,  1924a).  They  have  reported  this  phe- 
nomenon in  seven  species,  including  Puccinia  coronota,  P.  graminis,  P. 
podophijlli,  Gytnnoconia  peckiana,  and    Uromyces  pisi.     The  maximum 


SPORE  DISCHARGE  AND  DISSEMINATION 


349 


distance  above  the  aecia  to  which  the  spores  were  shot  varied  from  6  to 
15  mm.,  being  about  8  mm.  for  P.  graminis.  The  exact  mechanism  of 
aeciospore  discharge  is  not  known,  but  it  probably  depends  largely  upon 
turgor  pressure  of  the  mature  aeciospores,  which  are  formed  in  chains. 
The  double  wall  between  two  aeciospores  is  at  first  flat,  but  near  maturity 
of  the  spores  the  osmotic  pressure  increases  and  tends  to  make  these  w^alls 
convex.  It  is  assumed  that  the  adhesive  force  between  the  two  walls  is 
suddenly  overcome  by  the  increased  osmotic  pressure,  and  the  terminal 
spore  or  group  of  spores  is  thrown  outward  into  the  air. 


Fig.  72.  Discharge  of  the  third  basidiospore  from  basidium  of  Agaricus  campestris, 
following  the  formation  of  a  drop  of  water  at  the  tip  of  the  sterigma.  (After  Duller, 
Researches  on  Fungi,  Vol.  II,  p.  12,  1922.  Reproduced  by  permission  of  Longmans, 
Roberts  and  Green.) 

Liberation  of  zoospores.  In  most  of  the  aquatic  Phycomycetes  and  in 
some  terrestrial  forms  which  show  definite  aquatic  affinity,  zoospores  are 
the  primary  means  of  reproduction.  The  characteristic  motility  of  the 
zoospores  may  be  more  correctly  considered  as  a  means  of  local  dissemina- 
tion, but  motility  is  also  involved  in  the  liberation  from  such  large  spo- 
rangia as  those  of  Saprolegnia  and  related  fungi. 

INFLUENCE  OF  EXTERNAL  CONDITIONS 

The  effect  of  light  upon  the  discharge  of  spores  by  Piloholus,  Ascoholus, 
Sordaria,  and  Sphaeroholus  has  been  discussed  briefly.  Light  is  necessary 
for  the  production  of  spores  in  a  number  of  fungi  in  which  it  plays  no 
direct  part  in  spore  discharge. 

Favorable  temperature  is  a  prerequisite  for  all  biological  activity. 
Its  effect  upon  spore  formation  and  discharge  is  often  not  clearly  defined. 
The  maximum  temperature  permitting  spore  discharge  is  appreciably 


350 


PHYSIOLOGY  OF  THE  FUNGI 


Fig.  73.  Stills  from  ultra-high-speed  film  showing  impact  of  drop  of  water  with  a  fruit 
body  of  Lycoperdon  perlatum  and  the  subsequent  puff  of  spores.  The  drop,  5.0  mm.  in 
diameter,  fell  with  a  velocity  of  -440  cm.  per  sec.  The  time  elapsed  between  the  con- 
tact of  drop  with  peridiimi  until  the  last  photograph  was  0.046  sec.  (Courtesy  of 
Gregory,  Trans.  Brit.  Mycol.  Soc.  32 :  14,  1949.  Published  by  permission  of  Cam- 
bridge University  Press.) 


SPORE  DISCHARGE  AND  DISSEMINATION  351 

lower  than  that  for  viabiht}^  of  the  fungus.  Low  temperatures  usually 
merely  slow  down  spore  production  and  discharge.  Duller  (1909)  found 
that  Daedalea  unicolor,  Lenzites  hetulinus,  Polyporus  versicolor,  and  P. 
hirsutus  discharged  spores  when  the  air  temperature  was  0°C.  Schizo- 
phyllum  commune  shed  spores  vigorously  at  5°C.,  but  not  at  0°C. 

Andersen  et  al.  (1947)  showed  that  few  conidia  of  Piricularia  orijzae 
were  liberated  when  the  host  plants  were  dry.  Continued  wetting  greatly 
increased  the  secondary  spread  of  the  fungus  in  experimental  trials. 
Sporidia  of  rusts  are  formed  and  discharged  only  during  periods  of  high 
humidity.  The  need  of  the  downy  mildew  fungi  for  changing  conditions 
of  humidity  has  been  pointed  out.  The  hygroscopic  character  of  the 
capillitium  of  certain  slime  molds  aids  in  pushing  the  spores  to  the  surface, 
where  they  may  be  disseminated  by  various  agents.  Many  other  species 
of  fungi  appear  to  be  independent  of  the  air  moisture,  as  long  as  there  is 
sufficient  moisture  in  the  fruit  body.  Gravity  is  believed  to  have  little 
effect  upon  spore  discharge,  except  in  the  proper  orientation  of  the  fruit 
bodies. 

The  spores  of  some  puffballs  are  enclosed  within  the  nearly  spherical 
peridium,  which  opens  by  an  apical  pore.  When  sudden  pressure  is 
applied  to  the  peridium,  the  spores  are  puffed  out  of  the  pore  in  clouds  of 
"smoke."  Gregory  (1949)  has  shown,  by  use  of  ultraspeed  photography, 
that  raindrops  falling  on  the  thin  peridium  of  Lycoperdon  perlatum  cause 
a  puffing  of  spores  (Fig.  73).  The  velocity  of  the  puff  as  the  spores 
emerged  from  the  ostiole  was  approximately  100  cm.  per  sec.  Under  the 
conditions  of  the  experiment,  it  was  estimated  that  a  drop  of  water  falling 
130  cm.  caused  the  ejection  of  approximately  15  million  spores.  The 
endoperidium  and  the  spores  inside  remain  dry,  and  the  puffing  is  not 
hindered,  even  during  a  rain.  The  impact  of  raindrops  is  believed  to 
be  an  important  means  of  spore  discharge  from  the  ostiolate  puffballs. 

SPORE  DISSEMINATION 

Many  other  fungi  do  not  have  any  special  method  of  spore  discharge 
and  must  depend  upon  physical  or  biotic  agents  in  nature  for  getting  their 
spores  away  from  the  fruit  body  where  they  are  produced.  Some  of  these 
possess  some  special  adaptations  for  dissemination  by  certain  agents. 
The  most  important  agent  of  dissemination  is  air  currents.  The  uredio- 
spores  of  the  rusts  are  not  violently  discharged.  They  accumulate  in 
the  sorus  and  must  be  dislodged  by  the  wind  or  movement  of  the  host 
plant.  Many  of  the  Monihales  which  produce  dry  conidia  also  depend 
upon  air  movement  to  shake  them  loose  from  the  conidiophores.  Con- 
vection currents  are  responsible  for  local  spread  of  dry  spores,  while 
splashing  rain  is  important  in  the  dissemination  of  spores  with  matrix. 


352  PHYSIOLOGY  OF  THE  FUNGI 

It  is  well  known  that  spores  of  some  of  these  fungi  may  be  blown  for 
hundreds  of  miles  and  remain  viable. 

Dissemination  of  the  zoospores  of  the  aquatic  fungi  may  be  accom- 
plished locally  by  means  of  the  flagella  or  for  greater  distances  by  the 
movement  of  water,  which  may  carry  the  spores  both  in  the  motile  and 
in  the  encysted  stages.  The  condition  of  diplanetism,  which  involves 
two  motile  stages  separated  by  an  encysted  stage,  may  be  advantageous 
for  the  greater  dissemination  in  water.  The  translocation  of  mud  and 
moist  soil  by  means  other  than  water  may  also  be  important.  Although 
the  motility  of  zoospores  of  certain  parasitic  fungi,  such  as  Phytophthora 
infestans,  can  scarcely  be  considered  an  important  means  of  dissemina- 
tion, it  does  enable  the  spore  to  move  short  distances  in  a  drop  of  water 
on  the  host  and  facilitate  penetration  through  stomata. 

Insects  are  likewise  important  agents  of  dissemination  of  fungus  spores. 
The  insect  may  be  attracted  to  spore  masses  by  odor  or  color  and  feed 
upon  the  spores,  or  the  insect  may  be  merely  an  incidental  carrier  of 
spores  adhering  to  the  external  parts  of  the  body.  Leach  (1940)  states 
that  "in  the  majority  of  cases  where  the  question  has  been  investigated, 
spores  have  been  found  to  pass  through  the  intestinal  tract  of  insects 
uninjured."  The  conidia  of  Claviceps  purpurea,  being  produced  in  sweet 
droplets  of  liquid,  also  emit  an  odor  which  attracts  insects.  Similarly, 
the  stinkhorn  fungi  produce  their  spores  in  a  malodorous  matrix  and  are 
frequently  visited  by  flies.  These  adaptations  ensure  insect  dissemina- 
tion. The  blue-stain  fungi  {Ceratostomella  spp.)  produce  spores  in  sticky 
droplets  in  the  tunnels  of  bark  beetles,  which  act  as  the  principal  agents 
of  dissemination. 

The  symbiotic  relationship  between  species  of  Septohasidium  and  scale 
insects  represents  a  highly  evolved  adaptation  for  the  dissemination  of 
the  fungus  spores.  Although  these  fungi  produce  basidiospores,  they  are 
unlike  most  Basidiomycetes  in  that  the  spores  are  not  forcibly  discharged 
from  the  basidium.  Couch  (1938)  has  made  a  comprehensive  study  of 
this  genus  and  has  described  the  life  history  in  detail.  The  fungus  forms 
a  layer  over  the  bodies  of  scale  insects,  some  of  which  are  parasitized, 
while  others  are  not.  The  uninfected  female  insects  under  the  fungus 
may  produce  young,  many  of  which  crawl  over  the  sporulating  surface 
at  the  time  of  sporulation.  These  may  become  infected,  crawl  about,  and 
settle  down  some  distance  away.  Such  infected  young  insects  are  solely 
responsible  for  the  dissemination  of  the  fungus. 

Spore  dissemination  is  also  unique  in  the  Tuberales  (truffles),  whose 
fruit  bodies  are  formed  entirely  underground.  There  are  no  direct  means 
of  getting  the  spores  up  to  the  air  for  dissemination.  These  fruit  bodies, 
which  give  off  an  odor,  are  dug  up  and  eaten  by  rodents.  In  this  process, 
pieces  of  the  fruit  bodies  are  dropped,  and  the  spores  are  thus  disseminated. 


SPORE  DISCHARGE  AND  DISSEMINATION  353 

SUMMARY 

Many  fungi  have  no  means  of  forcibly  discharging  their  spores  but  must 
depend  upon  the  physical  and  biotic  factors  in  nature  for  liberation  and 
dissemination  of  spores.  Others  possess  special  mechanisms  for  dis- 
charging their  spores  away  from  the  fruiting  structures  which  bear  them. 
In  most  cases  this  violent  discharge  depends  upon  high  osmotic  pressure 
within  certain  cells  of  the  fungus.  Increased  osmotic  pressure  usually 
is  a  result  of  the  digestion  of  glycogen  to  soluble  sugars. 

Many  coprophilous  fungi,  such  as  Pilobolus,  Ascobolus,  Sordaria,  and 
Pleurage,  forcibly  discharge  their  spores  for  some  distance  toward  the 
source  of  light.  In  nature  this  adaptation  is  of  great  advantage  to  the 
fungus  in  its  dissemination  by  animals,  which  ingest  the  spores  with  the 
vegetation.  Some  Discomycetes  exhibit  a  puffing  of  the  spores  when 
many  asci  discharge  their  spores  simultaneously. 

In  Pilobolus,  Basidiobolus,  and  Entomophfhora  the  sporangia  are  forcibly 
abjected  from  the  sporangiophores.  Ascospores  may  be  ejected  either 
simultaneously  or  successively  from  the  ascus.  The  ascus  may  elongate 
to  reach  the  surface  of  the  ascocarp  and  discharge  its  spores,  or  the  asci 
may  become  detached  in  some  Pyrenomycetes  and,  after  being  forced 
through  the  ostiole,  may  explode  to  release  the  spores.  In  other  species 
the  ascus  walls  are  deliquescent,  and  the  ascospores  ooze  out  of  the 
ostiole. 

The  basidiospores  of  most  Basidiomycetes  are  forcibly  discharged  by  a 
mechanism  which  is  not  well  understood.  A  drop  of  liquid  is  extruded  at 
the  tip  of  the  sterigma  just  prior  to  discharge  and  is  believed  to  affect  the 
process  in  some  way.  The  peridioles  of  Cijathus  and  Sphaerobolus  may 
be  thrown  several  feet  away  from  the  fruit  bodies.  The  latter  fungus 
exhibits  a  unique  catapult  action  by  a  portion  of  the  fruit  body.  Other 
mechanisms  act  in  the  forcible  discharge  of  the  aeciospores  of  many  rusts 
and  the  sporangia  of  Peronospora. 

Air  currents  are  the  most  common  agent  of  dissemination  of  dry  spores. 
Spores  borne  in  a  sticky,  malodorous,  or  sweet  matrix  are  well  adapted  to 
insect  dissemination.  Some  fungi,  such  as  Septobasidium,  have  estab- 
lished a  symbiotic  relationship  with  certain  insects,  which  are  the  sole 
agents  of  dissemination. 

REFERENCES 

Andersen,  A.  L.,  W.  B.  Henry,  and  E.  C.  Tullis:  Factors  affecting  infectivity, 

spread  and  persistence  of  Piricularia  oryzae,  Phytopathology  37:  9-4-110,  1947. 
Brefeld,  O.:  Botanische  Untersuchungen  iiber  Schimmelpilze,  Heft  4,  Verlag  von 

Arthur  Felix,  Leipzig,  1881. 
*Buller,  A.  H.  R.:  Researches  on  Fungi,  Longmans,  Roberts  and  Green,  London. 

Vol.  I,  1909;  Vol.  II,  1922;  Vol.  Ill,  1924;  Vol.  VI,  1934. 
Couch,  J.  N. :  The  Genus  Septobasidium,  The  University  of  North  Carolina  Press, 

Chapel  Hill,  1938. 


354  PHYSIOLOGY  OF  THE  FUNGI 

De  Bary,  a.:  Comparative  Morphology  and  Biology  of  the  Fungi,  Mycetozoa  and 

Bacteria  (trans.  H.  E.  F.  Garrney,  rev.  I.  B.  Balfour),  Oxford  University  Press, 

New  York,  1887. 
Dodge,  B.  O.:  Aecidiospore  discharge  as  related  to  the  character  of  the  spore  wall, 

Jour.  Agr.  Research  27:  749-75G,  1921. 
Dodge,  B.  O.:  Expulsion  of  aeciospores  by  the  may  apple  rust,  Puccinia  podophylii 

Schw.,  Jour.  Agr.  Research  28:  923-92(i,  1924a. 
Dodge,  B.  O.:  Discharge  of  the  sporangioles  of  bird's  nest  fungi,  Mycologia  33: 

650-654,  1941. 
FiTZPATRiCK,  H.  M.:  The  Lower  Fungi,  McGraw-Hill  Book  Company,  Inc.,  New 

York,  1930. 
Gregory,  P.  H. :  The  operation  of  the  puff-ball  mechanism  of  Lijcoperdon  perlatum 

by   raindrops   shown   by   ultra-high-speed    Schlieren   cinematography,    Trans. 

Brit.  Mycol.  Soc.  32:  11-15,  1949. 
Heald,  F.  D.,  and  R.  C.  Walton:  The  expulsion  of  ascospores  from  the  perithecia 

of  the  chestnut  blight  fungus,  Endothia  parasitica,  Am.  Jour.  Botany  1 :  499-522, 

1914. 
*Ingold,  C.  T.:  Spore  discharge  in  the  ascomycetes,  New  Phytologist  32:  178-196, 

1933. 
Ingold,    C.   T.:   The   spore-discharge   mechanism   in    Basidiobolus  ranarum,    New 

Phytologist  33:  274-277,  1934. 
*Leach,  J.  G.:  Insect  Transmission  of  Plant  Diseases,  McGraw-Hill  Book  Company, 

Inc.,  New  York,  1940. 
*PiNCKARD,  J.  A.:  The  mechanism  of  spore  dispersal  in  Peronospora  tabacina  and 

certain  other  downy  mildew  fungi.  Phytopathology  32:  505-511,  1942. 
Sawyer,  W.  H.  :  Studies  in  the  morphology  and  development  of  an  insect-destroying 

fungus,  Entoniophthora  sphaerosperma,  Mycologia  23:  411-432,  1931. 
Seaver,  F.  J.:  The  North  American  Cup  Fungi.     Operculates,  published  by  the 

author,  New  York,  1928. 
*  Walker,  L.  B.:  The  development  and  mechanism  of  the  discharge  in  Sphaerobolus 

iowensis  and  S.  stellatus,  Jour.  Elisha  Mitchell  Set.  Soc.  42 :  151-178,  1927. 
Walker,  L.  B.,  and  E.  N.  Andersen:  Relation  of  glycogen  to  spore  ejection, 

Mycologia  17:  154-159,  1925. 
Weimer,  J.  L.:  Some  observations  on  the  spore  discharge  of  Pleurage  curvicolla, 

Am.  Jour.  Botany!:  75-77,  1920. 
Wolf,  F.  A.,  and  F.  T.  Wolf:  The  Fungi,  Vol.  II,  John  Wiley  &  Sons,  Inc.,  New 

York,  1947. 


CHAPTER   16 
SPORE    GERMINATION 

Spore  germination,  in  general,  implies  a  change  from  an  inactive  to  an 
actively  growing  condition.  This  is  accomplished  in  most  fungi  b}^  the 
formation  of  a  germ  tube,  which  continues  to  elongate  and  ultimately 
leads  to  the  formation  of  the  vegetative  body  of  the  fungus.  In  the 
Phycomycetes  the  germination  of  oospores  and  of  some  sporangia  may 
take  place  by  the  internal  formation  of  zoospores.  Certain  higher  fungi 
also  produce  secondary  spores  externally  without  the  formation  of 
mycelium. 

Among  the  universal  requirements  for  the  germination  of  spores  are 
(1)  suitable  temperature,  (2)  adequate  moisture  supply,  (3)  adequate 
oxygen  supply,  (4)  suitable  hydrogen-ion  concentration,  and  (5)  viable 
spores. 

Some  of  these  factors  may  be  measured  quantitatively,  and  for  each 
there  is  an  optimum  for  germination  for  a  given  fungus.  The  maximum 
percentage  of  spore  germination  in  the  shortest  time  will  occur  when  all 
the  influencing  factors  are  at  or  near  the  optimum.  This  is  a  situation 
which  might  seldom  occur  in  nature.  Germination  will  occur  or  not 
depending  upon  the  number  and  relative  importance  of  the  favorable 
factors. 

Clayton  (1942)  suggests  that  the  differences  in  nutritional  require- 
ments for  germination  shown  by  the  various  species  may  be  due  to  dif- 
ferences in  the  spore  wall  and  in  the  composition  and  quantity  of  reserve 
foods.  Some  spores  contain  stored  food  in  the  form  of  oil,  while  others 
contain  glycogen.  It  is  believed  that  water  enters  the  spore  by  imbibition 
or  osmosis  and  activates  the  enzyme,  glycogenase,  which  hydrolyzes 
glycogen  to  sugars. 

Spores  which  do  not  germinate  after  being  exposed  to  the  usually 
favorable  conditions  for  a  reasonable  length  of  time  are  said  to  have  a 
period  of  dormancy,  which  may  be  broken  by  the  presence  of  a  special 
set  of  conditions.  Among  these  are  (1)  exposure  to  high  or  low  tempera- 
ture, (2)  the  presence  of  certain  nutrients  or  stimulants  in  natural  prod- 
ucts, (3)  exposure  to  chemical  stimulants,  (4)  alternate  wetting  and  dry- 
ing, and  (5)  aging.  These  same  factors  may  also  influence  the  percentage 
of  spore  germination  of  many  species  which  do  not  have  a  definite  dormant 
period.     Dormancy  may  be  due,  at  least  in  some  cases,  to  the  failure  of 

355 


356  PHYSIOLOGY  OF  THE  FUNGI 

the  usual  favorable  conditions  to  activate  a  certain  enzyme.     Some 
special  stimulus  may  then  be  required  to  perform  this  function. 

The  present  discussion  deals  primarily  with  the  factors  which  influence 
some  physiological  phase  of  spore  germination.  The  main  emphasis 
will  be  placed  upon  a  discussion  of  these  factors  and  how  they  act,  or 
what  part  they  play  in  germination.  Doran  (1922)  gives  a  good  discus- 
sion of  the  more  important  factors  affecting  germination  and  lists  many 
references  to  the  earlier  work.  A  more  recent  discussion  of  this  subject 
is  given  in  the  excellent  review  by  Gottlieb  (1950). 

PHYSICAL  FACTORS 

Temperature.  Temperature  is  one  of  the  most  important  external 
factors  which  influence  germination.  It  not  only  affects  the  percentage 
of  germination  but  also  the  length  of  time  required  for  germination,  and, 
in  certain  fungi,  it  often  determines  the  method  of  germination.  The 
literature  contains  a  great  many  references  to  spore  germination  at  dif- 
ferent temperatures,  but  none  of  these  have  attempted  to  explain  the 
intricate  effects  or  responses  within  the  spore.  We  may  assume  that  a 
favorable  temperature  permits  certain  enzymatic  activities  essential  to 
germination.  Different  species  of  fungi  have  different  temperature 
requirements  for  germination.  The  cardinal  temperatures  (minimum, 
optimum,  and  maximum)  for  spore  germination  may  be  found  for  many 
fungi  in  scattered  reports.  These  are  based  mainly  on  casual  observa- 
tions made  during  the  study  of  other  problems.  Few  comprehen- 
sive studies  have  been  made  of  the  effects  of  temperature  upon  spore 
germination. 

The  cardinal  temperatures  of  a  few  selected  species  are  presented  as 
examples:  AUernaria  solani,  1  to  3°C.,  26  to  28°C.,  37  to  45°C. ;  Cronartium 
rihicola  aeciospores,  5°C.,  12°C.,  19°C.;  Phyllosticta  antirrhini,  18°C., 
25°C.,  'i7°C.  ■,Phytophthora  infestans,  2  to  3°C.,  12  to  13°C.,  24  to  25°C.; 
Puccinia  graminis  teliospores,  5  to  9°C.,  20  to  22°C.,  23  to  25°C. ;  Venturia 
inaequalis  conidia,  3°C.,  14  to  15°C.,  31°C.  For  other  examples,  see 
Doran  (1922). 

The  general  optimum  temperature  for  spore  germination  of  certain 
species  of  Agaricaceae  and  Nidulariaceae  is  near  30°C.  (Kauffman,  1934). 
Walker  and  Wellman  (1926)  found  that,  when  the  soil  temperature  was 
above  25°C.,  there  was  low  percentage  of  "chlamydospore"  germination 
of  Urocystis  cepulae,  while  spore  germination  ceased  entirely  at  29  or 
30°C.  They  attribute  the  low  percentage  of  infection  above  25°C.  to  the 
direct  inhibitory  effect  of  the  higher  temperature  upon  the  parasite. 

The  presence  of  nutrients  and  the  supply  of  oxygen  caused  variation 
in  the  minimum  temperature  for  spore  germination  in  Colletotrichum 
lagenarium  (Gardner,  1918).     The  minimum  was  14°C.  in  water  with  the 


SPORE  GERMINATION  357 

hanging-drop  technique,  7°C.  in  exposed  drops  of  water,  and  4°C.  in 
prune  decoction.  This  may  serve  to  exphiin,  in  part,  the  differences  in 
cardinal  temperatures  reported  by  different  authors. 

The  method  of  spore  germination  may  be  determined  by  temperature. 
Below  20°C.  the  sporangia  of  Phytophthora  infestans  germinate  more 
frequently  by  the  formation  of  zoospores,  while  above  this  temperature 
production  of  a  germ  tube  is  more  common. 

The  optimum  temperatures  for  germination  for  the  various  species  of 
Myxomycetes  were  from  22  to  30°C.,  with  an  over-all  range  of  2  to  36°C. 
Certain  species  had  narrower  ranges.  Below  10  and  above  30°C.  the 
rate  of  germination  was  greatly  reduced.  When  the  spores  of  Enteridium 
rozeanum  were  held  for  a  time  at  a  temperature  above  maximum  and  then 
returned  to  optimum,  the  spores  germinated  explosively  through  a  thin 
area  in  the  wall  (Smart,  1937).  Under  uniform  optimum  temperature 
an  irregular  pore  was  formed  and  the  protoplast  slowly  squeezed  out. 

It  must  be  emphasized  that  temperature  affects  the  time  required  for 
germination,  as  well  as  the  percentage  of  germination  and  the  growth 
rate  of  the  germ  tubes.  All  three  have  been  used  as  measurements  of 
spore  germination.  It  seems  likely  that  temperature  might  have  a 
greater  effect  upon  the  time  of  germination  than  upon  the  percentage 
of  germination. 

Heat  treatment  and  breaking  of  dormancy.  The  effectiveness  of  pre- 
heating ascospores  of  Neurospora  tetrasperma  in  breaking  their  dormancy 
has  been  discussed  by  several  authors.  Heating  the  spores  to  50°C.  for 
a  few  minutes  induced  germination  2  or  3  hr.  after  they  were  returned  to 
a  favorable  temperature.  Goddard  (1935)  found  that  spores  thus 
"activated"  could  be  "inactivated"  (returned  to  the  dormant  condition) 
by  placing  them  under  anaerobic  conditions  for  a  short  time.  They 
remained  dormant  when  brought  back  into  air.  The  respiration  rate  of 
the  activated  spores  was  greatly  increased  and  germination  occurred 
only  after  2  to  3  hr.  of  continuous  high  respiration.  Further  work  by 
Goddard  and  Smith  (1938)  led  to  the  conclusion  that  the  heat  activated 
carboxylase,  which  is  latent  in  the  dormant  spores,  and  that  two  different 
respiratory  systems  are  in  operation :  one,  the  dormant  system,  functions 
in  the  absence  of  carboxylase;  and  the  second,  the  active  system,  functions 
after  the  spores  are  heated.  Similar  stimulation  of  germination  of 
Ascoholus  ascospores  by  heat  was  found  by  Dodge  (1912).  Only  a  few 
spores  germinated  in  water  without  being  preheated.  Heating  the  spores 
to  65  to  75°C.  for  approximately  15  min.  and  then  returning  them  to 
favorable  temperature  allowed  good  germination  of  most  species. 

Moisture.  The  spores  of  many  species  of  fungi  will  not  germinate 
unless  they  are  in  contact  with  liquid  water.  Others  are  capable  of 
germination  on  dry  surfaces  in  an  atmosphere  of  high  humidity,  usually 


358  PHYSIOLOGY  OF  THE  FUNGI 

95  per  cent  or  above.  A  third  group  is  represented  by  some  of  the  pow- 
dery mildews,  whose  spores  are  able  to  pat  out  short  germ  tubes  under 
conditions  of  extremely  low  relative  humidity.  Comparatively  little 
careful  work  has  been  done  to  determine  the  moisture  requirements  for 
spore  germination.  Doran  (1922)  reviewed  some  of  the  earlier  reports 
and  gives  the  results  of  his  own  experiments.  Among  the  species  whose 
spores  have  been  reported  as  requiring  contact  with  liquid  water  for 
germination  are  the  following:  Sderotinia  Jnicticola,  Peronospora  pygmaea, 
Phyllosticta  antirrhim,  teliospores  of  Gymno sporangium  juniperi-vir- 
ginianae,  Cylindrocladium  scoparium,  and  Plasmopara  viticola. 

Numerous  fungi  whose  spores  may  germinate  in  the  absence  of  liquid 
water  have  been  reported.  Some  of  these  have  been  germinated  on  a  dry 
glass  slide  in  a  moist  chamber,  where  the  humidity  is  assumed  to  be  at 
100  per  cent,  the  saturation  point.  However,  Clayton  (1942)  showed 
that  a  humidity  of  100  per  cent  sometimes  gave  visible  condensation  of 
water  vapor,  w^hereas  a  relative  humidity  below  99.85  per  cent  gave  no 
condensation  at  constant  temperature.  The  spores  of  this  group  of 
fungi  usually  show  a  much  higher  percentage  of  germination  in  liquid 
w'ater  if  a  plentiful  supply  of  oxygen  is  present.  Some  representative 
fungi  reported  in  this  group  with  the  approximate  minimum  humidity 
are  Puccinia  glumarum  urediospores,  99  per  cent;  Venturia  inaequalis 
ascospores  and  conidia,  98.7  per  cent;  Ustilago  nuda,  95  per  cent;  and 
Penicillium  glaucum,  84  per  cent.  The  minimum  relative  humidity  for 
Aspergillus  niger  is  near  70  per  cent  (Bonner,  1948).  Figure  74  shows 
germination  curves. 

The  germination  of  the  conidia  of  certain  species  of  the  Erysiphaceae 
in  relative  humidity  at  or  near  zero  has  been  reported  by  several  investi- 
gators (Brodie,  1945;  Brodie  and  Neufeld,  1942;  Yarwood,  1936;  and 
Clayton,  1942).  These  species  are  Erysiphe  polygoni,  E.  graminis,  and 
Microsphaera  alni.  The  mechanism  for  spore  germination  under  these 
very  dry  conditions  must  be  quite  different  from  that  of  other  spores, 
which  require  liquid  water  or  high  humidity  for  germination.  The 
"apparent  osmotic  pressure"  of  the  cell  sap  of  the  conidia  is  reported  by 
Brodie  (1945)  as  about  63  atm.  for  E.  polygoni  and  68  atm.  for  E.  graminis 
hordei.  It  is  likely  that  these  high  osmotic  pressures  may  be  an  aid  in 
absorbing  moisture  from  a  relatively  dry  atmosphere.  Brodie  believes 
that  the  conidia  contain  little  free  water  but  that  imbibition  may  be 
partly  brought  about  by  hydrophilic  colloids. 

Brodie  and  Neufeld  (1942)  offer  a  tentative  theory  to  explain  "germi- 
nation" under  conditions  of  0  per  cent  humidity.  They  believe  that,  as 
germination  begins,  free  water  is  released  by  respiration  and  by  changes 
in  colloidal  materials  containing  bound  water.  No  changes  in  the  length 
or  width  of  the  conidia  could  be  detected  during  germination.     The 


SPORE  GERMINATION 


359 


formation  of  the  germ  tube  was  calculated  to  add  approximately  2  or  3 
per  cent  to  the  volume  of  the  ungerminated  conidium.  It  is  believed 
possible  that  this  slight  increase  in  volume  might  be  accounted  for  by 
one  or  both  of  the  above  factors.  Yarwood  (1936)  offers  no  explanation 
of  the  process  of  germination  at  such  a  low  humidity  but  reports  a  decrease 


o 


3 
•*- 
O 

a. 
£ 


10 
15 

/< 

20 
25 
30 
35 
40 
45 

IPl— 

/ 

1 

,^ 

-^ 



<7 

/ 

/ 

A 

^ 

/ 

< 

1 1 

'/ 

f\ 

/> 

(1 

'^\ 

[1 

/ 

L 

[\ 

{ 

v[ 

o 

^ 

V. 

~^^^ 

•z ^ 

-^ — ' 

50 

Asper 

grown 

gillus  n 
on  cellc 

ger 
Dphane 

dipped  1 

n  malt 

70 


95 


100 


75  80  85  90 

Relative  humidity  in  percent 

Fig.  74.  Germination  curves  for  Aspergillus  niger  under  variable  temperature  and 
humidity.  Note  that  the  optimum  temperature  for  germination  varied  with  the 
relative  humidity,  being  near  30°C.  at  relative  humidity  of  100  per  cent  and  near  40°C. 
at  93  per  cent.  As  the  temperature  or  humidity  digressed  from  the  optimum, ^the 
time  required  for  germination  increased.  (Courtesy  of  Bonner,  Mycologia  40 :  733, 
1948.) 

of  about  24  per  cent  in  volume  of  the  Erijsiphe  conidia  during  germina- 
tion. Spores  of  all  other  fungi  (except  other  powdery  mildews)  which  he 
tested  showed  increases  in  volume  during  germination. 

Dormancy  of  some  spores  may  be  broken  by  alternate  wetting  and 
drying.  This  treatment  apparently  makes  the  thick  resistant  wall  more 
permeable  to  water. 

Oxygen  supply.  Since  respiration  is  greatly  accelerated  during  spore 
germination,  it  follows  thg^t  an  adequate  supply  of  oxygen  is  a  prerequisite 


360  PHYSIOLOGY  OF  THE  FUNGI 

for  germination.  Brief  reports  of  a  number  of  observers  on  oxygen 
requirements  are  given  by  Doran  (1922).  It  is  generally  agreed  that 
reduced  oxygen  supply  decreases  spore  germination.  Spores  germinate 
better  on  or  near  the  surface  of  a  li(iuid  than  when  submerged  deep  in  the 
liquid.  In  some  cases  the  spores  may  germinate  under  water,  but  only 
abnormal  germ  tubes  are  formed.  Aerated  water  gives  better  germina- 
tion than  nonaerated  water.  The  spore  load  in  a  drop  of  water,  whether 
all  of  the  same  species  or  of  mixed  spores,  influences  greatly  the  percent- 
age of  germination.  This  is  believed  to  be  due  primarily  to  the  competi- 
tion for  the  limited  supply  of  oxygen,  rather  than  to  toxic  substances 
produced  by  other  germinating  spores. 

According  to  Jones  (1923),  spore  germination  of  Ustilago  avenae  is 
greatest  in  soil  with  30  per  cent  of  water-holding  capacity  and  is  greatly 
reduced  at  80  per  cent.  This  was  probably  due  to  the  amount  of  avail- 
able oxygen.  The  spores  failed  to  germinate  in  water  when  exposed  to 
an  oxygen-free  atmosphere.  The  "  chlamydospores"  of  Ustilago  zeae 
do  not  germinate  in  the  absence  of  oxygen,  and  at  least  5  per  cent  oxygen 
must  be  present  to  allow  germination  as  high  as  in  the  open  air  (Platz 
et  at.,  1927). 

The  supply  of  oxygen  may  influence  the  method  of  spore  germination 
(Uppal,  1926).  Germination  by  zoospores  was  possible  in  the  absence 
of  oxygen  for  the  sporangia  of  Phytophthora  mfcstans,  P.  colocasiae,  P. 
palmivora,  and  P.  parasitica.  Germination  by  germ  tubes  does  not  take 
place  in  these  species  in  the  absence  of  oxygen.  However,  the  presence 
of  oxygen  is  essential  for  zoospore  formation  by  sporangia  of  Alhiigo 
Candida,  Plasmopara  viticola,  and  Sclerospora  graminicola.  The  two 
methods  of  germination  are  different  processes,  the  direct  method  more 
nearly  resembling  vegetative  growth. 

Hydrogen-ion  concentration.  Under  natural  conditions  acidity  is  not 
usually  a  limiting  factor  for  spore  germination.  In  general,  spores  will 
germinate  within  a  wide  pH  range.  It  seems  significant  that,  in  most 
species  of  fungi,  germination  is  favored  by  an  acid  medium,  often  at  a 
pH  considerably  lower  than  the  optimum  for  vegetative  growth  or  sporu- 
lation.  The  effects  of  acidity  of  the  medium  upon  a  number  of  species, 
including  Botrytis  cinerea,  Aspergillus  niger,  Penicillium  cyclopium,  P. 
italicum,  Puccinia  graminis  urediospores,  Lenzites  saepiaria,  Colleto- 
trichum  gossypii,  and  Fusarium  sp.,  are  reported  by  Webb  (1921).  The 
spores  of  the  Fusarium  germinated  equally  well  in  alkaline  and  acid 
media,  while  CoUetotrichuni  gossypii  was  the  only  species  of  the  group 
studied  in  which  germination  was  better  in  an  alkaline  medium.  At 
pH  2.5  spore  germination  was  prevented  in  all  species,  and  the  optimum 
for  most  species  was  3.0  to  4.0.  In  sucrose-nitrate  (Czapek's)  solution, 
two  maxima  usually  occurred,  the  primary  one  at  pH  3.0  to  4.0  and  a 


SPORE  GERMINATION  361 

secondary  one  between  G.O  and  7.0.  Of  all  the  media  tested,  beet  decoc- 
tion gave  the  maximum  germination  imder  the  widest  range  of  conditions. 
Webb  also  clearly  demonstrated  that  the  range  of  pH  favoring  germina- 
tion is  influenced  by  temperature  and  by  the  constituents  of  the  medium. 

All  the  species  of  Myxomycetes  studied  by  Smart  (1937)  germinated 
within  a  pH  range  of  4.0  to  8.0.  Spores  of  Fuligo  septica  germinated 
from  pH  2.0  to  10.0.  Optimum  for  all  species  ranged  from  4.5  to  7.0, 
with  some  germinating  better  near  4.5  and  others  near  7.0.  The  spores 
of  Urocystis  occulta  germinated  between  pH  5.0  and  8.9,  with  the  optimum 
at  6.8  (Ling,  1940).  This  optimum  is  higher  than  those  for  most  fungi. 
Kauffman  (1934)  found  the  range  for  spore  germination  of  several  species 
of  Basidiomycetes  (Agaricaceae  and  Nidulariaceae)  to  be  pH  5.0  to  8.5 
with  the  optimum  near  7.5. 

It  is  interesting  that  Doran  (1922)  in  his  review  of  spore  germination 
makes  no  mention  of  acidity  as  a  factor.  It  would  appear  that  acidity 
is  of  more  or  less  importance  as  a  modifying  factor,  even  though  it  is 
seldom  a  limiting  factor  for  spore  germination.  This  may  explain,  at 
least  in  part,  the  fact  that  we  often  find  abundant  ungerminated  spores 
in  fruiting  liquid  cultures.  Some  fungi  sporulate  only  in  neutral  or 
alkaline  media,  which,  in  general,  are  not  favorable  to  spore  germination. 

NUTRIENTS  AND  STIMULANTS 

The  constituents  of  the  substrate  are  known  to  influence  spore  germina- 
tion of  some  species  of  fungi.  Some  species  germinate  well  in  distilled 
or  tap  water,  while  others  require  certain  special  nutrients  such  as  sugar, 
salts,  or  even  a  particular  nitrogen  source.  No  one  medium  has  been 
found  which  will  allow  good  germination  of  all  fungi,  although  certain 
natural  media,  such  as  beet  or  bean  decoction  and  soil  infusion,  seem  to 
favor  germination  in  a  large  number  of  fungi.  When  such  media  con- 
taining natural  products  are  used,  it  is  difficult  to  determine  whether  the 
higher  percentage  of  spore  germination  is  due  to  the  nutrients  or  to  some 
stimulant  which  is  not  used  in  the  metabolism  of  the  fungus. 

Duggar  w^as  one  of  the  foremost  American  workers  interested  in  spore 
germination  as  a  primary  subject  of  experimentation.  Prior  to  his  work, 
most  of  the  study  on  spore  germination  was  only  incidental  to  other 
problems.  Duggar  (1901)  demonstrated  that  species  differ  in  their 
nutrient  requirements  for  germination  by  placing  spores  in  water,  bean 
decoction,  nutrient-salt  solution,  and  cane-sugar  solution.  A  portion  of 
his  data  showing  the  percentage  of  germination  after  15  hr.  is  given  in 
Table  59. 

Among  some  of  the  compounds  Duggar  found  to  influence  sporulation 
of  Aspergillus  flavus  and  A.  niger  were  varying  amounts  of  peptone, 
ammonium  nitrate,  and  magnesium  sulfate.     Ammonium  nitrate  at  a 


362 


PHWSIOLOGY  OF  THE  FUNGI 


particular  concentration  gave  abundant  germination  of  A .  flavus  but  had 
no  effect  upon  ^4.  niger. 

Brefeld  (1905)  was  perhaps  the  first  to  observe  the  germination  of  the 
spores  of  various  smuts  in  culture.  He  noted  that  the  spores  germinated 
poorly  or  not  at  all  in  water,  while  excellent  germination  occurred  in 
nutrient  solutions  (probably  dung  infusion).  Brefeld  expressed  surprise 
at  the  vigorous  saprophytic  development  which  followed,  especially  since 
the  species  had  previously  been  known  only  as  obligate  parasites. 

More  recently  it  was  noted  that  pretreatment  with  dung  infusion 
markedly  stimulated  germination  of  spores  of  Ustilago  striiformis  (Cheo 

Table  59.     Percentage  of  Spore  Germination  after  15  Hours 
(Duggar,  Botan.  Gaz.  31,  1901.) 


Spores  of 


Aspergillus  niger 

Penicillium  glaucum 

Monilia  fructigena 

Mucor  spinosus 

Phycomyces  nitens 

Coprinus  jimetarius 

C.  comatus 

C.  micaceus 

Uromyces  caryophyllinus 


Water 


0 
0 
75 
0 
0 
0 
0 
0 
100 


Bean 
decoction 


100 

100 

100 

100 

100 

5-10 

0 

100 

75 


Xutrient-salt 
solution 


100 

100 

100 

100 

100 

0 

0 

0 


Sucrose 
solution 


75 
1 

100 
1 

2-10 
0 
0 
0 

100 


and  Leach,  1950).  Untreated  spores  in  distilled  water  germinated  only 
after  5  to  8  days,  and  the  total  germination  was  less  than  1  per  cent. 
Spores  soaked  in  a  concentrated  horse-dung  infusion  for  15  days  or  more, 
then  placed  in  distilled  water,  germinated  within  5  hr.,  with  a  total 
germination  of  50  per  cent  or  higher.  The  exposure  to  the  dung  infusion 
is  believed  to  increase  the  permeability  of  the  spore  wall,  allowing  the 
more  rapid  absorption  of  water.  It  might  also  be  pointed  out  that  the 
dung  infusion  evidently  contains  substances  which  prevent  spore  germina- 
tion until  highly  diluted  or  removed  entirely. 

Although  the  spores  of  the  Myxomycetes  germinate  in  distilled  water, 
the  percentage  may  be  greater  in  weak  decoctions  of  the  natural  sub- 
strate, such  as  rooting  wood,  bark,  leaves,  or  humus  (Smart,  1937). 
Similarly,  the  conidia  of  PhyUosticia  solitaria  germinate  more  profusely 
in  apple-bark  decoction  and  potato-dextrose  broth  than  in  distilled  water 
(Burgert,  1934).  While  it  is  possible  that  increased  spore  germination 
is  due  primarily  to  some  stimulating  substance,  it  seems  likely  that  certain 
nutrients  are  also  involved. 

The  conidia  of  Glomerella  cingulata  apparently  have  special  nutritional 
requirements  for  germination.     There  was  little  or  no  germination  in 


SPORE  GERMINATION 


363 


distilled  water  and  in  dextrose  solution  lacking  minerals  (Lin,  1945). 
From  his  experiments  involving  various  inorganic  compounds,  Lin 
concluded  that  carbon,  magnesium,  nitrogen,  and  phosphorus,  are 
required  (Table  60).  The  need  for  sulfur  was  not  so  evident  as  that  for 
the  other  elements,  and  sulfur  was  not  essential.  The  minimum  require- 
ments of  nitrogen  and  phosphorus  were  calculated  to  be  of  the  order  of 
10"'  Mg  per  spore.  No  evidence  was  found  that  an  external  supply 
of  any  organic  substance,  other  than  sugar,  is  necessary  for  spore 
germination. 

Table  60.     The  Essentiality  of  Various  Ions  for  the  Germination  of  the 

CoNiDiA  OF  GlomereUa  cingulata 
(Lin,  Am.  Jour.  Botany  32,  1945.) 


Chemical  substance  applied* 

Element 
lacking 

Germination, 

% 

None  (redistilled  water) 

dnrosR                                              

Carbon  and  minerals 

Minerals 

None 

None 

Nitrogen 

Potassium 

Phosphorus 

Sulfur 

Magnesium 

Carbon 

0.0 
0.0 

Glucose,  KNO3,  KH2PO4,  MgS04 

Glucose,  NH4CI,  KH2PO4,  MgS04 

Glucose,  KCl,  KH.,P04,  MgS04 

Glucose,  NaNOs,  NaH,P04,  .MgS04 

Glucose,  KNO3,  KCl,  MgS04 

Glucose,  KNO3,  KH2PO4,  MgCh 

Glucose,  KNO3,  KH.POj,  ^aSO, 

KXO,,  KH2PO4,  MgS04 

80.4 
92.8 

3.9 
84.1 

1.5 
79.3 

0.9 

0.7 

*  In  all  cases,  the  concentration  of  glucose  is  0.01  per  cent,  that  of  each  of  the  mineral  salts    1.0   milli- 
mole. 

The  constituents  of  the  medium  may  modify  the  effects  of  pH  on  spore 
germination.  This  is  illustrated  in  Fig.  75  by  the  germination  of  Lenzites 
saepiaria  on  2  per  cent  bacto-peptone,  in  sucrose-nitrate  (Czapek's) 
solution,  and  in  beet  decoction  (Webb,  1921). 

Emerson  (1948)  showed  that  D-xylose  as  a  carbon  source  gave  a  high 
percentage  of  germination  of  ascospores  of  Neurospora  crassa  without 
heat  treatment.  Xylose  was  more  effective  when  autoclaved  than  when 
filtered.  This  was  believed  to  be  due  to  the  slight  conversion  to  furfural, 
which  was  also  shown  to  be  active  in  increasing  spore  germination. 

From  this  brief  discussion  it  is  evident  that  little  is  known  about  the 
effects  of  nutrition  upon  spore  germination.  This  is  no  doubt  due,  in 
part,  to  the  lack  of  planned  experimental  work  along  this  line.  Many 
of  the  favorable  effects  of  natural  products  may  in  fact  be  due  to  the 
presence  of  stimulants  rather  than  to  the  nutrients.  At  the  present  time 
we  have  no  conclusive  evidence  that  spores  require  an  external  source  of 
vitamins  for  germination.     In  the  light  of  the  recent  discovery  of  Ryan 


364 


PHYSIOLOGY  OF  THE  FUNGI 


r 

A 

1 

1 

\ 

\ 

\ 

■^V 

■s       \ 

V^       ■ 

J 

\ 

1 

';  / 
•  / 

c' 

\ 
^ 

\ 

I 

J/ 

/ 

X 

V, 

\ 

(1948)  that  the  amino  acids  leucine,  lysine,  and  proline  favored  spore 
germination  in  mutants  of  Neurospora  deficient  for  those  amino  acids,  it 
also  seems  likely  that  spore  germination  in  certain  vitamin-deficient  fungi 
may  be  aided  by  the  addition  of  the  vitamins  in  question.  A  careful 
study  of  the  effects  of  vitamins  is  needed. 

The  spores  of  some  fungi,  such  as  Botrytis  cinerca,  germinate  much 
better  Avhen  in  contact  with  plant  tissue  than  in  distilled  water  (Brown, 
1922).  It  was  concluded  that  certain  substances  diffuse  out  of  the  host 
plant  into  the  infection  drop  containing  the  spores  and  stimulate  germina- 
tion and  infection.     Leach  (1923)  believes  that  a  similar  situation  may 

100 

80 

c 
,o 

o 

■|  60 

a> 
o> 

§  40 
I 
20 


»  23456789 

pH  of  medium 

Fig.  75.  The  effect  of  the  pH  and  kind  of  medium  on  the  percentage  of  germination 
of  spores  of  Lenzitcs  sacpiaria  at  20  to  23°C.  A,  in  sugar-beet  decoction;  B,  in  2  per 
cent  bacto-peptone  sohition;  C,  in  Czapek's  full  nutrient  solution.  (Redrawn  from 
Webb,  Ann.  Missouri  Botan.  Garden  8:  325-327,  1921.) 

exist  with  Colletotrichum  Undemuthianum.  The  spores  of  this  fungus 
germinated  poorly  in  distilled  water  alone,  but  distilled  water  plus  a 
piece  of  fresh  bean  tissue  gave  a  high  percentage  of  germination.  Fresh 
bean  juice  was  equally  effective,  but  boiled  bean  decoction  did  not  stimu- 
late germination.  However,  green-bean  agar  made  from  a  similar  decoc- 
tion gave  excellent  germination,  as  did  potato-dextrose  agar.  These 
results  led  Leach  to  conclude  that  two  distinct  stimulating  factors  may 
be  involved.     A  portion  of  Leach's  data  is  summarized  in  Table  61. 

Some  know^n  stimulants  may  eliminate  the  need  for  certain  factors 
ordinarily  supplied  by  natural  media  for  the  germination  of  spores  of 
Phycomijces  (Robbins  et  al.,  1942).  Germination  of  spores  was  about 
12  per  cent  or  less  on  mineral-dextrose  agar  with  thiamine.  The  addition 
of  an  extract  of  potatoes,  or  of  other  natural  products,  of  hypoxanthine, 
acetate,  or  some  other  organic  acids  increased  germination  to  nearly  100 
per  cent.  Treatment  of  spores  with  aqueous  pyridine  had  the  same 
favorable  effect.  These  authors  believe  that  certain  factors  (called  Z 
factors)   are  essential  in  spore  germination.     One  of  these   (factor  Zi) 


SPORE  GERMINATION  365 

has  been  identified  as  hypoxanthine,  while  the  identity  of  factor  Z2  i» 
still  unknown.  An  explanation  of  the  effects  of  these  stimuli  is  given  by 
these  authors: 

The  dormant  spores  are  considered  to  lack  sufficient  available  Z  factors  for 
germination.  The  extracts  of  natural  products  or  the  Z  factors  furnished  in  the 
medium  supply  this  deficiency,  which  may  also  be  met  by  treatment  with  heat, 
cold,  acetate  or  pyridine.  These  treatments  are  thought  to  change  the  Z  factors 
in  the  spores  from  an  unavailable  to  an  available  form. 

The  effects  of  certain  gases  and  volatile  compounds  upon  germination 
have  also  been  demonstrated.  It  has  been  observed  that  spores  of  num- 
erous fungi  germinate  better  in  a  container  in  which  some  living  plant 
part  is  also  present.  This  was  demonstrated  for  Basisporium  gallarum 
by  Durrell  (1925),  who  also  found  that  the  introduction  of  carbon  dioxide 

T.^ELE  61.     The  Effect  of  Various  Media  and  Plant  Tissues  on  Spore  Germina- 
tion OF  Colletotrichuni  lindemuthianum 
(Leach,  Minn.  Agr.  Expt.  Sta.  Bull.  14,  1923.) 

Medium  Germination,  % 

Distilled  water 3-6 

Sucrose-nitrate  (Czapek's)  solution 5-11 

Sucrose-nitrate  (Czapek's)  solution  plus  bean  decoction 10 

Bean  decoction 8 

Distilled  water  plus  fresh  bean  tissue 83-95 

Distilled  water  plus  sunflower  tissue 5 

Distilled  water  plus  wheat  tissue 12 

Distilled  water  plus  corn  tissue 10 

Distilled  water  plus  tomato  tissue 2 

Sucrose-nitrate  (Czapek's)  solution  plus  bean  tissue 95 

Green-bean  agar 97 

Potato-glucose  agar 98 

into  the  container  enclosing  the  spores  gave  the  same  increase  in  germina- 
tion. The  same  effect  was  demonstrated  for  Ustilago  zeae  (Platz  et  at., 
1927).  An  atmosphere  containing  15  per  cent  carbon  dioxide  was  found 
to  be  optimum  for  spore  germination.  Such  a  condition  gave  a  pH  of 
the  medivmi  from  4.9  to  5.6.  These  authors  conclude  that  the  stimulating 
effect  is  apparently  due  to  "a  definite  action  of  carbonic  acid."  Is  it 
possible  that  this  is  an  example  of  heterotrophic  utilization  of  carbon 
dioxide? 

While  the  release  of  carbon  dioxide  into  the  atmosphere  by  various 
living  plant  parts  may  explain  the  stimulation  of  spore  germination  in 
many  cases,  the  presence  of  carbon  dioxide  alone  will  not  explain  certain 
results  obtained  by  some  workers.  For  instance,  spore  germination  of 
Botrijtis  cinerea  was  stimulated  by  the  presence  of  living  tissues  of  apples 
or  leaves  of  Ruta  or  Eucalyptus  in  the  same  container,  while  tissues  of 
potato  tuber  and   onion  scales  inhibited  germination    (Brown,    1922). 


366  PHYSIOLOGY  OF  THE  FUNGI 

Distillates  of  these  leaves  increased  germination  four  to  ten  times.  Ethyl 
acetate  likewise  gave  similar  results.  The  possibility  of  specific  activity 
was  suggested  by  the  fact  that  apple  tissues  distinctly  stimulated  germina- 
tion of  B.  cinerea  spores,  while  they  inhibited  germination  of  spores  of 
Colletotrichum  lindemuthianum.  The  stimulation  was  greater  with  old 
spores. 

Presoaking  and  the  subsequent  addition  of  a  stimulating  volatile  agent 
gave  optimum  germination  of  Urocystis  tritici  spores  (Noble,  1923). 
The  expressed  sap  of  wheat  placed  in  the  same  container  with  germinating 
spores,  but  in  separate  dishes,  proved  to  be  a  good  stimulating  agent. 
Uninjured  seedlings  of  certain  nonsusceptible  hosts  likewise  stimulated 
spore  germination.  Benzaldehyde,  salicylaldehyde,  butyric  acid,  and 
acetone  in  certain  concentrations  stimulated  germination  of  presoaked 
spores.  Noble  believed  that  presoaking  increased  permeability  of  the 
spore  and  allowed  the  more  rapid  intake  of  the  stimulatory  volatile  sub- 
stance, which  increased  the  permeability  of  the  protoplasmic  membrane 
by  changing  its  physical  condition. 

Likewise,  a  solution  of  benzaldehyde  (3/2,000,000)  stimulated  germina- 
tion of  Urocystis  occulta  spores,  which  germinated  very  poorly  in  water 
(Ling,  1940).  Ethyl  alcohol  stimulated  spore  germination  in  Aspergillus 
flavus;  methyl  alcohol  was  slower  and  less  effective  (Duggar,  1901).  A. 
niger  was  stimulated  by  oxalic  acid,  whereas  .4.  fiavus  was  not.  It  is 
understood  that  the  stimulatory  power  of  these  chemicals  depends  upon 
the  concentration. 

An  interesting  situation  exists  in  the  germination  response  of  some 
spores  to  the  presence  of  other  fungi,  or  even  to  the  medium  in  which 
other  fungi  have  grown.  The  few  experiments  conducted  along  this  line 
suggest  that  the  constant  association  with  other  organisms  may  be  highly 
beneficial  to  spore  germination  as  well  as  subsequent  growth  of  some  fungi 
in  nature. 

The  germination  of  a  number  of  species  of  Myxomycetes  was  increased 
by  the  addition  of  the  filtrate  of  a  medium  in  which  spores  had  previously 
been  germinated  (Smart,  1937).  Smart  calls  the  stimulatory  factor  an 
"autocatalytic  agent."  A  portion  of  Smart's  data  is  presented  in  Table 
62. 

Fries  (1941,  1943)  obtained  almost  phenomenal  results  with  spores  of  a 
number  of  Hymenomycetes,  which  previously  had  germinated  poorly 
or  not  at  all,  by  sowing  the  spores  on  malt  agar  with  living  cultures  of 
Torulopsis  sanguinea.  Spores  of  ten  species  of  Tricholoma,  which  ger- 
minated only  with  difficulty  without  the  yeast,  were  found  to  germinate 
readily  in  its  presence.  One  species  of  Tricholoma  gave  only  negative 
results.  In  Amanita  mappa,  A.  porphyria,  and  A.  rubescens  germination 
occurred  only  when  Torulopsis  was  present.     Germination  of  two  other 


SPORE  GERMINATION 


367 


species  of  Amanita  was  considerably  improved  by  the  presence  of  the 
yeast.  None  of  the  seven  species  of  Boletus  germinated  on  malt  agar 
without  the  yeast.  On  the  same  medium  and  in  the  presence  of  Torulopsis 
sanguinca,  germination  was  obtained  with  spores  of  B.  bovinus,  B.  elegans, 
B.  flavidus,  B.  granulatus,  B.  luteus,  B.  variegatus,  and  B.  viscidus.  Some 
germination  of  Boletus  spores  was  also  obtained  in  the  presence  of  living 
colonies  of  certain  other  fungi,  but  none  was  so  effective  as  Torulopsis. 
Spores  of  certain  other  fungi  {Hijdnum  repandum,  H.  imhricatum,  Craterel- 
lus  lutescens,  Lycoperdon  umhrinum,  L.  echinatum.,  L.  nigrescens,  L.  pra- 
tense,  L.  pyriforme,  and  Scleroderma  aurantium)   germinated  in  Fries's 

Table  62.     Germination  of  Single  Myxomycete  Spores 
(Smart,  Aryi.  Jour.  Botany  24,  1937.) 


' 

Number  of  spores  germinating 

Species 

Lot  1  (10  spores) 
(fresh  medium) 

Lot  2  (10  spores) 

(previous  germination 

medium) 

Vulino  sevtica, 

9  after  3  hr. 
6  after  3  days 

3  after  2  days 

4  after  1  day 
0  in  2  weeks 
0  in  2  weeks 
3  after  2  days 
6  after  6  days 
2  after  18  days 
9  after  7  days 

10  in  45  min. 

Physarum  polycephahini 

Stemonitis  fusca 

S  ciTifero, 

8  in  15  hr. 

9  in  1  day 
8  in  8  hr. 

Enteridium  rozeanum 

Reticularia  lycoperdon 

Lycogala  epidendrum 

Arcyria  denudata 

Dictydium  cancellaium 

Physarum   cinereum 

10  in  30  min. 
8  in  15  min. 

8  in  3  hr. 

G  in  6  days 
2  after  18  days 

9  in  6  days 

experiments  only  in  the  presence  of  T.  sanguinea.  He  also  tested  the 
effects  of  mycelial  extracts  on  spore  germination  and  found  that  extracts 
of  certain  species  of  Boletus  stimulated  germination  of  spores  of  the  same 
species.  Many  of  the  fungi  studied  by  Fries  are  believed  to  be  mycorhizal 
and  may  require  the  presence  of  a  special  set  of  conditions,  perhaps  the 
roots  of  certain  plants  (or  conditions  which  simulate  their  presence), 
before  germination  will  occur. 

The  time  required  for  a  spore  to  germinate  after  being  subjected  to 
favorable  conditions  is  a  reflection  of  the  interaction  and  relative  impor- 
tance of  all  the  various  influencing  factors.  The  nearer  all  these  factors 
are  to  the  optimum,  the  shorter  will  be  the  time  required  for  germination. 
Time  is  an  important  factor  for  the  subsequent  infection  of  the  host.  In 
nature  the  near-optimum  environmental  conditions,  principally  tempera- 
ture and  moisture,  may  persist  for  bvit  a  short  time,  and  a  change  in  but 
one  of  these  factors  may  inhibit  spore  germination. 


3G8  PHYSIOLOGY  OF  THE  FUNGI 

LONGEVITY  OF  SPORES 

The  length  of  Hfe  of  spores  is  usually  measured  by  their  ability  to 
germinate  after  various  periods  of  time.  It  is  affected  by  environmental 
conditions,  principally  temperature  and  moisture.  The  greatest  period 
of  longevity  reported  for  fungus  spores  appears  to  be  among  the  Myx- 
omycetes.  Smith  (1929)  succeeded  in  germinating  spores  from  herbarium 
specimens  of  Myxomycetes  5  to  32  years  after  they  were  collected.  A 
few  of  the  common  species  whose  spores  germinated  after  approximately 
30  years  are  Physarum  cinereum,  Fuligo  septica,  Hemiirichia  clavata,  and 
Stemonitis  ferruginea.  Smut  spores  also  have  a  long  period  of  viability 
(Lowther,  1950).  Spores  of  Aspergillus  orijzae  germinated  after  22  years 
in  a  sealed  tube  at  room  temperature  (McCrea,  1923). 

In  contrast  to  long  periods  of  longevity,  some  fungus  spores  die  very 
soon  after  they  are  liberated.  The  sporidia  of  Cronartium  rihicola  lived 
less  than  10  min.  at  room  temperature  with  a  humidity  of  90  per  cent 
(Spaulding,  cited  by  Doran,  1922).  Sporidia  of  Gymno sporangium 
juniperi-virginianae  lived  no  longer  than  6  days  in  dry  air.  Eight  weeks 
is  reported  as  the  maximum  longevity  of  aeciospores  of  C.  rihicola,  with 
only  5  per  cent  germination  after  7  weeks.  In  general,  aeciospores  of  the 
rust  fungi  remain  viable  about  50  per  cent  longer  than  the  urediospores, 
whose  average  longevity  ranged  from  30  to  60  days  (Doran,  1922). 

Other  factors  have  been  reported  to  influence  longevity  of  spores. 
Ascospores  of  Endothia  parasitica  remained  viable  for  a  year  when  dried 
in  the  bark,  but  when  removed  from  the  bark,  they  lost  the  ability  to 
germinate  within  5  months  (Anderson  and  Rankin,  1914).  Similarly, 
conidia  in  dry  spore  horns  retained  viability  for  at  least  a  year,  but  when 
placed  in  water,  separated,  and  then  dried,  the  time  was  less  than  1 
month.  It  seems  likely  that  one  of  the  functions  of  the  gelatinous  matrix 
of  the  conidia  of  certain  fungi,  such  as  Gloeosporium,  Colletotrichum,  and 
Cytospora,  is  to  increase  the  longevity  of  the  spores  through  its  water- 
holding  capacity. 

Light  is  apparently  only  of  minor  importance  as  a  factor  influencing 
longevity.  No  doubt  ultraviolet  light  in  nature  plays  an  important  part 
in  reducing  the  period  of  viability  and  even  in  killing  many  of  the  hyaline 
spores.  Spores  having  dark  walls  are  protected  somewhat  against  the 
penetration  of  the  ultraviolet  rays. 

SUMMARY 

Spore  germination  represents  a  change  from  an  inactive  to  an  active 
phase  in  the  life  cycle  of  a  fungus.  Since  it  involves  the  first  stages  of 
growth,  it  is  reasonable  to  expect  that  many  of  the  factors  which  influence 
vegetative  growth  also  affect  spore  germination.  On  the  other  hand, 
the  spore,  being  a  resting  cell,  may  contain  stored  materials  not  usually 


SPORE  GERMINATION  369 

present  in  appreciable  quantities  in  vegetative  cells.  Since  the  metabolic 
activity  of  a  resting  spore  is  at  a  minimum  in  contrast  with  that  of  actively- 
growing  vegetative  cells,  the  internal  responses  to  the  environmental 
factors  may  be  quite  different. 

The  variability  of  the  needs  of  spores  of  different  fungi  for  germination 
is  adequately  illustrated  in  the  literature.  Certain  general  conditions 
are  essential  for  all  spores,  while  some  require  a  special  set  of  conditions. 
Water  is  essential  to  activate  certain  enzyme  systems,  to  initiate  other 
internal  chemical  changes,  and  to  increase  the  volume  of  the  germinating 
spore.  WTien  the  temperature  is  near  the  optimum,  the  enzymatic 
activity  and  the  rate  of  spore  germination  are  increased.  The  supply  of 
oxygen  must  be  adequate  to  meet  the  demands  of  the  greatly  increased 
rate  of  respiration.  The  acidity  of  the  substrate  must  be  favorable. 
Variability  in  the  period  of  viability  of  spores  is  striking,  but  longevity 
is  greatly  influenced  by  the  environment.  Much  information  is  yet  to  be 
gained  regarding  the  longevity  of  spores,  particularly  of  the  plant 
pathogens. 

Certain  special  conditions  are  required  for  germination  of  some  spores. 
These  may  act  as  a  stimulant  in  breaking  dormancy  or  may  supply  needed 
nutrients.  The  effects  of  other  living  organisms,  or  even  of  the  substrate 
upon  which  they  have  grown,  are  of  particular  interest,  for  such  associa- 
tion is  the  usual  condition  under  which  germination  occurs  in  nature 
One  might  suppose  that  the  secretions  of  certain  plants  would  exert  a 
selective  action  on  spore  germination  and  affect  the  pathogenicity  of 
certain  fungi,  but  evidence  on  this  point  is  lacking. 

The  production  of  short  germ  tubes  by  spores  of  some  species  of  Ery- 
siphe  in  an  absolutely  dry  atmosphere  is  unusual.  If  this  is  to  be  con- 
sidered as  true  germination,  it  must  represent  a  unique  method  among 
fungi.  The  Erysiphales,  however,  are  excellent  examples  of  fungi  whose 
spores  germinate  in  atmospheres  of  lower  relative  humidity  than  most 
fungi  can  endure. 

Under  the  changing  conditions  of  nature,  the  period  of  time  during 
which  a  factor  is  active  is  of  utmost  importance.  Germination  is  the 
result  of  the  action  of  all  the  influencing  factors  operating  at  the  same 
time.  Most  of  these  factors  vary  in  intensity  or  concentration,  so  that 
the  combined  optima  of  all  factors  are  seldom,  if  ever,  reached  at  any 
given  time  in  nature.  As  a  result,  an  extremely  low  percentage  of  the 
spores  formed  by  a  fungus  ever  germinate,  while  still  fewer  give  rise  to 
extensive  mycelium. 

REFERENCES 

Anderson,  P.  J.,  and  W.  H.  Rankin:  Endothia  canker  of  chestnut,  Cornell  Univ. 

Agr.  Expt.  Sta.  Bull.  347,  1914. 
Bonner,  J.  T. :  A  study  of  the  temperature  and  humidity  requirements  of  Aspergiltus 

niger,  Mycologia  40:  728-738,  1948- 


370  PHYSIOLOGY  OF  THE  FUNGI 

Brefeld,  O.:  Die  Brandpilze  IV.  in  Untersuchungen  aus  dem  Gesammtgebiete  der 

Mykologic,  Heft  13,  Muenster,  Heinrich  Schoningh,  1905. 
*Brodie,   H.  J.:  Further  investigations  on  the  mechanism  of  germination  of  the 

conidia  of  various  species  of  powdery  mildew  at  low  humidity,   Can.  Jour. 

Research  23:  198-211,  1945. 
Brodie,  n.  J.,  and  C.  C.  Neufeld:  The  development  and  structure  of  the  conidia  of 

Enjsiphe  polygoni  DC.   and  their  germination  at  low  humidity.   Can.  Jour. 

Research  20:  41-61,  1942. 
*Brown,  W. :  Studies  in  the  physiology  of  parasitism.     IX.  The  effect  on  the  germi- 
nation of  fungal  spores  of  volatile  substances  arising  from  plant  tissues,  Ann. 

Botany  36:  285-300,  1922. 
BuRGERT,   I.   A.:   Some  factors  influencing  germination  of  spores  of  Phyllosticta 

solitaria,  Phytopathology  24:  384-396,  1934. 
*Chbo,  p.  C,  and  J.  G.  Leach:  The  stimulating  effect  of  dung  infusion  on  the  germi- 
nation of  spores  of  Ustilago  striiformis,  Phytopathology  40 :  584-589,  1950. 
*Clayton,  C.  N.:  The  germination  of  fungous  spores  in  relation  to  controlled  humid- 
ity, Phytopathology  32:  921-934,  1942. 
Dodge,  B.   O.:   Methods  of  culture  and  morphology  of  the  archicarp  in  certain 

species  of  the  Ascobolaceae,  Bull.  Torrey  Boian.  Club  39:  139-197,  1912. 
DoRAN,  W.  L. :  Effect  of  external  and  internal  factors  on  the  germination  of  fungous 

spores,  Bull.  Torrey  Botan.  Club  49:  313-336,  1922. 
DuGGAR,   B.    M.:  Physiological  studies  with  special  reference  to  germination  of 

certain  fungous  spores,  Botan.  Gaz.  31:  38-66,  1901. 
DuRRELL,  L.  W. :  Basisporium  dry  rot  of  corn,  Iowa  Agr.  Expt.  Sta.  Bull.  84,  1925. 
Emerson,    M.   A.:   Chemical  activation   of  ascospore   germination   in   Xeurospora 

crassa,  Jour.  Bad.  55 :  327-330,  1948. 
Fries,   N.:   Ueber  die   Sporenkeimung  bei  einigen   Gasterromyceten  und  mykor- 

rhizabildenen  Hymenomyceten,  Arch.  Mikrobiol.  12:  266-284,  1941. 
Fries,   N.:   Untersuchungen  iiber   Sporenkeimung  und   Mycelentwicklung  boden- 

bewohnender  Hymenomj'ceten,  Symholae  Botan.  Upsaliensis  6:  1-81,  1943. 
Gardner,  M.  W.  :  Anthracnose  of  cucurbits,  U.S.  Dept.  Agr.  Bull.  727,  1918. 
GoDDARD,    D.    R. :    The    reversible    heat    activation    inducing    germination    and 

increased  respiration  in  the  ascospores  of  Neurospora  tetrasperma,  Jour.  Gen. 

Physiol.  19  :  45-60,  1935. 
GoDDARD,  D.  R.,   and  P.  E.  Smith:  Respiratory  block  in  the  dormant  spores  of 

Neurospora  tetrasperma,  Plant  Physiol.  13:  241-264,  1938. 
*GoTTLiEB,  D.:  The  physiology  of  spore  germination  in  fungi,  Botan.  Rev.  16  :  229-257, 

1950. 
Jones,  E.  S.  :  Influence  of  temperature,  moisture  and  oxygen  on  spore  germination 

of  Ustilago  avenae,  Jour.  Agr.  Research  24:  577-591,  1923. 
Kauffman,  F.  H.  O.  :  Studies  on  the  germination  of  the  spores  of  certain  Basidio- 

mycetes,  Botan.  Gaz.  96 :  282-297,  1934. 
Leach,  J.  G.:  The  parasitism  of  Colletotrichum  lindemuthianum,  Minn.  Agr.  Expt. 

Sta.  Bull.  14,  1923. 
*LiN,  C.  K.:  Nutrient  requirements  in  the  germination  of  the  conidia  of  Glomerella 

cingulata,  Am.  Jour.  Botany  32 :  296-298,  1945. 
Ling,  L.  :  Factors  affecting  spore  germination  and  growth  of  Urocystis  occulta  in 

culture.  Phytopathology  30:  579-591,  1940. 
Lowther,  C.  v.:  Chlamydospore  germination  in  physiologic  races  of  Tilletia  caries 

and  Tilletia  foetida,  Phytopathology  40 :  590-603,  1950. 
McCrea,  a.:  Spores  of  Aspergillus  oryzae  alive  after  22  years  stored  in  test  tube. 

Science  (N.S.)  58:  426,  1923. 


SPORE  GERMINATION  371 

Noble,  R.  J.:  Studios  on  the  parasitism  of  Urocystis  tritici  Koern,  the  organism 

causing  flag  smut  of  wheat,  Jour.  Ayr.  Research  27  :  451-489,  1924. 
Platz,  G.  a.,  L.  W.  Durrell,  and  M.  E.  Howe:  Effect  of  carbon  dioxide  upon  the 

germination  of  chlamydosporcs  of   Ustilago  zeae  (Beckm.)   Ung.,  Jour.  Agr. 

Research  34 :  137-147,  1927. 
RoBBiNS,   W.  J.,  V.  W.  Kavanagh,  and  F.  Kavanagh:  Growth  substances  and 

dormancy  of  spores  of  Phycomyces,  Botan.  Gaz.  104 :  224-242,  1942. 
Ryan,  F.  J.:  The  appHcation  of  Neurospora  to  bioassay,  Fed.  Proc.  3 :  365-369,  1946. 
*Ryan,  F.  J.:  The  germination  of  conidia  from  biochemical  mutants  oi  Neurospora, 

Am.  Jour.  Botany  35:  497-503,  1948. 
Smart,  R.  F.:  Influence  of  certain  external  factors  on  spore  germination  in  the 

Myxomycetes,  Am.  Jour.  Botany  24:  145-159,  1937. 
Smith,  E.  C:  The  longevity  of  Alyxomycete  spores,  Mycologia  21:  321-323,  1929. 
Uppal,  B,  N.:  Relation  of  oxygen  to  spore  germination  in  some  species  of  Perono- 

sporales,  Phytopathology  16 :  285-292,  1926. 
Walker,  J.  C.,  and  F.  L.  Wellman:  Relation  of  temperature  to  spore  germination 

and  growth  of  Urocystis  cepulae,  Jour.  Agr.  Research  32:  133-146,  1926. 
^Webb,  R.  W.:  Studies  in  the  physiology  of  fungi.     XV.  Germination  of  the  spores 

of   certain   fungi  in   relation   to   hydrogen-ion   concentration,   Ann.   Missouri 

Botan.  Gardens:  282-341,  1921. 
Wolf,  F.  A.,  and  F.  T.  Wolf:  The  Fungi,  Vol.  II,  John  Wiley  &  Sons,  Inc.,  New 

York,  1947. 
Yarwood,  C.  E.:  The  tolerance  of  Erysiphe  polygoni  and  certain  other  powdery 

mildews  to  low  humidity,  Phytopathology  26 :  845-859,  1936. 


CHAPTER  17 
THE  PHYSIOLOGY  OF  PARASITISM  AND  RESISTANCE 

A  discussion  of  parasitism  occupies  an  important  position  in  any 
treatise  on  the  physiology  of  fungi,  particularly  for  those  students  who  are 
interested  in  plant  diseases  or  the  fungi  which  cause  them.  This  phase 
of  study  offers  many  challenging  unsolved  problems.  Parasitism  involves 
primarily  two  living  organisms,  the  parasite,  whose  actions  are  offensive, 
and  the  host,  whose  reactions  are  defensive.  If  the  defenses  of  the  host 
plant,  either  before  or  after  penetration  by  the  parasite,  are  successful, 
the  plant  is  resistant;  if  not,  it  is  susceptible.  To  be  successful,  a  parasite 
must  find  the  nutritional  and  environmental  conditions  favorable  for  its 
development.  If  even  a  single  important  factor  is  unfavorable  to  the 
parasite,  the  fungus  may  fail  to  establish  a  parasitic  relationship  with  its 
proposed  host.  Such  factors  may  exert  their  influence  either  before  or 
after  penetration  by  the  fungus.  Environmental  factors  acting  before 
penetration  may  in  reality  bring  about  an  escape  from  a  disease  rather 
than  true  resistance  to  it. 

The  present  discussion  is  divided  into  three  main  parts:  (1)  penetra- 
tion; (2)  parasitism,  the  action  of  the  parasite  in  becoming  established  and 
obtaining  its  food ;  (3)  resistance  of  the  host  to  penetration  or  against  the 
parasite  after  penetration.  The  comprehensive  reviews  of  the  physiology 
of  the  host-parasite  relationship  given  by  Brown  (1936,  19-48)  should  be 
read  by  all  students.  Similar  reference  is  made  to  Arthur  et  al.  (1929), 
who  give  an  excellent  discussion  of  the  parasitic  relations  of  the  rusts, 
and  to  the  treatise  of  Gaumann  (1946,  1950)  on  the  principles  of  plant 

infection, 

PENETRATION 

A  parasite  may  gain  entrance  into  the  host  (1)  through  the  natural 
openings,  such  as  stomata  or  lenticels,  (2)  by  direct  penetration  through 
the  uninjured  epidermis,  or  (3)  through  wounds. 

Through  stomata.  Viable  spores  may  fall  upon  a  host  plant  and  pro- 
duce germ  tubes,  which  by  chance  grow  over  or  near  stomata.  The 
outer  walls  of  the  epidermal  cells  of  aerial  plant  parts  are  covered  with 
cutin,  which  is  somewhat  resistant  to  penetration  by  some  fungi.  The 
germ  tube  which  enters  through  a  stoma  may  then  be  favored  by  the  moist 
atmosphere  in  the  substomatal  cavity.  In  some  cases,  the  unspecialized 
hyphae  may  penetrate  the  host  cells;  in  other  fungi,  haustoria,  which 

372 


PARASITISM  AND  RESISTANCE  373 

arise  from  the  intercellular  mycelium,  penetrate  the  host  cells  and  absorb 
food.  Water  vapor  has  been  suggested  as  the  stimulus  which  causes  the 
germ  tube  to  turn  inward  and  enter  a  stoma.  This,  however,  cannot  be 
the  case  with  zoospores  which  are  immersed  in  water  and  which  have  been 
noted  to  cluster  around  stomata.  The  fungi  which  normally  enter  the 
host  plant  through  stomata  include  the  cereal  rusts  (aeciospore  and 
urediospore  stages),  Cercospora  heticola,  Phytophthora  infestans  (zoospore 
stage),  the  Peronosporales,  Albugo  Candida,  and  others. 

The  cereal  rusts  have  received  a  great  deal  of  attention  in  resistance 
studies.  It  has  been  reported  (Hart,  1929)  that  Puccinia  graminis 
apparentl}^  requires  the  open  stomata  of  wheat  plant  for  penetration. 
On  the  other  hand,  Caldwell  and  Stone  (1936)  have  shown  that  the  germ 
tubes  of  Puccinia  triticina  are  able  to  force  their  way  between  the  guard 
cells  of  closed  stomata  of  wheat  leaves.  A  germ  tube  from  a  urediospore 
may  start  to  enter  an  open  stoma,  but  as  it  forms  an  appressorium,  the 
stoma  closes.  Further  penetration  is  accomplished  between  the  guard 
cells  by  a  slender  hypha.  Allen  (192G)  believes  that  the  appressorium 
probably  secretes  some  toxin  which  harms  or  even  kills  the  guard  cells, 
causing  the  stoma  to  close.  Caldwell  and  Stone,  however,  do  not  believe 
that  this  injury  to  the  guard  cells  is  necessary  for  entry  of  germ  tubes. 
The  appressorium  seems  to  function  as  a  special  organ  to  apply  the  pres- 
sure needed  for  the  forced  entry  between  the  closed  guard  cells. 

Penetration  through  lenticels  more  often  occurs  in  the  underground 
parts  of  the  host  which  are  in  a  more  or  less  moist  situation.  Potato 
tubers  may  become  infected  by  Actinomyces  scabies  and  by  germ  tubes 
from  sporangia  of  Phytophthora  infestans,  chiefly  through  the  lenticels. 

Direct  penetration.  A  large  number  of  fungi  are  capable  of  penetrating 
the  unbroken  epidermis  of  a  plant,  directly  through  the  cutinized  outer 
walls.  The  spore  may  germinate  on  the  surface  of  the  plant  in  a  drop  of 
water.  The  germ  tube  grows  over  the  epidermis  and  by  some  stimulus 
is  caused  to  turn  inward  and  penetrate  the  cell.  Brown  (1922)  demon- 
strated that  there  is  a  certain  amount  of  exosmosis  of  materials  from  host 
tissue  into  a  drop  of  liquid  on  the  surface.  In  some  cases  this  may  lead 
to  a  chemotropic  response  by  the  fungus.  However,  in  most  cases  the 
stimulus  of  contact  is  believed  to  initiate  appressorial  formation  and 
penetration.  The  formation  of  appressoria  is  common  among  many  fungi 
when  the  germ  tubes  come  in  contact  with  the  epidermal  cells.  The 
fact  that  the  appressoria  are  often  formed  on  a  glass  slide  is  further 
evidence  that  their  formation  is  in  response  to  contact  with  a  solid  sur- 
face. Appressoria  are  usually  bulb-like  or  disk-like  in  shape  and  are 
believed  to  serve  as  an  adhesive  disk  against  which  the  slender  infection 
hypha  may  push  in  penetrating  the  cell  wall.  Brown  (1915,  1922) 
presents  evidence  that  the  host  cells  are  not  killed  by  Botrytis  cinerea, 


374  PHYSIOLOGY  OF  THE  FUNGI 

Sclcrotinia  sclerotiorum,  and  Collctoirichum  lindemuthianum  until  after  the 
fungus  penetrates  the  cuticle  of  the  host.  In  other  words,  there  is  little 
or  no  diffusion  of  the  toxic  materials  through  the  cuticle.  Direct  penetra- 
tion through  cutinized  walls  is  believed  to  be  entirely  by  mechanical 
pressure,  since  no  cutin-dissolving  enzyme  has  been  demonstrated  in  the 
fungi. 

The  rhizomorphs  of  Armillaria  mellea  usually  gain  entrance  directly 
through  the  sound  cork  layer  of  comparatively  old  roots  (Thomas,  1934). 
Penetration  is  believed  to  be  accomplished  partly  by  mechanical  pressure 
and  partly  by  chemical  means.  There  is  evidence  that  a  suberin-dis- 
solving  enzyme  aids  in  the  destruction  of  some  of  the  cork  cells.  Some 
fungi  may  enter  the  same  host  by  more  than  one  method.  Fusarium 
lint  may  enter  through  young  epidermal  cells  of  the  root,  root  hairs, 
stomata  of  seedlings,  and  perhaps  through  wounds. 

The  penetration  of  noncutinized  cell  walls  may  be  either  by  mechanical 
pressure  or  by  the  dissolving  action  of  enzymes  secreted  by  the  fungus. 
Hawkins  and  Harvey  (1919)  concluded  that  the  hyphae  of  Pythium 
debaryanum  penetrated  the  cell  walls  of  susceptible  potato  tubers  by 
mechanical  pressure,  and  that  the  resistant  varieties  in  general  showed 
greater  resistance  to  mechanical  puncture.  They  found  no  evidence  of 
cellulases  which  might  aid  in  penetration  by  dissolving  the  cellulose  cell 
wall.  Using  cane  sugar  as  the  plasmolyzing  solution,  they  found  that 
the  hyphae  of  P.  debaryanum  were  capable  of  exerting  as  high  as  54  atm. 
osmotic  pressure.  These  hyphae  would  have  a  strong  tendency  to  absorb 
water,  and  as  a  result  greater  internal  pressure  would  be  exerted  against 
the  hyphal  wall.  Apparently  the  hyphal  wall  is  capable  of  withstanding 
this  pressure  at  all  points  except  its  tip,  where  growth  occurs.  The 
pressure  exerted  by  the  growing  tip  is  believed  to  be  sufficient  to  cause 
penetration  of  the  host  cell  wall.  By  direct  microscopic  examination 
Hawkins  and  Harvey  observed  that,  just  after  the  hyphal  tip  came  in 
contact  with  the  host  cell  wall,  it  formed  a  swelling,  back  of  which  a  bend 
developed.     This  was  followed  by  penetration  of  the  wall  by  a  small  tube. 

Penetration  through  noncutinized  cell  walls  by  chemical  means  has 
been  described  for  Spongospora  suhterranea  by  Kunkel  (1915).  It  seems 
likely  that  other  nonfilamentous  fungi  penetrate  cell  walls  in  the  same 
way.  Likewise,  wood  rot  fungi  penetrate  the  cellulose  and  lignified  cell 
walls  by  enzymatic  action,  as  evidenced  by  the  boreholes  in  decaying 
wood.  It  may  be  significant  that  the  hyphal  walls  of  Pythium,  as  well 
as  of  other  Oomycetes,  contain  cellulose,  while  the  hyphal  walls  of  other 
fungi  are  composed  principally  of  chitin,  which  would  not  be  acted  upon 
by  cellulases. 

It  must  be  emphasized  that  penetration  of  the  host  in  itself  does  not 
necessarily  lead  to  the  establishment  of  the  fungus  in  the  host  and  the 


PARASITISM  AND  RESISTANCE  375 

production  of  a  disease.  In  some  cases  it  is  known  that  a  fungus  may 
enter  resistant  or  immune  plants,  as  well  as  susceptible  ones,  but  find  the 
conditions  unfavorable  for  its  establishment  and  further  development. 
Through  wounds.  A  number  of  fungi  apparently  are  unable  to  pene- 
trate a  healthy  plant  except  through  wounds.  These  may  be  insect 
wounds,  broken  branches  of  trees,  broken  roots,  etc.  In  addition,  some 
fungi  which  are  capable  of  entering  the  host  by  other  means  may  also 
penetrate  through  wounds.  Phymatotrichum  omnivorum,  the  cause  of 
numerous  root  rots,  commonly  enters  roots  through  wounds,  although 
these  are  not  necessary.  Fusarium,  causing  dry  rot  of  potato,  apparently 
enters  the  tubers  only  after  they  have  been  wounded.  Likewise  most 
of  the  wood-rotting  Basidiomycetes  enter  the  host  only  through  wounds, 
principally  at  broken  or  dead  branches  and  at  pruning  or  lightning  and 
fire  scars.  Here,  the  air-borne  basidiospores  must  find  suitable  moisture 
for  germination  and  for  penetration  of  the  wood.  Endothia  parasitica 
is  said  to  enter  the  chestnut  tree  only  through  wounds  that  extend  through 
the  corky  layer.  Ceratostomella  ulmi  is  transmitted  by  the  European 
bark  beetle,  which  introduces  the  spores  into  its  feeding  wounds.  Bruises 
and  wounds  of  fruits  and  vegetables  are  common  ports  of  entry  for  numer- 
ous rot-producing  fungi,  such  as  Rhizopus  nigricans  on  sweet  potato. 
Monilinia  fructicola  on  stone  fruits,  Penicillium  expansum  on  apple,  and 
P.  italic^im.  and  P.  digitatum  on  citrus  fruits. 

PARASITISM 

A  discussion  of  the  action  of  the  parasite  after  it  enters  the  host  is  so 
closely  correlated  with  the  defense  of  the  host  that  it  is  difficult  to  discuss 
each  topic  separately.  For  the  sake  of  convenience,  however,  it  seems 
desirable  to  discuss  some  of  the  outstanding  effects  of  fungi  upon  their 
hosts  and  the  methods  by  which  the  parasites  obtain  their  food  under  a 
separate  heading  of  parasitism. 

Parasitism  in  plants.  Parasitism  may  begin  as  soon  as  a  fungus  hypha 
enters  the  host.  The  primary  consideration  is  the  securing  of  suitable 
nutrients  and  water  by  the  fungus.  This  may  be  accomplished  by  two 
general  methods,  (1)  by  killing  the  cells  of  the  host  and  obtaining  food 
from  the  dead  cells,  or  (2)  by  establishing  a  close  nutritional  relationship 
with  the  living  host  cells  and  absorbing  the  soluble  nutrients  without 
causing  necrosis.  The  fungi  falling  in  the  first  group  are  the  destructive 
parasites,  while  those  belonging  to  the  second  group  have  been  called  the 
balanced  parasites  (Bessey,  1935).  The  latter  group  includes  those  fungi 
known  at  present  as  obligate  parasites  (such  as  the  Uredinales,  Erysiphales, 
and  Peronosporaceae),  and  some  other  fungi  (such  as  the  Ustilaginalesand 
Taphrina)  which  in  their  hosts  obtain  food  only  from  living  cells. 

The  destructive  parasites,  as  a  whole,  are  strong  producers  of  enzymet; 


376  PHYSIOLOGY  OF  THE  FUNGI 

and  toxins  but  may  be  weak  in  mechanical  action.  Some  of  these  cause 
rapid  rots  of  fruits  or  vegetables  but  are  unable  to  penetrate  the  unbroken 
epidermis  and  must  depend  on  wounds  for  their  entrance.  Others,  which 
are  seldom,  if  ever,  found  as  pathogens  in  nature,  may  cause  rot  when 
artificially  inoculated  into  succulent  plant  tissues. 

Rotting  of  the  tissue  is  due  to  two  distinct  effects  of  the  fungus  on  the 
host:  (1)  death  of  the  cells,  and  (2)  dissolution  of  the  middle  lamellae. 
The  separation  of  the  cells  is  due  to  the  action  of  the  enzymes  proto- 
pectinase,  pectinase,  and  pectase  on  the  middle  lamella.  These  three 
enzymes  are  often  collectively  referred  to  as  pectinase.  There  is  some 
evidence  that  pectinase  may  also  cause  a  change  in  permeability  of  the 
cell  membranes  and  the  death  of  the  cells,  but  it  is  possible  that  some 
other  toxic  substance  may  be  closely  associated  with  pectinase.  How- 
ever, no  such  substance  has  been  isolated.  Extracts  of  rotted  tissues 
have  been  shown  to  cause  the  same  effects  as  the  fungi  themselves.  These 
effects  are  described  by  De  Bary  (1886)  for  Sclerotinia  sclerotiorum  and 
by  Brown  (1915)  for  Botrytis  cinerea.  Higgins  (1927)  believes  that 
oxalic  acid  produced  by  Sclerotium  rolfsii  is  the  principal  agent  of  destruc- 
tion. The  death  of  the  host  cells  well  in  advance  of  the  invading  hyphae 
indicates  rapid  diffusion  of  the  toxic  substance  in  the  case  of  fungi  produc- 
ing soft  rot.  Brown  (1948)  believes  that  the  enzyme  pectinase  acts  as  a 
cytolytic  toxin.  For  a  discussion  of  the  identity  of  enzymes  and  toxins 
of  species  of  Clostridium,  see  Smith  (1949). 

Thatcher  (1942)  has  shown  that  B.  cinerea  and  S.  sclerotiorum  cause 
a  fourfold  increase  in  the  permeability  to  water  of  the  host  cells  just 
beyond  the  discolored  necrotic  zone.  Some  substance  other  than  pecti- 
nase may  bring  about  this  change  in  permeability  and  be  a  contributing 
factor  to  the  "action  in  advance"  of  many  fungi.  PhytophtJwra  infestans 
caused  a  change  in  permeability  in  host  cells  beyond  the  extent  of  the 
hyphae  which  penetrated  the  living  tissue.  The  identity  of  the  substance 
causing  a  change  in  permeability  is  unknown,  but  it  is  likely  a  weak  toxin 
or  an  enzyme  which  alters  the  structure  or  activity  of  the  plasma  mem- 
brane. The  increase  in  permeability  may  concern  water  alone  or  both 
nutrients  and  water. 

An  osmotic  pressure  higher  in  the  fungus  cells  than  in  the  surrounding 
host  cells  is  apparently  characteristic  of  the  host-parasite  relationship 
(Table  63) .  This  is  necessary  before  the  parasite  can  absorb  water  from 
the  host  cells. 

The  production  of  pectinase  and  its  activity  under  different  conditions 
were  studied  by  Vasudeva  (1930)  and  Chona  (1932),  who  showed  that 
the  amount  produced  by  Botrytis  allii  varied  with  the  medium  in  which 
the  fungus  was  grown.  B.  allii  did  not  secrete  a  demonstrable  amount  of 
pectinase  when  grown  on  apple  extract,  but  when  asparagine,  potassium 


PARASITISM  AND  RESISTANCE 


377 


nitrate,  or  ammonium  sulfate  was  added  to  the  apple  extract,  there  was  a 
decided  increase  in  the  amount  produced.  This  is  not  surprising,  for 
there  are  numerous  reports  that  the  available  nutrients  and  the  pH  of  the 
culture  medium  affect  both  the  kind  and  amount  of  metabolic  products  of 
a  fungus. 

The  activity  of  the  enzymes  produced  by  a  pathogen  varies  with  the 
conditions  of  the  host  cells.  It  seems  probable  that  the  inhibition  of  the 
fungus  enzymes  by  the  host  cells  is  an  important  factor  in  resistance. 
Klotz  (1927)  proposed  this  hypothesis  to  explain  the  greater  resistance 
of  sour  orange  and  the  greater  susceptibility  of  lemon  to  Pythiacystis 
citrophthora  and  Phomopsis  calif  or  nica,  the  causes  of  certain  bark  diseases. 

Table  63.     The  Osmotic  Pressures  of  Host  and  Parasite 
(Thatcher,  Can.  Jour.  Research  20,  1942.) 


Fungus 

Ave. 
osmotic 
pressure, 

atm. 

Host 

Ave. 
osmotic 
pressure, 

atm. 

Uromyces  fabae,  germ  tubes .  .  . 
haustoria     

44.25 

21.9 

18.6 

29.8 

23.5 

18.9 
18.0 

17.4 
15.5 
18.1 
41.3 

Pisum  sativum,  leaf 

petiole 

Dianthus,  leaf  base 

9.15 
10.1 

U.  caryophyUinus,  haustoria.  .  . 

Botrytis  cinerea,  hyphae 

Sclerotinia  sderoliorum,  hyphae. 
Puccinia    graminis,    haustoria 
(race  21) 

11.2 

Apium  graveolens,  petiole 

A.  graveolens,  petiole 

Mindum  wheat,  leaf 

Brassica,  leaf 

Solanum  tuberosum,  tuber 

netiole       

8.3 

13.4 

9.4 

Erysiphe  polygoni,  hyphae 

Phytophihora  infestans,  hj^phae 
(aerial) 

10.6 
10.6 

hyphae  (intercellular) 

sporangia 

8.9 

Brassica,  root 

Phoma  lingam,  hyphae 

11.3 

The  greater  pathogenic  action  of  a  destructive  fungus  occurs  in  the  host 
whose  cells  are  favorable  for  the  activity  of  the  enzymes  of  the  fungus. 
Further  evidence  of  enzyme  inhibition  of  certain  plant  tissues  was 
presented  by  Chona  (1932),  who  studied  the  rotting  action  of  B.  cinerea, 
the  cause  of  a  soft  rot  of  various  plant  tissues,  and  Pythium  sp.,  a  rot 
producer  of  potato  tubers.  Vigorous  germination  of  spores  of  Botrytis 
and  even  some  sporulation  took  place  in  artificial  wounds  in  potato 
tubers,  but  no  decomposition  followed.  The  pectinase  produced  by  B. 
cinerea  was  active  against  apple  tissue,  but  the  presence  of  potato  tissue 
inhibited  its  activity.  It  was  then  found  that  the  mineral  salts,  particu- 
larly KH2PO4  and  MgS04,  in  the  potato  were  the  inhibiting  factors.  On 
the  other  hand,  Pythium  spores  germinated  well  on  apple  tissue  but  failed 


378  PHYSIOLOGY  OF  THE  FUNGI 

to  rot  it.  The  inhibition  in  this  case  was  traced  to  the  mahc  acid  in 
the  apple.  The  pectinase  produced  by  Pythium  was  most  active  in  an 
alkahne  medium,  near  pH  8.0,  while  that  of  B.  cinerea  was  more  active 
in  an  acid  medium,  at  pH  5.0  to  5.5. 

In  contrast  with  the  destructive  fungi  which  rot  the  host  tissue  are 
those  which  cause  wilting  and  certain  types  of  necrosis  without  disintegra- 
tion of  the  host  cells.  These  fungi  produce  little  or  no  pectinase.  Some 
common  fungi  which  cause  wilting  of  mature  plants  are  species  of 
Fusarium,  Verticillium,  Cephalosporium,  and  Ceratostomella.  It  is  now 
generally  believed  that  in  most  cases  wilting  caused  by  fungi  is  due  to 
toxins  or  to  the  plugging  of  the  vessels  by  polysaccharides  or  other  similar 
metabolic  products  of  the  fungus,  rather  than  to  plugging  by  the  excessive 
mycelial  growth  in  the  vessels.  Extracts  of  the  mycelium  or  the  culture 
filtrate  of  a  number  of  these  fungi  cause  effects  that  are  the  same  as  or 
similar  to  those  caused  by  the  fungi  themselves  in  their  respective  hosts. 

A  definite  correlation  between  the  pathogenecity  of  two  strains  of 
Fusarium  lycopersici  and  the  toxicity  of  their  metabolic  products  was 
demonstrated  by  Haymaker  (1928).  There  was  similarity  of  symptoms 
and  of  the  effect  of  temperature  on  wilting.  The  culture  filtrate  was  more 
toxic  when  the  fungus  was  grown  at  28°C.  than  that  obtained  at  any  other 
temperature.  The  toxic  substance  was  not  identified.  Other  workers 
(Plattner  and  Clausson-Kaas,  1945;  Woolley,  1946)  have  reported  that 
the  wilt-inducing  compound  produced  by  F.  lycopersici  is  lycomarasmin, 
a  peptide  of  aspartic  acid.  Gaumann  and  Jaag  (1947)  reported  that 
clavacin  exerted  a  wilting  effect  on  detached  tomato  shoots  similar  to 
that  of  lycomarasmin.  But,  whereas  lycomarasmin  acted  chiefly  on  the 
cells  of  the  leaf  blade,  clavacin  is  toxic  mainly  to  the  phloem  and  paren- 
chyma of  wood  and  cortex  of  the  stem  and  petiole.  The  action  of  both 
compounds  is  believed  to  be  similar,  destroying  the  semipermeability  of 
the  plasma  membranes,  thereby  decreasing  the  water-holding  capacity  of 
the  cells  and  inducing  wilting. 

Various  polysaccharides  have  been  shown  to  produce  wilting  in  tomato 
cuttings  (Hodgson  et  al.,  1949).  Since  there  was  a  direct  relationship 
between  molecular  weight  and  wilt-inducing  action  of  these  compounds, 
it  was  concluded  that  their  action  was  mainly  by  mechanically  interfering 
with  the  transportation  of  water.  Dimond  (1947)  also  reported  wilting 
of  elm  leaves  due  in  part  to  a  polysaccharide  produced  by  Ceratostomella 
ulmi  in  culture.  Its  action  is  believed  to  be  similar  in  naturally  infected 
elm  trees. 

More  recently,  Feldman  et  al.  (1950)  have  presented  evidence  to  show 
that  the  primary  wilt-inducing  agent  produced  by  C.  ulmi  is  not  the 
polysaccharide,  but  a  toxin.  The  production  of  toxin  in  liquid  culture 
filtrate  was  greatly  influenced  by  the  pH  of  the  medium,  being  greater  in 


PARASITISM  AND  RESISTANCE 


379 


buffered  media  at  pH  4.25  than  at  5.25,  although  growth  was  more  rapid 
in  the  less  acid  medium  (Fig.  76).  The  toxin  was  shown  to  be  irreversibly- 
inactivated  at  pH  G  or  above.  The  introduction  of  calcium  hydroxide 
into  trees  and  the  application  of  basic  chemicals  to  the  soil  have  been 
somewhat  successful  in  retarding  the  disease.  Presumably,  these  chem- 
icals act  by  raising  the  pH  of  the  sap  of  the  tree. 


Days 
Fig.  76.  Growth  of  Ceratostomella  ulmi  and  production  of  toxin,  as  measured  by  wilt 
of  tomato  seedlings  induced  by  culture  filtrate,  in  buffered  media  at  different  pH 
levels.  Note  that  toxin  production  is  favored  by  the  more  acid  medium,  while 
growth  is  greater  in  the  less  acid  medium.  (Courtesy  of  Feldman,  Caroselli,  and 
Howard,  Phytopathology.  40:  348,  1950.) 

The  varieties  of  oats  susceptible  to  toxic  culture  filtrates  of  Helmin- 
thosporium  victoriae  were  the  same  that  were  susceptible  to  the  fungus  in 
nature  (Meehan  and  Murphy,  1947).  Plants  of  Boone  variety  were 
killed,  but  Clinton  plants  were  unaffected  when  grown  in  the  same  con- 
centrations of  the  filtrate.  The  toxic  substance,  which  was  not  identified, 
was  produced  when  the  media  contained  either  organic  or  inorganic 
nitrogen.  This  species  differs  from  H.  sacchari,  which  was  reported  by 
Lee  (1929)  to  reduce  nitrates  to  nitrites,  which  were  toxic  to  sugar  cane. 


380  PHYSIOLOGY  OF  THE  FUNGI 

The  toxicity  of  the  metabohc  products  of  Fusarium  vasinfectum  was 
found  to  be  dependent  upon  the  medium  on  which  the  fungus  was  cultured 
(Rosen,  192G).  Filtrates  of  cultures  grown  in  a  medium  containing 
potassium  nitrate  and  sucrose  were  highly  toxic  to  cotton  plants,  while 
filtrates  from  cultures  grown  in  a  medium  containing  ammonium  lactate, 
sodium  asparaginate,  and  glycerin  were  not  toxic.  The  filtrate  of  the 
nitrate-sucrose  medium  contained  nitrites.  Solutions  of  chemically  pure 
sodium  nitrite  were  also  decidedly  toxic  to  cotton  plants.  We  may 
assume  that  the  action  of  this  fungus  in  converting  nitrates  to  nitrites 
is  the  same  within  the  host  plant  as  it  is  in  the  culture  vessel. 

Thus,  there  seems  to  be  abundant  evidence  that  the  metabolic  products, 
including  enzymes  and  toxins,  of  a  given  fungus  vary  both  in  kind  and  in 
amount  with  the  pH  and  composition  of  the  culture  medium.  On  the 
other  hand,  the  evidence  that  the  same  situation  exists  in  nature  is 
extremely  scarce.  One  may  speculate,  however,  that  the  types  of  nutri- 
ents furnished  by  the  host  cells  and  the  pH  of  the  cell  sap  may  also  influ- 
ence the  metabolic  products  of  the  fungus  in  the  host  plant.  If  this  is 
true,  a  given  fungus  may  find  the  nutrients  and  environment  supplied  by 
one  host  particularly  favorable  for  the  production  of  a  disease-inducing 
toxin  or  enzyme.  If  the  host  is  unable  to  inhibit  the  action  of  these 
substances,  disease  may  result.  The  natures  of  both  the  pathogen  and 
the  host  determine  the  severity  of  the  disease.  This  hypothesis  may 
help  to  explain,  in  part,  the  variation  in  intensity  of  parasitism  of  a 
fungus  on  its  different  hosts.  While  there  is  little  evidence  to  support 
this  idea  at  present,  it  is  hoped  that  experimental  work  will  be  conducted 
to  test  its  merits. 

The  possibility  that  the  presence  of  vitamins  may  affect  pathogenicity 
has  been  suggested  (Pehrson,  1948;  Prasad,  1949).  There  is  no  evidence 
that  deficiencies  for  vitamins  are  correlated  with  either  parasitism  or 
pathogenicity,  and  vitamin  deficiency  may  be  excluded  as  a  factor  leading 
to  the  parasitic  habit.  Likewise,  there  seems  to  be  little  or  no  correlation 
between  the  nitrogen  requirements  of  fungi  and  the  parasitic  habit. 
Nonliving  organic  materials  in  nature  are  sources  of  vitamins  and  organic 
nitrogen  just  as  are  the  living  plants.  For  example,  Ustilago  striiformis, 
a  highly  parasitic  fungus,  is  self-sufficient  with  respect  to  vitamins,  and 
some  isolates  are  capable  of  utilizing  nitrate  nitrogen,  while  Phycomyces 
hlakesleeanus,  an  obligate  saprophyte,  is  deficient  for  thiamine  and  is 
unable  to  utilize  nitrate  nitrogen. 

Opposed  to  the  destructive  parasites  discussed  above  are  the  balanced 
parasites,  which,  in  general,  have  a  strong  power  to  penetrate  mechan- 
ically but  whose  chemical  actions  on  the  host  are  relatively  weak.  Most 
of  the  filamentous  balanced  parasites  produce  intercellular  mycelium, 
sending  haustoria  into  the  host  cells.     These  serve  as  food-absorbing 


PARASITISM  AND  RESISTANCE  381 

structures,  but  the  exact  mechanism  of  the  transfer  of  food  is  not  so  well 
understood.  The  haustorium  of  the  filamentous  parasite  is  very  similar 
in  its  behavior  to  the  intracellular  nonfilamentous  parasite,  being  sur- 
rounded by  the  protoplasm  of  the  host  cell.  Haustoria  may  be  of  several 
forms,  simple  and  nearly  spherical,  coiled,  and  branched  in  various  ways. 
Most  cytologists  agree  that  there  is  a  cellulose  wall,  or  sheath,  around  the 
older  haustoria.  It  is  presumably  formed  by  the  host  cell  and  suggests 
a  weak  mechanism  of  defense  against  the  invading  parasite,  yet  it  does 
not  prevent  the  diffusion  of  soluble  food  into  the  haustorium. 

The  haustorium  commonly  comes  into  contact  with  the  nucleus  of  the 
host  cell.  In  23  of  the  35  cases  (host-parasite  combinations)  reported 
(Rice,  1927,  1935),  habitual  contact  was  observed  between  haustorium 
and  nucleus.  Two  theories  as  to  the  meaning  of  this  contact  have  been 
suggested.  One  is  that  the  haustorium  seeks  out  the  region  of  the  cell 
nucleus  in  order  to  facilitate  the  absorption  of  food  from  the  cell.  The 
second  theory  is  that  the  action  of  the  cell  nucleus  is  defensive  and  that 
in  some  cases  it  may  cause  the  death  and  degeneration  of  the  haustorium. 

In  the  case  of  Synchytrium  (Chrysophylyctis)  endohioticum  the  swarm 
cells  migrate  into  close  proximity  w^th  the  nucleus  of  the  host  cell  (Orton 
and  Kern,  1919).  In  the  majority  of  cases  the  nucleus  is  engulfed  at  the 
time  or  soon  after  the  swarm  cells  unite  to  form  the  vegetative  body  of  the 
parasite.  The  host  nucleus  disappears  as  the  sporangia  develop.  The 
exact  significance  of  this  close  relationship  between  parasite  and  host 
nucleus  is  not  clear,  but  it  apparently  represents  a  more  or  less  unique 
method  of  parasitism  among  the  fungi. 

It  is  generally  believed  that  the  balanced  parasite  causes  harm  to  a 
susceptible  host  primarily  through  its  demand  upon  food  and  water. 
There  is  little  or  no  evidence  that  the  protoplast  is  attacked  chemically, 
although  host  cells  may  be  killed  by  growth  pressure.  There  are  numer- 
ous reports  of  the  disappearance  of  food  in  the  region  of  haustoria.  Butler 
(1918)  reported  that  starch  is  absent  in  cells  containing  haustoria  of 
Sclerospora  graminicola,  and  at  the  time  of  sporulation  the  host  cells 
collapse  and  die.  The  only  abnormal  effect  observed  by  Mains  (1917) 
on  the  cells  of  corn  parasitized  by  Puccinia  sorghi  was  the  absence  of 
starch  in  the  bundle  sheaths  near  the  rust  pustules.  He  interpreted  this 
to  mean  that  the  parasite  uses  the  food  materials  before  they  reach  the 
bundle  sheath  where  they  are  normally  stored.  Similar  disappearance 
of  starch  in  the  host  cells  near  infection  by  Synchytrium  endohioticum  has 
been  reported  by  Orton  and  Kern  (1919). 

On  the  other  hand,  starch  may  accumulate  in  the  infected  tissues  dur- 
ing early  stages  of  development  of  rusts  but  usually  disappears  in  later 
stages  of  development.  This  may  be  due  to  some  disruption  of  the  host's 
physiology.     The  physiological  reactions  of  the  host  are  known  to  involve 


382  PHYSIOLOGY  OF  THE  FUNGI 

translocation  of  food,  transpiration,  respiration,  and  photosynthesis. 
Increased  respiration  has  been  reported  for  some  hosts,  while  a  decrease 
has  been  found  for  others.  The  rate  of  transpiration  is  usually  increased. 
An  early  infection  of  orange  rust  on  Ruhus  may  even  cause  the  formation 
of  stomata  on  the  upper  epidermis,  where  they  are  normally  lacking 
(Dodge,  1923). 

The  reactions  of  the  chloroplasts  of  the  host  cells  are  believed  to  indicate 
the  degree  of  adjustment  between  the  host  and  parasite  (Rice,  1935). 
Local  chlorosis  and  streaking  are  common  symptoms  of  a  number  of 
diseases  caused  by  haustoria-forming  parasites. 

Thatcher  (1939,  1942)  has  shown  that  certain  obligate  parasites  cause 
an  increase  in  permeability  of  the  cell  membranes  of  susceptible  hosts. 
There  was  a  decided  reduction  in  osmotic  pressure  of  the  tissues  of  Pisum 
surrounding  the  rust  hyphae.  If  the  fungus  is  unable  to  bring  about  an 
increase  in  permeability  so  that  it  can  obtain  its  necessary  nutrients,  the 
host  is  resistant.  Thatcher  found  evidence  that  the  plasma  membranes 
of  some  resistant  varieties  of  wheat  may  actually  become  less  permeable 
as  a  reaction  to  the  rust  hyphae,  and  starvation  of  the  fungus  may  result. 

The  change  in  permeability  incited  by  the  balanced  parasites  seems 
to  be  similar  to  the  action  of  the  destructive  parasites,  except  for  the 
matter  of  degree.  Thatcher  (1939)  believes  that  parasitism  in  the  rusts 
has  become  highly  specialized,  and  the  intensity  of  the  effect  on  permea- 
bility of  the  cell  membranes  has  been  reduced.  The  substance  involved 
is  apparently  a  metabolic  product  of  the  fungus. 

If  the  conditions  afforded  by  a  certain  variety  of  host  are  favorable 
for  the  production  of  a  comparatively  large  amount  of  toxin  (assuming 
that  this  substance  is  a  toxin) ,  the  host  cells  may  be  killed  and  the  further 
development  of  the  obligate  parasite  would  be  prevented.  The  sudden 
death  of  the  host  cells  is  the  condition  described  by  Stakman  (1914)  as 
hyper  sensitiveness.  Hypersensitive  hosts  are  highly  resistant  or  immune 
to  the  pathogenic  action  of  the  obligate  parasites.  Stakman  reported 
that,  in  varieties  of  wheat  resistant  to  Pucciiiia  graminis  tritici,  when 
the  hyphae  of  the  fungus  come  in  contact  with  the  host  cells,  the  latter 
often  show  plasmolysis,  disintegration,  and  finally  death.  After  the 
death  of  a  few  surrounding  cells  the  tips  of  the  hyphae  die.  However,  it 
was  discovered  that  in  some  cases  the  hyphae  may  die  before  the  host  cells 
are  killed.  Stakman  concluded  that  the  problem  of  resistance  to  rvists 
is  one  of  toxins  of  the  parasite  or  the  host,  or  both,  and  can  best 
be  explained  by  what  he  terms  the  "toxin  or  enzyme  theory."  Brooks 
(1948)  also  concluded  that  the  death  of  the  parasite  is  due  to  the  lethal 
action  of  the  host  rather  than  to  starvation. 

Opponents  of  the  toxin  (or  enzyme)  theory  of  parasitism  in  the  rusts 


PARASITISM  AND  RESISTANCE  383 

point  out  that  no  toxin  has  ever  been  demonstrated  experimentally. 
Leach  (1919)  believes  that  each  physiologic  race  of  Puccinia  graminis 
has  its  own  characteristic  food  requirements  which  are  met  by  only  a  few 
varieties  of  the  host.  According  to  this  hypothesis,  if  a  race  of  rust 
enters  a  host  which  does  not  meet  its  specific  nutritional  requirements,  it 
dies,  and  enzymes  which  are  injurious  to  the  host  cells  are  released.  This 
hypothesis  is  supported  by  Wellensiek  (1927)  who  worked  with  Puccinia 
sorghi. 

While  it  is  evident  that  the  food  supply  varies  with  the  varieties  of  the 
host,  it  seems  equally  possible  that  the  difference  in  nutrients  may  have  a 
more  indirect  effect  in  determining  whether  the  fungus  survives.  Is  it 
merely  that  the  fungus  starves  if  the  host  does  not  provide  the  appropriate 
food,  or  are  the  conditions  in  the  host  unfavorable  for  the  production  of 
certain  metabolic  products  which  are  essential  to  the  pathogenic  actions 
of  the  fungus? 

The  type  of  host-parasite  relationship  found  in  Phyllachora  graminis 
seems  to  be  unique  (Orton,  1924) .  This  fungus  apparently  has  the  power 
of  digesting  and  absorbing  the  tissues  within  the  leaf,  producing  cavities 
in  which  the  ascocarps  later  form.  The  hyphae  bore  their  way  through 
the  cell  walls  of  any  of  these  tissues  and,  in  doing  so,  absorb  a  portion  of 
the  wall.  The  parenchyma  cells  become  disorganized,  and  their  contents 
disintegrate.  The  vascular  cells  may  be  invaded  and  partially  absorbed 
and  become  filled  with  hyphae.  The  most  striking  physiological  charac- 
teristic of  this  fungus  is  its  ability  to  absorb,  replace,  and  engulf  the 
tissues  of  the  host  leaf  without  any  external  evidence  of  necrosis  of  the 
host.  This  would  seem  to  indicate  the  presence  of  highly  active  cellulo- 
iytic  enzymes  (and  perhaps  others)  confined  to  the  area  near  the  fungus, 
without  the  presence  of  toxic  substances,  which  would  cause  necrosis  of 
the  leaf  tissue. 

Actually,  comparatively  little  is  known  about  the  activities  which 
lead  to  parasitism,  particularly  of  the  balanced  parasites.  It  is  hoped 
that  more  planned  experiments  will  be  conducted  in  an  attempt  to  gain 
more  knowledge  regarding  the  mode  of  parasitism  of  plant  pathogens. 
Only  by  understanding  the  action  of  the  parasite  can  we  understand  the 
basic  facts  underlying  resistance  and  susceptibility. 

Parasitism  and  symbiosis  with  insects.  There  are  numerous  reports 
of  the  parasitic  and  symbiotic  relations  of  fungi  with  insects.  For  a 
more  complete  discussion  than  this  text  offers,  see  Leach  (1940)  and 
Steinhaus  (1946).  In  many  cases  the  relationship  is  solely  to  the  advan- 
tage of  the  fungus  (true  parasitism),  but  a  number  of  cases  of  mutualistic 
symbiosis  do  exist.  The  fungi  may  be  disseminated  by  the  insects  which 
serve  as  their  hosts.     One  can  only  speculate  regarding  the  basic  nutri- 


384  PHYSIOLOGY  OF  THE  FUNGI 

tional  requirements  of  these  fungi,  since  very  little  is  known.  We  may 
assume  that  rather  specific  nutritional  needs,  either  for  growth  or  for 
reproduction,  are  satisfied  by  the  relation  with  insects. 

Among  the  fungi  parasitic  on  insects  the  genus  Entomophthora  is  the 
most  common.  Various  common  species  attack  houseflies,  grasshoppers, 
and  other  insects.  A  direct  correlation  between  the  amount  of  precipita- 
tion and  the  number  of  infections  on  houseflies  was  reported  by  Yeager 
(1939).  Massospora  cicadina  infects  the  seventeen-year  cicada  and 
produces  spores  inside  the  abdomen.  The  posterior  portion  of  the  abdo- 
men sloughs  away,  exposing  the  spores  w^hile  the  insect  is  still  able  to  crawl 
about.  This  is  apparently  the  chief  method  of  dissemination  of  the 
spores.  The  mode  and  time  of  infection  are  unknown.  Species  of 
Cordyceps  are  common  on  pupae  and  larvae  of  certain  insects.  The  fact 
that  C.  militaris  produces  abundant  mycelial  growth  on  a  variety  of 
synthetic  media  in  the  laboratory  suggests  the  possibility  that  in  nature 
this  fungus  may  grow  on  other  substrata,  requiring  the  insect  association 
only  to  fruit. 

Fawcett  (1910)  described  the  use  of  a  fungus,  which  he  named  Aegerita 
webberi,  in  controlling  whitefly  in  the  orange  groves  of  Florida.  Ascher- 
sonia  aleyrodis  has  also  been  used  for  the  same  purpose.  A  chytrid, 
Myrophagus  ucrainicus,  is  reported  (Karling,  1948)  as  a  parasite  on  scale 
insects  in  Bermuda,  Louisiana,  and  Ontario.  In  severe  outbreaks  as 
many  as  45  per  cent  of  the  female  insects  may  be  killed.  It  has  also  been 
transmitted  to  mealy  bugs.  Another  group  of  fungi  parasitic  on  insects 
is  the  Laboulbeniales.  These  are  minute  fungi  developing  almost  entirely 
on  the  surface,  sending  short  haustoria  into  the  insect  to  obtain  food. 

The  symbiotic  relationship  between  Septobasidium  and  scale  insects 
is  interesting  because  of  the  high  degree  of  specialization  on  the  part  of 
the  fungus  (Couch,  1938) .  The  dependence  of  the  fungus  for  its  distribu- 
tion upon  the  migrating  young  scale  insects  was  previously  mentioned  in 
Chap.  15  under  Spore  Dissemination.  The  fungus  forms  a  crust  over 
scale  insects,  some  of  which  are  parasitized  while  others  are  not.  The 
uninfected  females  give  rise  to  young  insects,  which  may  remain  under 
the  fungus  crust,  crawl  out  through  tunnels  under  the  fungus,  or  crawl 
out  over  the  sporulation  surface  of  the  fungus.  The  young  insects  are 
infected  only  by  the  bud  cells  from  the  basidiospores,  never  by  the  older 
fungus  hyphae.  The  bud  cells  germinate  on  the  surface  of  the  insect  and 
apparently  enter  principally  through  the  natural  openings.  The  fungus 
then  produces  coiled  haustoria,  which  absorb  food  directly  from  the 
circulatory  system  of  the  insect,  which  in  turn  sucks  its  food  from  the 
host  tree.  Some  infected  insects  may  settle  down  on  the  bark,  while 
others  crawl  under  a  nearby  protective  fungus  colony.  Only  the  former 
are  responsible  for  distributing  the  fungus,  while  the  latter  are  responsible 


PARASITISM  AND  RESISTANCE  385 

for  the  survival  of  the  ah-eady  formed  fungus  colonies.  Connections  are 
then  made  by  anastomoses  of  the  hyphae  from  the  insect  and  the  hyphae 
of  the  fungus  crust  under  which  the  insect  has  come  to  rest.  Thus,  the 
fungus  colony  does  not  originate  from  one  individual  but  from  the  aggre- 
gation of  several  individuals  by  anastomosis,  or  grafting.  The  parasitized 
insects  are  dwarfed  and  do  not  reproduce  but  may  live  as  long  as  the 
uninfected  insects.  The  fungus  covers  the  insect's  body  but  is  in  contact 
with  it  only  by  the  numerous  coiled  haustoria.  The  insect  in  turn 
receives  protection  from  severe  weather  conditions,  from  parasitic  wasps, 
other  insects,  and  birds.  Certain  species,  particularly  S.  hurtii,  are  easily 
cultivated  on  liquid  or  agar  media.  Couch  believes  that  failure  of  the 
fungus  to  fruit  in  culture  may  be  due  to  lack  of  proper  nutrition,  which 
is  furnished  by  insects  in  nature,  or  to  a  complicated  heterothallic  condi- 
tion of  the  fungus. 

Fungi  parasitic  on  other  fungi.  The  parasitic  habit  of  many  of  the 
chytrids  upon  other  aquatic  or  semiaquatic  fungi  and  algae  is  apparently 
quite  common.  A  number  of  these  genera  are  described  and  illustrated 
by  Fitzpatrick  (1930)  and  Karling  (1942).  Practically  nothing  is  known 
regarding  their  nutritional  requirements.  There  appear  to  be  fewer 
filamentous  fungi  parasitic  upon  fungi.  The  mention  of  only  a  few  of 
these  will  serve  as  examples.  Species  of  Piptocephalis,  Chaetocladium, 
and  Syncephalis  are  parasitic  on  other  Mucorales.  A  number  of  fungi  are 
reported  to  be  parasitic  on  members  of  the  Agaricaceae  and  other  higher 
fungi.  Among  these  are  species  of  Spinellus,  Mycogone,  Hypomyces, 
Nyctalis,  and  some  Myxomycetes.  A  species  of  Penicillium  is  parasitic 
upon  an  Aspergillus  (Thorn  and  Raper,  1945).  Of  particular  interest 
are  the  hyperparasites,  fungi  parasitic  upon  other  parasitic  fungi.  Cicin- 
noholus  cesatii  is  parasitic  on  the  Erysiphales,  and  Darlucafilum  is  parasitic 
on  Uredinales.  So  far  as  is  known,  no  study  of  the  basic  nutritional 
requirements  of  these  fungi  has  been  attempted. 

Fungi  parasitic  on  man  and  animals.  Many  of  the  fungi  which  cause 
disease  of  man  and  animals  show  distinct  differences  in  morphology  when 
grown  under  different  conditions.  The  spore  forms  produced  on  artificial 
media  may  be  quite  different  from  those  developed  in  the  host.  This 
may  be  a  response  to  certain  nutritional  factors,  to  temperature  differ- 
ences, or  to  the  presence  of  certain  chemical  substances  which  inhibit 
or  limit  the  production  of  certain  spore  forms. 

In  general,  the  pathogens  of  man  and  animals  have  no  unique  nutri- 
tional requirements.  Some  are  able  to  utilize  inorganic  nitrogen,  while 
others  are  not;  some  are  deficient  for  certain  vitamins.  Nickerson  (1947) 
points  out  that  there  is  no  direct  correlation  between  nutritional  require- 
ments and  pathogenicity.  In  fact,  there  is  little  concrete  evidence 
regarding  the  mode  of  action  of  these  fungi  in  causing  disease. 


38G  PHYSIOLOGY  OF  THE  FUNGI 

In  the  case  of  the  dermatophytes,  Nickerson  has  suggested  that  growth 
and  sporulation  in  vivo  may  be  affected  by  a  chemical  supplied  to  the 
hair  and  scales  of  the  skin  by  diffusion  from  the  adjacent  resistant  tissues. 
There  is  some  evidence  that  resistance  of  skin  to  fungus  infection  may  be 
influenced  by  the  nutrition  of  the  individual. 

For  more  complete  discussions  of  the  fungus  diseases  of  man  and 
animals,  the  student  is  referred  to  Nickerson  (1947),  Wolf  and  Wolf 
(1947),  Conant  et  al.  (1944),  Emmons  (1940),  and  Dodge  (1935).  The 
pioneering  work  of  Sabouraud  (1910)  should  also  be  consulted. 

Cultural  characteristics  and  pathogenicity.  Studies  of  numerous 
isolates  of  a  given  species  or  genus  have  indicated  a  possible  correlation 
between  pathogenicity  and  some  particular  cultural  characteristic.  The 
recognition  of  such  relations  and  definite  knowledge  regarding  them  would 
be  of  great  value  to  plant  pathologists.  One  such  study  was  made  by 
Houston  (1945)  on  52  isolates  of  Corticium  solani  from  various  hosts. 
These  isolates  fell  into  three  culture  types  based  upon  the  characteristics 
of  the  mycelium  and  sclerotia.  There  was  a  certain  degree  of  correlation 
between  culture  type  and  pathogenicity  and  symptoms  on  certain  hosts. 
He  concluded  that  the  culture  type  of  C.  solani  is  more  important  in 
predicting  the  pathogenicity  of  an  isolate  than  the  host  from  which  it  was 
isolated. 

During  a  study  of  the  physiological  characteristics  and  pathogenicity 
of  143  isolates  of  Actinomyces,  mostly  from  scabbed  potato  tubers,  it  was 
found  (Taylor  and  Decker,  1947)  that  certain  isolates  produced  a  dark 
ring  of  growth  at  the  surface  of  separated  milk.  This  characteristic  was 
correlated  perfectly  with  the  ability  to  produce  typical  scab  lesions  on 
potato  tubers.  No  attempt  was  made  to  explain  the  basic  relation  of 
these  two  apparently  unrelated  physiological  characteristics. 

RESISTANCE 

Resistance  is  the  ability  of  a  host  to  prevent  or  oppose  the  entrance  or 
subsequent  growth  and  development  of  a  parasite.  It  may  be  effective 
either  before  or  after  penetration  of  the  host.  A  host  which  cannot 
successfully  prevent  such  actions  of  the  parasite  is  susceptible.  Studies 
in  the  nature  of  resistance  have  been  only  partially  enlightening,  and  in 
many  cases  the  nature  of  disease  resistance  is  still  obscure.  Some  of  the 
present  theories  are  based  on  what  might  be  termed  "circumstantial 
evidence,"  such  as  a  general  correlation  between  resistance  and  some 
characteristic  of  the  host.  There  is  sufficient  evidence,  however,  that  it 
is  dangerous  to  generalize  about  the  nature  of  resistance.  It  seems  likely 
that  in  many  cases  the  cause  of  resistance  may  be  specific,  being  common 
perhaps  to  but  one  or  only  a  few  host-parasite  combinations. 

The  types  of  resistance  may  be  placed  for  convenience  into  three 


PARASITISM  AND  RESISTANCE  387 

groups:  (1)  mechanical,  the  prevention  of  penetration  or  of  unlimited 
spread  by  the  structure  of  the  host;  (2)  functional,  the  prevention  of 
penetration  by  stomatal  action  of  the  host;  (3)  physiological,  chemical 
action  against  the  parasite  or  incompatible  food  relations.  The  relative 
importance  of  these  factors  is  difficult  to  determine,  but  Butler  (1918) 
states  that  physiological  characters  are  much  more  important  as  a  factor 
for  resistance  than  the  anatomical  characters  of  the  host. 

Mechanical  resistance  might  be  considered  as  the  first  line  of  defense 
by  the  host.  According  to  Melander  and  Craigie  (1927)  resistance  of 
species  of  Berheris  to  infection  by  sporidia  of  Puccinia  graminis  is  due 
to  the  thickness  of  the  cuticle.  B.  thunbergii,  which  is  immune,  has  a 
heavy  layer  of  cutin,  while  in  general  the  susceptible  species  have  a 
thin  layer.  These  conclusions  were  reached  after  anatomical  studies  and 
after  using  a  mechanical  device  to  measure  the  resistance  of  the  epidermis 
to  puncture.  The  thickness  of  the  cuticle  increases  with  age,  as  does  the 
resistance  to  mechanical  puncture  and  to  infection.  The  same  is  true 
with  the  apple  scab  fungus  and  powdery  mildew  of  barley;  young  leaves 
are  susceptible  but  become  more  resistant  with  age. 

Resistance  in  some  cases  is  apparently  due  to  layers  of  cork  cells  formed 
by  the  host  in  advance  of  the  invading  parasite.  Varieties  of  flax  resistant 
to  wilt  {Fusarium  lini)  and  of  cotton  to  black  root  rot  {Thielaviopsis 
hasicola)  seem  to  be  successful  in  walling  off  the  parasite  by  forming  such 
a  layer  of  cork  which  it  cannot  penetrate.  Varieties  of  potatoes  resistant 
to  scab  {Actinomyces  scabies)  form  cork  more  quickly  when  wounded  than 
do  susceptible  varieties  and  are  believed  to  owe  their  resistance  to  this 
characteristic.  Thomas  (1934)  found  that  the  newly  formed  layer  of 
cork  cells  was  penetrated  by  invading  hyphae  of  Armillaria  mellea  and 
that  the  cork  layer  did  not  successfully  stop  the  advance  of  this  fungus. 
Brown  (1936)  states  that  there  is  some  doubt  as  to  whether  the  cork  layer 
really  functions  at  all  or  whether  it  is  formed  after  the  fungus  has  been 
stopped  by  some  chemical  means. 

Lignified  tissues  offer  more  mechanical  resistance  than  nonlignified 
cells.  Certain  varieties  of  wheat  resistant  to  stem  rust  have  a  compara- 
tively greater  amount  of  sclerenchyma  and  a  correspondingly  lesser 
amount  of  collenchyma  and  parenchyma  in  the  stem,  as  compared  with 
susceptible  varieties.  The  maturity  of  host  tissue  may  be  a  factor  in 
resistance,  even  though  the  tissue  does  not  become  lignified  or  suberized. 
Some  of  the  systemic  smut  fungi  in  cereals  are  able  to  grow  and  penetrate 
the  cell  walls  in  meristematic  tissue  but  are  apparently  unable  to  pene- 
trate the  cellulose  walls  of  mature  parenchyma  cells.  After  infection 
in  the  embryo  or  seedling  stage,  the  fungus  must  continue  to  grow  in  the 
growing  tip  of  the  shoot  if  it  is  to  reach  the  flower  parts.  Conditions 
which  favor  slow  growth  and  delay  the  maturity  of  the  host  favor  the 


388  PHYSIOLOGY  OF  THE  FUNGI 

fungus,  while  conditions  which  favor  rapid  maturity  of  the  host  cells  may 
cause  the  fungus  to  be  left  behind  in  the  mature  tissues  which  it  cannot 

penetrate. 

Hart  (1929)  studied  the  nature  of  resistance  of  wheat  varieties  to  stem 
rust  and  described  a  type  of  resistance  that  she  terms  funciio7ial  resistance, 
which  is  dependent  upon  the  stomatal  movements  of  the  host,  and  con- 
cluded that  the  parasite  enters  the  wheat  only  through  open  stomata. 
There  has  been  frequent  discussion  regarding  the  importance  of  the 
acidity  of  the  cell  sap  of  the  host  and  its  effect  upon  resistance.     The 
effects  of  cell-sap  acidity  may  be  threefold:  (1)  an  increase  in  the  hydrogen 
ions;  (2)  the  toxicity  of  the  organic  acids;  (3)  the  influence  upon  the 
chemical  changes  and  the  possible  formation  of  toxic  products  by  the  host 
cells.     In  some  cases  these  effects  have  not  been  satisfactorily  distin- 
guished.    Numerous  examples  may  be  found  in  the  literature  in  which 
resistance  has  been  attributed  to  the  acidity  of  the  host  or  host  part. 
Butler  (1918)  refers  to  investigations  showing  that  the  leaves  of  varieties 
of  grape  resistant  to  powdery  mildew  contain  three  to  five  times  as  much 
acid  as  the  nonresistant  varieties.     He  also  showed  that  the  red  rot 
fungus  of  sugar  cane,  Colletotrichum  falcatum,  was  present  in  infected 
canes  from  sowing  time  but  usually  did  not  develop  severely  until  matu- 
rity of  the  canes.     He  attributed  this  to  either  the  increase  in  sugar  or  the 
decrease  in  acid.     The  more  acid  lemons  are  less  attacked  by  the  fruit- 
rotting  fungi.     The  amount  of  acid  in  the  fruit,  as  indicated  by  chemical 
analysis,  may  be  greater  than  the  amount  necessary  to  check  the  growth 
of  the  fungus  in  culture  (Cook  and  Taubenhaus,  1911).     A  number  of 
workers  have  considered  cell-sap  acidity  as  a  possible  cause  for  resistance 
of  wheat  to  stem  rust,  but  this  factor  now  is  believed  to  be  of  little  impor- 
tance.    No  correlation  was  found  between  resistance  and  acidity  of  the 
expressed  sap  (Hurd,  1924).     Similarly,  there  was  no  correlation  between 
resistance  and  hydrogen-ion  values  or  the  titratable-acid  values  of  the 
juice  of  wheat  plants  resistant  to  Ustilago  tritici  (Tapke,  1929). 

Some  of  the  most  complete  experimental  evidence  showing  the  correla- 
tion between  acidity  and  resistance  is  presented  by  Reddy  (1933)  for 
different  inbred  lines  of  corn  in  relation  to  Basisporium  gallarum.  Briefly, 
he  found  that  when  the  pH  of  the  cob  was  below  5.0,  resistance  to  cob 
infection  was  high.  Resistance  was  notably  lower  at  high  pH  values. 
Table  64  gives  a  summary  of  some  of  Reddy's  experiments.  Reddy  also 
believes  that  the  influence  of  pH  may  explain  why  the  seedlings,  which 
are  acid,  are  resistant  to  infection  by  B.  gallarum,  while  the  dry  kernels, 
which  are  neutral  or  alkahne,  are  susceptible.  On  the  basis  of  evidence 
previously  discussed,  it  is  likely  that  the  pathogenic  activities  of  certain 
enzymes  produced  by  B.  gallarum  are  inhibited  in  media  having  pH  of 
5.0  or  less. 


PARASITISM  AND  RESISTANCE 


389 


On  the  other  hand,  greater  acidity  of  the  cell  sap  may  favor  the  develop- 
ment of  some  diseases.  The  susceptibility  of  certain  varieties  of  grape  to 
Guignardia  bidwellii  has  been  correlated  with  a  greater  amount  of  tartaric 
acid  (Butler,  1918).  This  author  points  out  that  leaves  are  susceptible 
only  while  they  are  young  and  rich  in  tartaric  acid. 

Table  64.     Hydrogen-ion  Readings  of  Apparently  Healthy  Cobs  of  75  Inbred 

Lines  of  Corn  and  Incidence  of  Basisporium  Ear  Infection  Following 

Both  Natural  and  Artificial  Inoculation 

(Redd3^  loim  Agr.  Expt.  Sta.  Research  Bull.  167,  1933.) 


CobpH 

No.  of  inbreds  in 

No.  of  in- 

No. of  ears 

Ears 

class  interval 

breds  infected 

observed 

infected,  % 

4.4-4.7 

5 

0 

116 

0 

4.8 

6 

1 

121 

2.5 

4.9-5.0 

14 

7 

312 

7.4 

5.1-5.2 

16 

12 

313 

22.7 

5.3-5.4 

12 

11 

258 

38.0 

5.5-5.6 

7 

7 

175 

41.7 

5.7-5.8 

8 

7 

185 

33.5 

5.9-6.3 

7 

6 

173 

48.6 

According  to  Smith  et  al.  (1946),  there  is  evidence  that  slight  variation 
in  pH  may  have  a  greater  influence  upon  disease  resistance  of  a  plant  than 
is  generally  believed.  Such  resistance  is  not  due  directly  to  the  number 
of  hydrogen  ions.     These  authors  state: 

The  observed  behavior  of  hydroquinone  and  catechol,  representatives  of  the 
widely  occurring  ortho-  and  para-dihydroxyphenolic  compounds,  suggested  that 
hydrogen  ion  differences  also  may  influence  toxicity  by  affecting  the  rate  or 
extent  of  conversion  to  the  more  toxic  quinones  on  invasion  by  pathogens  or  by 
other  injur5^ 

The  possibility  that  the  presence  of  the  pathogen  may  alter  the  pH 
of  the  host  cells,  making  it  more  favorable  to  extensive  invasion,  should 
not  be  overlooked.  Apparently  this  situation  exists  in  the  relation  of 
Erwinia  carnegieana  to  its  host,  the  giant  cactus  of  Arizona.  Boyle 
(1949)  reported  that  the  freshly  expressed  sap  from  uninfected  plants 
gave  pH  readings  of  5.0  to  5.5,  while  the  healthy-appearing  tissue  from 
infected  plants  had  pH  values  of  7.0  to  7.4,  and  the  pH  of  discolored  tissue 
not  yet  broken  down  was  8.7  to  9.0.  These  differences  could  not  be 
attributed  to  genetic  variation  and  were  believed  to  be  a  result  of  the 
pathogen.  The  possibility  that  similar  relations  exist  between  fungus 
pathogens  and  their  hosts  seems  to  merit  greater  consideration  than  it 
has  received. 


390  PHYSIOLOGY  OF  THE  FUNGI 

That  resistance  is  due  to  the  presence  of  some  toxic  substance,  perhaps 
an  organic  acid  or  some  related  compound,  in  the  living  host  cell  is  one 
of  the  most  popular  theories.  However,  detailed  proof  of  the  effective- 
ness of  such  a  compound,  even  though  present  in  the  plant,  is  often  diffi- 
cult to  obtain.  Cook  and  Taubenhaus  (1911)  list  some  organic  acids  in 
order  of  their  toxicity  as  follows:  tannic,  gallic,  malic,  tartaric,  and  citric. 
They  state  that  vegetable  juices  contain  an  enzyme  which  acts  upon 
gallic  acid  to  produce  tannin  or  a  tannin-like  compound  which  is  toxic 
to  fungi.  The  amount  of  the  enzyme  decreases  with  maturity  and 
ripening  of  the  fruits  (apples,  pears,  persimmons,  etc.),  which  accordingly 
become  more  susceptible  to  rot  fungi. 

An  outstanding  example  of  chemical  resistance  is  that  described  by 
Link  and  Walker  (1933)  for  onion  smudge  caused  by  CoUetotrichum 
circinans.  The  cell  sap  of  the  colored  varieties  (resistant)  is  much  more 
toxic  to  the  fungus  than  the  cell  sap  of  the  white-skinned  varieties  (sus- 
ceptible) .  Furthermore,  the  sap  of  the  colored  varieties  contains  catechol 
and  protocatechuic  acid  in  amounts  that  would  account  for  the  resistance 
of  these  varieties  to  the  fungus.  The  action  of  volatile  and  nonvolatile 
antibiotics  in  the  fleshy  scales  of  the  onion  is  believed  to  be  a  definite 
factor  in  relative  resistance  of  onion  varieties  to  C.  circinans,  Aspergillus 
niger,  and  Botrytis  allii  (Hatfield  et  al.,  1948).  Reynolds  (1931)  explains 
resistance  of  flax  varieties  to  Fusarium  lini  as  being  due  to  the  higher 
amounts  of  glucosides,  which  upon  hydrolysis  yield  hydrocyanic  acid. 
Similarly,  the  resistance  of  species  of  Solamim  to  Cladosporiuni  fulvum 
is  believed  to  be  due  to  the  presence  of  higher  amounts  of  solanine 
(Schmidt,  1933;  cited  by  Brown,  1936).  Rochlin  (1933)  believes  that 
there  is  a  direct  connection  betw^een  resistance  of  crucifers  to  clubroot 
and  the  amount  of  glucosides,  which  on  fermentation  give  rise  to  pungent 
mustard  oils.  The  isolation  of  2-methoxy-l,4-naphthoquinone  from 
Impatiens  balsamina  was  reported  by  Little  et  al.  (1948).  This  substance 
had  a  high  antibiotic  activity  against  several  fungi  and  was  nontoxic  to 
tomato  and  bean  plants.  This  may  be  an  example  of  a  naturally  occur- 
ring antibiotic  as  a  factor  in  resistance.  Fontaine  et  al.  (1947)  suggest 
that  tomatin  may  be  a  factor  in  the  resistance  of  certain  tomato  varieties 
to  Fusarium  lycopersici. 

An  interesting  theory  of  resistance  to  obligate  parasites  is  presented 
by  Dufrenoy  (1936).  He  divides  the  hosts  into  three  groups:  (1)  highly 
resistant,  (2)  moderately  susceptible,  (3)  extremely  susceptible.  He 
believes  that,  when  a  fungus  enters  the  highly  resistant  host,  it  kills  the 
cells  it  penetrates  and  that  the  death  of  these  cells  alters  the  metabolism 
of  the  surrounding  cells,  so  that  their  cell  sap  becomes  rich  in  phenolic 
compounds,  which  prevents  the  further  invasion  by  the  pathogen.  In 
the  moderately  susceptible  host  the  host  cells  and  their  living  contents 


PARASITISM  AND  RESISTANCE  391 

are  so  modified  that  they  revert  to  the  embryonic  condition  and  may  even 
divide.  When  the  obhgate  parasite  enters  the  extremely  susceptible 
host,  it  causes  so  little  disturbance  that,  at  least  in  the  first  stages  of 
infection,  the  metabolism  of  the  host  is  afTected  but  little  or  not  at  all. 
Walker  and  Link  (1935)  caution  against  jumping  at  conclusions  regard- 
ing the  importance  of  phenolic  compounds  as  factors  in  resistance.  They 
point  out  that 

.  .  .  the  mere  piesence  of  phenolic  substances  in  a  host  plant  does  not  warrant 
the  conclusion  that  they  play  a  role  in  the  resistance  of  that  host  to  a  given 
parasite  or  parasites.  Toxic  phenolic  substances  might  be  present  in  concen- 
trations so  low  that  their  inhibitory  effects  are  negligible,  or  they  might  also  be 
present  in  concentrations  that  have  a  stimulative  effect.  When  a  phenolic  sub- 
stance with  a  specific  toxicity  toward  a  given  organism  is  present  in  the  host  in  an 
appropriate  concentration,  it  may  be  regarded  as  a  part  of  the  disease  resisting 
mechanism  of  that  host. 

The  four  fungi  studied  by  Walker  and  Link  {Colletotrichum  circinans, 
Gibber ella  saubinetii,  Botrytis  allii,  and  Aspergillus  niger)  reacted  quite 
differently  to  the  various  phenolic  compounds.  Protocatechuic  acid 
inhibited  C.  circinans  at  1/800  and  retarded  growth  at  1/12,800,  while 
it  did  not  affect  A.  niger  at  1/200.  Colored  onions  containing  this  acid 
are  resistant  to  C.  circinans  but  quite  susceptible  to  A.  niger. 

The  immunity  of  monocotyledonous  plants  to  Phymatotrichum  omniv- 
orum  is  due  to  certain  unidentified  toxic  materials  present  in  monocots 
but  apparently  absent  in  most  or  all  dicots  (Ezekiel  and  Fudge,  1938). 
Growth  of  the  pathogen  was  prevented  by  the  expressed  juices  from  a 
number  of  monocots  but  not  by  juices  of  susceptible  dicots.  Ether 
fractions  of  monocot  roots,  or  other  underground  parts,  were  highly 
potent  against  the  pathogen,  while  similar  extracts  from  susceptible  dicot 
plants  were  uniformly  nonpotent. 

In  some  other  highly  parasitic  fungi  the  action  of  the  fungus  causes 
the  death  of  the  surrounding  cells,  which  then  prevents  the  further  spread 
of  the  parasite.  Leach  (1923)  found  that  in  a  highly  resistant  variety 
of  bean  the  hyphae  of  Colletotrichum  lindemuthianum  seldom  attack  more 
than  one  or  two  cells  of  the  host.  Both  the  host  cells  and  the  fungus 
hyphae  then  die,  and  the  entire  cell  contents  are  stained  a  reddish  brown. 
In  less  resistant  varieties  the  parasite  attacks  more  host  cells,  but  sooner 
or  later  the  mycelium  disintegrates.  Leach  interprets  this  as  "a  nutri- 
tional phenomenon,"  which  results  in  death  of  the  fungus  by  starvation, 
and  the  products  of  autolysis  then  kill  and  stain  the  host  cells. 

It  has  been  pointed  out  previously  that  certain  fungi  are  able  to  pene- 
trate some  plants  but  are  then  unable  to  establish  themselves  (Stakman, 
1914;  Jones,  1919;  Salmon,  1905).     These  plants  may  be  either  closely 


392  PHYSIOLOGY  OF  THE  FUNGI 

related  or  unrelated  to  those  which  serve  as  the  natural  host  of  the  fungus. 
In  such  cases  the  failure  to  cause  disease  may  be  due  to  unfavoraVjle 
nutritional  relations.  The  theory  of  a  toxin-antitoxin,  or  toxin-counter- 
toxin,  between  parasite  and  host  has  been  suggested  by  a  number  of 
investigators  (Ward,  1905;  Marryat,  1907;  Stakman,  1914;  Allen,  1923; 
Walker,  1924)  as  a  possible  explanation  for  resistance  to  the  rusts. 

Cytological  studies  of  Puccinia  graminis  tritici  infections  of  both 
susceptible  and  resistant  varieties  of  wheat  were  made  by  Allen  (1923), 
who  concluded  that  secretions  from  the  fungus  stimulate  the  metabolic 
activities  of  the  susceptible  host  to  produce  more  food,  while  in  the 
resistant  host  the  same  secretions  cause  disintegration  and  death  of  the 
host  cells  near  the  infection.  More  distant  cells  may  be  stimulated. 
The  haustoria  usually  die  soon  after  the  host  cells  are  killed.  Leach 
(1919)  believes  resistance  to  P.  graminis  tritici  and  P.  graminis  tritici- 
compacti  can  best  be  explained  on  the  basis  of  specific  food  requirements 
of  the  parasite  and  specific  food  production  by  the  host.  It  was  sug- 
gested that  the  injury  to  the  host  cells  might  be  due  to  an  excess  in  amount 
of  enzymes  stimulated  by  a  limited  supply  of  food  in  resistant  hosts. 
Similarly,  Wellensiek  (1927)  believes  that  this  theory  best  explains  the 
resistance  of  corn  to  strains  of  P.  sorghi.  He  suggests  that  the  difference 
between  susceptibility  and  resistance  is  of  a  quantitative  nature  and  that 
the  amount  of  the  specific  nutrient  determines  resistance  or  susceptibility. 

Walker  (1924)  points  out  that  resistance  may  be  due  to  the  action  of  a 
number  of  factors  and  that  a  clear  understanding  of  resistance  must  be 
based  upon  a  thorough  understanding  of  parasitism.  Walker's  excellent 
discussion  of  the  nature  of  disease  resistance  gives  many  references  to  the 
literature  on  this  subject. 

Host  nutrition  and  its  effect  on  the  development  and  severity  of  disease 
is  a  relatively  new  phase  of  study,  and  much  more  investigation  is  neces- 
sary before  general  conclusions  can  be  drawn.  The  fungi  vary  widely 
in  their  reactions  to  differences  in  host  nutrition,  the  type  of  parasitism 
apparently  being  a  determining  factor.  The  action  on  the  pathogen  is 
believed  to  be  principally  indirectly  through  the  effects  of  nutrition  on 
the  host,  although  it  is  possible  that  some  of  the  vascular  parasites  may 
be  directly  affected  by  the  nutrients  which  pass  through  the  xylem.  An 
increase  in  the  salt  concentration  of  the  nutrient  solution  increased  the 
development  of  clubroot,  w^hile  it  decreased  the  severity  of  cabbage 
yellows  (Walker,  1946).  The  development  of  Fusarium  wilt  of  tomato 
was  affected  in  a  way  similar  to  cabbage  yellows.  More  recently,  Gallegly 
(1949)  reported  that  the  development  of  Verticillium  wilt  of  tomato  was 
reduced  with  a  reduction  in  salt  concentration  of  the  balanced  solution 
used  to  grow  the  tomato  plants.  Stakman  (1914)  and  Ward  (1902)  came 
to  the  conclusion  that  deficiencies  in  nitrogen  and  phosphorus  salts  avail- 


PARASITISM  AND  RESISTANCE  393 

able  to  the  host  had  no  appreciable  direct  effect  upon  the  resistance  to 
rusts.  A  summary  of  the  work  on  the  effect  of  soil  nutrients  and  environ- 
ment upon  resistance  to  disease  has  been  presented  by  Wingard  (1941). 

The  carbon  metabolism  of  a  plant  likewise  influences  resistance  to 
certain  rusts.  Waters  (1926)  found  that  urediospores  of  Uroniyces  fabae 
developed  on  detached  leaves  floating  on  5  per  cent  sucrose  solution  in  the 
dark,  while  none  formed  when  leaves  were  floated  on  water.  These 
observations  were  confirmed  by  Yarwood  (1934)  for  rust  and  powdery 
mildew  of  clover.  It  follows  that  active  carbon  assimilation  increases 
susceptibility  of  the  host  to  the  obligate  parasites. 

Although  the  environmental  factors  are  of  great  importance  in  deter- 
mining the  resistance  or  susceptibility  to  a  disease,  their  effects  are  usually 
upon  the  host  and  only  indirectly  upon  the  parasite.  Abundant  refer- 
ences on  this  subject  can  be  found  in  the  literature.  The  effect  of  tem- 
perature upon  the  metabolism  and  resistance  of  certain  hosts  may  be 
illustrated  by  Gihherella  zeae  on  wheat  and  corn  (Dickson,  1923).  Seed- 
ling infection  of  wheat  was  found  to  occur  at  high  temperatures  and  of 
corn  at  low  temperatures;  i.e.,  the  temperatures  unfavorable  to  host 
development.  In  the  germination  of  w^heat  at  low  soil  temperatures  the 
starch  of  the  endosperm  is  hydrolyzed  more  rapidly  than  the  proteins, 
which  results  in  abundant  sugar  but  little  available  nitrogen  for  seedling 
growth.  Thus,  the  cell  walls  are  thickened  and  more  resistant.  At 
higher  temperatures  both  starch  and  proteins  are  rapidly  hydrolyzed; 
there  is  a  greater  supply  of  available  nitrogen,  and  growth  is  more  rapid. 
The  cell  walls  remain  longer  in  the  pectic  condition  and  are  more  suscepti- 
ble to  attack.  In  corn  the  situation  is  reversed.  At  high  temperatures, 
which  favor  the  corn,  the  cell  w'alls  develop  more  rapidly  and  are  more 
resistant. 

Sharvelle  (1936)  concludes  that  the  resistance  of  flax  to  flax  rust  cannot 
be  attributed  to  any  single  factor  but  probably  results  from  a  number  of 
factors  operating  together.  Doubtless,  the  same  statement  could  be 
applied  to  many  other  diseases  to  which  the  nature  of  resistance  is  not  well 
understood. 

SUMMARY 

Some  of  the  different  types  of  parasitism  may  be  summarized  as  fol- 
lows: (1)  The  parasite  produces  extracellular  enzymes,  particularly 
pectinase,  w'hich  dissolves  the  middle  lamellae  of  the  host  cells,  allowing 
the  cells  to  separate  (rotting).  This  may  or  may  not  be  accompanied 
by  toxic  substances  but  results  in  the  death  of  the  cells.  The  soluble 
food  materials  are  then  free  to  be  absorbed  by  the  fungus.  The  insoluble 
foods  stored  in  the  host  cells  may  be  digested  by  other  extracellular 
enzymes.     This  type  is  illustrated  by  the  rots  of  fruits  and  vegetables. 


394  PHYSIOLOGY  OF  THE  FUNGI 

(2)  The  parasite  may  produce  toxic  materials  or  other  substances  whi(  ii 
may  be  active  at  some  distance  from  the  fungus,  but  it  usually  does  not 
cause  the  rotting  of  the  tissue.  This  is  illustrated  by  a  number  of  wilt 
diseases  and  by  some  others.  (3)  The  third  type  depends  upon  a  con- 
genial nutritional  relationship  between  the  parasite  and  the  host  cells. 
In  susceptible  hosts  of  this  type  there  is  little  or  no  apparent  effect  upon 
the  host  cells.  The  resistant  hosts  may  show  a  high  degree  of  sensitivity 
to  the  parasite,  which  may  result  in  the  death  of  the  invaded  cells  and 
starvation  of  the  parasite.  This  type  of  parasitism  is  characteristic  of 
the  balanced  parasites.  The  balanced  parasite  enters  the  susceptible 
host  cell  and  establishes  a  compatible  food  relationship,  absorbing  the 
soluble  nutrients  elaborated  by  the  host,  without  disturbing  the  metabolic 
activity  of  the  host  in  the  early  stages.  In  this  respect,  the  relationship 
of  parasite  and  susceptible  host  represents  the  most  highly  specialized 
type  of  parasitism. 

The  destructive  parasites,  as  a  rule,  are  strong  producers  of  toxins 
and  exoenzymes,  while  the  balanced  parasites  must  be  quite  weak  in  this 
respect.  In  many  host-parasite  relations  studied,  there  is  a  change  in 
the  permeability  of  the  host  cells  surrounding  the  invading  hyphae.  This 
is  believed  to  be  a  direct  response  to  substances  secreted  by  the  parasite. 
Increased  permeability  would  allow  greater  diffusion  of  water  and  nutri- 
ents from  the  host  cells  to  the  parasite.  The  metabolic  products  of  the 
fungi  involved  in  parasitism  are  for  the  most  part  undetermined,  but  they 
are  known  to  include  toxins,  enzymes,  and  polysaccharides.  Since  the 
kind  and  amounts  of  such  products  are  known  to  vary  with  the  composi- 
tion of  the  medium  in  the  laboratory,  it  is  believed  that  like  variation 
may  occur  in  different  hosts  in  nature. 

The  basis  of  resistance  to  disease  may  be  mechanical,  functional,  or 
physiological.  Some  of  the  known  or  proposed  causes  of  physiological 
resistance  are  (1)  cell-sap  acidity;  (2)  toxic  substances  of  the  host;  (3) 
inhibition  of  the  activity  of  certain  enzymes  of  the  parasite  by  the  host ; 
(4)  hypersensitiveness ;  (5)  incompatible  nutritional  relationship;  (6) 
decreased  permeability  of  the  cell  membranes  of  the  host,  resulting  in 
partial  or  complete  starvation  of  the  parasite;  (7)  a  combination  of 
various  factors  acting  together. 

The  obligate  parasites,  principally  the  rusts,  offer  some  challenging 
unsolved  problems  for  the  future  students  of  parasitism.  Probably  the 
principal  one  involves  the  culturing  of  such  fungi  under  controlled  condi- 
tions on  media  of  known  composition.  All  of  the  many  attempts  to  solve 
this  problem  have  met  with  failure,  yet  few  investigators  doubt  that  it 
can  be  solved.  The  phenomenon  of  heteroecism  among  the  rusts  is  of 
great  interest  from  the  standpoint  of  food  relationships.  For  instance, 
sve  must  either  assume  that  the  wheat  and  the  barberry  furnish  the  same 


PARASITISM  AND  RESISTANCE  395 

nutrients  for  Puccinia  graminis  tritici,  and  the  white  pine  and  Ribes  for 
Cronartium  ribicola,  or  that  the  nutrient  requirements  of  the  haploid 
mycelium  are  different  from  those  of  the  diploid  mycelium. 

Much  more  investigation  is  needed  to  increase  our  knowledge  of  possi- 
ble correlations  between  pathogenicity  and  metabolic  products.  This 
should  lead  to  a  better  understanding  of  parasitism.  The  possible  role 
of  antibiotics  occurring  naturally  in  host  plants  as  a  factor  in  disease 
resistance  has  received  some  attention  recently,  but  much  more  knowl- 
edge of  this  type  is  desired.  Many  of  the  problems  of  today  may  come 
nearer  to  solution  with  a  clearer  understanding  of  the  enzyme  systems  of 
the  parasitic  fungi  and  the  basic  principles  of  specific  enzymatic  action. 

REFERENCES 

Allen,  R.  F.:  Cytological  studies  of  infection  of  Baart,  Kanred  and  Mindum  wheats 

by  Puccinia  graminis  tritici  forms  III  and  XIX,  Jour.  Agr.  Research  26 :  571-604, 

1923. 
Allen,  R.  F.:  A  cytological  study  of  Puccinia  triticina  physiologic  form  11  on  Little 

Club  Wheat,  Jour.  Agr.  Research  33:  201-222,  1926. 
♦Arthur,  J.  C,  F.  D.  Kern,  C.  R.  Orton,  F.  D.  Fromme,  H.  S.  Jackson,  E.  B. 

Mains,  and  G.  R.  Bisby:  The  Plant  Rusts,  John  Wiley  &  Sons,  Inc.,  New  York, 

1929. 
Bessey,  E.  a.:  a  Textbook  of  Mycology,  The  Blakiston  Company,  Philadelphia, 

1935. 
Boyle,  A.  M.:  Further  studies  of  the  bacterial  necrosis  of  the  giant  cactus,  Phyto- 
pathology 39 :  1029-1052,  1949. 
*Brooks,  F.  T.:  Host  resistance  to  fungi,  chiefly  in  relation  to  obUgate  parasites, 

Proc.  Roy.  Sac.  (London),  Ser.  B,  135:  180-186,  1948. 
Brown,  W.:  Studies  in  the  physiology  of  parasitism.     I.  The  action  of  Botrytis 

cinerea,  Ann.  Botany  29:  313-348,  1915. 
Brown,  W.:  Studies  in  the  physiology  of  parasitism.     VHI.  On  the  exosmosis  of 

nutrient  substances  from  the  host  tissues  into  the  infection  drop,  Ann.  Botany 

36:  101-119,  1922. 
•*Brown,  W.:  The  physiology  of  the  host-parasite  relation,  Botan.  Rev.  2:  236-281, 

1936. 
Brown,  W.:  Physiology  of  the  facultative  type  of  parasite,  Proc.  Roy.  Soc.  (London), 

Ser.  B,  135:  171-179,  1948. 
Butler,  E.  J.:  Immunity  and  disease  in  plants,  Agr.  Jour.  India,  Special  Indian 

Sci.  Congr.  Issue,  1918 :  10-28. 
Caldwell,  R.  M.,  and  G.  M.  Stone:  Relation  of  stomatal  function  of  wheat  to 

invasion  and  infection  by  leaf  rust  (Puccinia  triticina),  Jour.  Agr.  Research  52: 

917-932,  1936. 
*Chona,  B.  L.  :  Studies  in  the  physiology  of  parasitism.     XIII.  An  analysis  of  factors 

underlying  specialization  of  parasitism,  with  special  reference  to  certain  fungi 

parasitic  on  apple  and  potato,  Ann.  Botany  46:  1033-1050,  1932. 
CoNANT,  N.  F.,  D.  S.  Martin,  D.  T.  Smith,  R.  D.  Baker,  and  J.  M.  Callaway: 

Manual  of  Clinical  Mycology,  W.  B.  Saunders  Company,  Philadelphia,  1944. 
Cook,  M.  T.,  and  J.  J.  Taubenhaus:  The  relation  of  parasitic  fungi  to  the  contents 

of  the  cells  of  the  host  plants,  Delaware  Agr.  Expt.  Sta.  Bull.  91,  1911. 


396  PHYSIOLOGY  OF  THE  FUNGI 

Couch,  J.  N.:  The  Genus  Septobasidium,  The  University  of  North  Carolina  Press 

Chapel  Hill,  1938. 
De  Bary,  a.:  Ueber  einige  Sclerotinien  uiid  Sclerotien  Krankheiten,  Bolan.  Zig. 

44:377-474,  188G. 
Dickson,  J.  G.:  Influence  of  soil  temperature  and  moisture  on  the  development  of 

the  seedling-blight  of  wheat  and  corn  caused  by  Gibberella  sauhinettii,  Jour.  Agr. 

Research.  23 :  837-870    1923. 
DiMOND,  A.  E.:  Symptoms  of  Dutch  elm  disease  reproduced  by  toxins  of  Graphium 

ulmi  in  culture.  Phytopathology  37 :  7,  1947. 
Dodge,  B.  O.:  Effect  of  the  orange-rusts  of  Rubus  on  the  development  and  distribu- 
tion of  stomata,  Jour.  Agr.  Research  25 :  495-500,  1923. 
Dodge,  C.  W.:  Medical  Mycology,  The  C.  V.  Mosby  Company,  St.  Louis,  1935. 
DuFRENOY,  J.:  Cellular  immunity,  Am.  Jour.  Botany  23:  70-79,  1936. 
Emmons,  C.  W.:  Medical  mycology,  Botan.  Rev.  6:  474-514,  1940. 
EzEKiEL,  W.  N.,  and  J.  F.  Fudge:  Studies  on  the  cause  of  immunity  of  monocotyle- 

donous  plants  to  Phymatotrichum  root  rot.  Jour.  Agr.  Research  56 :  773-786, 

1938. 
Fawcett,  H.  S.  :  An  important  entomogenous  fungus,  ]\Iycologia  2:  164-168,  1910. 
*Fbldman,  a.  W.,  N.  E.  Caroselli,  and  F.  L.  Howard:  Physiology  of  toxin  produc- 
tion by  Ceratostomella  ulmi,  Phytopathology  40:  341-354,  1950. 
Fitzpatrick,  H.  M.:  The  Lower  Fungi,  Phycomycetes,  McGraw-Hill  Book  Com- 
pany, Inc.,  New  York,  1930. 
Fontaine,  T.  D.,  G.  W.  Irving,  Jr.,  and  S.  P.  Doolittle:  Partial  purification  and 

properties  of  tomatin,  an  antibiotic  agent  from  the  tomato  plant.  Arch.  Biochem. 

12 :  395-404,  1947. 
Gallegly,  M.  E.  :  Host  nutrition  in  relation  to  development  of  Verticillium  wilt  of 

tomato.  Phytopathology  39:  7,  1949. 
Gaumann,  E.  :  Pflanzliche  Infektionslehre,  Verlag  Birkhauser,  Basel,  1946.     Trans. 

by  W.  B.  Brierly  as  Principles  of  Plant  Infection,  Hafner  Publishing  Co.,  New 

York,  1950. 
Gaumann,  E.,  and  O.  Jaag:  Die  physiologischen  Grundlagen  des  parasitogenen 

Welkens.     II,  III.     Ber.  schweiz.  botan.  Ges.  52:  132-148,  227-241,  1947. 
Hart,  H.:  Relation  of  stomatal  behavior  to  stem-rust  resistance  in  wheat.  Jour.  Agr. 

Research  39:  929-948,  1929. 
*Hatfield,  W.  C,  J.  C.  Walker,  and  J.  H.  Owen:  Antibiotic  substances  in  onion 

in  relation  to  disease  resistance.  Jour.  Agr.  Research  77:  115-135,  1948. 
Hawkins,  L.  A.,  and  L.  B.  Harvey:  Physiological  study  of  the  parasitism  oiPythium 

debaryanum  on  potato  tuber.  Jour.  Agr.  Research  18:  275-297,  1919. 
Haymaker,  H.  H.  :  Relation  of  toxic  excretory  products  from  two  strains  of  Fusariu  m 

lycopersici  to  tomato  wilt.  Jour.  Agr.  Research  36:  697-719,  1928. 
HiGGiNS,  B.  B.:  Physiology  and  parasitism  of  Sclerotmm  rolfsii,  Phytopathology  17: 

417-448,  1927. 
Hodgson,  R.,  W.  H.  Peterson,  and  A.  J.  Riker:  The  toxicity  of  polysaccharides 

and  other  large  molecules  to  tomato  cuttings.  Phytopathology  39:  47-62,  1949. 
Houston,  B.  R.:  Culture  types  and  pathogenicity  of  isolates  of  Corticium  solani, 

Phijtopathology  35:  371-393,  1945. 
Hurd,  a.  M.:  The  course  of  acidity  changes  during  the  growth  period  of  wheat  with 

special  reference  to  stem-rust  resistance.  Jour.  Agr.  Research  27:  725-735,  1924. 
Jones,  F.  R.  :  The  leaf-spot  disease  of  alfalfa  and  red  clover  caused  by  the  fungi 

Pseudopeziza  medicaginis  and  Pseudopeziza  trifolii,   respectively,    U.S.   Dept. 

Agr.  Bull.  759,  1919. 
Karling,  J.  S.:  Plasmodiophorales,  published  by  the  author,  New  York,  1942. 


PARASITISM  AND  RESISTANCE  397 

Karling,  J.  S.:  Chj'trifliosis  of  scale  Insects,  Am.  Jour.  Botany  35:  246-254,  1948. 
Klotz,  L.  J.:  Inhibition  of  enzymatic  action  as  a  possible  factor  in  the  resistance  of 

plants  to  disease,  Science  66:  631-632,  1927. 
KuNKEL,  L.  O.:  A  contril)ution  to  the  life  history  of  Spongospora  subterranea,  Jour. 

Agr.  Research  4:  26.5-278,  1915. 
Leach,  J.  G.:  The  parasitism  of  Puccinia  grnminis  tritici  Erikss.  and  Ilenn.  and 

Puccinia  graminis  tritici-compacti  Stak.  and  Piem.,  Phytopathology  9:  59-88, 

1919. 
*Leach,  J.  G.:  The  parasitism  of  Colletotrichum  lindemuthianum,  Minn.  Agr.  Expt. 

Sta.  Bull.  14,  1923. 
Leach,  J.  G.:  Insect  Transmission  of  Plant  Diseases,  McGraw-Hill  Book  Company, 

Inc.,  New  York,  1940. 
Lee,  a.:  The  toxic  substance  produced  by  the  eye-spot  fungus  of  sugar  cane,  Helmin- 

thosporium  sacchari,  Plant  Physiol.  4:  193-212,  1929. 
*LiNK,  K.  P.,  and  J.  C.  Walker:  The  isolation  of  catechol  from  pigmented  onion 

scales  and  its  significance  in  relation  to  disease  resistance  in  onions,  Joiir.  Biol. 

Chem.  100:  379-383,  1933. 
Little,  J.  E.,  T.  J.  Sproston,  and  M.  W.  Foote:  Isolation  and  antifungal  action  of 

naturally  occurring  2-methoxy-l,4-naphthoquinone,   Jour.   Biol.   Chem.    174: 

335-342,  1948. 
Mains,  P>.  B.  :  The  relations  of  some  rusts  to  the  physiology  of  their  hosts,  Am.  Jour. 

Botany  ^:  179-220,  1917. 
Marryat,  D.  C.:  Notes  on  the  infection  and  histology  of  two  wheats  immune  to  the 

attacks  of  Puccinia  glumarum,  yellow  rupt,  Jour.  Agr.  Sci.  2:  129-138,  1907. 
Meehan,  F.,  and  H.  E.  Murphy:  Differential  phytotoxicity  of  metabohc  by-prod- 
ucts of  Helminthosporium  victoriae,  Science  106:  270-271,  1947. 
Melander,  L.  W.,  and  J.  H.  Craigie:  Nature  of  resistance  of  Berberis  spp.  to 

Puccinia  graminis,  Phytopathology  17:  95-114,  1927. 
NicKERsoN,  W.  J.:  Biology  of  Pathogenic  Fungi,  Chronica  Botanica  Co.,  Waltham, 

1947. 
Orton,  C.  R.:  Studies  in  the  morphology  of  the  Ascomycetes.     I.     The  stroma  and 

the  compound  fructification  of  the  Dothideaceae  and  other  groups,  Mycologia 

16:  49-95,  1924. 
Orton,  C.  R.,  and  F.  D.  Kern:  The  potato  wart  disease,  Penna.  Agr.  Expt.  Sta. 

Bull.  156,  1919. 
Pehrsox,  S.  O.:  Studies  of  the  growth  physiology  of  Phacidium  infestans  Karst., 

Physiologia  Plantarum  1:  38-56,  1948. 
Plattner,  p.  A.,  and  N.  Clausson-Kaas:  Ueber  ein  Welke  erzeugendes  Stoffwech- 

selprodukt  von  Fusarium  lycopersici  Sacc,  Helv.  Chim.  Acta  28:  188-195,  1945. 
Prasad,   N.:  Variability  of  the  cucurbit  root-rot  fungus,  Fusarium  (H ijpo7nyces) 

solani  f.  cucurbitae,  Phytopathology  39:  133-141,  1949. 
*Reddy,  C.  S.:  Resistance  of  dent  corn  to  Basisporium  gallarum  Moll.,  Iowa  Agr. 

Expt.  Sta.  Research  Bxdl.  167,  1933. 
Reynolds,  E.  S.:  Studies  on  the  physiology  of  plant  disease,  Ann.  Missouri  Botan. 

Garden  18:  57-95,  1931. 
Rice,  M.  A.:  The  haustoria  of  certain  rusts  and  the  relation  between  host  and 

pathogenes.  Bull.  Torrey  Botan.  Club  54:  63-153,  1927. 
*Rice,  M.  a.:  The  cytology  of  the  host-parasite  relations,  Botan.  Rev.  1:  327-353, 

1935. 
Rochlin,  E.  J.:  On  the  question  of  non-susceptibility  of  Cruciferae  to  Plasmodio- 

phora  hrassicae  Wor.,  Bull.  Plant  Protection,  (U.S.S.R.),  Ser.  II,  Phytopathology, 

Leningrad,  1933.     (Abst.  in  Rev.  Applied  Mycol.  13:  140-141,  1934.) 


398  PHYSIOLOGY  OF  THE  FUNGI 

Rosen,  H.  R.:  Efforts  to  determine  the  means  by  which  the  cotton  wilt  fungus, 
Fusarivm  vasinfedum,  produces  wilting,  Jour.  Agr.  Research  33:   1143-1162, 
1926. 
Sabouraud,  R.:  Maladies  du  cuir  chevelu.     Vol.  Ill,  Les  Maladies  cryptogamiques, 
Les  Teignes,  Masson  et  Cie,  Paris,  1910. 

Salmon,  E.  S.:  On  the  stages  of  development  reached  by  certain  biological  forms  of 
Erysiphe  in  cases  of  non-infection,  Neiv  Phytologist  4:  217-222,  1905. 

Schmidt,  M.:  Zur  Entwicklungsphysiologie  von  Cladosporium  Jidvxim  und  liber  die 
Widerstandfjihigkeit  von  Solanum  racemigerum  gegen  diesen  Parasiten,  Planta 
20:  407-439,  1933. 

Sharvelle,  E.  J.:  The  nature  of  resistance  of  flax  to  Melampsora  lint,  Jour.  Agr. 
Research  53:  81-127,  1936. 
*Smith,  F.  G.,  J.  C.  Walker,  and  W.  J.  Hooker:  Effect  of  hydrogen-ion  concentra- 
tion on  the  toxicity  to  Colletotrichum  circinans  (Berk.)  Vogl.  of  some  carboxylic 
acids,  phenols,  and  crucifer  extracts.  Am.  Jour.  Botany  33:  351-356,  1946. 

Smith,  L.  D.  S.:  Clostridia  in  gas  gangrene.  Bad.  Revs.  13:  233-254,  1949. 

Stakman,  E.  C.  :  a  study  in  cereal  rusts.  Physiological  races,  Minn.  Agr.  Expt. 
Sta.  Bull.  138,  1914. 

Steinhaus,  E.  A.:  Insect  Microbiology,  Comstock  Publishing  Company,  Inc.,  Ithaca. 
1946. 

Tapke,  V.  F.:  Influence  of  varietal  resistance,  sap  acidity  and  certain  environ- 
mental factors  on  the  occurrence  of  loose  smut  of  wheat.  Jour.  Agr.  Research 
39:313-319,  1929. 

Taylor,  C.  F.,  and  P.  Decker:  A  correlation  between  pathogenicity  and  cultural 
characteristics  in  the  genus  Actinomyces,  Phytopathology  37 :  49-58,  1947. 

Thatcher,   F.   S.  :   Osmotic  and  permeability  studies  in  the  nutrition  of  fungus 
parasites.  Am.  Jour.  Botany  26 :  449-458,  1939. 
*Thatcher,  F.  S.:  Further  studies  of  osmotic  and  permeability  relations  in  parasi- 
tism. Can.  Jour.  Research  20:  283-311,  1942. 

Thom,  C,  and  K.  B.  Raper:  A  Manual  of  the  Aspergilli,  The  Williams  &  Wilkins 
Company,  Baltimore,  1945. 

Thomas,  H.  E.:  Studies  on  Armillaria  mellea  (Vahl.)  Quel.,  infection,  parasitism  and 
host  resistance.  Jour.  Agr.  Research  48:  187-218,  1934. 

Vasudeva,  R.  S.  :  Studies  in  the  physiology  of  parasitism.  XL  An  analysis  of  the 
factors  underlying  specialism  of  parasitism,  with  special  reference  to  the  fungi 
Botrytis  allii  and  Monilia  frudigena,  Ann.  Botany  44:  469-493,  1930. 

Walker,  J.  C:  On  the  nature  of  disease  resistance  in  plants,  Trans.  Wisconsin  Acad. 
Sci.  21 :  225-247,  1924. 
*Walker,  J.  C. :  Soil  management  and  plant  nutrition  in  relation  to  disease  develop- 
ment. Soil  Sci.  61 :  47-54,  1946. 

Walker,  J.  C,  and  K.  P.  Link:  Toxicity  of  phenolic  compounds  to  certain  onion 
bulb  parasites,  Botan.  Gaz.  96 :  468-484,  1935. 

Ward,  M.  :  On  the  relations  between  host  and  parasite  in  the  bromes  and  their 
brown  rust,  Puccinia  dispersa  Erikss.,  Ann.  Botany  16:  233-315,  1902. 

Ward,  M.:  Recent  researches  on  the  parasitism  of  fungi,  Ann.  Botany  19:  1-54, 
1905. 

Waters,  C.  W.:  The  relations  of  bean  rust  grown  on  leaves  in  solution.  Papers  Mich. 
Acad.  Sci.  5:  163-177,  1926. 

Wellensiek,  S.  J. :  The  nature  of  resistance  in  Zea  mays  L.  to  Puccinia  sorghi  Schw., 
Phytopathology  17:  815-825,  1927. 
* Wingard,  S.  a.  :  The  nature  of  disease  resistance  in  plants,  Botan.  Rev.  7 :  59-109, 
1941. 


PARASITISM  AND  RESISTANCE  399 

Wolf,  F.  A.,  and  F.  T.  Wolf:  The  Fungi,  Vol.  II,  John  Wiley  &  Sons,  Inc.,  New 

York,  1947. 
WooLLEY,  D.  W. :  Strepogenin  activity  of  seryl  glycyl  glutamic  acid,  Jour.  Biol. 

Chem.  166:  783-784,  1946. 
Yarwood,  C.  E.  :  The  comparative  behavior  of  four  clover-leaf  parasites  on  excised 

leaves,  Phytopatholgoy  24 :  797-806,  1934. 
Yeager,  C.  C.  :  Empusa  infections  of  the  house-fly  in  relation  to  moisture  conditions 

of  northern  Idaho,  Mycologia  31:  154-156,  1939. 


CHAPTER  18 

PHYSIOLOGICAL  VARIATIONS  AND  INHERITANCE  OF 
PHYSIOLOGICAL  CHARACTERS 

Variation  in  the  results  of  experimental  work  with  fungi  is  of  frequent 
occurrence;  it  is  perhaps  even  more  frequent  than  uniformity.  Different 
investigators  conducting  the  same  experiments  with  the  same  species  of 
fungus  have  often  failed  to  obtain  the  same  results.  Such  variation  may 
be  attributed  to  (1)  genetic  differences  in  the  strains  or  isolates  used,  (2) 
slight  nutritional  differences  in  the  experiments,  or  (3)  differences  in  the 
physical  environment.  Examples  of  the  second  and  third  groups  of 
factors  have  been  pointed  out  frequently  in  the  earlier  chapters.  A  brief 
discussion  of  the  genetic  differences  involving  physiological  expression 
and  the  general  mode  of  inheritance  of  these  factors  (in  so  far  as  they  are 
known)  will  be  given. 

PHYSIOLOGICAL  VARIATION 

Variation  in  physiological  behavior  of  different  species  of  fungi  has 
been  noted  in  the  preceding  chapters.  The  present  discussion  emphasizes 
the  physiological  variation  within  a  species,  i.e.,  between  different  isolates, 
strains,  or  races,  which  show  little  or  no  morphological  difference. 

Nutritional  requirements.  Variations  in  the  nutritional  requirements 
of  different  isolates  of  the  same  species  are  numerous.  Differences  in 
vitamin  requirements  or  in  carbon  and  nitrogen  utilization  may  serve  as 
examples. 

Differences  in  deficiencies  for  one  or  more  vitamins  have  been  reported 
for  different  isolates  of  Fusarium  avenaceuni  (Robbins  and  Ma,  1941), 
Sclerotinia  minor  (Barnett  and  Lilly,  19-47),  Saccharomyces  cerevisiae 
(Leonian  and  Lilly,  1942;  Burkholder  and  Moyer,  1942),  Sordaria  fimicola 
(Hawker,  1939;  Barnett  and  Lilly,  1947a),  and  numerous  others.  For 
example,  certain  isolates  of  Sordaria  fimicola  from  nature  are  totally 
deficient  for  biotin  alone,  while  others  are  deficient  for  both  biotin  and 
thiamine.  A  somewhat  different  type  of  variation  is  reported  by  Thren 
(1941)  for  Ustilago  nuda.  The  haploid  mycelium  showed  no  deficiency 
for  vitamins,  while  the  diploid  mycelium  required  an  external  supply  of 
thiamine  or  pyrimidine.  The  plus  and  minus  strains  were  also  found  to 
differ  in  their  nutritional  requirements. 

400 


VARIATION  AND  INHERITANCE  401 

Different  isolates  of  Ustilago  striiformis  have  shown  strikingly  different 
responses  to  sources  of  carbon  and  nitrogen  (Cheo,  1949).  The  isolates 
from  bluegrass  segregated  into  two  groups  based  on  mycelial  type,  ''frag- 
menting" and  "mycelial."  The  "fragmenting"  type  grew  well  only 
on  media  containing  sucrose  and  organic  nitrogen,  while  the  "mycelial" 
type  could  utilize  a  number  of  sugars  and  nitrate  nitrogen.  Single-spore 
(haploid)  isolates  from  the  same  fruit  body  of  Lenzites  trdbea  collected  in 
nature  varied  nearly  fourfold  in  their  ability  to  synthesize  thiamine 
(Lilly  and  Barnett,  1948). 

Induced  deficiencies  for  a  number  of  vitamins  and  amino  acids  have 
been  demonstrated  by  Beadle  (1946)  in  mutants  of  Neurospora  and  by 
Bonner  (1946)  in  mutants  of  Penicilliiim.  The  mutations  were  induced 
by  exposure  of  spores  of  these  fungi  to  ultraviolet  and  X-ray  radiation. 
Mutants  that  showed  deficiencies  for  thiamine  and  differences  in  nitrogen 
requirements  were  also  reported  for  Aspergillus  terreus  (Thorn  and  Raper, 
1945).  One  mutant  differed  from  most  species  of  Aspergillus  in  its 
inability  to  utilize  nitrate  nitrogen.  Fries  (1948)  describes  spontaneous 
mutations  of  Ophiostoma  which  yield  the  same  strains  and  in  the  same 
proportion  as  those  induced  by  X  rays.  These  results  lead  us  to  conclude 
that  similar  mutations  are  the  principal  cause  of  variation  in  the  isolates 
obtained  from  nature. 

Response  to  environment.  Isolates  of  the  same  species  frequently 
vary  in  their  physiological  responses  to  some  environmental  factors, 
among  which  are  temperature  and  light.  For  example,  isolates  of 
Phytophthora  infestans  were  found  to  vary  in  their  resistance  to  high 
temperature  (Martin,  1949).  Of  the  eight  isolates  studied,  four  from 
Louisiana  withstood  exposure  to  36°C.  for  6  days,  while  three  isolates 
from  Minnesota  were  killed  after  4  days  and  one  isolate  from  New  York 
was  killed  in  less  than  6  hr.  at  the  same  temperature.  The  presence  of 
the  high-temperature  strain  is  believed  to  be  responsible  for  the  prevalence 
of  late  blight  in  Louisiana  during  the  past  few  years.  Houston  (1945) 
found  that,  for  one  group  of  isolates  of  Corticiiim  solani,  the  optimum 
temperature  for  growth  was  24  to  25°C.  and  the  maximum  was  33°C. 
For  two  other  groups  the  optimum  and  maximum  temperatures  were  28 
to  29°C.  and  40°C.,  respectively.  The  three  groups  also  varied  in  growth 
rates. 

Variation  in  response  to  light  is  illustrated  by  Choanephora  cucur- 
hitarum.  This  was  indicated  first  by  Wolf  (1917)  for  two  isolates.  The 
isolate  used  by  Christenberry  (1938)  produced  conidia  in  continuous 
total  darkness,  while  two  isolates  used  in  our  laboratory  failed  to  produce 
conidia  in  continuous  darkness  (Barnett  and  Lilly,  1950). 

Metabolic  products.  Both  qualitative  and  quantitative  variations  in 
the  metabolic  products  of  different  isolates  of  the  same  species  are  com- 


402  PHYSIOLOGY  OF  THE  FUNGI 

mon.  Industries  involved  in  the  commercial  production  of  alcohols, 
certain  organic  acids,  and  antibiotics  are  in  constant  search  for  higher 
yielding  "strains"  of  the  species  in  present  use,  as  well  as  of  other  species 
of  fungi.  Such  a  search  led  to  the  discovery  of  PenicilHum  chrijsogenum 
Q176  and  its  variants,  which  are  high  producers  of  penicillin.  Brewer's 
yeast  is  said  to  grow  in  media  with  an  alcohol  content  as  high  as  14  to 
17  per  cent,  while  the  baker's  yeast  is  checked  in  about  4  per  cent  alcohol 
(Wolf  and  Wolf,  1947).  Both  yeasts  belong  to  the  species  Saccharomyces 
cerevisiae. 

A  different  type  of  variation,  apparently  linked  with  sexuality,  is 
reported  in  Mucor  racemosus  (Harris,  1948).  Here,  the  production  of  an 
undetermined  antibiotic  seems  to  be  confined  to  the  minus  strain.  Varia- 
tion in  bioluminescence  is  reported  for  Panus  stypticus  (Macrae,  1942). 
The  fruit  bodies  and  mycelium  of  the  strain  found  in  North  America 
are  luminescent,  while  those  found  in  Europe  are  not  (Fig.  77).  Variants 
of  the  same  species  commonly  differ  in  pigment  production  (Christensen 
and  Graham,  1934;  Leonian,  1929).  Mutants,  or  saltants,  are  commonly 
lighter  in  color  than  the  parent  type. 

Sporulating  ability.  Many  investigators  have  noted  the  spontaneous 
development  of  nonsporulating  cultures  or  sectors  from  a  sporulating 
mycelium.  Variation  in  abundance  of  spores  produced  by  different 
isolates  from  nature  is  also  common.  For  example,  some  of  the  species 
which  illustrate  this  variability  are  Fusarium  spp.,  Phytophthora  spp., 
Phoma  terrestris,  Gibber ella  zeae,  Glomerella  cingulata,  Lenzites  trabea, 
Monilinia  fructicola,  and  Ustilago  striiformis.  Variations  in  fruit  bodies 
of  Cyathus  stercoreus  produced  in  culture  are  described  by  Brodie  (1948). 
Variation  in  production  of  sclerotia  has  been  observed  in  isolates  of 
Sclerotinia  trifoliorum  by  Kreitlow  (1949)  and  of  S.  sclerotiorum  in  our 
laboratory. 

Pathogenicity.  Variability  in  the  metabolic  products  such  as  enzymes 
and  toxins  and  in  the  ability  to  establish  compatible  food  relations  with 
the  host  may  be  of  great  importance  in  determining  pathogenicity.  Das 
Gupta  (1936)  discusses  the  pathogenicity  as  well  as  other  characteristics 
of  "saltants."  Such  soil-inhabiting  fungi  as  Fusarium  spp.  are  notorious 
for  their  variability  in  pathogenicity  within  a  species.  Species  of  Helmin- 
thosporium  (Christensen,  1922;  Dickinson,  1932)  and  Cortidum  solani 
(Houston,  1945)  are  likewise  highly  variable.  In  the  highly  parasitic 
fungi,  such  as  the  smuts,  rusts,  and  powdery  mildews,  there  is  a  high 
degree  of  physiologic  specialization  of  races. 

The  determination  of  physiologic  races  is  based  on  infection  types  of 
several  varieties  or  species  of  the  host.  Dickson  (1947)  reports  the 
existance  of  189  known  physiologic  races  of  Puccinia  graminis  tritici  and 
128  physiologic  races  of  P.  rubigo-vera  tritici.     Genetic  studies  indicate 


VARIATION  AND  INHERITANCE 


403 


that  the  physiologic  races  may  vary  in  Init  a  single  gene  and  that  they  may 
arise  by  hybridization  or  by  mutation. 

There  is  abundant  evidence  that  the  haploid  and  diploid  stages  of  some 
fungi  may  differ  in  pathogenicity.     The  haploid  phase  of  a  number  of 


Fig.  77.  Panus  stypticus  grown  on  malt  agar.  A,  diploid  mycelium,  4  weeks  old,  from 
a  pairing  between  a  haplont  of  the  luminous  American  form  and  a  haplont  of  the 
nonluminous  European  form,  photographed  by  reflected  light;  B,  the  same  culture  as  A 
photographed  by  its  own  light;  C,  a  2-weeks-old  pairing  between  a  nonluminous 
haplont,  on  the  left,  and  a  luminous  haplont,  on  the  right,  photographed  by  reflected 
light;  D,  the  same  pairing  as  C  photographed  by  its  own  light.  (Courtesy  of  Macrae, 
Can.  Jour.  Research,  Sec.  C,  20:  424,  1942.) 

smuts  is  apparently  unable  to  cause  infection,  while  the  diploid  mycelium 
is  pathogenic.  Since  the  haploid  and  diploid  mycelia  of  the  heteroecious 
rusts  parasitize  different  hosts,  we  must  conclude  that  they  also  differ  in 
pathogenicity. 


404  PHYSIOLOGY  OF  THE  FUNGI 

INHERITANCE  OF  PHYSIOLOGICAL  CHARACTERS 

The  genetics  of  the  fungi  has  been,  in  general,  a  neglected  study. 
Numerous  papers  have  appeared  on  the  sexuality  of  the  fungi,  particu- 
larly with  regard  to  the  various  sexual  or  compatibility  groups  in  the 
Basidiomycetes.  The  sexuality  of  the  Mucorales  has  been  studied  to  a 
lesser  extent.  Genetic  studies  of  morphological  characters  have  been 
decidedly  fewer.  Perhaps  this  is  due  to  the  failure  to  recognize  definite 
morphological  differences  betw^een  individuals  of  opposite  sex  but  of  the 
same  species.  An  equally  great  handicap  to  such  studies  lies  in  the 
difficulty  in  obtaining  the  perfect  stage  of  many  of  the  fungi  which  other- 
wise might  be  suitable.  Studies  dealing  with  inheritance  of  physiological 
characters  (if  sexuality  is  excluded)  are  comparatively  few  and  recent. 

The  basis  of  inheritance.  The  physical  basis  of  inheritance  is  the 
gene,  located  at  a  specific  position  on  a  certain  chromosome.  In  mitosis 
the  chromosomes  and  their  genes  divide,  and  half  of  each  goes  to  each 
daughter  nucleus.  With  the  exception  of  parthenogenesis,  all  perfect 
stages  of  the  fungi  arise  as  a  result  of  the  union  of  two  nuclei.  These 
two  nuclei  may  arise  from  the  same  haploid  individual  (homothallism) 
or  from  separate  haploid  thalli  (heterothallism).  The  union  of  the  two 
haploid  nuclei,  each  with  a  single  set  of  chromosomes,  initiates  the  diploid 
nucleus,  or  the  syncaryotic  stage,  in  which  the  chromosomes  are  paired. 
The  syncaryotic  stage  in  fungi  is  usually  short  in  duration,  being  followed 
closely  by  meiosis,  which  involves  the  separation  of  the  two  chromosomes 
(and  genes)  of  each  pair.  Certain  pairs  of  chromosomes  may  separate  in 
the  first  division,  while  others  separate  in  the  second.  Therefore,  two 
successive  nuclear  divisions  are  necessary  to  complete  the  reduction  of 
all  pairs  of  chromosomes  (and  likewise  all  the  pairs  of  genes).  In  the 
Ascomycetes  and  the  Eubasidiomycetes  karyogamy  and  meiosis  occur 
in  the  same  cell,  the  ascus  and  the  basidium,  respectively.  In  the  smuts 
and  rusts,  meiosis  typically  takes  place  in  a  promycelium,  while  kary- 
ogamy occurs  in  the  teliospore.  When  a  single  pair  of  genes  is  considered, 
half  the  haploid  ascospores  or  basidiospores  carry  one  gene  and  half  carry 
the  other  gene. 

Inheritance  in  the  Ascomycetes.  Some  of  the  outstanding  genetic 
work  has  been  done  by  Dodge  (1927,  1928)  and  others  with  Neurospora, 
by  Ames  (1934)  and  Doweling  (1931)  with  Pleurage  anserina,  by  Edgerton 
et  at.  (1945),  Chilton  and  Wheeler  (1949),  and  their  associates  with 
Glomerella,  and  by  Lindegren  (1945,  1948)  and  his  colleagues  with  yeasts. 
Most  of  these  investigations  have  been  concerned  primarily  with  sexual 
or  morphological  characters.  The  life  cycle  of  Neurospora  is  shown 
diagrammatically  in  Fig.  78. 

Beadle  and  his  associates  have  contributed  much  to  our  knowledge 


VARIATION  AND  INHERITANCE 


405 


of  the  inheritance  of  physiological  characters  in  the  Ascomycetes.  Beadle 
(1946)  believed  that,  if  the  ability  to  synthesize  a  certain  amino  acid  or 
growth  factor  were  due  to  the  action  of  a  single  gene,  it  should  be  possible 
to  modify  the  gene  in  such  a  way  that  the  fungus  could  no  longer  syn- 
thesize that  compound.  Previous  work  of  other  geneticists  with  corn, 
Drosophila,  and  other  organisms  had  shown  that  exposure  to  X  rays  or 
ultraviolet  radiation  caused  mutations  by  either  destroying  the  gene  or 
modifying  it  so  that  it  could  no  longer  function  normally.  Beadle  found 
that  exposure  of  conidia  of  Neurospora  crassa  and  A^.  sitophila  to  X  rays 
or  ultraviolet  rays  had  the  similar  effect  of  causing  mutations  that  were 


Germinating 

a  SCO  spore-  .-^ 


Germinating 
ascospore 


Conidia 


Conidia 


Protoperittiecium  A     — 


~-  Protoperithecium  a 


Hypnal  fusion 

Fig.  78.  Diagram  of  life  cycle  of  Neurospora.  (Courtesy  of  Beadle,  Am.  Scientist 
34 :  36,  1946,  and  Science  in  Progress,  1947.  Published  by  permission  of  the  Society  of 
the  Sigma  Xi.) 

expressed  in  the  inability  of  the  fungus  to  synthesize  vitamins,  amino 
acids,  and  other  essential  metabolites. 

The  'Svild  type"  of  Neurospora  is  deficient  for  biotin  but  is  self-suffi- 
cient for  all  other  vitamins  and  for  its  necessary  amino  acids.  The 
conidia  were  exposed  to  the  ultraviolet  rays  of  a  Sterilamp  for  such  a 
time  that  most  of  the  spores  were  killed.  The  spores  were  then 
sown  over  the  surface  of  agar  plates  in  such  concentration  as  to  give 
individual  "colonies,"  which  were  isolated  and  allowed  to  grow.  When 
these  were  transferred  to  a  minimal  medium,  containing  sucrose,  nitrate, 
mineral  salts,  and  biotin,  the  failure  of  an  isolate  to  grow  showed  an 
induced  variation  from  the  wild  type  in  its  capacity  to  synthesize  essen- 
tial metabolites. 

The  variant  cultures  were  then  selected  and  crossed  with  the  wild 


406 


PHYSIOLOGY  OF  THE  FUNGI 


strain  of  the  opposite  sex  to  determine  if  the  changes  were  inherited.  The 
ascospores  from  these  crosses  were  planted  on  both  the  minimal  medium 
and  a  complete  medium.  The  appearance  of  the  deficiency  in  half  of 
the  cultures  was  considered  as  evidence  that  the  change  was  of  genetic 
origin;  i.e.,  a  mutation.  Transfers  of  the  mutant  to  four  different  media 
(minimal,  with  amino  acids,  with  vitamins,  and  complete)  then  deter- 
mined whether  the  deficiency  was  for  an  amino  acid  or  a  vitamin.  All 
media  contained  biotin.  For  a  diagrammatic  scheme  of  the  procedure 
see  Fig.  79. 


X-ravs  or 
ultraviolet 


0 


-^  ©  ©  ©     - 

Coniolict 
(asexuoil  spores) 


Wild  fv/pe 


O 


J- 


^ 


\.y 


Crossed 
with  wild 
t^pe  of 
opposite 
sex 


Frui+inq  body 

I 


-<r- 


V. 


Sexut^l  spore 


Complete 

meolium 

(with  vitoimins, 

oimino  cicids, 

etc.) 


T 

O 


T 


T" 

n 


A 


\y 


Minim(7il      Complete 


MinirriCTl  Vitoimins        Amino 

medium  acids 

Fig.  79.     Experimental  procedure  by  which  biochemical  mutants  are  produced  and 

detected  in  Neurospora.     (Courtesy  of  Beadle,  Am.  Scientist  34 :  37,  1946,  and  Science 

in  Prug'-ess,  1947.     Published  by  permission  of  the  Society  of  the  Sigma  Xi.) 

7  he  identification  of  the  specific  deficiency  involved  the  growth  of  the 
iiiyt2,nt  upon  the  minimal  medium  plus  each  of  the  amino  acids  and 
vitamins  added  singly.  This  procedure  is  shown  in  Fig.  80.  Figure  81 
shows  the  proof  of  inheritance  of  the  deficiency  for  pantothenic  acid. 

Mutations  involving  tli'^  following  vitamins  have  been  described: 
thiamine,  riboflavin,  pyridoxin'^,  niacin,  pantothenic  acid,  p-amino- 
ben.^oic  acid,  inositol,  and  choline.  All  mutants,  as  well  as  the  wild 
type,  are  deficient  for  biotin.  In  addition,  mutations  have  appeared 
which  cannot  synthesize  tlis  following  amino  acids:  arginine,  isoleucine, 


VARIATION  AND  INHERITANCE 


40; 


Q  Q  Q   Q  Q 


Thiamin 


^ 


Complete 
medium 


Ribo- 
flavin 


Pyri- 
doxin 


Panto-    Niacin 
thenif 
acid 


p-Amino- 

benzoic 

acid 


Inositol    Choline 


Folic 
acid 


I        © 


Nucleic 
acid 


I 

Minimal 
control 


Fig.  80.  Tests  of  mutant  on  individual  vitamins  or  growth  factors.  Growth  only  on 
pantothenic  acid  indicates  a  single  deficiency  for  this  vitamin.  (Courtesy  of  Beadle, 
Am.  Scientist  34:  39,  1946,  and  Science  in  Progress,  1947.  Published  by  permission 
of  the  Society  of  the  Sigma  Xi.) 


o 


I 
f 


Fruiting 


D 


O 


QQOOQOQO 


Uhu)  Mb*  iHi  iki  iSS 

|r    jF    I    I    p 

KJ        kJ        \J        \J        kJ 

With  pantothenic  oicid 

i     }     1     I     1     1     1     1 

QOOOQOQQ 


Panfothenicless 


p 


¥ 


B 


Wild 
t:ype 

Without  pantothenic  acid 
Fig.  81.     Scheme  by  which  the  inheritance  of  a  mutant  type  is  determined.     The  1  to 
1  ratio  with  regard  to  need  for  pantothenic  acid  indicates  simple  Mendelian  inheritance. 
(Courtesy  of  Beadle,  Am.  Scientist  34:  40,  1946,  and  Science  in  Progress,  1947.     Pub- 
lished by  permission  of  the  Society  of  the  Sigma  Xi.) 


408  PHYSIOLOGY  OF  THE  FUNGI 

leucine,   lysine,   methionine,   phenylalanine,   proline,   threonine,   trypto- 
phane, and  valine. 
Beadle  (1946)  states: 

The  list  of  compounds  that  Neurospora  can  be  made  to  require  from  an  external 
source  is  remarkably  similar  to  a  list  of  chemicals  that  we  cannot  make  and 
require  in  our  food  suppl3^  It  is  clear,  therefore,  that  the  substances  the  bread 
mold  needs  in  its  metabolism  are  very  much  the  same  as  those  we  need.  The 
difference  is  only  an  apparent  one  and  results  from  the  fact  that  bread  mold 
makes  them  whereas  we  let  some  other  organism  make  them  for  us.  By  inacti- 
vating the  right  genes  the  bread  mold  can  be  made  very  similar  to  man  in  its 
nutritional  requirements. 

Using  a  technique  similar  to  that  described  above  for  Beadle's  work, 
Bonner  (1946)  exposed  conidia  oiPeniciUium.  notatum  and  P.  chrysogcmnn 
to  X  rays  and  ultraviolet  rays.  Of  a  total  of  85,595  "strains"  tested, 
398  were  found  to  be  deficient  in  synthetic  ability.  Since  these  species 
of  PeniciUium  are  imperfect,  the  genetic  basis  for  the  biochemical  changes 
cannot  be  proved,  but  it  seems  likely  that  this  is  the  case,  just  as  in 
Neurospora. 

Inheritance  in  the  Basidiomycetes.  In  the  life  cycle  of  the  Basidio- 
mycetes  there  exists  a  distinct  diploid  (dicaryotic)  vegetative  phase  of 
extended  duration,  in  which  the  cells  usually  contain  two  haploid  con- 
jugate nuclei.  Buller  (1941)  cites  the  results  of  numerous  experiments 
by  himself  and  others  to  furnish  ample  proof  that  one  nucleus  of  the 
conjugate  haploid  pair  in  the  diploid  mycelium  may  affect  the  expression 
of  the  other  nucleus  and  thus  exhibit  dominance.  This  is  true  for  physio- 
logical as  well  as  morphological  characters.  Experimental  evidence 
indicates  that  the  genetic  behavior  of  a  cell  containing  two  conjugate 
haploid  nuclei  is  similar  to  that  of  the  diploid  nucleus,  if  the  two  were 
fused.  Buller  believes  that  the  term  "diploid  cell"  can  apply  equally 
well  to  a  cell  containing  two  conjugate  haploid  nuclei  and  to  a  cell  con- 
taining one  diploid  nucleus.  He  prefers  to  use  the  terms  "haploid"  and 
"diploid"  in  describing  mycelium  or  cells  to  the  terms  "monocaryotic" 
and  "dicaryotic"  which  are  also  in  use.  We  prefer  to  follow  Buller  in 
the  use  of  these  terms. 

In  the  higher  Basidiomycetes,  principally  the  Agaricales,  genetic  studies 
have  been  chiefly  limited  to  the  inheritance  of  sex  factors  or  compatibility 
factors.  Fewer  studies  have  dealt  with  the  more  strictly  physiological 
characters.  In  a  heterothallic  species,  two  compatible  haploid  mycelia 
unite  to  initiate  the  diploid  mycelium,  which  in  many  species  is  recog- 
nized by  the  presence  of  clamp  connections.  Usually,  the  formation  of 
the  diploid  mycelium  is  a  prerequisite  to  the  production  of  fruit  bodies. 

Exidia,  in  the  Tremellales,  Avill  serve  as  an  example  of  the  higher 
Basidiomycetes.     In  four  species  studied    (Barnett,    1937)   the  single- 


VARIATION  AND  INHERITANCE  409 

spore  haploid  cultures  from  the  same  fruit  body  fell  into  two  compatibility- 
groups.  Such  a  condition  is  described  as  bipolar.  Diploid  mycelium 
was  formed  only  when  two  haploid  mycelia  of  different  compatibility 
groups  were  paired.  If  A  and  a  represent  the  genes  for  compatibility, 
the  combination  of  Aa  would  be  necessary  for  the  formation  of  diploid 
mycelium.     AA  and  aa  would  be  incompatible. 

^\^lile  the  single-spore  cultures  of  a  single  fruit  body  of  Exidiaglandu- 
losa  give  rise  to  two  compatibility  groups  A  and  a,  a  second  fruit  body 
collected  at  some  distance  away  may  give  rise  to  haploid  mycelia  which 
apparently  fall  into  the  same  groups  {A  and  a).  Yet  we  may  find  that 
all  the  haploid  mycelia  of  the  first  fruit  bod}^  are  compatible  with  all  the 
haploid  mycelia  of  the  second  fruit  body.  In  other  words,  the  two  groups 
of  the  second  fruit  body  are  slightly  different  from  the  two  groups  of  the 
first  fruit  body.  It  is,  therefore,  likely  that  numerous  compatibility 
groups  exist,  even  though  only  two  occur  in  any  one  fruit  body.  Com- 
patibility, in  this  case,  is  apparently  determined  by  multiple  alleles.  The 
existence  of  geographic  races  has  been  described  for  a  number  of  fungi 
by  Buller  (1941)  and  others. 

A  somewhat  different  situation  exists  in  Collybia  velutipes  and  a  number 
of  other  species.  It  has  been  found  that  each  spore  on  a  basidium  may 
differ  in  its  compatibility  factors.  Compatibility  in  this  case  is  deter- 
mined by  two  pairs  of  genes  on  different  chromosomes.  These  groups 
are  usually  designated  as  AB,  Ah,  aB,  and  ah.  The  combination  of 
AaBb  is  then  necessary  for  compatibility  and  formation  of  diploid 
mycelium. 

Compatibility  in  itself  does  not  necessarily  indicate  that  fertile  fruit 
bodies  will  be  formed.  For  instance,  some  of  the  single-spore  isolates 
of  Lenzites  trahea  were  found  to  produce  fertile  fruit  bodies,  while  other 
cultures  failed  to  do  so  (Barnett  and  Lilly,  1949).  By  pairing  compatible 
fruiting  isolates  and  also  the  compatible  nonfruiting  isolates,  it  was  possi- 
ble to  establish  a  correlation  between  the  fertility  of  the  diploid  mycelia 
with  that  of  the  haploid  "parents."  It  seems  probable,  therefore,  that 
the  ability  to  produce  fertile  fruit  bodies  has  a  genetic  basis,  in  addition 
to  that  of  compatibility. 

In  Schizophyllum  commune  the  ability  to  produce  normal  fruit  bodies  is 
dominant  over  the  formation  of  abnormal,  knot-like  fruit  bodies  (Zatler, 
in  Buller,  1941).  If  G  represents  the  factor  for  normal  fruit  bodies  and 
g  the  factor  for  knot-like  fruit  bodies,  the  results  could  be  expressed  as 
follows:  G  crossed  with  G  or  G  crossed  with  g  gives  normal  fruit  bodies, 
while  g  crossed  with  g  gives  knot-like  fruit  bodies.  Zatler  also  showed 
that  in  Collybia  velutipes  inheritance  of  pigmentation  of  his  cultures  was 
due  to  two  genes  located  on  different  chromosomes.  A  combination  of 
the  two  dominant  factors  in  the  haploid  mycelium  resulted  in  a  deep 


410  PHYSIOLOGY  OF  THE  FUNGI 

brown  color;  one  dominant  and  one  recessive  factor  gave  lighter  shades 
of  brown ;  while  a  combination  of  the  two  recessive  factors  gave  pure  white 
mycelium. 

The  normal  haplont  of  Peniophora  allescheri  is  reported  (Nobles,  1935) 
as  slow-growing  with  scant  mycelium  bearing  conidia.  A  mutant  grew 
rapidly  with  abundant  mycelium  but  bore  no  conidia.  The  combination 
of  normal  haplont  and  mutant  haplont  yielded  diploid  mycelium  which 
grew  rapidly  and  abundantly  and  produced  conidia.  Thus,  rapid  growth 
and  conidial  production  were  dominant  over  slow  growth  and  nonproduc- 
tion  of  conidia. 

Bioluminescence  of  the  North  American  race  of  Panus  stypticus  was 
found  (Macrae,  1942)  to  be  dominant  over  nonluminescence  of  the 
European  race  when  the  two  haplonts  were  crossed  (Fig.  77). 

The  single-spore  isolates  from  a  single  fruit  body  of  Lenzites  trahea  were 
found  (Lilly  and  Barnett,  1948)  to  vary  nearly  fourfold  in  their  ability  to 
synthesize  thiamine.  WTien  a  haplont  of  low  synthetic  ability  was 
crossed  with  one  of  high  synthetic  ability,  no  definite  evidence  of  domi- 
nance was  observed.  By  making  other  types  of  crosses  it  was  found  that, 
in  general,  the  synthetic  ability  of  the  Fi  haplonts  was  similar  to  that  of 
the  "parent"  haplonts.  Yet,  when  the  "parent"  haplonts  differed 
widely  in  synthetic  ability,  the  Fi  haplonts  did  not  segregate  into  the 
1  to  1  ratio,  as  would  be  the  case  if  inheritance  were  due  to  a  single  gene. 
Single-spore  cultures  from  haploid  fruit  bodies  produced  by  certain 
haplonts  were  more  uniform  in  their  ability  to  synthesize  thiamine  than 
were  single-spore  cultures  from  diploid  fruit  bodies  of  known  origin.  It 
was  concluded  that  the  ability  to  synthesize  thiamine  by  L.  trahea  is 
genetically  controlled,  and  that  the  mode  of  inheritance  is  complex  and 
not  due  to  a  single  gene. 

The  smuts  have  received  much  attention  in  genetic  studies  by  Stakman, 
Christensen,  and  their  associates  at  the  University  of  Minnesota.  Such 
characters  as  sex,  pigmentation,  pathogenicity,  and  morphological  fea- 
tures of  the  mycelium  and  spores  have  been  included  in  the  study.  Little 
is  known  regarding  the  factors  governing  the  more  strictly  physiological 
or  nutritional  processes.  An  excellent  review  of  the  genetics  of  the  smuts 
is  given  by  Christensen  and  Rodenhiser  (1940).  These  authors  discuss 
the  work  of  Goldschmidt,  who  found  that  the  diploid  mycelium  derived 
from  two  haplonts  of  different  races  of  Ustilago  violaceae  was  able  to 
attack  the  hosts  which  were  susceptible  to  each  parent  race.  This 
indicates  that  the  diplont  contained  the  combined  pathogenic  characters 
of  the  two  haplonts.  Hanna  (1932)  made  an  interspecific  cross  between 
Tilletia  levis,  with  smooth  spores  and  an  odor  of  trimethylamine,  and  T. 
tritici,  which  had  rough  spores  and  no  odor  of  trimethylamine.  The 
Fi  "chlamydospores"  had  smooth  walls  and  emitted  an  odor  of  tri- 
methylamine, showing  that  both  characters  were  dominant. 


VARIATION  AND  INHERITANCE  411 

Numerous  articles  on  the  inheritance  of  the  rusts  may  be  cited.  Among 
the  characters  commonly  studied  are  color  of  urediospores  and  patho- 
genicity. One  striking  example  of  inheritance  of  pathogenicity  of  races 
of  Puccinia  graminis  tritici  is  reported  by  Johnson  and  Newton  (1940). 
Using  pathogenically  homozygous  mycelia  of  race  9  and  race  36,  it  was 
found  that  Kanred  wheat  was  not  attacked  by  race  9  but  was  highly 
susceptible  to  race  36.  Urediospores  were  obtained  from  a  hybrid  of 
these  two  races  and  were  sown  on  different  varieties  of  wheat.  No  infec- 
tion occurred  on  Kanred,  indicating  that  the  nonpathogenicity  of  race  9 
w^as  dominant  over  the  pathogenicity  of  race  36.  In  the  r2  uredio- 
spores the  pathogenicity  to  Kanred  wheat  segregated  in  a  1  to  3  ratio, 
indicating  true  ]\Iendelian  inheritance. 

The  basis  of  variation  in  the  imperfect  fungi.  Any  change  or  variation 
in  the  imperfect  fungi  may  be  either  temporary  or  permanent.  Tem- 
porary variations  do  not  involve  gene  changes,  ^vhile  the  permanent 
variations  are  believed  to  have  their  basis  in  the  gene,  or  at  least  in  the 
nucleus.  The  Mendelian  inheritance  of  these  variations  cannot  be  proved 
in  those  fungi  with  no  sexual  stage. 

Most  investigators  studying  the  permanent  variations  which  arise  in 
culture  or  which  are  recognized  in  different  isolates  of  many  of  the  imper- 
fect fungi  would  explain  the  origin  of  these  variants  as  mutations.  For 
example,  Dickinson  (1932)  studied  "saltation"  in  the  genera  Fusarium 
and  Helminthosporium  and  noted  frequent  anastomoses  between  hyphae 
of  different  "saltants."  This  author  discussed  the  possibility  of  cyto- 
plasmic inheritance  but  concluded  that  the  permanent  variations  were 
due  to  actual  mutations. 

Hansen  (1938)  would  explain  many  such  variations  in  the  imperfect 
fungi  on  a  different  basis  and  presents  abundant  evidence  to  substantiate 
his  argument.  Only  the  essential  features  of  Hansen's  "dual  phenom- 
enon" will  be  presented  below.  The  conidia  and  mycelial  cells  of  many 
of  the  imperfect  fungi  contain  two  or  more  nuclei.  These  nuclei  may 
not  all  be  alike.  Considering  the  nucleus  rather  than  the  cell  as  the 
basic  unit  of  the  individual,  an  isolate  may  be  composed  of  two  culturally 
distinct  individuals.  This  condition  is  referred  to  as  the  dual  phenom- 
enon. A  heterocaryotic  fungus,  when  single-spored,  would  give  rise  to 
homotypes  of  each  of  the  individuals  and  the  heterotype  like  the  parent 
isolate.  One  homotype  is  characterized  by  abundant  mycelium  and  few 
or  no  conidia  and  is  called  the  M  (mycelial)  type.  The  other  is  charac- 
terized by  abundant  conidia  and  often  a  lesser  amount  of  mycelium  and  is 
called  the  C  (conidial)  type.  The  heterotype  is,  in  general,  intermediate 
between  the  M  and  C  types  and  is  called  the  MC  type.  Cultures  of  the 
M  and  C  types  give  only  the  parent  type  when  single-spored.  The 
frequency  w4th  w^hich  the  dual  phenomenon  is  encountered  in  the  imper- 
fect fungi  suggests  that  this  is  the  natural  condition  for  many  fungi. 


412  PHYSIOLOGY  OF  THE  FUNGI 

In  a  later  paper,  Hansen  and  Snyder  (1943)  state  that  the  change  from 
the  C  to  the  M  type  is  a  true  mutation.  In  Hypomyces  solani  f .  cucurhitae 
this  change  also  involved  the  change  from  the  hermaphroditic  phase  to 
the  unisexual  male  phase.  They  conclude  that  the  M  and  C  genes  are 
alleles  and  are  inherited  independently  of  the  factors  for  compatibility. 
On  the  contrary,  Robbins  and  McVeigh  (1949)  have  presented  evidence 
that  variants  of  Trichophyton  mentagrophytes  arise  as  mutations  and  that 
the  dual  phenomenon  does  not  exist  in  this  fungus. 

Nutritional  adaptations.  Most  of  our  present  knowledge  regarding 
nutritional  adaptations  has  come  from  experiments  with  yeasts,  and  to  a 
lesser  extent  with  Neurospora  and  bacteria.  Leonian  and  Lilly  (1942) 
were  able  to  "train"  eight  strains  of  Saccharomyces  cerevisiae  so  that  they 
grew  without  the  addition  of  one  or  more  of  the  vitamins  which  they 
formerly  required.  The  techniciue  employed  consisted  in  increasing  the 
amount  of  initial  inoculum,  prolongation  of  the  incubation  period,  and 
repeated  subculturing  on  media  deficient  for  one  of  the  necessary  vita- 
mins. Reversions  occurred  in  most  of  the  strains  after  being  cultured 
continuously  on  media  containing  all  the  vitamins.  Such  adaptations 
as  these  may  or  may  not  involve  gene  changes. 

Lindegren  and  his  associates  have  written  numerous  articles  on  the 
genetics  and  adaptations  of  yeasts.  From  over  400  isolates  of  Sac- 
charomyces cerevisiae,  Skoog  and  Lindegren  (1947)  found  12  which  could 
not  utilize  glucose.  Eleven  of  these  isolates  reverted  to  glucose  utiliza- 
tion within  a  period  of  a  few  days.  One  isolate  remained  glucose-negative 
for  a  period  of  3  months  when  grown  on  lactate  medium.  They  believe 
that  the  reversion  to  glucose  utilization  involves  more  than  a  single-step 
change. 

Spiegelman  (1950)  points  out  that  "a  basic  assumption  of  modern 
biology  is  that  genes  function  by  controlling  enzyme  synthesis.  From 
this  point  of  view  it  is  obvious  that  enzymatic  adaptation  has  profound 
implications  for  one  of  the  central  themes  of  biological  thinking."  This 
does  not  mean  that  the  presence  of  the  gene  is  always  accompanied  by 
the  presence  of  the  enzyme  in  the  cell,  but  merely  that  the  potentialities 
for  the  production  of  the  enzyme  are  present.  The  synthesis  of  the 
specific  enzyme,  as  well  as  its  subsequent  activity,  depends  upon  other 
factors,  a  major  one  being  the  type  of  substrate.  From  numerous  experi- 
ments it  is  evident  that  the  specific  enzyme  either  is  produced,  or  becomes 
detectable,  only  when  its  corresponding  substrate  is  present  in  the 
medium.  It  appears,  however,  that  the  specific  enzyme  may  be  formed 
even  when  the  corresponding  substrate  is  not  present  but  that,  under 
such  conditions,  the  adaptive  enzymes  are  usually  not  detected.  In  the 
course  of  a  "long-term  adaptation,"  there  is  not  only  synthesis  of  the 
specific  enzyme  but  also  an  increase  in  the  rate  of  enzyme  formation.     In 


VARIATION  AND  INHERITANCE  413 

this  respect,  Spiegelman  states,  "In  particular,  the  rate  of  formation 
of  a  given  enzyme  is  an  autocatalytic  function  of  the  amount  of  that 
particular  enzyme  present  in  the  cytoplasm." 

Ryan  (1946)  found  that  certain  adaptations  of  the  "  prolineless "  and 
"  thiamineless  "  mutants  of  Neurospora  are  not  inherited.  For  example, 
the  "thiamineless"  mutant  may  not  grow  for  several  days  after  being 
placed  on  a  thiamine-free  agar  medium  and  may  finally  begin  to  grow. 
This  is  explained  on  the  basis  of  adaptive  enzyme  formation.  For  further 
discussion  of  adaptive  enzymes,  see  Chap.  4.  For  a  more  complete 
discussion  of  cytoplasmic  inheritance  and  adaptive  enzymes  in  yeast,  the 
reader  is  referred  to  Lindegren  (1945,  1949)  and  Spiegelman  (1950). 

Back  mutations.  The  mutations  studied  by  Beadle  and  others  were 
those  involving  deficiencies  for  growth  factors  which  the  wild  type  was 
able  to  synthesize.  It  has  been  shown  that,  under  certain  conditions, 
there  may  be  a  reversion  from  the  deficient  type  to  the  wild  type.  This 
may  be  clue  to  a  noninherited  condition  (an  adaptation),  such  as  that 
described  above,  or  it  may  involve  a  gene  change  (a  back  mutation). 
Ryan  (1946)  has  discussed  at  some  length  the  topic  of  back  mutation  and 
adaptation  in  certain  organisms.  Only  a  few  of  his  ideas  will  be  presented 
here,  omitting  the  detailed  results  of  his  experiments.  He  points  out 
that  the  change  from  a  deficient  to  a  self-sufficient  habit  for  growth  fac- 
tors may  be  induced  experimentally  in  both  the  fungi  and  the  bacteria. 
In  some  cases  these  nutritional  changes  are  inherited,  indicating  gene 
changes.  In  the  case  of  the  "leucineless"  mutant  of  Neurospora  the 
adaptation  back  to  the  autotrophic  habit  was  determined  to  be  due  to  a 
reverse  mutation  of  the  leucineless  factor  to  the  wild  type.  Ryan  believes 
that  the  ultraviolet  rays,  in  causing  the  original  "leucineless"  mutation, 
caused  a  change  in  the  wild-type  gene  so  that  it  still  retained  the  ability 
to  reproduce  but  was  unable  to  act  in  the  synthesis  of  leucine.  He  found 
that  the  "adaptation  frequency"  varied  inversely  with  the  amount  of 
leucine  present  in  the  medium.  This  theory  assumes  merely  the  inactiva- 
tion  and  reactivation  rather  than  the  destruction  of  a  gene  by  the  ultra- 
violet rays. 

Lindegren  (1949)  found  that  a  mutation  from  pantothenate  deficiency 
to  pantothenate  independence  was  at  a  different  locus,  and  that  the 
synthesis  of  pantothenate  by  the  mutant  was  by  a  different  route  than 
in  the  original  wild  type. 

Giles  and  Lederberg  (1948)  have  recently  studied  the  effects  of  various 
mutagenic  agents  in  inducing  adaptations  (reversions)  of  deficient 
mutants  of  Neurospora  crassa.  They  found  that  the  frequency  of  adapta- 
tion of  certain  mutants  vaay  be  greatly  increased  by  ultraviolet  radiation. 
This  was  true  with  the  "inositolless,"  "cholineless,"  "methionineless," 
and ' '  ribofla vinless ' '  mutants.     The ' '  pantothenicless ' '  mutants  remained 


414  PHYSIOLOGY  OF  THE  FUNGI 

unchanged  by  the  same  treatment.  Indications  are  that  these  changes 
represent  mutations  to  the  wild  type.  These  adaptations  may  also  be 
initiated  by  X  rays,  nitrogen  mustard,  and  radiophosphorus. 

Chemically  induced  mutations.  Nitrogen  and  sulfur  mustard  gases 
have  been  used  to  induce  mutations  in  various  fungi.  The  method  of 
treatment  is  simple  and  consists  in  exposing  spores  or  mycelium  to  a 
buffered  solution  of  the  chemical  for  30  min.  or  longer.  The  spores  or 
mycelium  are  then  washed  and  plated  out.  The  methods  used  in  detect- 
ing mutants  are  then  the  same  as  when  X  rays  or  ultraviolet  irradiations 
are  used.  Treatment  of  young  conidia  or  germinating  conidia  of  Neuros- 
pora  crassa  with  nitrogen  mustard  produced  more  mutants  than  treat- 
ment of  old  or  ungerminated  conidia  (McElroy  et  al.,  1947).  The  tech- 
nique of  using  the  vapor  of  mustard  gas  to  induce  mutation  is  described 
by  Hockenhull  (1948).  Mustard  gas  in  a  buffered  solution  (pH  6.9  to 
7.0)  was  used  by  Hockenhull  (1949)  to  produce  mutants  of  Aspergillus 
nidulans. 

The  mustard  gases,  in  common  with  nitrous  acid,  react  with  proteins. 
For  example,  casein  which  has  been  treated  with  mustard  gas  no  longer 
supports  the  growth  of  the  chick  or  rat.  This  is  due  to  the  inactivation 
of  certain  essential  amino  acids  (Kinsey  and  Grant,  1946). 

Mutation-inducing  chemicals  may  be  encountered  by  fungi  under 
natural  conditions.  It  was  shown  that  toxic  metabolic  products  of 
Bacillus  mesentericus  affected  the  production  of  mutants  by  certain  strains 
of  Helminthosporium  sativum  (Christensen  and  Davis,  1940).  The 
filtrate  of  B.  mesentericus  cultures  induced  sectoring  of  H.  sativum.  These 
mutants  differed  from  the  parent  in  morphology,  pathogenicity,  and  other 
physiological  characters. 

Among  the  chemical  compounds  which  induce  mutations  in  fungi,  the 
action  of  nitrous  acid  has  been  especially  studied  (Thorn  and  Steinberg, 
1939;  Steinberg  and  Thom,  1940,  1942).  Mutants  of  Aspergillus  niger, 
A.  amstelodami,  A.  variecolor,  A.  fumigatus,  A.  fischeri,  A.  flavus,  A. 
alliaceus,  and  A.  nidulans  w^ere  produced  with  ease  by  growing  fungi  on 
mannitol-nitrite  medium.  Some  of  these  mutants  w^ere  stable  in  culture 
for  over  20  years.  In  addition  to  morphological  changes,  these  nitrous 
acid-induced  mutants  were  characterized  by  reduced  ability  to  sporulate 
and  other  physiological  changes,  especially  ability  to  utilize  certain  amino 
acids  and  a  reduced  rate  of  growth.  It  was  postulated  that  nitrous  acid 
reacted  with  free  amino  groups  of  the  proteins  of  the  genes.  Evidence 
supporting  this  hypothesis  was  obtained  when  it  w^as  found  that  other 
chemicals  which  also  react  with  free  amino  groups  (ninhydrin,  chlora- 
mine-T,  potassium  iodide,  and  hexamethylenetetramine)  induced  similar 
mutations  in  A.  niger. 

Certain  amino  acids,  when  added  to  the  medium  on  which  these 


VARIATION  AND  INHERITANCE  415 

mutants  were  cultured,  induced  partial  or  complete  reversion  to  the 
morphology  of  the  parent  type.  Of  the  single  amino  acids  tested,  only 
lysine,  cystine,  /3-phenyl-/3-alanine,  threonine,  and  vahne  induced  com- 
plete reversion  with  the  mutant  of  A.  nigcr  studied.  Nicotinic  acid, 
lysine,  and  valine  in  combination  gave  the  best  results.  Complete 
reversion  of  a  mutant  of  A.  amstelodami  was  obtained  only  with  a  mixture 
of  lysine  and  threonine. 

It  should  be  pointed  out  that,  although  the  mutants  of  Aspergillus 
have  been  apparently  stable  for  a  long  period  of  time,  proof  of  the  inherit- 
ance of  these  characters  by  crossing  the  "mutants"  with  the  parent  type 
is  not  available. 

SUMMARY 

Some  knowledge  of  physiological  variation  and  of  the  inheritance  of 
the  underlying  factors  is  of  great  importance  to  the  investigator  who 
cultures  fungi.  The  plant  pathologist  is  greatly  interested  in  knowing 
the  stability  of  the  pathogenicity  of  the  plant  pathogenic  fungi  and  in 
learning  whether  the  variations  which  he  finds  are  apt  to  be  permanent 
or  only  temporary.  Considerable  effort  has  been  made  toward  an  under- 
standing of  the  genetics  of  the  smuts,  the  rusts,  and  the  yeasts.  A  few 
other  fungi,  such  as  Neurospora,  Glomerella,  and  Phycomyces,  have  received 
attention  because  they  are  particularly  adapted  to  genetic  studies. 
However,  the  work  up  to  the  present  leaves  much  to  be  desired  regarding 
the  relation  between  genetics  and  physiology. 

The  genetics  of  sex  and  compatibility  has  been  most  frequently  studied. 
It  is  only  natural  that  such  is  the  case,  for  much  of  the  understanding 
of  a  fungus  depends  upon  the  completion  of  its  sexual  life  cycle  and  a 
clear  knowledge  of  the  various  stages.  Studies  on  the  inheritance  of 
morphological  characters  are  represented  by  numerous  isolated  reports 
on  characters  more  or  less  clearly  defined.  The  reasons  for  the  limited 
number  of  investigations  on  the  genetics  of  physiological  characters  are 
numerous.  Perhaps  the  principal  reason  is  our  meager  knovvledge  of  the 
intricate  physiological  processes  of  the  fungi.  The  difficulty  in  obtaining 
strains  of  the  different  sexes  of  suitable  fungi  which  have  clear-cut  physio- 
logical differences  is  great.  The  actual  difficulty  in  carrying  out  the 
physiological  tests  has  no  doubt  discouraged  work  along  this  line. 

Studies  on  the  inheritance  of  induced  deficiencies  for  a  number  of 
vitamins  and  amino  acids  have  indicated  that  such  deficiencies  may  be 
inherited  in  a  simple  Mendelian  fashion.  On  the  other  hand,  the  inherit- 
ance of  partial  thiamine  deficiency  in  Lenzites  trabea  is  complex  and  does 
not  follow  the  simple  Mendelian  pattern.  Studies  of  the  yeasts  have 
indicated  that  some  physiological  characters  are  inherited  through,  or 
influenced  by,  the  cytoplasm.  This  possibility  should  not  be  overlooked 
in  the  filamentous  fungi,  where  anastomoses  of  hyphae  are  common. 


416  PHYSIOLOGY  OF  THE  FUNGI 

The  basis  for  permanent  variation  in  imperfect  fungi  (as  well  as  in  other 
groups)  is  believed  by  many  to  be  true  mutation.  There  is  circumstantial 
evidence  that  many  such  sudden  changes  are  mutations,  but  the  inherit- 
ance of  such  a  change  in  the  imperfect  fungi  cannot  be  proved.  Other 
variations  in  the  imperfect  fungi  are  explained  on  the  basis  of  hetero- 
caryosis.  Separation  of  such  nuclei  into  different  spores  would  result  in 
the  segregation  into  different  mycelial  or  sporulating  types.  Thus,  the 
dual  phenomenon  would  explain  many  of  the  variations  in  the  imperfect 
fungi. 

The  type  of  variation  known  as  physiological  specialization  is  of  prac- 
tical interest  and  importance  among  the  pathogenic  fungi.  Pathogenicity 
on  a  number  of  varieties  of  hosts  offers  a  rigorous  test  to  distinguish  races 
which  differ  perhaps  only  slightly  in  their  physiology.  The  future  of  the 
work  dealing  with  physiology  and  genetics  of  the  fungi  lies  principally 
in  the  hands  of  competent,  well-trained  investigators  who  may  be  inter- 
ested enough  to  spend  much  time  and  effort  in  this  narrow  field  of  investi- 
gation.    Many  problems  in  this  field  are  waiting  to  be  solved. 

REFERENCES 

Ames,  L.   M.:  Hermaphroditism  involving  self-sterility  and  cross-fertility  in  the 

ascomycete  Pleurage  anserina,  Mycologia  26:  392-414,  1934. 
Barnett,  H.  L.  :  Studies  in  the  sexuaUty  of  the  Heterobasidiae,  Mycologia  29 :  626- 

649,  1937. 
Barnett,  H.  L.,  and  V.  G.  Lilly:  Vitamin  deficiencies  in  the  Sclerotiniaceae,  Phyto- 
pathology 37:  2,  1947. 
Barnett,  H.  L.,  and  V.  G.  Lilly:  The  effects  of  biotin  upon  the  formation  and 

development  of  perithecia,  asci  and  ascospores  by  Sordaria  fimicola,  Am.  Jour. 

Botany  34:-.  196-204,  1947a. 
Barnett,  H.  L.,  and  V.  G.  IjIlly:  The  production  of  haploid  and  diploid  fruit  bodies 

of  Lenzites  trabea  in  culture,  Ptoc.  West  Va.  Acad.  Sci.  19:  34-39,  1949. 
Barnett,  H.  L.,  and  V.  G.  Lilly:  The  influence  of  nutritional  and  environmental 

factors  upon   asexual   reproduction  of   Choanephora  cucurbitarum  in   culture. 

Phytopathology  40:  80-89,  1950. 
"^Beadle,  G.  W.:  Genes  and  the  chemistry  of  the  organism,  Am.  Scientist  34:  31-53, 

1946. 
*BoNNER,  D.:  Production  of  biochemical  mutants  in  Penicillium,  Am.  Jour.  Botany 

33:  788-791,  1946. 
Brodie,  H.  J.:  Variation  in  fruit  bodies  of  Cyathus  stercoreus  produced  in  culture, 

Mycologia  40:  614-626,  1948. 
'Buller,  a.  H.  R.  :  The  diploid  cell  and  the  diploidization  process  in  plants  and 

animals,  with  special  reference  to  the  higher  fungi.     I,  II.     Botan.  Rev.  7 :  335- 

431,  1941. 
BuRKHOLDER,  P,  R.,  and  D.  Moyer:  Vitamin  deficiencies  of  fifty  yeasts  and  molds, 

Bull.  Torrey  Botan.  Club  70:  372-377,  1942. 
Cheo,  p.  C.:  Stripe  smut  of  blue  grass  (Ustilago  striiformis  forma  poae-pratensis) : 

Spore  germination,  artificial  inoculation,  pathological  histology,  and  growth  in 

artificial  media,  thesis.  West  Virginia  University,  1949. 


VARIATION  AND  INHERITANCE  417 

Chilton,   S.  J.   P.,  and  II.  E.  Wheeler:  Genetics  of  Glomerella.     VI.     Linkage, 

Aw.  Jour.  Botany  36:  270-276,  1949. 
Christenberry,  G.  a.  :  a  study  of  the  effect  of  light  of  various  periods  and  wave 

lengths  on  the  growth  and  asexual  reproduction  of  Choanephora  cucurbitarum 

(Berk,  and  Rav.)  Thaxter.,  Jour.  Elisha  Mitchell  Sci.  Soc.  54:  297-310,  1938. 
Christensen,  J.  J.:  Studies  on  the  parasitism  of  Helminthosporium  sativum,  Minn. 

Agr.  Expt.  Sta.  Bull.  11,  1922. 
Christensen,  J.  J.,  and  F.  R.  Davis:  Variation  in  Helminthosporium  sativum  induced 

by  a  toxic  substance  produced  by  Bacillus  mesentericus,  Phytopathology  30 : 

1017-1033,  1940. 
Christensen,  J.  J.,  and  T.  W.  Graham:  Physiologic  specialization  and  variation  in 

Helminthosporium  gramineum  Rab.,  Minn.  Agr.  Expt.  Sta.  Bull.  95,  1934. 
*Christexsen,  J.  J.,  and  H.  A.  Rodenhiser:  Physiologic  specialization  and  genetics 

of  the  smut  fungi,  Botan.  Rev.  6:  398-425,  1940. 
Das  Gupta,  S.  N.:  Saltation  in  fungi,  Lucknow  Univ.  Studies  5,  1936. 
*DiCKiNSON,  S.:  The  nature  of  saltation  in  Fusarium  and  Helminthosporium,  Minn. 

Agr.  Expt.  Sta.  Bull.  88,  1932. 
Dickson,  J.  G.:  Diseases  of  Field  Crops,  McGraw-Hill  Book  Company,  Inc.,  New 

York,  1947. 
Douge,  B.  O.:  Nuclear  phenomena  associated  with  heterothallism  and  homothaUism 

in  the  ascomycete  Neurospora,  Jour.  Agr.  Research  35:  289-305,  1927. 
Dodge,  B.  O.  :  Production  of  fertile  hybrids  in  the  ascomycete  Neurospora,  Jour. 

Agr.  Research  36:  1-14,  1928. 
Dodge,  B.  O.  :  Some  problems  in  the  genetics  of  the  fungi.  Science  90 :  379-385,  1939. 
DowDiNG,  E.  S.:  The  sexuality  of  the  normal,  giant  and  dwarf  spores  of  Pleurage 

anserina  (Ces.)  Kuntze.,  Ann.  Botany  45:  1-15,  1931. 
Edgerton,  C.  W.,  S.  J.  P.  Chilton,  and  G.  B.  Lucas:  Genetics  of  Glomerella.     II. 

Fertilization  between  strains.  Am.  Jour.  Botany  32:  115-118,  1945. 
Fries,  N.:  Spontaneous  physiological  mutations  in  Ophiostoma,  Hereditas  34:  338- 

350,  1948. 
*GiLES,  N.  H.,  and  E.  Z.  Lederberg:  Induced  reversions  of  biochemical  mutants  in 

Neurospora  crassa,  Am.  Jour.  Botany  35:  150-157,  1948. 
Hanna,  W.  F.:  The  odor  of  bunt  spores.  Phytopathology  22:  978-979,  1932. 
Hansen,  H.  N.:  The  dual  phenomenon  in  imperfect  fungi,  Mycologia  30:  442-455 

1938. 
*Hansen,  H.  N.,  and  W.  C.  Snyder:  The  dual  phenomenon  and  sex  in  Hypomyces 

solanii.  cucurbitae,  Am.  Jour.  Botany  30:  419-422,  1943. 
Harris,  H.  A.:  Heterothallic  antibiosis  in  Mucor  racemosus,  Mycologia  40:  347-351^ 

1948. 
Hawker,  L.  E.  :  The  nature  of  the  accessory  growth  factors  influencing  growth  and 

fruiting  of  Melanospora  destruens  Shear  and  of  some  other  fungi,  Ann.  Bot. 

(N.S.)  3 :  657-676,  1939. 
Hockenhull,  D.  J.  D.:  Mustard-gas  mutations  in  Aspergillus  nidulans.  Nature  161 : 

100,  1948. 
Hockenhull,  D.  J.  D.:  The  sulphur  metaboHsm  of  mould  fungi:  The  use  of  "bio- 
chemical mutant"  strains  of  Aspergillus  nidulans  in  elucidating  the  biosynthesis 

of  cystine,  Biochim.  et  Biophys.  Acta  3:  326-335,  1949. 
Houston,  B.  R.:  Culture  types  and  pathogenicity  of  isolates  of  Corticium  solani, 

Phytopathology  35:  371-393,  1945. 
Johnson,  T.,  and  M.  Newton:  Mendelian  inheritance  of  certain  pathogenic  charac- 
ters of  Puccinia  graminis  tritici,  Can.  Jour.  Research  18:  599-611,  1940. 


418  PHYSIOLOGY  OF  THE  FUNGI 

KiNSEY,  V.  E.,  and  W.  M.  Grant:  The  reaction  of  mustard  gas  with  proteins.     II. 

Biological  assay  of  amino  acids  affected,  Arch.  Biochem.  10:  311-322,  1946. 
Kreitlow,  K.  W.:  Sclerotinia  trifoliorum,  a  pathogen  of  ladino  clover,  Phytopath- 

ologijZd:  158-166,  1949. 
Leonian,  L.  H.:  Studies  in  the  variability  and  dissociations  in  the  genus  Fusarium, 

Phijtopathologij  19:  753-868,  1929. 
Leonian,  L.  H.,  and  V.  G.  Lilly:  The  effects  of  vitamins  on  ten  strains  of  Sac- 

charoinyces  cerevisiae,  Am.  Jour.  Botany  29:  459-464,  1942. 
Lilly,  V.  G.,  and  H.  L.  Barnett:  The  inheritance  of  partial  thiamine  deficiency  in 

Lenzites  trabea,  Jour.  Agr.  Research  77:  287-300,  1948. 
LiNDEGREN,  C.  C.:  Mcndelian  and  cytoplasmic  inheritance  in  yeasts,  Ann.  Missouri 

Botan.  Garden  32  :  107-123,  1945. 
LiNDEGREN,  C.  C.'.  Gcuetics  of  the  fungi,  Ann.  Rev.  Microbiol.  1948:  47-70. 
LiNDEGREN,  C.  C.:  The  Yeast  Cell,  Its  Genetics  and  Cytology,  Educational  Pub- 
lishers, St.  Louis,  1949. 
McElroy,  W.  D.,  J.  E.  CusHiNG,  and  H.  Miller:  The  induction  of  biochemical 

mutations  in  Neurospora  crassa  by  nitrogen  mustard,   Jour.   Cellular  Comp. 

Physiol.  30:  331-346,  1947. 
Macrae,  R.  :  Interfertility  studies  and  inheritance  of  luminosity  in  Panus  stypticus, 

Can.  Jour.  Research,  Sec.  C,  20:  411-434,  1942. 
Martin,  W.  J.:  Strains  of  Phytophthora  infestans  capable  of  surviving  high  tempera- 
tures, Phytopathology  39:  14,  1949. 
Nobles,    M.   K.:  Conidial  formation,  mutation  and  hybridization  in  Peniophora 

allescheri,  Mycologia  27:  286-301,  1935. 
RoBBiNS,  W.  J.,  and  R.  Ma:  Biotin  and  the  growth  of  Fusarium  avenaceuni,   Bull. 

Torrey  Botan.  Club  68:  446-462,  1941. 
RoBBiNS,  W.  J.,  and  L  McVeigh:  The  dual  phenomenon  and  Trichophyton  menta- 

grophytes,  Mycologia  41 :  128-140,  1949. 
Ryan,  F.  J.:  Back  mutation  and  adaptation  of  nutritional  mutants.  Cold  Spring 

Harbor  Symposia  Quant.  Biol.  11:  215-227,  1946. 
Skoog,  F.  K.,  and  C.   C.  Lindegren:  Adaptation  in  a  yeast  unable  to  ferment 

glucose.  Jour.  Bad.  53:  729-742,  1947. 
Spiegelman,  S.  :  in  J.  B.  Sumner  and  K.  Myrback  (Editors),  The  Enzymes,  Vol.  I, 

Academic  Press,  Inc.,  New  York,  1950. 
Spiegelman,  S.,  C.  C.  Lindegren,  and  G.  Lindegren:  Maintenance  and  increase  of 

a  genetic  character  by  a  substrate-cytoplasmic  interaction  in  the  absence  of  the 

specific  gene,  Proc.  Natl.  Acad.  Sci.  U.S.  31 :  95-102,  1945. 
Steinberg,  R.  A.,  and  C.  Thom:  Mutations  and  reversions  in  reproductivity  of 

Aspergilli  with  nitrite,  colchicine  and  d-lysine,  Proc.  Xatl.  Acad.  Sci.  U.S.  26: 

363-366,  1940. 
Steinberg,   R.  A.,  and  C.  Thom:  Reversions  in  morphology  of  nitrite-induced 

"Mutants"  of  Aspergilli  grown  upon  amino  acids,  Jour.  Agr.  Research  64: 

645-652,  1942. 
Thom,  C,  and  K.  B.  Raper:  A  Manual  of  the  Aspergilli,  The  WiUiams  &  Wilkins 

Company,  Baltimore,  1945. 
Thom,  C,  and  R.  A.  Steinberg:  The  chemical  induction  of  genetic  changes  in  fungi, 

Proc.  Natl.  Acad.  Sci.  U.S.  25 :  329-335,  1939. 
Thren,   R.:  Zur   Entwicklungsphysiologie  der  Dikaryophase  von    Ustilago  nuda 

(Jensen)  Kellerm.  et  Sw.,  Arch.  Mikrobiol.  12 :  192-228,  1941. 
Wolf,  F.  A.:  A  squash  disease  caused  by  Choanephora  cucurbitarum,  Jour.  Agr. 

Research  B:  319-328,  1917. 
Wolf,  F.  A.,  and  F.  T.  Wolf:  The  Fungi,  Vol,  II,  John  Wiley  &  Sons,  Inc.,  New 

York,  1947. 


SUGGESTED  LABORATORY  EXERCISES 

Each  of  the  laboratory  exercises  is  designed  to  ilKistrate  one  or  more 
important  principles  regarding  the  phj'siological  activities  of  the  fungi. 
Each  represents  a  more  or  less  complete  unit  or  phase,  but  there  is  neces- 
sarily some  overlapping  and  duplication  of  the  techniques.  Laboratory 
exercises  requiring  greenhouse  space  and  living  plants  have  been  omitted. 
This  does  not  mean  that  an  understanding  of  the  cultural  activities  should 
be  the  only  aim  of  the  laboratory  work.  Rather,  it  is  believed  that  such 
information  regarding  the  nutritional  and  environmental  requirements  of 
fungi  in  pure  culture  will  aid  in  the  better  understanding  of  the  behavior 
of  both  parasitic  and  nonparasitic  fungi  in  nature. 

The  laboratory  exercises  are  planned  so  as  to  allow  a  high  degree  of 
flexibility.  The  instructor  may  wish  to  omit,  change,  or  add  to  some  of 
the  suggested  exercises  to  suit  the  equipment  and  other  facilities  available. 
The  exercises  are  outlined  to  require  a  minimum  of  laboratory  equipment. 
Any  or  all  of  the  exercises  may  be  expanded  or  shortened  as  desired  for  a 
large  or  small  number  of  students. 

Each  student  should  select  one  or  more  laboratory  exercises  in  which 
he  has  a  particular  interest.  He  may  then  be  designated  as  the  leader 
of  that  exercise  or  exercises.  His  duties  would  include  (1)  the  general 
planning  of  the  exercise  in  consultation  ^^^th  the  instructor,  (2)  the  direct- 
ing of  the  execution  of  the  exercise,  (3)  collecting  and  organization  of  the 
data  from  other  students,  and  (4)  the  writing  of  a  comprehensive  report 
on  the  subject.  References  should  be  read  and  discussed  in  comparison 
with  the  results  of  the  experiments  in  the  laboratory. 

The  supply  of  test  fungi  is  almost  unlimited.  In  most  exercises  each 
student  will  use  two  or  more  species  of  fungi.  The  total  number  for  the 
class  should  be  sufficiently  large  to  emphasize  the  differences  as  well  as 
the  similarities  that  exist  among  various  fungi.  In  many  instances  our 
knowledge  is  limited,  and  little  or  nothing  has  been  published  regarding 
the  physiology  of  some  of  the  species  used.  In  this  sense,  much  of  the 
work  done  in  the  laboratory  will  be  experimental  and  should  be  carried 
out  with  all  the  care  of  a  research  problem.  Fungi  should  be  selected 
so  that  both  growth  and  sporulation  may  be  studied  in  the  same  experi- 
ment. In  each  exercise  are  listed  a  few  fungi,  some  of  which  have  been 
used  in  our  laboratory  and  have  given  quite  satisfactory  results.  Other 
species  may  be  added  or  substituted. 

Notes  should  be  recorded  in  a  full-sized  permanent  notebook  reserved 

419 


420  PHYSIOLOGY  OF  THE  FUNGI 

only  for  the  laboratory  work  in  this  course.  The  student  should  take 
notes  and  record  data  so  that  anyone  familiar  with  the  subject  could 
organize  the  data  and  write  an  intelligent  summary.  Data  taken  during 
the  experiments  are  important  and  in  themselves  may  be  quite  convincing 
at  that  time,  but  data  alone  without  organization  and  discussion  of  their 
meaning  are  often  sterile  and  soon  forgotten. 

The  paper  should  be  written  as  soon  as  possible  after  the  termination 
of  the  laboratory  exercise.  It  is  suggested  that  all  students  read  the 
excellent  discussion  of  Riker  (1946)  on  the  preparation  of  manuscripts. 

The  student  should  become  familiar  with  the  microscopic  character- 
istics of  the  fungi  used  in  the  experiments.  He  should  make  frequent 
microscopic  observations  during  the  experiment  and  look  for  microscopic 
changes,  such  as  abundance  and  maturity  of  spores.  Contaminants  are 
often  more  easily  recognized  under  the  microscope.  Drawings  are  often 
desirable  as  records  of  differences  in  microscopic  characteristics. 

EXERCISE  1 
General  Laboratory  Procedure 

Exercise  1  is  suitably  carried  out  by  the  instructor  as  a  demonstration 
of  general  laboratory  techniques.  This  affords  the  instructor  an  oppor- 
tunity to  discuss  the  details  of  various  procedures  and  to  acquaint  the 
students  with  the  laboratory  facilities.  It  is  convenient  to  prepare  suffi- 
cient stock  culture  medium,  e.g.,  medium  1  (Ex.  2),  for  growing  the 
inoculum  required  by  the  class  for  the  next  exercise. 

General  directions  for  preparing  media.  The  culture  medium  should 
be  selected  with  the  purpose  of  the  experiment  in  mind.  The  precautions 
to  be  observed  may  be  elemental  or  elaborate,  depending  upon  the  pur- 
pose for  which  the  medium  is  to  be  used.  The  accuracy  of  one  measure- 
ment should  be  consistent  with  the  accuracy  of  the  others.  The  weights 
of  each  constituent  of  a  given  lot  of  medium  should  be  written  in  a  note- 
book. As  each  constituent  is  measured,  make  a  check  mark  against  this 
constituent. 

Never  weigh  chemicals  directly  on  the  balance  pans.  Use  a  clean  piece 
of  paper  or  watch  glass.  The  weights  are  placed  on  the  right-hand  pan 
as  you  face  the  balance.  Be  sure  the  spatula  is  cleaned  between  weigh- 
ings of  different  chemicals.  If  you  remove  more  of  a  chemical  than 
necessary,  discard  the  excess.  (Material  still  on  the  spatula  may  be 
returned  to  the  stock  bottle.)  Keep  the  stock  bottles  closed.  This 
prevents  the  entrance  of  dust  and  atmospheric  moisture.  Malt  extract, 
yeast  extract,  and  peptone  quickly  absorb  water  from  the  air,  and  when 
these  substances  have  done  so,  an  intractable  mass  results. 

For  work  of  ordinary  accuracy  use  a  graduated  cylinder  for  measuring 
liquids.     Volumetric  glassware  should  be  used  for  precise  work.     The 


SUGGESTED  LABORATORY  EXERCISES  421 

chemicals  should  be  dissolved  in  less  than  the  desired  volume  of  water, 
and  after  they  are  in  solution,  the  medium  should  be  made  up  to  volume 
by  the  addition  of  distilled  water. 

Constituents  of  media.  The  salts  used  in  making  media  should  be  of 
c.p.  grade.  These  will  generally  contain  enough  of  the  micro  essential 
elements  to  satisfy  the  needs  of  most  fungi.  Iron,  zinc,  and  manganese 
should  be  added  routinely  to  all  synthetic  media.  It  is  convenient  to 
make  up  a  solution  containing  the  above  micro  essential  elements  in  such 
strength  that  1  ml.  of  the  master  solution  will  contain  0.1  mg.  Fe+ + +, 
0.1  mg.  Zn+  +,  and  0.05  mg.  Mn+  +.  The  following  amounts  of  the  given 
salts  have  been  found  convenient  to  use:  Fe(N03)3-9H20,  723.5  mg. ; 
ZnS04-7H20,  439.8  mg.;  MnS04-4H20,  203.0  mg.  Dissolve  these  three 
salts  in  600  ml.  of  distilled  water,  add  sufficient  c.p.  sulfuric  acid  to  yield 
a  clear  solution,  and  make  up  the  volume  to  1  liter  with  distilled  water. 
Use  2  ml.  of  this  solution  per  liter  of  medium.  Add  the  source  of  carbon 
at  the  rate  of  10  to  25  g.  per  liter,  depending  on  the  use  of  the  medium. 
Add  the  nitrogen  source  at  the  rate  of  2  g.  per  liter,  or  the  amount  which 
will  give  an  amount  of  nitrogen  equivalent  to  that  furnished  by  2g  .  of 
anhydrous  asparagine  (0.425  g.  of  N). 

The  activated  carbon  used  in  Ex.  9  to  absorb  residual  traces  of  vitamins 
is  a  commercial  preparation,  Norit  A  (pharmaceutical  grade).  Use  at 
the  rate  of  5  g.  per  liter,  or  more  if  necessary.  The  vitamins  used  rou- 
tinely are  thiamine  and  biotin.  Inositol  and  pyridoxine  are  also  used  in 
Ex.  9.  These  vitamins  may  be  purchased  from  JMerck  and  Co.,  Rahway, 
N.J.,  and  many  other  pharmaceutical  houses.  It  is  convenient  to  make 
up  master  solutions  of  these  four  vitamins.  Make  the  master  solutions 
in  20  per  cent  alcohol,  store  in  100-ml.  volumetric  flasks,  and  keep  in  a 
refrigerator  when  not  in  use.  It  is  convenient  to  make  the  master  solu- 
tions of  thiamine  and  pyridoxine  to  contain  100  /xg  per  ml.  (10  mg.  per 
100  ml.).  The  master  solution  of  inositol  contains  5  mg.  per  ml.  Biotin 
is  used  at  the  rate  of  5  Mg  per  liter.  Either  the  contents  of  an  ampule  may 
be  made  up,  or  a  microbalance  used  to  weigh  this  vitamin.  A  master 
solution  containing  either  1  or  5  jug  per  ml.  is  convenient. 

For  a  semisolid  medium,  agar  is  commonly  used  at  the  rate  of  20  g. 
per  liter.  Agar  is  not  a  pure  compound,  and  its  use  introduces  various 
unsuspected  constituents  into  media.  No  medium  may  be  said  to  be  of 
known  composition  if  agar  is  used,  although  such  media  are  very  useful. 
Some  of  the  impurities  in  agar  may  be  removed  by  leaching  with  aqueous 
pyridine  solution.  Place  1  lb.  of  agar  in  a  6-liter  flask,  add  5  liters  of 
distilled  water  and  500  ml.  of  pyridine.  Allow  to  stand  24  hr.  Insert 
a  piece  of  6-mm.  glass  tubing  of  sufficient  length  to  admit  air  to 
the  flask,  tie  a  piece  of  cheesecloth  over  the  neck  of  the  flask,  invert  the 
flask,  and  allow  the  pyridine  solution  to  drain.     Wash  the  agar  three 


422  PHYSIOLOGY  OF  THE  FUNGI 

times  with  distilled  water.  Wash  the  agar  twice  with  95  per  cent  alcohol, 
allowing  the  alcohol  to  stand  on  agar  overnight  before  draining.  Dry 
the  agar  in  thin  layers  between  cheesecloth.  This  procedure  takes  about 
10  days  (see  Robbins,  1939).  In  some  instances  the  agar  and  other  con- 
stituents of  the  medium  may  be  autoclaved  separately  and  the  two  solu- 
tions  mixed,  using  antiseptic  precautions.  This  should  be  done  when  it 
is  required  to  have  a  very  acidic  agar  medium.  A  known  amount  of 
sterile  acid  may  be  added  to  the  agar  medium  after  sterilization. 

pH  or  reaction  of  the  medium.  See  Chap.  8  for  a  discussion  of  pH. 
An  approximate  method  of  determining  pH  is  sufficiently  accurate  for 
many  purposes.  On  a  white  porcelain  spot  plate  place  one  drop  of 
Hellige  (or  other)  wide-range  indicator  in  each  depression.  Have  the 
drops  of  indicator  of  equal  size.  Then  add  one  drop  of  the  medium  to 
a  drop  of  indicator  in  one  of  the  depressions  of  the  spot  plate.  The  color 
of  the  mixed  drops  indicates  the  pH  of  the  medium.  Thus,  red  indicates 
a  pH  in  the  neighborhood  of  4,  light  green  7,  purple  10.  Standard  buffers 
(solutions  of  known  pH)  may  be  provided  so  that  the  student  may  have 
standards  with  which  to  make  comparisons. 

The  buffers  used  in  testing  pH  are  most  conveniently  made  from  buffer 
tablets  (Coleman).  Dissolve  one  tablet  in  100  ml.  of  distilled  water. 
Add  a  crystal  of  thymol  as  a  preservative.  Thymol  aids  in  preventing 
contamination  and  does  not  appreciably  affect  the  pH  of  the  buffer.  It 
is  convenient  to  store  the  buffers  in  brown-glass  dropping  bottles  fitted 
with  pipettes. 

Unless  otherwise  specified,  media  used  in  the  laboratory  should  be 
adjusted  to  pH  6  before  autoclaving.  This  may  be  done  by  the  use  of 
the  spot  plate,  adding  a  drop  of  pH  6  buffer  to  a  drop  of  indicator.  This 
is  the  standard  color  to  which  the  media  should  be  adjusted.  Add  either 
QN  NaOH  or  QN  HCl  to  the  medium  until  the  color  produced  by  one  drop 
of  medium  matches  the  color  produced  by  the  standard  buffer.  Always 
agitate  the  medium  after  each  addition  of  acid  or  base  and  then  test  the 
pH.  The  use  of  concentrated  acid  and  alkali  is  recommended  so  that 
dilution  of  the  medium  may  be  avoided.  More  precise  methods  of 
measuring  pH  may  be  used  if  desired. 

Autoclaving  usually  lowers  the  pH  of  a  medium.  In  general,  this 
change  will  not  be  great,  but  the  student  should  never  assume  that  the 
pH  will  remain  unchanged  in  autoclaving. 

Sterilization  of  media  and  glassware.  Except  in  special  instances,  the 
autoclave  may  be  used  to  sterilize  both  media  and  glassware.  Fifteen 
minutes  at  15  lb.  steam  pressure  is  adequate  for  test  tubes  and  flasks 
which  do  not  contain  over  150  ml.  of  medium.  Larger  lots  of  media 
should  be  autoclaved  20  min.  at  15  lb.  steam  pressure.  Petri  dishes  may 
be  sterilized  20  min.  in  the  autoclave.     It  is  convenient  to  wrap  two  Petri 


SUGGESTED  LABORATORY  EXERCISES  423 

dishes  in  a  paper  towel  or  several  in  a  paper  bag  before  placing  in  the 
autoclave.  This  wrapping  should  remain  on  the  Petri  dishes  until  they 
are  used,  to  prevent  contamination.  Do  not  remove  wrapped  glassware 
from  the  autoclave  until  several  minutes  after  the  pressure  is  down. 
Pipettes  should  be  wrapped  and  placed  in  pipette  cases.  Sterilize  in  the 
same  way  as  Petri  dishes. 

Use  either  a  water  bath  or  the  autoclave  for  melting  agar.  Never  melt 
agar  in  or  sterilize  flasks  which  are  more  than  half  full.  Test  tubes  should 
not  be  more  than  one-fourth  full.  The  reason  for  this  lies  in  the  fact 
that  the  autoclave  cools  quicker  than  the  medium.  This  leaves  the 
medium  superheated,  and  under  this  condition  it  is  likely  to  boil  violently. 
Never  remove  a  flask  of  melted  agar  from  the  autoclave  as  soon  as  the 
pressure  is  down.  Agitation  may  cause  violent  boiling.  Your  instructor 
will  give  you  full  instructions  for  operating  the  autoclave.  Follow  these 
instructions  carefully. 

On  handling  cultures.  It  will  be  necessary  for  each  student  to  main- 
tain the  identity  of  his  cultures.  This  may  be  done  by  name  or  by  stock- 
culture  number.  Each  medium  used  in  the  course  will  receive  a  number. 
If  a  medium  is  used  more  than  once,  it  will  be  given  another  number. 
The  composition  of  each  medium  should  be  entered  in  the  laboratory 
notebook.  The  name  of  the  fungus  (or  stock-culture  number)  and  the 
number  of  the  medium  should  be  also  written  on  each  culture  vessel. 
The  date  of  inoculation  and  the  kind  of  inoculum  used  should  be  entered 
in  the  notebook.  It  is  convenient  to  fasten  together  duplicate  or  tripli- 
cate cultures  in  test  tubes  with  a  rubber  band. 

Preservation  of  stock  cultures.  The  maintenance  of  a  stock  culture 
collection  of  filamentous  fungi  for  class  use  is  highly  desirable.  Such  a 
collection  need  not  be  extensive  but  should  include  a  sufficient  number  of 
selected  species  of  known  phj^siological  reaction  and  any  others  which 
may  be  desired  for  general  use.  The  method  of  preserving  cultures  in 
our  laboratory  has  proved  quite  satisfactory  when  frequent  transfers  are 
needed  for  research  or  class  use.  Test  tubes  with  constricted  tops  and 
plastic  screw  caps  are  used.  Malt  extract  or  any  other  suitable  agar  may 
be  used.  After  inoculation  the  tubes  are  allowed  to  remain  at  room  tem- 
perature for  a  few  days  until  the  inoculum  starts  to  grow.  Then  the  caps 
are  screwed  down  tightly  and  the  cultures  stored  at  5  to  10°C.  Most 
species  continue  to  grow  slowly,  and  under  these  conditions  the  tube? 
remain  free  from  contamination  and  the  agar  dries  out  very  slowly.  Thi.= 
method  also  excludes  mites.  Some  vegetative  cultures  have  remained 
viable  for  a  period  of  more  than  2  years  without  being  transferred.  How- 
ever, it  is  suggested  that  all  cultures  be  transferred  every  year,  and  the 
entire  stock  should  be  looked  over  carefully  every  few  months,  as  some 
species  may  require  more  frequent  transfers.     The  first  transfer  from 


424  PHYSIOLOGY  OF  THE  FUNGI 

stock  culture  should  be  to  another  stock-culture  tube,  and  the  old  tube 
should  be  kept  until  the  new  culture  begins  to  grow  free  from  contamina- 
tion. Other  methods  of  storing  stock  cultures  of  fungi  are  described  by 
Greene  and  Fred  (1934),  Thom  and  Raper  (1945),  Fennell  et  al.  (1950), 
and  Buell  and  Weston  (1947). 

Methods  of  inoculation.  It  is  customary  to  use  a  bit  of  mycelium  from 
a  growing  culture  to  inoculate  fresh  media.  For  ordinary  uses  this  is 
satisfactory,  if  only  a  few  cultures  are  to  inoculated  at  a  time  and  no 
special  precautions  need  be  taken.  Some  fungi  produce  a  tough  mat  of 
mycelium  difficult  to  cut  with  a  needle.  Often  this  can  be  overcome  by 
growing  the  mycelium  for  inoculum  on  an  agar  medium  quite  low  in 
sugar.  A  small  cork  borer  may  be  used  for  cutting  out  uniform  disks  of 
mycelium  and  agar  from  Petri  dishes. 

Spores  alone  may  be  transferred  by  a  dry  needle,  or  they  may  be 
suspended  in  water  and  inoculated  by  use  of  a  loop  or  a  sterile  pipette 
with  a  cotton  plug  at  the  upper  end.  The  use  of  a  pipette  fitted  w^ith  a 
small  rubber  bulb  greatly  decreases  the  inoculation  time  when  many 
cultures  of  the  same  fungus  are  made.  It  is  preferable  to  use  spores  as 
inoculum  in  studies  of  vitamins  or  micro  elements,  where  none  of  the 
previous  medium  should  be  added. 

Nonsporulating  mycelium  may  be  fragmented  by  placing  it  with  about 
50  ml.  water  in  a  sterile  Waring  Blendor  jar  for  about  30  sec.  Either 
agar  or  liquid  medium  may  be  used  if  the  addition  of  the  medium  is  of  no 
consequence.  In  vitamin  studies  the  mycelium  may  be  grown  in  liquid 
medium  and,  w^hen  ready  for  use,  w'ashed  in  sterile  distilled  water  and 
fragmented  in  the  Blendor.  Either  a  loop  or  a  pipette  may  be  used  to 
dispense  the  mycelial  suspension. 

Methods  of  obtainmg  single-spore  cultures.  In  certain  physiological 
studies  it  is  desirable  to  use  single-spore  cultures.  These  may  be  obtained 
by  a  number  of  different  methods.  A  review  of  the  literature  on  these 
techniques  has  been  given  by  Hildebrand  (1938).  Other  modified  tech- 
niques are  described  by  Georg  (1947)  and  Thom  and  Raper  (1945). 
Still  another  modification  may  be  worthy  of  brief  mention.  In  this 
laboratory  we  have  used  a  specially  prepared  small  sewing  needle  as  a 
tool  for  picking  out  single  germinated  spores.  The  eye  of  the  needle  is 
rounded  and  the  thick  metal  portion  filed  down,  making  a  rather  thin 
edge  for  cutting  agar.  The  pointed  end  is  fastened  in  a  convenient 
holder,  and  the  needle  bent  in  such  a  way  that,  when  held  over  a  Petri 
dish,  the  eye  portion  will  be  parallel  with  the  surface  of  the  medium  so 
that  it  can  be  pushed  straight  down  into  the  agar.  An  isolated,  germi- 
nated spore  is  located  on  a  dilution  plate  by  use  of  a  microscope.  The 
needle  is  then  held  in  place  under  the  objective  so  that  the  spore  is  visible 
through  the  eye  of  the  needle.     The  eye  is  pressed  down  around  the  spore 


SUGGESTED  LABORATORY  EXERCISES  425 

and  is  lifted  up  with  a  bit  of  agar  and  the  germinated  spore.  Another 
needle  may  then  be  used  to  transfer  the  bit  of  agar  and  spore  to  a  tube  or 
plate.  This  method  requires  a  steady  hand  but  has  the  advantage  of 
being  rapid  and  simple.  It  may  be  employed  with  high  magnification 
of  the  stereoscopic  microscope  or  the  low-power  objective  of  the  com- 
pound microscope. 

EXERCISE  2 

The  Influence  of  Temperature  on  Growth  and  Sporulation 

This  exercise  is  outlined  to  illustrate  three  main  points:  (1)  the  general 
effects  of  different  temperatures  upon  growth;  (2)  the  approximate 
optimum  temperature  for  growth  of  a  few  fungi ;  (3)  that  the  temperature 
limits  for  sporulation  are  narrower  than  those  for  vegetative  growth. 
Inasmuch  as  the  composition  of  the  medium  may  influence  temperature 
limits,  only  one  medium  should  be  used  in  this  experiment. 

Mediimi 

1.   Malt  extract,  20  g.;  j^east  extract,  2  g. ;  agar,  20  g.;  distilled  water,  1,000  ml. 

Adjust  the  pH,  if  necessary,  to  approximately  6,  and  autoclave.  Pour 
into  sterile  Petri  dishes,  about  20  ml.  per  plate,  and  inoculate  them  at  the 
center.  Inoculate  plates  in  duplicate  or  in  triplicate  for  each  condition. 
Incubate  the  plates  at  a  range  of  temperatures  with  con^^enient  intervals. 
The  range  and  the  exact  temperatures  used  will  depend  upon  the  facilities 
available.  Suggested  temperatures  are  10,  15,  20,  25,  30,  35,  and  40°C. 
Since  light  affects  some  of  the  species  listed  below,  it  is  desirable  to  have 
the  cultures  illuminated  for  a  part  of  each  day. 

Records.  Observations  of  growth  should  be  made  daily  or  at  least 
every  2  days.  For  the  purpose  of  this  exercise  the  radial  extension  of  the 
mycelium  may  be  used  as  a  measurement  of  growth.  This  usually  can  be 
easily  measured  by  placing  a  rule  (preferable  one  calibrated  in  millimeters) 
on  the  bottom  of  the  Petri  dish  and  looking  through  it  towai'd  a  light.  In 
order  to  have  an  accurate  measure  of  the  effect  of  different  temperatures, 
the  cultures  must  be  compared  before  the  mycelium  reaches  the  edge  of 
the  Petri  dish.  A  more  accurate  measure  of  growth  may  be  obtained  by 
determining  the  dry  weight  of  mycelium  grown  in  liquid  medium.  Rec- 
ords may  be  made  in  table  form  for  each  fungus,  giving  the  days  of  incuba- 
tion and  the  diameter  of  the  colony.  The  average  daily  radial  extension 
of  the  mycelium  may  then  be  calculated  for  each  temperature.  The 
amount  of  sporulation  should  likewise  be  recorded  each  time. 

List  of  test  fungi:  Alternaria  sp.,  Aspergillus  rugidosus,  Botryis  cinerea, 
Cephalothecium  roseum.,  Ceratosiomella  fimhriala,  Choanephora  cucur- 
hitarum,  Glomerella  cingulata,  Guignardia  hidwellii,  Monilinia  fructicola, 
Penicillium  expansum,  Phytophthora  infestans,  Sclerotinia  sclcrotiorum, 
Septoria  nodorum. 


426  PHYSIOLOGY  OF  THE  FUNGI 

EXERCISE  3 
The  Influence  of  Light  on  Growth  and  Sporulation 

This  experiment  is  designed  to  illustrate  the  variable  effects  of  visible 
and  ultraviolet  light,  particularly  upon  reproduction  of  some  common 
fungi.  Exposure  to  light  is  essential  to  spore  formation  in  some  fungi, 
while  other  fungi  may  sporulate  abundantly  in  total  darkness.  For  a 
more  complete  discussion,  see  Chaps.  3  and  15. 

Media 

2.  Same  as  medium  1 

3.  .\ny  semisynthetic  medium,  such  as  the  basal  medium  in  Ex.  4 

Adjust  the  pH  of  the  media  to  approximately  6.0,  autoclave,  and  pour 
into  sterile  Petri  dishes.  Inoculate  plates  in  triplicate  for  each  condition. 
Place  the  inoculated  plates  under  the  following  conditions: 

A.  Continuous  total  darkness  at  25°C. 

B.  Continuous  artificial  light  at  25°C. 

C.  Alternating  light  and  darkness  at  25°C. 

D.  Same  as  C,  but  expose  to  ultraviolet  three  times  for  2  min  each  time,  at  intervals 
of  2  days 

The  exposure  to  ultraviolet  light  should  be  made  in  an  inoculating 
chamber,  with  the  lids  of  the  dishes  removed,  at  a  distance  of  approxi- 
mately 10  to  12  in.  from  the  source  (the  G.E.  germicidal  lamp  is  quite 
satisfactory).  Wearing  of  spectacles  or  sunglasses  will  protect  the  eyes 
from  the  ultraviolet  rays.  The  first  exposure  should  be  made  when  the 
fungus  colony  is  approximately  1  in.  in  diameter.  Subsequent  exposures 
should  be  timed  so  that  the  last  one  is  made  before  the  mycelium  reaches 
the  edge  of  the  plate.  With  rapidly  growing  species,  the  interval  between 
exposures  may  be  shortened.  At  the  time  of  each  exposure,  use  a  wax 
pencil  to  outline  the  extent  of  the  mycelium  by  marking  the  bottom  of  the 
Petri  dish.  The  plates  kept  in  total  darkness  should  be  wrapped  in  paper 
or  stored  in  a  lighttight  cardboard  box.  Examination  of  most  fungi 
should  be  made  after  about  7  days.  The  rapidly  growing  species  should 
be  placed  in  a  separate  box  which  can  be  opened  earlier. 

List  of  test  fungi:  Botrytis  cinerea,  Cephalothecium  roseum,  Cerato- 
stomella  fimbriata,  Choanephora  cucurhitarum,  Dendrophoma  obscurans, 
Endothia  parasitica,  Moniliniafructicola,  Neocosmopara  vasinfecta,Penicil- 
lium  expansum,  Septoria  nodorum,  Trichoderma  lignorum. 

EXERCISE  4 
The  Effect  of  the  Carbon  Source  on  Growth  and  Sporulation 

This  exercise  is  designed  to  show  that  fungi  differ  in  their  ability  to 
utilize  certain  compounds  as  a  source  of  carbon.     For  discussion  of  this 


SUGGESTED  LABORATORY  EXERCISES  427 

topic  see  Chaps.  7  and  14.  In  this  study,  the  nitrogen  source  in  the 
media  should  be  simple  and  available  to  as  many  fungi  as  possible.  For 
this  purpose,  asparagine  is  quite  satisfactory. 

Malt  extract-yeast  extract  agar  may  be  used  as  a  standard  controi 
medium,  since  this  is  an  excellent  natural  medium  for  most  fungi.  If 
desired,  any  other  standard  natural  medium  may  be  used  as  a  control 
medium.  All  other  media  used  in  the  exercise  will  have  the  same  basal 
composition,  with  the  carbon  source  as  the  only  variable.  Adjust  the 
pH  of  all  media  to  approximately  6.0  before  autoclaving. 

Basal  Semisynthetic  Medium 

Carbon  source 10  g. 

Asparagine 2  g. 

KH2PO4 1  g. 

MgSOrTH.0 0.5  g. 

Fe+  +  + 0.2  mg. 

Zn++ 0.2  mg. 

Mn++ 0.1  mg. 

Biotin 5  Mg 

Thiamine 100  ^g 

Distilled  water  to  make 1  liter 

Agar  (for  solid  media) 20  g. 

For  the  most  accurate  measure  of  growth,  liquid  media  should  be  used, 
so  that  the  mycelium  may  be  filtered  off,  washed,  dried,  and  weighed. 
However,  agar  media  are  often  more  satisfactory  for  reproduction.  It  is 
suggested  that  this  exercise  be  carried  out  on  agar  slants  in  test  tubes. 

Media 

4.  Same  as  medium  1 

5.  Glucose.     Media  5  to  13  will  all  contain  the  basal  medium  above. 

6.  Sucrose  7.  Sorbose 
8.  Lactose  9.  Maltose 

10.  Galactose  11.  Starch 

12.  Cellulosei  13.  No  sugar 

Each  student  should  select  two  or  more  species  of  fungi  and  inoculate 
with  each  fungus  three  tubes  of  each  of  the  10  media  listed  above.  Incu- 
bate at  25°C.,  or  at  room  temperature.  Use  for  inoculum  spores  or  small 
bits  of  mycelium  with  as  little  agar  as  possible. 

Records.  The  student  will  be  responsible  for  taking  notes  or  data  on 
the  growth  and  sporulation  of  the  fungi  he  selects,  but  he  should  follow 
the  form  suggested  by  the  leader  of  the  exercise.  He  should  also  observe 
the  results  of  students  who  use  other  species.  Records  on  growth  may  be 
made  at  the  end  of  3  to  7  days,  depending  upon  the  fungus,  while  a  greater 
time  should  be  allowed  before  making  final  records  on  sporulation.  For 
most  purposes  a  record  of  the  relative  amount  of  growth  or  sporulation, 

'  Use  good  grade  of  filter  paper,  add  water,  and  cut  to  a  pulp  in  a  Waring  Blendor. 


428  PHYSIOLOGY  OF  THE  FUNGI 

when  compared  with  that  on  a  control  medium,  is  cjuite  satisfactory. 
Thus,  if  the  growth  and  sporulation  (if  any)  of  each  fungus  on  medium  4 
are  arbitrarily  given  the  values  of  4  +  ,  the  estimated  abundance  on  other 
media  may  be  designated  as  greater  or  less  than  4,  as  the  case  may  be. 
Such  a  rough  method  has  been  found  satisfactory  for  illustrating  principles 
and  determining  the  availability  of  carbon  sources  utilized  by  the  fungi. 
If  a  more  accurate  measurement  is  desired,  liquid  media  should  be  used 
and  dry  weights  of  the  mycelia  obtained.  Observations  and  records 
should  be  made  on  any  other  characteristics  which  are  affected  by  changes 
in  carbon  source. 

List  of  test  fungi:  Aspergillus  rugulosus,  Ceratostomella  fimbriata, 
Dendrophoma  obsmrans,  Endothia  parasitica,  Glomerella  cingulata,  Guig- 
nardia  bidwellii,  Melanospora  sp.,  Monilinia  friicticola,  Phycomyces 
blakesleeanus,  Pleurage  curvicolla,  Sordaria  Jimicola,  Sphaeropsis  malonan, 

Ustilago  striiformis. 

EXERCISE  5 

The  Effect  of  the  Nitrogen  Source  on  Growth  and  Sporulation 

The  purpose  of  this  exercise  is  to  illustrate  the  utilization  of  different 
sources  of  nitrogen  by  different  fungi.  This  exercise  should  follow  soon 
after  Ex.  4,  and  the  procedure  should  be  the  same.  The  use  of  some  of 
the  same  test  fungi  in  this  exercise  should  emphasize  the  importance  of  a 
suitable  semisynthetic  medium  for  growth  and  reproduction.  In  this 
case  the  carbon  source  (glucose)  shall  be  kept  constant  and  the  nitrogen 
source  varied  with  each  medium.  The  other  constituents  of  the  basal 
medium  will  be  the  same  as  listed  under  Ex.  4.  Medium  1  may  again 
be  used  as  a  control,  but  if  a  different  lot  is  made,  it  must  carry  a  new 
number. 

Media 


14. 

Malt  extract-yeast 

extract 

15. 

Potassium  nitrate 

16. 

Ammonium  sulfate 

17. 

Ammonium  tartrate 

18. 

Asparagine 

19. 

Glutamic  acid 

20. 

Glycine 

21. 

Urea 

22. 

Casein  hydrolysate 

23. 

No  nitrogen 

The  amount  of  nitrogen  should  be  kept  constant.  The  weights  of  the 
compounds  used  should  be  calculated  to  contain  a  weight  of  nitrogen 
equivalent  to  that  in  2  g.  of  asparagine. 

EXERCISE  6 
Special  Nutritional  Conditions  Which  Influence  Growth  and  Sporulation 

The  two  previous  exercises  have  dealt  mainly  with  the  effect  of  the 
constituents  of  the  medium  on  a  qualitative  basis  (see  Chaps.  3  and  14). 
This  exercise  is  outlined  to  emphasize  some  of  the  effects  of  quantitative 


SUGGESTED  LABORATORY  EXERCISES  429 

differences  in  media.  This  ma}^  be  illustrated  by  altering  the  concentra- 
tion of  one  or  more  components  of  the  medium.  It  is  suggested  that  this 
exercise  be  carried  out  in  Petri  dishes,  or  in  flasks  if  liquid  media  are  used. 
Media:  This  exercise  may  be  divided  into  four  parts  based  upon  the 
variations  in  media. 

A.  Dilutions  of  the  entire  medium.  Either  liquid  or  agar  media  may  be 
used. 

24.  Basal  medium,  containing  asparagine  and  20  g.  glucose  per  liter 

25.  Medium  24  diluted  to  one-half  strength 

26.  Medium  24  diluted  to  one-fourth  strength 

27.  Medium  24  diluted  to  one-sixteenth  strength 

28.  Medium  24  diluted  to  one  sixty-fourth  strength 

B.  Varying  concentrations  of  sugar.  Use  either  liquid  or  agar  media. 
The  same  controls  as  in  A  may  be  used,  if  the  same  fungi  are  tested. 

29.  Basal  medium,  with  40  g.  glucose 

30.  Basal  medium,  with  10  g.  glucose 

31.  Basal  medium,  with  5  g.  glucose 

32.  Basal  medium,  with  2  g.  glucose 

C.  Change  in  medium  during  incubation.  Use  liquid  media  in  flasks 
(15  ml.  in  125-ml.  Erlenmeyer  flasks,  or  25  ml.  in  250-ml.  flasks). 

33.  Basal  medium,  same  as  medium  24  above.     Inoculate  10  flasks;  after  growth  is 

near  maximum  (see  instructor),  separate  the  flasks  into  three  groups. 
a.  Replace  old  medium  with  fresh  medium 
h.  Replace  medium  with  sterile  distilled  water 
c.  Leave  as  control 

D.  Different  natural  products  as  media.     Use  as  agar  media. 

34.  V-8  juice  (diluted  to  one-half  strength) 

35.  Potato  extract  (200  g.  potatoes  per  liter) 

36.  Malt  extract  (20  g.  per  liter) 

37.  Stems  of  bean,  pea,  etc.,  in  water  agar 

Each  student  should  select  one  fungus  for  use  in  this  exercise.  It  is 
suggested  that  the  fungi  used  should  ordinarily  produce  considerable 
mycelium  before  fruiting. 

Records;  Careful  notes  must  be  taken  regarding  time  of  appearance 
of  fruit  bodies  and  spores.  The  amount  of  growth  and  sporulation  may 
be  compared  to  that  on  medium  24.  Consult  the  leader  or  your  instruc- 
tor for  further  details  on  recording  data. 

List  of  test  fungi:  Aspergillus  riigidosus,  Ceratostomella  Jimhriata, 
Choanephora  cucurhitarum,  Endothia  parasitica,  Glomerclla  cingulata, 
Guignardia  hidwellii,  Helminthosporium  victoriae,  Melanospora  sp.,  Moni- 
linia  fructicola,  N eocosmopara  vasinfecta,  Phoma  betae,  Sordaria  fimicola, 
and  Sphaeropsis  malorum. 


430  PHYSIOLOGY  OF  THE  FUNGI 

EXERCISE  7 
The  Influence  of  Hydrogen-ion  Concentration  on  Growth  and  Sporulation 

This  experiment  is  outlined  to  demonstrate  (1)  that  the  pH  require- 
ments for  optimum  growth  and  reproduction  vary  with  the  different 
species  of  fungi,  (2)  that  the  pH  of  the  culture  medium  changes  during  the 
growth  of  the  fungus,  (3)  the  techniques  by  which  pH  changes  may  be 
followed  during  growth,  (4)  that  the  pH  at  the  time  of  sporulation  may  be 
considerably  different  from  that  during  most  rapid  growth,  (5)  that  the 
pH  changes  are  also  influenced  by  the  composition  of  the  medium.  For  a 
discussion  of  pH,  see  Chap.  8.  The  pH  of  liquid  media  is  more  easily 
tested  than  that  of  solid  media  by  colorimetric  methods;  hence  liquid 
media  should  be  used  in  this  exercise.  Twenty-five  milliliters  of  medium 
in  a  250-ml.  Erlenmeyer  flask  gives  rapid,  even  growth  of  many  fungi  and 
has  been  found  to  be  quite  satisfactory.  Two  different  media  are  given 
below,  each  set  at  four  different  pH  values.  For  convenience  each  is 
given  a  separate  number. 

Media 

38.  Glucose-asparagine  (otherwise,  basal  medium  as  given  in  Ex.  4),  pH  3.0 

39.  As  above,  pH  4.0 

40.  As  above,  pH  6.0 

41.  As  above,  pH  8.0 

42.  Sucrose-nitrate  (otherwise,  basal  medium  as  given  in  Ex.  4),  pH  3.0 

43.  As  above,  pH  4.0 

44.  As  above,  pH  6.0 

45.  As  above,  pH  8.0 

Prepare  these  media,  adjust  the  pH  of  each,  and  distribute  to  flasks 
before  autoclaving.  Then  use  one  flask  of  each  medium  to  determine  the 
pH  after  autoclaving.  This  value  should  be  considered  the  "initial  pH." 
Each  student  should  select  one  fungus  and  inoculate  eight  flasks  of  each 
medium  listed  above.  Incubate  the  flasks  at  25°C.  or  at  room  tempera- 
ture. The  pH  of  the  culture  filtrate  should  be  determined  at  three  differ- 
ent times  during  the  period  of  active  vegetative  growth  (about  the  fourth 
and  eighth  days)  and  at  about  the  time  or  shortly  after  maximum  growth 
is  reached  (sporulation  of  some  fungi  will  occur  at  this  time).  Duplicate 
cultures  should  be  used  for  each  determination.  See  your  instructor 
regarding  the  method  of  determining  pH  of  culture  medium.  The 
sucrose-nitrate  media  may  not  be  favorable  for  the  growth  of  some  of  the 
fvmgi  selected  for  use  in  this  exercise. 

Records.  The  relative  amounts  of  growth  should  be  recorded  at  the 
time  of  each  pH  determination.  If  more  accurate  growth  measurements 
are  desired,  the  mycelium  can  be  dried  and  weighed.  Also  record  the 
time  of  the  earliest  sporulation  and  the  amount  at  subsequent  intervals. 


SUGGESTED  LABORATORY  EXERCISES  431 

Consult  the  leader  of  the  exercise  or  the  instructor  as  to  when  the  experi- 
ment should  be  terminated.  A  portion  of  these  data  may  be  presented 
in  the  form  of  a  graph,  plotting  changes  in  pH  against  time  for  each 
fungus  and  each  medium. 

List  of  test  fungi:  Aspergillus  rugulosus,  Cephalofhecium  roseum,  Cerato- 
stomella  fimhriata,  Glomerella  cingulata,  Monilinia  fructicola,  Neocosmo- 
para  vasinfecta,  Penicillium  spiculosporum,  Penicillium  expansum ,  Sordaria 
fimicola,  Sphaeropsis  malorum,  Phycomyces  blakesleeanus  (plus  and 
minus). 

EXERCISE  8 

Methods  of  Inoculating  Agar  Media  and  Their  Effect  upon   Growth 

and  Sporulation 

This  is  a  brief  and  simple  exercise,  but  it  is  outlined  to  demonstrate  a 
principle  which  seems  to  be  fundamental,  at  least  for  certain  fungi.  The 
most  common  way  of  inoculating  agar  media  is  to  place  a  bit  of  actively 
growing  mycelium  or  a  few  spores  at  the  center  of  the  medium  surface. 
For  most  purposes  this  is  entirely  satisfactory,  but  in  special  cases  other 
methods  may  be  used.  A  drop  of  spore  suspension  or  of  finely  cut 
mycelium  may  be  placed  at  the  center  of  the  agar  plate,  or  the  entire 
surface  may  be  flooded  with  heavy  spore  suspension  or  suspension  of  cut 
mycelium.  The  mycelium  may  be  fragmented  by  cutting  in  a  Waring 
Blendor  in  50  ml.  of  sterile  water  for  30  sec.  to  1  min. 

For  this  exercise  it  is  suggested  that  Petri  dishes  containing  glucose- 
asparagine-sucrose  agar  (such  as  the  basal  medium  in  Ex.  4)  be  used. 
Fungi  which  ordinarily  produce  considerable  mycelial  growth  before 
fruiting  abundantly  may  give  the  best  results  and  will  provide  spore-free 
inoculum  when  cultures  are  young.  Some  pycnidium-producing  species 
should  be  included. 

Each  student  should  choose  one  or  more  fungi  and  inoculate  plates  in 
triplicate  by  the  following  methods: 

A.  A  bit  of  mycelium  placed  at  the  center 

B.  A  few  spores  transferred  by  a  needle  placed  at  the  center 

C.  A  drop  of  suspension  of  cut  mycelium  placed  at  the  center. 

D.  A  drop  of  spore  suspension  placed  at  the  center. 

E.  Flooding  the  entire  plate  with  suspension  of  cut  mycelium 

F.  Flooding  the  entire  plate  with  a  heavy  spore  suspension 

G.  Diluting  the  spore  suspension  1/1,000  and  flooding  the  plate 

Observe  the  cultures  daily  and  note  the  abundance  of  vegetative  growth 
and  the  time  and  abundance  of  sporulation  in  each  case. 

List  of  test  fungi:  Alternaria  sp.,  Ceratostomella  fimhriata,  Dendrophoma 
ohscurans,  Endothia  parasitica,  Fusarium  sp.,  Glomerella  cingulata,  Guig- 
nardia  bidwellii,  Helminthosporium  sativum,  Monilinia  fructicola,  Neo- 


432  PHYSIOLOGY  OF  THE  FUNGI 

cosmopara  vasinfecta,  Phoma  bctae,  Septoria  nodorum,  Sordaria  fimicoia, 
Sphaeropsis  malorum. 

EXERCISE  9 
Vitamin  Deficiencies  in  the  Fungi 

This  exercise  is  oiithned  to  demonstrate  (a)  vitamin  deficiencies  in  the 
filamentous  fungi,  ih)  the  differences  in  the  needs  of  the  different  species 
of  fungi,  (c)  the  techniques  used  to  determine  these  deficiencies.  For  a 
discussion  of  vitamin  deficiencies  and  lists  of  vitamin-deficient  fungi,  see 
Chap.  9  and  Robbins  and  Kavanagh  (1942). 

In  studying  the  vitamin  requirements  of  the  fungi,  great  care  must  be 
taken  to  use  glassware  and  chemicals  which  are  free  from  vitamins.  The 
glucose-asparagine  medium  has  the  advantage  of  being  a  suitable  source 
of  carbon  and  nitrogen  for  most  fungi.  Casein  hydrolj'sate  may  be  used 
in  the  place  of  asparagine.  In  its  preparation,  the  medium  should  be 
boiled  with  activated  charcoal  (Norit,  5  g.  per  liter),  to  remove  any 
vitamins  present,  and  filtered.  Thus,  the  medium  is  "essentially  free" 
of  vitamins.  Contamination  may  occur  from  dust,  cotton  fibers  from  the 
plug,  dirty  glassware,  etc.  Micro  essential  elements  and  vitamins  must 
be  added  after  this  treatment. 

Media.  It  is  best  to  use  litiuid  media  for  this  exercise  so  the  mycelium 
can  be  dried  and  weighed. 

46.  Basal  glucose-asparagine  (vitamin-free) 

47.  As  above,  with  thiamine  (100  yug  per  liter) 

48.  As  above,  with  biotin  (5  ng  per  liter) 

49.  As  above,  with  thiamine  and  biotin 

50.  As  above,  with  thiamine,  biotin,  inositol  (5  mg.  per  liter),  and  pyridoxine  (100  pg 
per  liter) 

Adjust  the  pH  to  6.0  and  distribute  to  tubes  or  flasks  (25  ml.  per  250-ml. 
flasks  or  15  ml.  per  125-ml.  flasks)  before  autoclaving. 

A.  Screening  test  for  vitamin  deficiencies.  Simple  screening  tests  to 
determine  roughly  the  deficiencies  of  fungi  may  be  carried  out  in  either 
liquid  or  purified-agar  (see  Ex.  1)  media.  Test  tubes  may  be  used  for 
agar  media,  but  Erlenmeyer  flasks  are  suggested  for  liquid  media.  The 
student  should  select  four  or  five  species  from  the  stock-culture  collection, 
and  inoculate  tubes  or  flasks  of  each  of  the  above  media,  in  triplicate, 
with  each  species.  Either  spores  or  a  small  bit  of  mycelium  may  be  used 
as  inoculum.  Daily  observations  and  records  of  growth  should  be  made. 
Visual  estimates  of  relative  amounts  of  growth  are  sufficiently  accurate 
to  detect  most  deficiencies.  If  liquid  media  are  used,  the  mycelium  may 
be  weighed. 

B.  Growth  curves  of  vitamin-deficient  fungi.  Inoculate  10  flasks  each 
of  media  46  to  50  with  a  filamentous  fungus  shown  in  part  ^4.  to  be  defi- 


SUGGESTED  LABORATORY  EXERCISES  433 

cient.  The  fungus  will  be  harvested  at  intervals  and  the  amount  of 
growth  determined  by  obtaining  the  dry  weight  of  the  mycelium.  If 
there  is  sufficient  growth  in  the  flasks,  the  first  harvest  should  be  made 
after  3  or  4  days. 

Harvesting  of  the  mycelium  is  accomplished  by  filtering  the  medium 
through  a  fine  cloth  and  washing  the  mycelium  with  distilled  water. 
Harvest  duplicate  cultures  at  each  time.  The  mycelium  is  then  trans- 
ferred to  small  aluminum  pans  of  known  weight,  dried  for  12  to  24  hr.  at 
90°C.,  and  weighed.  The  subsequent  harvests  should  be  made  at  inter- 
vals of  2  to  4  days,  depending  upon  the  growth  rate  of  the  fungus  used. 
A  convenient  method  of  presenting  the  results  is  in  the  form  of  a  graph, 
ploting  time  against  weight  of  mycelium  for  each  of  the  four  media  used. 

EXERCISE  10 
The  Influence  of  Vitamin  Concentration  on  Growth  and  Sporulation 

This  exercise  is  designed  to  illustrate  a  few  important  principles  regard- 
ing the  need  for  an  adequate  supply  of  vitamins  in  the  medium  for  vita- 
min-deficient fungi.  Some  of  these  points  are  (1)  that  vegetative  growth 
may  be  limited  by  an  inadequate  supply  of  the  needed  vitamins ;  (2)  that 
higher  concentrations  of  vitamins  are  needed  for  reproduction  than  for 
vegetative  growth;  (3)  that  the  absolute  amount  of  a  vitamin  necessary 
to  induce  reproduction  varies  with  the  amount  of  sugar  in  the  medium; 
(4)  that  the  number  of  perithecia  (or  other  reproductive  structures)  is 
partiality  dependent  upon  the  concentration  of  the  vitamins  in  the 
medium ;  (5)  that  fungi  may  be  used  in  bioassays  for  the  vitamin  content 
of  various  products. 

The  following  experiments  are  suggested  (these  may  be  conducted  as 
demonstrations  before  the  whole  class,  if  desired) : 

A.   Thiamine  concentration  and  growth;  thiamine  assay 

Media 

51.  Basal  glucose-asparagine  medium,  no  vitamins 

52.  As  above,  but  with  100  ng  thiamine  per  liter 

53.  As  above,  25  fig  thiamine  per  liter 

54.  As  above,  12.5  ^g  thiamine  per  liter 

55.  As  above,  6.25  Mg  thiamine  per  liter 

56.  As  above,  3.12  ng  thiamine  per  liter 

57.  As  above,  1.56  ng  thiamine  per  liter 

58.  As  above,  add  0.5  g.  cake  flour  per  flask 

59.  As  above,  add  0.5  g.  whole-wheat  flour  per  flask 

Other  amounts  of  cake  and  whole-wheat  flour  may  be  used,  or  polished 
and  brown  rice  may  be  used  instead. 

Adiust  the  pH  of  the  above  media  to  6.0  and  distribute  25  ml.  each  to 


434  PHYSIOLOGY  OF  THE  FUNGI 

250-ml.  flasks.     Inoculate  four  flasks  of  each  medium  with  one  strain  of 
Phycomyces  blakesleeanus  or  Ceratostomella  Jlmbriata. 

B.  Thiamine  concentration  and  sporulation.  Use  media  51  to  53,  55, 
and  57,  but  sohdify  with  20  g.  purified,  vitamin-free  agar  (see  Ex.  1)  per 
Hter.  Pour  into  sterile  Petri  dishes  (about  20  ml.  each).  Inoculate  four 
plates  with  Ceratostomella  fimhriata  (be  sure  to  use  ascospores  or  mycelium 
producing  perithecia),  Phycomyces  blakesleeanus  (plus  and  minus  strains 
on  opposite  sides  of  plate),  Choanephora  ciicurhitarum,  and  Dendrophoma 
obscurans.  Incubate  Phycomyces  at  20  to  22°C.,  the  others  near  25°C. 
Cultures  of  Choanephora  must  be  adequately  aerated  and  must  receive 
alternate  light  and  darkness.  Observe  cultures  of  Ceratostomella  for 
production  of  perithecia,  Phycomyces  for  zygospores,  Choanephora  for 
conidial  heads,  and  Dendrophoma  for  pycnidia. 

Allow  sufficient  time  for  the  above  cultures  to  grow;  then  add  one  or 
two  drops  of  sterile  (autoclaved)  solution  of  thiamine  to  some  of  the 
thiamine-starved,  nonsporulating  cultures  (leave  controls).  Observe  the 
effects. 

C.  Effects  of  added  thiamine  on  thiamine-starved  mycelium 

Media 

60.  Distilled  water  and  purified  agar 

61.  Distilled  water,  100  ^g  thiamine  per  liter,  purified  agar 
Note:  The  addition  of  agar  in  media  60  and  61  is  not  essential. 

From  a  thiamine-starved  culture  of  C.  fimbriata  which  has  produced 
no  perithecia  cut  quarter-inch  disks  with  a  cork  borer  and  place  them  in 
tubes  of  media  60  and  61.  If  liquid  media  are  used,  make  sure  that  the 
disks  of  inoculum  float  on  the  surface.  Observe  the  results  after  a  few 
days. 

D.  Relation  of  required  thiamine  to  sugar  in  medium.  To  show  that  the 
concentration  of  thiamine  required  for  the  production  of  perithecia 
depends  upon  the  amount  of  sugar  in  the  medium,  this  short  experiment 
may  be  performed. 

Media.     Liquid  glucose-asparagine 

62.  Glucose  25  g.,  thiamine  1  ng  per  liter 

63.  Glucose  2.5  g.,  thiamine  1  fig  per  liter 

64.  Glucose  0.25  g.,  thiamine  1  ng.  per  liter 

Distribute  the  media  in  250-ml.  Erlenmeyer  flasks,  inoculate  with  C. 
fimbriata,  and  incubate  at  25°C.  Observe  the  rate  of  growth,  time  of 
perithecium  formation,  and  relative  number  of  perithecia  formed  in  each 
medium.  Harvest  and  weigh  the  mycelium  of  each  culture  after  peri- 
thecia have  formed. 


SUGGESTED  LABORATORY  EXERCISES  435 

E.  Effects  of  hiotin  starvation  on  a  biotinr-deficient  fungus 

Media 

65.  Glucose-asparagine,  purified  agar;  no  bio  tin 

66.  As  above,  5  ng  biotin  per  liter 

67.  As  above,  1  ng.  biotin  per  liter 

68.  As  above,  0.5  Mg  biotin  per  liter 

69.  As  above,  0.1  ng  biotin  per  liter 

Pour  into  Petri  dishes  and  inoculate  each  medium  in  tripHcate  with 
Sordaria  fimicola.  Observe  the  results  after  about  6,  8,  and  10  days. 
Note  the  amount  of  growth,  the  time  of  perithecium  formation,  and  the 
relative  numbers  of  perithecia.  Examine  microscopically  the  perithecia 
formed  in  the  low  concentrations  of  biotin,  and  look  for  deformed  asci 
and  ascospores.  Add  a  drop  or  two  of  sterile  biotin  solution  to  some  of 
the  nonfruiting,  biotin-starved  cultures.  Observe  the  effects  in  a  few 
days. 

Records.  Take  full  notes  on  all  observations  of  the  above  experiments. 
Write  out  a  full  explanation  of  the  results  with  interpretations  based 
upon  physiological  processes  in  fungi  (see  Chap.  14). 

EXERCISE  11 
Factors  Affecting  Spore  Germination 

This  exercise  demonstrates  the  effects  of  nutrients,  humidity,  pH,  and 
temperature  upon  the  time  and  percentage  of  germination  of  the  spores 
of  some  common  fungi.  While  the  germination  of  fungus  spores  may  be 
influenced  by  a  number  of  factors,  only  a  few  of  them  can  be  easily  studied 
in  the  laboratory.  See  Chap.  16  for  a  discussion  of  factors  which  influ- 
ence spore  germination. 

A.  Effects  of  nutrients  upon  spore  germination.  Place  filter  paper  in  the 
bottoms  of  Petri  dishes.  Cut  two  holes  in  the  filter  paper  about  }^  in. 
in  diameter  and  place  a  glass  slide  over  these.  Add  water  to  moisten 
the  paper,  and  autoclave.  Make  up  a  spore  suspension  in  media  70  and 
71,  and  place  drops  of  this  suspension  on  the  slide  over  the  holes  in  the 
paper,  which  must  be  kept  moist  w^ith  sterile  water.  Incubate  at  25°C. 
After  incubation  the  slide  may  be  examined  by  placing  the  Petri  dish  on 
the  microscope  and  removing  the  lid. 

Media 

70.  Distilled  water 

71.  Distilled  water  plus  2  g.  yeast  extract  per  liter 

72.  Same  as  medium  71  but  solidified  with  agar;  adjust  to  pH  8.0 

73.  Same  as  medium  71,  but  pH  7.0 

74.  Same  as  medium  71,  but  pH  6.0 

75.  Same  as  medium  71,  but  pH  5.0 

76.  Same  as  medium  71,  but  pH  4.0 


436  PHYSIOLOGY  OF  THE  FUNGI 

B.  Effect  of  acidity.  This  experiment  may  be  carried  out  in  liquid 
media  72  to  76,  using  drops  of  spore  suspension  as  described  above  in 
part  A,  or  the  media  may  be  solidified  with  agar  and  Petri  dishes  used. 
Observe  the  results  at  intervals  up  to  48  hr.  Your  records  should  include 
the  approximate  time  required  for  germination  of  the  first  spores  and  the 
percentage  of  germination  at  each  examination.  The  first  appearance  of 
a  germ  tube  may  be  considered  as  germination. 

C.  Effect  of  relative  humidity.  Place  drops  of  spore  suspension  in  dis- 
tilled water  on  sterile  glass  slides  to  serve  as  controls.  On  three  other 
dry  glass  slides,  place  dry  spores.  Place  these  in  desiccators  as  follows: 
(1)  spores  in  water  and  in  desiccator  which  will  maintain  saturated  atmos- 
phere; (2)  spores  on  dry  slide  in  saturated  atmosphere;  (3)  spores  on  dry 
slide  in  desiccator  with  relative  humidity  at  approximately  98  per  cent; 
(4)  spores  on  dry  slide  in  desiccator  with  relative  humidity  at  approxi- 
mately 92  per  cent.  To  maintain  98  per  cent  humidity,  use  a  1.00  molal 
solution  of  sucrose;  for  92  per  cent  humidity  use  a  saturated  solution 
of  K2HPO4  (see  Clayton,  1942).  Open  the  desiccators  after  24  to  30  hr. 
and  examine  the  spores  for  germination. 

D.  Effect  of  temperature.  Use  four  Petri  dishes  with  medium  74  or  75. 
On  each  place  three  or  four  drops  of  spore  suspension  and  mark  these 
spots  on  the  bottom  of  the  dish.  Incubate  these  as  follows:  (1)  in  an 
incubator  at  30°C.;  (2)  at  25°C.;  (3)  in  a  refrigerator  at  18°C.;  (4)  in  a 
refrigerator  at  10°C. 

Examination  of  spores.  The  time  required  for  spores  of  the  various 
species  to  germinate  under  the  usual  conditions  varies  from  2  to  24  hr.  or 
more.  For  some  fungi  all  the  experiments  in  the  exercise  may  be  exam- 
ined and  compared  after  12,  18,  or  24  hr.  One  examination  should  be 
made  after  48  hr.  For  most  species  the  experiments  may  be  concluded 
at  this  time. 

List  of  test  fungi:  Alternaria  sp.,  Cephaloihecium  roseum,  Choanephora 
cucurhitarum,  Glomerella  cingulata,  Guignardia  hidwellii,  Helminiho- 
sporium  sativum,  Monilinia  fructicola,  Penicillium  sp.,  Phytophthora 
infestans. 

EXERCISE  12 
The  Associative  Effects  among  Fungi 

Pure  cultures  of  a  single  organism  seldom  exist  in  nature.  Instead, 
each  organism  is  constantly  exposed  to  a  biotic  as  well  as  a  physical 
environment.  As  a  result,  there  is  usually  competition  between  different 
fungi  and  between  fungi  and  other  organisms  in  the  same  substrate, 
particularly  the  soil.  On  the  other  hand,  many  organisms  are  benefited 
by  their  association  with  others.  Often  the  metabolic  products  of  one 
favor  the  growth  of  another. 


SUGGESTED  LABORATORY  EXERCISES  437 

If  one  desires  to  demonstrate  these  principles  in  the  laboratory,  pure 
cultures  must  be  used,  and  by  combining  two  or  more  of  these  species  in  a 
culture  vessel,  the  associative  effects  may  be  studied.  This  exercise  is 
outlined  to  demonstrate  the  main  types  of  associative  reactions  between 
species  of  fungi  in  the  laboratory  under  controlled  conditions  and  to 
show  that  a  species  may  react  differently  in  its  association  with  different 
fungi. 

Media 

77.  Malt  extract-yeast  extract  agar 

78.  Glucose-asparagine  purified  agar  (vitamin-free) 

79.  Glucose-asparagine  liquid  (vitamin-free) 

A.  Each  student  should  select  three  pairs  of  the  fungi  and  test  their 
interaction  on  agar  plates  of  media  77  and  78.  Duplicate  plates  should 
be  inoculated  for  each  condition.  It  is  suggested  that  the  two  pairs  of 
fungi  be  inoculated  on  opposite  halves  of  the  agar  plates  1  to  2  in.  apart. 
This  will  allow  both  fungi  to  make  some  growth  before  they  come  in 
contact.  Incubate  all  cultures  at  25°C.  Notes  should  be  taken  on  about 
the  fifth  day  and  the  tenth  day,  and  for  most  cultures  the  final  observa- 
tions may  be  made  after  2  weeks.  Carefully  made  sketches  may  add 
greatly  to  the  value  and  clarity  of  your  notes. 

The  types  of  reactions  may  be  grouped  under  (1)  none,  (2)  stimulation, 
(3)  symbiotic,  (4)  antagonistic.  Each  type  of  reaction  should  be 
explained  on  the  basis  of  the  present  experiments. 

B.  Using  vitamin-free  liquid  medium  79,  inoculate  flasks  with  Phyco- 
viyces  blakesleeanus  and  Sordaria  fimicola  separately  and  with  both  species 
together.  After  a  few  days  observe  the  results.  This  part  of  the  exercise 
may  be  conducted  as  a  demonstration  for  the  entire  class. 

List  of  test  fungi:  Actinomyces  sp.,  Alternaria  sp.,  Aspergillus  rugulosus, 
Botrytis  sp.,  Cephalotheciimi  roseum,  Guignardia  hidwellii,  Helmintho- 
sporium  sativum,  Monilia  sp.,  Monilinia  fructicola,  Penicillium,  chryso- 
genum,  Phycomyces  blakesleeanus,  Sordaria  fimicola,  Trichodermalignorum. 

Suggested  Demonstrations 

In  addition  to  the  experiments  outlined  in  the  above  exercises,  the 
following  are  suggested  as  demonstrations  for  the  entire  class.  These 
may  be  expanded  into  complete  exercises  for  individual  student 
participation. 

1.  Need  for  micro  elements  for  growth  and  sporulation.  Steinberg  (1919) 
describes  the  procedure  for  preparing  a  medium  essentially  free  of  micro 
elements  to  which  the  desired  elements  may  be  added.  Use  Aspergillus 
niger  as  a  test  species.  See  Chap.  5  for  other  methods  of  removing  micro 
elements  from  media. 


438  PHYSIOLOGY  OF  THE  FUNGI 

2.  Influence  of  light  on  spore  discharge.  This  experiment  demonstrates 
the  phototrophic  response  of  sporangiophores  and  perithecial  beaks  and 
the  discharge  of  the  spores  toward  the  source  of  Hght.  For  a  discussion 
of  this  subject,  see  Buller  (1934),  Use  species  of  Piloholus,  Sordaria,  or 
Pleurage. 

3.  Influence  of  aeration  on  sporulation.  Choanephora  cucurbitarum  is 
an  excellent  species  to  use  in  demonstrating  the  need  for  adequate  aeration 
for  the  production  of  conidia.  Grow  the  fungus  on  agar  in  Petri  dishes. 
Some  of  the  dishes  may  be  sealed  with  Scotch  tape,  w^hile  the  lids  of  others 
may  be  raised  to  permit  free  exchange  of  gases. 

4.  Longevity  of  spores.  This  may  be  designed  as  a  long-time  experi- 
ment to  determine  the  longevity  of  spores  of  several  fungi  under  different 
conditions  of  storage.  It  may  be  continued  from  year  to  year,  tests  for 
the  ability  to  germinate  being  made  every  few  months. 

5.  Action  of  fungicides  and  fungistatic  agents.  These  experiments 
should  be  outlined  to  show  the  effectiveness  of  various  agents  in  prevent- 
ing spore  germination.  For  references,  see  Chap.  11  and  Horsfall  (1945). 
These  agents  include  the  action  of  sulfur,  copper,  8-hydroxyquinoline,  anti- 
vitamins,  and  ultraviolet  radiation. 

6.  Action  of  antibiotics  against  fungi.  The  specificity  of  the  action  of 
some  antibiotics  is  easily  tested  against  growth  or  spore  germination  of 
some  common  fungi  by  the  use  of  penicylinders  in  agar  plates  flooded 
with  spores. 

7.  Inheritance  of  physiologic  characters.  (A)  The  inheritance  of  vita- 
min or  amino-acid  deficiency  may  be  demonstrated  by  crossing  deficient 
mutants  of  Neurospora  with  a  self-sufficient  strain  (see  Beadle,  1946,  and 
Chap.  18).  {B)  The  inheritance  of  bioluminescence  may  be  demon- 
strated by  crossing  the  North  American  and  European  strains  of  Panus 
stypticus  (see  Macrae,  1942). 

REFERENCES 

Beadle,  G.  W.:  Genes  and  the  chemistry  of  the  organism,  Am.  Scientist  34:  31-53, 

1946. 
BuELL,  C,  and  W.  H.  Weston:  Application  of  the  mineral  oil  conservation  method 

to  maintaining  collections  of  fungus  cultures,  Am.  Jour.  Botany  34 :  555-561, 1947. 
Buller,  A.  H.  R.:  Researches  on  Fungi,  Vol.  VI,  Longmans,  Roberts  and  Green, 

London,  1934. 
Clayton,  C.  N.:  The  germination  of  fungous  spores  in  relation  to  controlled  humidity, 

Phytopathology  32:  921-943,  1942. 
Fennell,  D.  I.,  K.  B.  Rapeb,  and  M.  H.  Flickinger:  Further  investigations  on  the 

preservation  of  mold  cultures,  Mycologia  42  :  135-147,  1950. 
Georg,  L.  K.  :  a  simple  and  rapid  method  for  obtaining  monospore  cultures  of  fungi, 

Mycologia  39:  368-371,  1947. 
Greene,  H.  C.,  and  E.  B.  Fred:  Maintenance  of  vigorous  mold  stock  cultures,  Ind. 

Eng.  Chem.  26:  1297-1298,  1934. 


SUGGESTED  LABORATORY  EXERCISES  439 

HiLDEBRAND,  E.  M.:  Techniques  for  the  isolation  of  single  microorganisms,  Bolan. 

Rev.  4:  627-664,  1938. 
HoRSFALL,  J.  G.:  Fungicides  and  Their  Action,  Chronica  Botanica  Co.,  Waltham, 

1945. 
Macrae,  R.:  Interfertility  studies  and  inheritance  of  luminosity  in  Panus  stypticus, 

Can.  Jour.  Research,  Sec.  C,  20:  411-434,  1942. 
RiKER,  A.  J.:  The  preparation  of  manuscripts  for  Phytopathology,  Phytopathology 

36:953-977,  1946. 
RoBBiNs,  W.  J.:  Growth  substances  in  agar,  Am.  Jour.  Botany  26:  772-778,  1939. 
RoBBiNs,  W.  J.,  and  V.  Kavanagh:  Vitamin  deficiencies  of  the  filamentous  fungi, 

Botan.  Rev.  8:  411-471,  1942. 
Steinberg,  R.  A.:  A  study  of  some  factors  in  the  chemical  stimulation  of  the  growth 

of  Aspergillus  niger,  Am.  Jour.  Botany  6:  330-372,  1919. 
Thom,  C,  and  K.  B.  Raper:  Manual  of  the  Aspergilli,  The  Wilhams  &  Wilkins  Com- 
pany, Baltimore,  1945. 


I 


INDEX 


Abrams,  E.,  222,  223 
Absidia  coerulea,  100 

cylindrospora,  70,  100 

dubia,  100 

glauca,  100 

orchidis,  100 
Acetohader  stiboxydans,  119 
Achlya  bisexualis,  326 

conspicna,  92,  93 

oblongata,  93 

polyandra,  93 

prolifera,  93 
Aconiturn  velatum,  159 
Actinomyces,  286,  386,  437 

lavendulae,  287 

scabies,  373,  387 
Adaptations,  nutritional,  412-413 
Adaptive  enzymes,  59,  60,  412,  413 
Adermin  (see  Pyridoxine) 
Aegerita  icebberi,  384 
Aerobacter  aerogenes,  138 
Agar,  chemical  composition  of,  14 

purification  of,  14,  421 

vitamins  in,  15 
Agaricus  campestris,  269,  318,  348 

{See  also  Psalliota  campestris) 
Ajl,  S.  J.,  138,  146 
/3-Alanine,  198,  234,  239 
.Albrecht,  H.,  104,  112 
Albugo  Candida,  360,  373 
Aldoses,  117 
Aleuria  repanda,  346 

vesiculosa,  346 
Alexander,  D.  F.,  284,  301 
Allen,  R.  F.,  373,  392,  395 
Allomyces,  295,  327 

carbon  sources  used  by,  121 
Allomyces  arbiiscida,  121,  159,  308 

cystogenus,  121 
javanicus,  121 

moniliformis,  121 


441 


Alternaria,  319,  327,  425,  431,  436,  437 

oleracea,  261 

solani,  127,  247,  248,  262,  319,  320,  356 

tenuis,  69 
Amanita,  292 

chlorinosma,  292 

mappa,  366 

muscaria,  81,  292 

pantherina,  292 

phalloides,  292 

porphyria,  292,  366 

radicata,  292 

rubescens,  366 

spreta,  292 

strobiliformis,  292 

verna,  292 

virosa,  292 

toxin  (see  Phalloidin) 
Ames,  L.  M.,  222,  224,  404,  416 
Amino  acids,  105-110 

antagonists  of,  236-239 

as  carbon  sources,  127 

deamination  of,  108 

list  of,  106 
p-Aminobenzoic  acid,  201-202 

antagonists  of,  227,  229-236 

formula  of  p-aminobenzoate  ion,  227 
Ammonium  nitrogen,   list  of  fungi   uti- 
lizing, 100 
Amylase,  47,  48,  50,  274 
Andersen,  A.  L.,  163,  169,  308,  309,  317, 

335,  351,  353 
Andersen,  E.  N.,  348,  354 
Anderson,  A.  K.,  158,  169,  308,  335 
Anderson,  D.  B.,  37,  44 
Anderson,  H.  W.,  291,  302 
-4nderson,  P.  J.,  368,  369 
Aneurine  (see  Thiamine) 
Antagonists,  metabolite,  226-240 
Antibiotics,  control  of  plant  diseases  by, 
288 
detection  of,  284-285 
as  factor  in  resistance,  390 
production  of,  283-291 


442 


PHYSIOLOGY  OF  THE  FUNGI 


Antimetabolite,  definition  of,  22G 
Anti vitamins,  229-23(3 
Anwar,  A.  A.,  286,  296 
Aphanornyces,  93 

camptostylus,  92 
Apium  graveolens,  377 
Apodachyla  brachynema,  127 
Apoenzymes,  53,  228 
Appling,  J.  W.,  219,  223 
Appressoria,  373 
Arabinose,  formula  of,  123 
Arcyria  denudata,  367 
Area  Leao,  A.  E.,  197,  203 
Armillaria  mellea,  99,  134,  160,  374,  387 

shii-take,  269 
Armstrong,  E.  F.,  129,  146 
Armstrong,  J.  I.,  167,  168 
Arnon,  D.  I.,  21,  23,  73,  86 
Arsenic,  110 
Arthur,  J.  C,  372,  395 
Aschersonia  aleyrodis,  384 
Ascobolus,  345,  346,  349,  353,  357 

denudata,  99 

leveillei,  99 

rnagnificus,  346 
Ascochyta  nymphaeae,  311,  317 

pisi,  99 
Ascophanus  carnens,  88,  319 
Ashbya  gossypii,  273 
Ashley,  J.  N.,  295,  296 
Aspergillus,  9,  94,  99,  172,  271-275,  283, 
288,  385,  401 

alliaceus,  414 

amstelodami,  414,  415 

aureus,  279 

awaniori,  279 

carbonarius,  279 

cinnamomeus,  279,  280 

clavatus,  279,  286,  288 

fischeri,  414 

flavus,  70,  76,  78,  414 

metabolic  products,  272,  275,  283, 

287,  288 
spore  germination  of,  361,  362,  366 

fumaricus,  279 

jumigatus,  34,  76,  414 

fuscus,  280 

giganteus,  283 

glaums,  37,  275,  279 

herbariorurn,  318 

itaconicus,  281 


Aspergillus,  nidulans,  93,  283,  414 

niger,  10,  55,  92,  98,  103,  136,  140,  167, 
194,  229-231,  248,  258,  260,  261, 
294,  325-327,  360-362,  366,  414, 
415,  437 
nutrition  of,  carbon,  121-129 
metallic-element,  65-83 
nitrogen,  105 
phosphorus,  94-96 
organic  acids  produced  by,  279,  280 
resistance  of  onion  to,  390,  391 
spore  germination  of,  358,  359 
use  of,  in  assays,  217-222 
vitamin  synthesis  by,  171,  172 
oryzae,  74,  76,  88,  89,   103,  159,  274, 
280,  283,  368 
carbon  sources  utiUzed  by,  121,  123- 
125,  127,  129,  132,  134,  136 
repens,  307,  308 
rugulosus,  28,  220,  328,  329,  425,  428, 

429,  431,  437 
terreus,  70,  222,  282,  287,  401 
variecolor,  414 
wentii,  278 
Assays,  microbiological,  for  amino  acids, 
216-217 
certain  media  used  in,  210,  211 
for  essential  elements,  217-219 
general  discussion  of,  208-209 
procedure,  209-214 
for  vitamins,  214-216 
Assimilation,  definition  of,  87 
Association,  effect  of,  on  growth,  219,  220 
on  sporulation,  327-328,  332-333 
laboratory  exercise,  436-437 
Astbury,  W.  T.,  110,  112 
Atkin,  L.,  198,  204,  215,  223 
Auernheimer,  A.  H.,  219,  223 
Aureomycin,  291 

Autoclaving,  effect  of,  on  media,  16 
Axelrod,  A.  E.,  194,  204 
Ayres,  T.  T.,  328,  334 
Azotobader,  98 

chroococcum,  79,  218 
vinlandii,  98 

B 

Bacillus  mesentericus,  414 
subtilis,  286 
typhosujn,  229 
weidmaniensis,  327 


INDEX 


443 


BaUey,  A.,  316,  336 
Bailey,  J.  H.,  290,  296 
Barker,  B.  T.  B.,  248,  263 
Barnes,  T.  C,  88,  112 
Barnett,  H.  L.,  34,  38,  43,  44,  167,  169, 
178,  181-184,  196,  204,  205,  232,  242, 
307,   309,   310,   312,   313,   317,   319, 
329-331,  334,  336,  400,  401,  408-410, 
416,  418 
Barratt,  R.  W.,  259,  263 
Barron,  E.  S.  G.,  57,  62,  96,   112,  251, 

263,  264 
Basidiohlus,  343,  353 

ranarum,  100,  342 
Basisporium  gallarum,  27,  172,  365,  388 
Basu,  S.  N.,  329,  335 
Battarrea,  307 

Baumberger,  J.  P.,  60,  61,  64 
Bayliss,  W.  M.,  54,  62 
Beadle,  G.  W.,  36,   161,   196,   197,   199, 
202,   204,   207,   211,   215,   217,  223, 
224,  401,  405-408,  416,  438 
Behr,  G.,  68,  69,  85 
Behrens,  O.  K.,  289,  296 
Bellamy,  W.  D.,  290,  296 
Benedek,  T.,  327,  334 
Bennett,  C.  W.,  79,  86,  100,  115 
Bennett,  I.  G.,  36,  43 
1,4-Benzoqiunone  formula,  257 
Berberis,  387 

thunbergii,  387 
Berger,  J.,  51,  63 
Bergmann,  M.,  51,  62 
Bernhard,  K.,  104,  112 
Bernhauer,  K.,  279,  296 
Bertrand,  D.,  81,  83 
Bertrand,  G.,  119,  146 
Bessey,  E.  A.,  375,  395 
Biological  substitution  of  elements,  68 
Bioluminescence,  402,  403,  410 
Biotin,  192-195 

antagonists  of,  233 

formula  of,  192 

fungi  deficient  for,  192,  193 

mode  of  action  of,  194,  195 

specificity  of,  193,  194 
Birkinshaw,  J.  H.,  94,  112,  276,  283,  296 
Bisby,  G.  R.,  313,  334 
Black,  A.,  212,  224 

Blakeslea  trispora,  102,  122,  124,  125,  133, 
190 


Blank,  L.  M.,  76,  83 

Blastocladia  pringsheimii,   93,    102,    137, 
156,  159,  197,  281 

production  of  resistant  sporangia  by, 
319 

vitamin  deficiencies  of,  180 
Blastodadiella  simplex,  159 
Blastomyces  brasiliensis,  40 

dermatitidis,  40 
Block,  R.  J.,  268,  296 
Bock,  H.,  50,  63 
Boletus,  293,  367 

bovinus,  367 

jdulis,  267 

elegans,  367 

flavidus,  367 

granulatus,  367 

luridus,  52,  292 

hiteus,  367 

miniato-olivaceus  var.  sensibilis,  293 

satanus,  292 

variegatus,  367 

viscidus,  367 
Boiling,  D.,  268,  296 
Bonar,  L.,  309,  336 
Bonner,  D.,  93,  112,  197,  204,  217,  223, 

401,  408,  416 
Bonner,  J.,  50,  62 
Bonner,  J.  T.,  358,  359,  369 
Booer,  J.  R.,  250,  263 
Borchers,  R.,  273,  300 
Bordeaux  mixture,  247-248 
Bortels,  H.,  79,  84 
Bose,  S.  R.,  319,  335 
Botryotinia  convoluta,  99,  311 
Botrytis,  253-255,  314,  437 

alia,  99,  172,  250,  261,  285 

production  of  pectinase  by,  376 
resistance  of  onion  to,  390,  391 

cinerea,  37,  69,  99,  373,  425,  426 
parasitism  of,  376,  377 
production  of  pectinase  by,  377,  378 
spore  germination  of,  360,  364-366 

paeoniae,  247,  248 
Bourquelot,  E.,  47,  62 
Bovarnick,  M.  R.,  197,  204 
Boyle,  A.  M.,  288,  296,  389,  395 
Brand,  E.,  216,  223,  225 
Brassica,  377 
Bray,  C.  W.,  269,  296 


444 


PHYSIOLOGY  OF  THE  FUNGI 


Brefeld,  O.,  6,  7,  9,  11,  22,  266,  296,  310, 

335,  339,  353,  362,  370 
biographical  note,  6 
Brenner,  W.,  107,  112 
Brian,  P.  W.,  97,  103,  104,  108,  112,  165, 

168,  285,  287,  290 
Briarea,  35 

Brink,  N.  G.,  273,  297 
Brodie,  H.  J.,  324,  335,  358,  370,  402,  416 
Brooks,  F.  T.,  382,  395 
Brown,  W.,  364,  365,  370,  372,  373,  376, 

387,  390,  395 
Brown  rot,  fungi  causing,  134,  135 
Buchanan,  R.  E.,  17,  22,  25,  43,  65,  84, 

99,  112 
Buchner,  E.,  46,  62,  277,  297 
Buell,  C,  424,  438 
Buffer  capacity  of  media,  155 
Buffers,  152-156 
Bull,  H.  B.,  110,  112 
Buller,  A.  H.  R.,  6,  7,  267,  297,  339,  341, 

345-349,  351,  353,  408,  409,  416,  438 
biographical  note,  6 
Bunker,  H.  J.,  49,  64 
Blinning,  E.,  55,  62,  167,  168 
Burgert,  I.  A.,  362,  370 
Burk,  D.,  108,  112,  186,  204,  233,  241 
Burkholder,  P.  R.,  42,  43,  104,  112,  174, 
180,  181,  186,  190,  192,  195,  197-199, 

204,  272,  297,  400,  416 
Buston,  H.  W.,  329,  335 
Butler,  E.  J.,  381,  387-389,  395 


c 


Calam,  C.  T.,  282,  297 
Calcium,  79-81 
Caldwell,  R.  M.,  373,  395 
Calfee,  R.  K.,  77,  78,  85 
Camp,  W.  G.,  320,  335 
Campbell,  W.  G.,  134,  146 
Canavanine,  239 
Candida  albicans,  42,  186,  287 

flareri,  272 

guilliermondi,  219,  272 
Cantherella  cibarius,  267 
Cantino,  E.  C,  93,   102,   112,   137,   146, 
156,   159,   168,    169,   180,    197,   204, 
319,  335 
Caputto,  R.,  140,  146 


Carbon  dioxide,  effect  of,  on  spore  germi- 
nation, 365 
as  metabolite,  136-138 
Carbon  sources,  effect  of,  on  sporulation, 
321-325 
laboratory  exercise,  426-428 
utilized  by  fungi,  116-138 
Carbon  utilization,  138-144 
Carboxylase,  52 
Caroselli,  N.  E.,  379,  396 
Carroll,  W.  R.,  91,  92,  114 
Carter,  H.  E.,  291,  297 
Catechol,  390 

Cavallito,  C.  J.,  290,  296,  297 
Cellobiose  formula,  130 
Cellulase,  47 

Cellulolytic  fungi,  221-222 
Cellulose,  134-135 
Cephalosporium,  378 
Cephalotheciuni  roseum,  99,  425,  426,  431, 

436,  437 
Ceratostomella,  200,  293,  352,  378 
anipullacea,  344 

fimbriata,  31,  37,  38,  100,  101,  175,  178, 
181,  190,  232,  345,  425,  426,  428, 
429,  431,  434 
factors    affecting    sporulation,    307, 

329,  330 
methods  of  measuring  growth  of,  30 
ips,  34,  180,  190,  194,  200 
inicrospora,  200 
montium,  190,  200 
multiannulata,  200 

(See    also    Ophiostoma    multiannula- 
tum) 
penicillata,  232 
piceaperda,  200 
pilifera,  34,  35,  200 
pint,  190,  200,  229 
pluriannulata,  200,  329 
pseudotsugae,  200 

ulmi,  100,  175,  200,  234,  260,  314,  345, 
375 
causes  of  wilting  by,  378,  379 
{See  also  Ophiostoma  ulmi) 
Cercospora  apii,  99,  172 
beiicola,  99,  172,  373 
nicotianae,  80 
Chaetocladium,  385 
Chaetomium,  222,  345 


INDEX 


445 


Chaetomium,  cochlioides,  99 

convolutum,  38,  99,  178,  190,  193,  232 
dilution  of  medium  and  growtli  of,  38 

funicola,  96,  222 

globosum,  99,  127,  172,  173,  222,  318 
growth  curve  of,  173 
Chain,  E.,  290,  297 
Chalaropsis  thielavioides,  232 
Challenger,  F.,  111-113 
Chan,  S.  Y.,  319,  320,  336 
Charles,  J.  H.  V.,  294,  297 
CheldeKn,  V.  H.,  199,  207,  232,  243 
Cheo,  P.  C,  120,  146,  362,  370,  401,  416 
Chilean  Nitrate  Educational  Bureau,  71, 

84 
Chilton,  S.  J.  P.,  404,  417 
ChIai7iydopus,  307 
Chloromycetin,  291 
Chlorosplenium  aeruginosum,  293 
Choanephora  cucurbitarum,  33,   100,  190, 
232,  401,  425,  426,  429,  434,  436,  438 

factors  affecting  sporulation   of,   309, 
310,  312,  313,  321,  331,  332 
Chona,  B.  L.,  376,  377,  395 
Christenberry,  G.  A.,  312,  335,  401,  417 
Christensen,   J.   J.,   293,   297,   402,   410, 

414,  417 
Chrysophlydis    endohiotica    (see    Synchy- 

triuin  endohioticum) 
Ciboria  acerina,  27 
Cicinnobolus  cesatii,  385 
Citric  acid,  production  of,  279-280 
Citromyces,  275,  279 

citricus,  279 

glaber,  279 

pfefferianus,  279 

{See  also  Penicillium) 
Cladosporium  cucumerinum,  288 

fulvum,  390 
Clarke,  H.  T.,  290,  297 
Clausson-Kaas,  N.,  378,  397 
Claviceps,  332 

paspali,  293 

purpurea,  291-293,  296,  304,  352 
Clayton,  C.  N.,  355,  358,  370,  436,  438 
Clitocybe  illudens,  293 
Clostridium,  376 

acetobutylicum,  201,  272,  273 

pasteurianum,  79 

septicum,  198 
Cobalt,  81,  273 


Cocarboxylase   (see  Thiamine  pyrophos- 
phate) 
Coccomyces  hiemalis,  304 
Codehydrogenase,  197 
Coemansia  interrupta,  178,  190,  193 
Coenzyme,  53 
Coghill,  R.  D.,  282,  284,  289,  297,  300, 

301 
Colletolrichum,  368 

circinans,  390,  391 

falcatum,  388 

gossypii,  360 

lagenarium,  99,  288,  356 

lindemuthianum,  99,  160,  186,  196,  307, 
309,  332,  374 
resistance  to,  391 
spore  germination,  364-366 
CoUybia  tuberosa,  99 

velutipes,  99,  409 
Colwell,  C.  A.,  258,  263 
Committee    on    the    Standardization    of 

Fungicidal  Tests,  262,  263 
Conant,  N.  F.,  333,  335,  386,  395 
Concentration  of  nutrients,  17-20 

effect  of,  on  growth,  38,  39 
on  sporulation,  320-321,  330 
Coniophora  cerebeUa,  159 
Coniothyrium,  316 
Cook,  E.  S.,  251,  263 
Cook,  M.  T.,  388,  390,  395 
Coons,  G.  H.,  307,  310,  311,  313,  317, 

335 
Copper,  76-77,  217 
Coprinus,  310 

comatus,  362 

jimetarius,  362 

lagopus,  102 

micaceus,  362 
Cordyceps  militaris,  99,  172,  332,  384 
Coriicium  solani,  386,  401,  402 

(See  also  Rhizodonia  solani) 
Coryell,  C.  D.,  60,  62 
Couch,  J.  N.,  352,  353,  384,  385,  396 
Craigie,  J.  H.,  387,  397 
Craterellus  liitescens,  367 
Cristol,  S.  J.,  238,  241 
Cronartiurn  ribicola,  356,  368,  395 
Crosier,  W.,  34,  43 
Cryptococcus  neoformans,  260,  287 
Crystal  violet,  259 


446 


PHYSIOLOGY  OF  THE  FUNGI 


Cultures,  preservation  of,  423-424 
single-spore  isolation,  424 

Cunninghamella  elegans,  70 

Currie,  J.  N.,  278,  283,  297 

Curtin,  T.,  294,  297 

Cury,  A.,  197,  203 

Cyanide,  effect  of,  on  respiration,  56 

Cyathus,  353 
siercorius,  324 
striatus,  100,  348 

Cylindrosporium  scoparium,  358 

Cystine,  biosynthesis  of,  93 

Cytophaga,  16 

Cytospora,  368 

Cytosporella  mendax,  311 


D 


Daedalea  quercina,  134 

tinicolor,  351 
Daldinia  concenirica,  172 
Dalphin,  C,  191,  204 
Darluca  filum,  385 
Das  Gupta,  S.  N.,  402,  417 
Davis,  A.  R.,  98,  113 
Davis,  F.  R.,  414,  417 
Davis,  J.  B.,  137,  146,  282,  298 
Dawson,  C.  R.,  52,  63,  77,  85 
Day,  D.,  14,  22,  29,  43 
Deamination,  107,  108 
De  Bary,  A.,  6,  7,  283,  343,  354,  376,  396 

biographical  note,  6 
Debar ijomyces,  309 
Decker,  P.,  386,  398 
Dendrophoma    obscurans,    99,    190,    426, 

428,  431,  434 
Denny,  F.  E.,  317,  335 
Desoxypyridoxine  formula,  234 
Desthiobiotin,  antagonists  of,  233 

formula  of,  193 
Dianthus,  377 
Diaporthe  phaseolorum  var.  batatatis,  312, 

316,  321,  324 
Diastase  (see  Amylase) 
Dichtyuchus  monosporus,  93 
Dickinson,  S.,  402,  411,  417 
Dickson,  J.  G.,  393,  396,  402,  417 
Dicranophora  juiva,  295 
Didydium  cancellatum,  367 
Dillon- Weston,  W.  A.  R.,  250,  263 
Dimond,  A.  E.,  164,  168,  378,  396 


Dinitrophenol,  61,  91 

Diplodia  macrospora,  17,  120-122,  193 
natalensis,  28,  34,  122 
tubericola,  37 

Disaccharides,  130-131 

Dissimilation,  definition  of,  87 

Dittmer,  K.,  233,  238,  241 

Dixon,  M.,  45,  62,  90,  113 

Dodge,  B.  O.,  1,  7,  348,  354,  357,  370, 
382,  396,  404,  417 

Dodge,  C.  W.,  386,  396 

Doermann,  A.  H.,  216,  224 

Dole,  M.,  158,  168 

Donovick,  R.,  290,  301 

Doran,  W.  L.,  356,  358,  360,  361,  368,  370 

Dormancy  of  spores,  355,  357 

Dorrell,  W.  W.,  91,  113 

Dothidella  quercus,  99 

Doudoroff,  M.,  324,  335 

Dowding,  E.  S.,  401,  417 

Dox,  A.  W.,  95,  113,  129,  146 

Drayton,  F.  L.,  311,  335 

Drosophila,  405 

Dual  phenomenon,  411-412 

Dubos,  R.  J.,  6,  7 

Dufrenoy,  J.,  290,  300,  390,  396 

Duggar,  B.  M.,  98,  113,  270,  291,  297, 
361,  362,  366,  370 

Dulaney,  E.  L.,  127,  146,  290,  297 

Dung  infusion,  effect  of,  on  spore  germi- 
nation, 362 
as  medium,  9 

Dunn,  C.  G.,  142,  147,  266,  277,  278,  300 

Dunn,  M.  S.,  216,  224 

Durrell,  L.  W.,  365,  370 

Duschinsky,  R.,  233,  241 

Duthie,  E.  S.,  290,  297 

Du  Vigneaud,  V.,  192,  194,  204,  233,  237, 
241 

Dyctidiurn  cancellatum,  367 


E 


Eakin,  E.  A.,  194,  199,  204 
Eakin,  R.  E.,  194,  204,  233,  241 
Eastcott,  E.  v.,  195,  204 
Echinodontium  tinctorium,  293 
Edgerton,  C.  W.,  404,  417 
Edwards,  G.  A.,  40,  44 
Ehrlich,  J.,  291,  297 
Elfving,  F.,  35,  43,  314,  335 


A 


I 


INDEX 


447 


Elliott,  E.  S.,  184,  204 

Elvehjem,  C.  A.,  51,  62 

Emerson,  M.  A.,  363,  370 

Emerson,  R.,  159,  169,  295,  297,  319,  327, 

335 
Emerson,  S.,  26,  43 
Emmons,  C.  W.,  386,  396 
Enantiomorphs  of  sugars,  117,  119 
Endoconidiophora  monilifonnis,  283 
Endomyces  vernalis,  231,  232,  235,  239, 

240,  271 
Endothia  -parasitica,   100,   180,   190,    193, 
368,  375,  426,  428,  429,  431 
ascospore  discharge  by,  344 
factors  affecting  sporulation  by,   311, 
314,  321 
Energy  utilization,  60 
heat  evolved  in,  61 
Englis,  D.  T.,  16,  22 
Enteridium  rozeanum,  357,  367 
Entomophthora,  343,  353,  384 

sphaerosperma,  343 
Enzymes,  45-62 
activators  of,  55 
adaptive,  59,  60,  412-413 
chemical  nature  of,  52,  53 
classification  of,  47-51 
effect  of  radiation  on,  57,  58 
factors  affecting  activity  of,  53-58 
inhibition  of,  55-57,  227,  228,  377-378 
lock-and-key  relationship  of,  228 
mechanism  of  action  of,  58 
naming  of,  47 

pH  and  rate  of  reaction  of,  55,  161 
production  of,  274-275 
role  of,  in  parasitism,  376,  377,  382 
suberin-dissolving,  374 
temperature  and  rate  of  reaction  of,  53 
toxicity  of,  376,  382 
Epichloe  typhina,  196 
Epidermophyton  floccosum,  70,  238 
Epstein,  S.,  291,  297 
Erb,  N.  M.,  274,  297 
Eremothecium  ashbyi,  273 
Ergot,  291 

Erwinia  carnegieana,  288,  289 
Erysiphe,  359 
graminis,  358 
graminis  hordei,  358 
polygoni,  311,  358,  377 


Escherichia  coli,  166,  201,  234,  238,  240, 
251,  285 

Essential  elements,  functions  of,  66 
list  of,  67,  82 

Esters,  production  of,  283 

Ethyl  pyridoxine  formula,  234 

Eucalyptus,  365 

Exidia,  408 

glandulosa,  409 

Experimental    results,    presentation    of, 
31-32 

External    factors,     effect    of,     on    mor- 
phology, 39-42 

Ezekiel,  W.  N.,  391,  396 


Fastness,  240 

Fat  production,  270-272 

Fawcett,  H.  S.,  28,  33,  34,  43,  384,  396 

Feldman,  A.  W.,  378,  379,  396 

Felix,  E.  L.,  235,  243,  257,  265 

Fellows,  H.,  12,  22,  89,  113 

Fennell,  D.  I.,  424,  438 

Ferger,  M.  F.,  237,  241 

Fermentation,  89,  142,  275-277 

Fieser,  L.  F.,  235,  241 

Fildes,  P.,  57,  63,  230,  241,  251,  263 

Fischer,  E.,  219,  224 

Fischer,  O.  E.,  292,  293,  298 

Fitzpatrick,   H.   M.,  343,  354,  385,  396 

Fitzpatrick,  W.  H.,  267,  298 

Fleming  A.,  220,  224,  290,  298 

Fleury,  C,  261,  263 
Florey,  H.  W.,  290,  297 
Fomes  applanatus,  41 

roseus,  159 
Fontaine,  T.  D.,  390,  396 
Food  from  fungi,  267-270 
Foote,  M.  W.,  258,  263 
Formaldehyde,  256 
Foster,  A.  A.,  249,  263 
Foster,  J.  W.,  39,  43,  65,  71,  76,  84,  86, 
103, 113, 136-138,  146,  156,  169,  266, 
272,  273,  277,  278,  280-283,  290,  298, 
302 
Fourneau,  E.,  231,  241 
Fox,  D.  L.,  295,  297,  327,  335 
Frear,  D.  E.  H.,  246,  247,  256,  263 
Fred,  E.  B.,  424,  438 
Frey-WyssUng,  A.,  82,  84 


448 


PHYSIOLOGY  OF  THE  FUNGI 


Friedemunn,  T.  E.,  142,  146 
Fries,  N.,  29,  31,  42,  43,  i)3,    113,    139, 
146,    190,   193,    195,  204,   205,   219, 
224,  366,  367,  370,  401,  417 
Fromageot,  C,  184,  204 
Fructose  formula,  119 
Fudge,  J.  F.,  391,  396 
Full  go  se plica,  361,  367,  368 
Fuller,  W.  H.,  49,  63,  134,  147 
Fulmer,  E.  I.,  17,  22,  25,  43,  65,  84,  99, 

112 
Fumago  vagans,  280 
Fumaric  acid,  production  of,  282-283 

scheme  for  biosynthesis  of,  137 
Fungi,  as  food,  267-270 

list  of  vitamin-sufficient,  172 
role  of,  in  nature,  1,  266,  267 
Fungicide,  meaning  of,  245,  246 
Fungicides,  copper,  246-249 
dithiocarbamate,  259 
evaluation  of,  261-262 
mercury,  249-251 
mode  of  action  of,  245,  249 
organic,  256-261 
quinone,  257-258 
Fungistatic  agent,  definition  of,  245 
Fusarium,   99,    140,    141,    172,   271,   273, 
275,   293,   313,   360,   375,   378,   402, 
411,  431 
avenaceum,  184,  400 
cepae,  316 
coeruleurn,  70 
discolor  sulphureum,  313 
eumartii,  89,  316 
graminearum,  91,  261,  295 
{See  also  Gibberella  zeae) 
lini,  96,  102,  103,  108,  138,  141,  186, 
191,  192,  270,  275,  295,  374,  387, 
390 
lycopersici,  122,  133,  160,  378,  390 
{See  also  oxysporum  var.  lycopersici, 
below) 
moniUforme,  127 
niveum,  188,  189 
orthoceras,  287 
oxysporum,  80,  88,  89,  222 
var.  lycopersici,  127,  260 

{See  also  lycopersici,  above) 
var.  nicotianae,  80 
sambucinum,  139 
solani,  295 


Fusarium,  trichothecioides,  96 
vasinfectum,  380 

G 

Galactitol  formula,  124 
Galactose  formula,  1 19 
Gallegly,  M.  E.,  392,  396 
Gallium,  81 

Ganoderma  lobatum,  134 
Gardner,  M.  W.,  356,  370 
Gastrock,  E.  A.,  280,  298 
Gaumann,  E.,  372,  378,  396 
Geiger,  W.  B.,  258,  260,  263 
Georg,  L.  K.,  424,  438 
Gerhardt,  P.,  279,  298 
Gibberella  saubinetti,  159,  391 

{See  also  zeae,  below;  Fusarium  grami- 
nearum) 
zeae,  293,  393,  402 

{See  also  saubineiii,  above; F usarium 
graminearum) 
Giles,  N.  H.,  413,  417 
Gillespie,  J.  M.,  201,  206,  230,  242 
Gilman,  H.,  117,  130,  146 
Gimingham,  C.  T.,  248,  263 
Gingrich,  W.,  197,  204 
Gliomastix  convoluta.  Til 
Gloeosporium,  368 
Glomerella,  404,  415 

cingulata,  36,   99,    172,   247,   248,   254, 
255,  261,  262,  402,  425,  428,  429, 
431,  436 
nutrients,  and  spore  germination  of, 
362,  363 
and  sporulation  of,  322,  324,  325 
ultraviolet  light  and  sporulation  of, 
315 
gossypii,  98 
Gluconic-acid  production,  280-281 
Glucose,  fermentation  of,  141,  142 
formula  of,  119 
a-D-gkicose,  128 
|3-D-glucose,  128 
fungi  not  utiUzing,  120 
Glutathione,  93 
Glycolysis,  141,  142 
Glycosides,  128-129 
Glynne,  M.  D.,  253,  264 
Gnomonia  ulmi,  304,  344 
Goddard,  D.  R.,  357,  370 


I 


INDEX 


449 


Goepp,  R.  M.,  Jr.,  14,  22,  117,  119,  128, 

133,  147 
Goering,  H.  L.,  238,  241 
Goldberg,  M.  W.,  194,  204 
Goldsworthy,  M.  C,  247-249,  2G3 
Gonatorrhodiella  highlei,  328 
Goodman,  I.,  238,  241 
Gorcica,  H.  J.,  268,  298 
Gordon,  M.  A.,  283,  298 
Gortner,  R.  A.,  37,  43,  47,  51,  63,  70, 

84,  88,  105,  113,  154,  169 
Gottlieb,  D.,  127,  146,  286,  298,  356,  370 
Gould,  B.  S.,  96,  113 
Graham,  T.  W.,  402,  417 
Grant,  W.  M.,  414,  418 
Grassmann,  W.,  134,  146 
Greathouse,  G.  A.,  222,  224 
Green,  E.  L.,  247-249,  263 
Greene,  H.  C,  424,  438 
Greene,  R.  D.,  212,  224 
Gregory,  P.  H.,  350,  351,  354 
Gries,  G.  A.,  257,  263 
Grob,  E.  C,  236,  243 
Grossbard,  E.,  286,  298 
Growth,  abnormal,  39-41 
definition  of,  24 

effect  of  depth  of  medium  on,  38 
factors  affecting,  32-38 
measurement  of,  27-31 
phases  of,  25-26 

of  filamentous  fungi,  26 
of  unicellular  organisms,  25 
rate  of,  26 
tube,  29 
Growth  factors  {see  Vitamins) 
Guignardia  bidwellii,  304,  344,  389,  425, 

428,  429,  431,  436,  437 
Guilloud,  M.,  186,  196,  206,  219,  225 
Guirard,  B.  M.,  201,  206 
Gunness,  M.,  193,  207 
Gupta,  B.  M.,  288,  298 
Gustafson,  F.  G.,  70,  84 
Gymnoascits  setosus,  99 
Gymnoconia  peckiana,  348 
Gymnosporangiurn    juniperi-virginianae, 

358,  368 
Gyromitra  esculenta,  293 

H 

Haag,  E.,  191,  204 
Haagen-Smit,  A.  J.,  326,  337 


Haehn,  H.,  108,  113 

Haenseler,  C.  M.,  32,  43 

Haldane,  J.  B.  S.,  55,  63,  161,  169 

Hamilton,  E.,  327,  336 

Hanahan,  D.,  16,  22 

Hanna,  W.  F.,  410,  417 

Hansen,  H.  N.,  17,  22,  305,  313,  332,  335, 

337,  411,  412,  417 
Hansenida  suaveolens,  219 
Hao,  L.  C.,  274,  298 
Harden,  A.,  46,  63,  96,  113,  277,  298 
Harris,  E.  E.,  270,  298 
Harris,  G.  C.  M.,  200,  205,  285,  303 
Harris,  H.  A.,  402,  417 
Harris,  S.  A.,  192,  205,  232,  233,  243 
Hart,  H.,  373,  388,  396 
Hartelius,  V.,  229-231,  239,  242 
Harter,  L.  L.,  50,  63 
Hartree,  E.  F.,  281,  299 
Harvey,  L.  B.,  374,  396 
Haskins,  R.  H.,  35,  43 
Hatfield,  W.  C.,  390,  396 
Haugen,  G.  E.,  142,  146 
Haustoria,  380-381 
Hawker,  L.  E.,  90,   113,   133,   147,  321, 

325,  329,  335,  400,  417 
Hawkins,  L.  A.,  37,  43,  374,  396 
Haymaker,  H.  H.,  378,  396 
Hazen,  E.  L.,  286,  301,  327,  334,  335 
Heald,  F.  D.,  344,  354 
Helicostylum  pyrifonne,  122 
Helminthosponum,  99,  294,  327,  402,  411 
cynodontis,  294 
euchlaenae,  294 
gramineinn,  172,  294,  311 
ravenellii,  294 
sacchari,  379 
sativum,  101,  122,  127,  221,  285,  286, 

414,  431,  436,  437 
turcicum,  294 
vidoriae,  172,  379,  429 
Helvellic  acid,  293 
Hemitrichia  clavata,  368 
Hemming,  H.  G.,  285,  296 
Hemophilus  parainjluenzae,  197 
Render sonia,  311 
Hendlin,  D.,  273,  298 
Henry,  A.  W.,  286,  298 
Henry,  B.  W.,   163,  169,  308,  309,  317. 

335 
Herbst,  R.  M.,  109,  113 


450 


PHYSIOLOGY  OF  THE  FUNGI 


Herrick,  J.  A.,  121,  123,  124,  147 

Hertz,  J.,  219,  224 

Hervey,  A.,  29,  43,  77,  78,  85 

Hesse,  A.,  270,  272,  298 

Hestrin,  S.,  46,  49,  63 

0-Heteribiotin  {see  Oxybiotin) 

Heuser,  E.,  135,  147 

Hevesy,  G.,  87,  113 

Hexoses,  119-122 

Hickey,  R.  J.,  73,  84,  272,  298 

Higgins,  B.  B.,  376,  396 

Hildebrand,  E.  M.,  424,  439 

Hill,  E.  G.,  16,  22 

Hirneola  polytricha,  269 

Hitchens,  A.  P.,  14,  22 

HockenhuU,  D.  J.  D.,  93,  113,  414,  417 

Hodgson,  R.,  378,  396 

Hok,  K.  A.,  32,  44 

Hollis,  J.  P.,  88,  113 

Holoenzymes,  53 

Holzappfel,  H.  H.,  139,  147 

Hopkins,  R.  H.,  50,  63 

Hormones,  effect  of,  on  reproduction,  326 

Horn,  M.  J.,  216,  224 

Horner,  C.  K.,  108,  112 

Horowitz,  N.  H.,  211,  216,  217,  224,  225, 

239,  242 
Horr,  W.  H.,  125,  126,  147 
Horsfall,  J.  G.,  235,  242,  246,  257,  259, 

260,  262-264,  438,  439 
Host-parasite  relationship,  372 
Host  penetration,  372-375 
Houlahan,  M.  B.,  182,  202,  205 
Houston,  B.  R.,  311,  335,  386,  396,  401, 

402,  417 
Howard,  B.  H.,  294,  299 
Howard,  F.  L.,  379,  396 
Humfield,  H.,  270,  299 
Hurd,  A.  M.,  388,  396 
Hutchings,  B.  L.,  216,  224 
Hydnum  imbricatum,  367 
repandum,  367 
Hydrogen,  87 
Hydrogen-ion  concentration,  changes  in 

culture  media  of,  160,  162-164 
effect  of,  on  cell  morphology,  42 

on  enzymes,  55,  161,  162 

on  growth,  158-161 

on  medium  composition,  164 

on  oxygen  supply,  165 

on  spore  germination,  360-361 


Hydrogen-ion    concentration,    effect   of, 
on  sporulation,  318-319 
on  utilization  of  nutrients,  165-167 

as  factor  in  resistance,  388-389 

laboratory  exercise,  430-431 

relation  of,  to  pH,  151,  152 

{See  also  pH) 
Hydrogen    sulfide,     production    of,    by 
spores,  254,  255 

toxicity  of,  254-256 
Hydrolases,  47 
Hydroxyproline  formula,  238 
8-Hydroxyquinoline,  260 
Hygrophorus  conicus,  292 
Hyper  parasites,  385 
Hypersensitiveness,  382 
Hypomyces,  385 

solani  f.  cucurbitae,  412 
Hypoxanthine,  effect  of,  on  spore  germi- 
nation, 364,  365 
Hypoxylon  pruinatum,  190,  193 


Imbibition,  37 

Impatiens  balsaminia,  257,  390 

Indicators,  pH,  157 

Ingold,  C.  T.,  342,  344,  354 

Inheritance,  404-411 

in  Ascomycetes,  404-408 

in  Basidiomycetes,  408-411 

basis  of,  404 

cytoplasmic,  413 
Inhibition  analysis,  299 
Inoculation,  effect  of,  on  sporulation,  332 

laboratory  exercise,  431-432 

methods  of,  424 
Inocybe  infelix,  293 

infida,  293 
Inositol,  195-196 

Internal  factors,  effect  of,  on  growth,  32 
Invertase  {see  Sucrase) 
Ion  antagonism,  70,  75,  77 
Ionization,  149,  150 
Ions  as  enzyme  activators,  55,  68,  69,  76, 

79 
Iron,  71,  74-76 
Isoachlya  monilifera,  92,  93 
Isomers  of  sugars,  117 
Itaconic-acid  production,  281-282 


INDEX 


451 


Jaag,  O.,  378,  396 

Jarvis,  F.  G.,  68,  84 

Javillier,  M.,  72,  77,  84 

Jillson,  O.  F.,  42,  43 

Johnson,  H.  W.,  159,  169 

Johnson,  J.,  288,  302 

Johnson,  M.  J.,  51,  63,  68,  73,  84,  86,  280, 

302 
Johnson,  T.,  411,  417 
Jones,  E.  S.,  360,  370 
Jones,  F.  R.,  391,  396 
Jones,  M.  J.,  228,  242 
Jones,  R.  C,  308,  335 
Juglone,  257 


K 


Kakeura,  M.,  136,  147 
Karhng,  J.  S.,  384,  385,  396 
Karlingia  rosea,  35 

{See  also  Rhizophlyctis  rosea) 
Karow,  E.  O.,  278,  279,  299 
Kauffman,  C.  H.,  6,  7,  305,  335 
Kauffman,  F.  H.  O.,  356,  361,  370 
Kavanagh,  F.,  182,  190,  206 
Kavanagh,  V.,  100,  172,   181,  206,  432, 

439 
Keilin,  D.,  281,  299 
Keitt,  G.  W.,  261,  263,  288,  299 
Kellerniania  yuccagena,  311 
Kelner,  A.,  58,  63 
Kern,  F.  D.,  381,  397 
Kernkamp,  H.  C.  H.,  293,  297 
Ketoses,  117 
Kinsel,  K.,  120,  147 
Kinsey,  V.  E.,  414,  418 
Kirkwood,  S.,  236,  242 
Klebs,  G.,  6,  7,  305-308,  320,  336 

biographical  note,  6 
Klebs's  laws  of  growth  and  reproduction, 

306 
Klimek,  J.  W.,  290,  296 
Kloeckera  brevis,  181 
Ivlotz,  L.  J.,  377,  397 
Knaudt,  J.  H.,  291,  300 
Knight,  B.  C.  J.  G.,  196,  205 
Knobloch,  H.,  279,  296 
Koch,  R.  S.,  289,  297 
Koffler,  H.,  75,  84 


Kogl,  F.,  192,  193,  195,  205,  209,  219,  224 

Kolthoff,  I.  M.,  157,  169 

Koser,  S.  A.,  194,  205 

Krampitz,  L.  O.,  231,  242 

Krebs  citric  acid  cycle,  143,  144 

Krehl,  W.  A.,  215,  224 

Kreitlow,  K.  W.,  402,  418 

Krieger,  L.  C.  C.,  292,  293,  299 

Kroemer,  K.,  37,  43 

Krumbholz,  G.,  37,  43 

Kubowitz,  F.,  52,  63 

Kuhn,  R.,  292,  299 

Kunkel,  L.  0.,  319,  336,  374,  397 


Laboratory  demonstrations,  437-438 
Laboratory  exercises,  419-428 
Laborey,  F.,  69,  70,  84 
Laccaria  amythestina,  293 
Lachaux,  M.,  273,  301 
Lachnea  scidellata,  346 
Ladarius  piper  at  us,  52 

torminosus,  293 
Lactase,  49,  133 
Lactic-acid  production,  281 
LadobaciUus,  202 

arabinosus,  193,  194,  215,  216,  232 

casei,  193,  194,  216,  218,  229,  232,  233 
Lactose  formula,  131 
La  Far,  F.,  39,  44,  111,  113 
Lambertella  corni-maris,  99 

pruni,  179,  180,  184,  190,  193 
Lampen,  J.  0  ,  228,  242 
Landerkin,  G.  B.,  199,  205 
Large,  E.  C.,  245,  252,  263 
Larsh,  H.  W.,  120,  148 
Lavollay,  J.,  69,  70,  84 
Leach,  J.  G.,  248,  249,  264,  269,  299,  352, 
354,   362,   364,   365,   370,   383,   391, 
392,  397 
Leaver,  F.  W.,  331,  336 
Leben,  C.,  261,  263,  288,  299 
Lederberg,  E.  Z.,  413,  417 
Lee,  A.,  379,  397 
Lee,  S.  B.,  277,  289,  290,  299 
Leibowitz,  J.,  46,  63 
Leikind,  M.  C.,  14,  22 
Le  Mense,  E.  H.,  274,  299 
Lemieux,  R.  U.,  290,  299 
Lentinus  lepideus,  123,  134,  135 

tigrinus,  99 


452 


PHYSIOLOGY  OF  THE  FUNGI 


Lenzites  hetulinus,  134,  351 

saepiaria,  135,  159,  300,  363,  364 
trabea,  100,  135,  191,  401,  402,  415 
inheritance  by,  of  fruiting  ability, 
409 
of  thiamine  deficiency,  410 
thiamine  deficiency  of,  177,  178 

Leonian,  L.  H.,  7,  14,  22,  42,  44,  60,  63, 
74,  84,  92,  102-105,  107,  113,  164, 
169,  183,  188,  189,  191-193,  195-199, 
205,  211,  214,  215,  224,  229,  233, 
242,  258,  264,  310,  317,  319,  321, 
326,  336,  400,  402,  412,  418 
biographical  note,  7 

Leopold,  H.,  108,  113 

Leptomitus  lactens,  120,  127,  145 

Letcher,  H.,  275,  299 

Levan,  A.,  39,  44 

Levine,  H.,  268,  272,  298,  299 

Lewis,  J.  C,  75,  84 

Lewis,  R.  W.,  144,  147 

Light,  effect  of,  on  growth,  35 
on  irradiated  spores,  58 
on  spore  discharge,  339-341,  344,  346 
laboratory  exercise,  426 

Lillie,  R.  J.,  273,  299 

Lilly,  V.  G.,  14,  22,  34,  38,  43,  44,  60, 
63,  74,  84,  92,  102-105,  107,  113, 
164, 167, 169, 178, 181-184,  188, 189, 
191-193,  195-199,  204,  205, 211, 214, 
215,  224,  229,  232,  233,  242,  307, 
309,  310,  312,  313,  317,  319,  326, 
329-331,  334,  336,  400,  401,  409, 
410,  412,  416,  418 

Lime  sulfur  as  fungicide,  256 

Liming,  O.  N.,  253,  264 

Lin,  C.  K.,  363,  370 

Lindeberg,  G.,  80,  84,  100,  113,  154,  159, 
169 

Lindegren,  C.  C.,  30,  44,  60,  63,  120,  148, 
212,  224,  327,  336,  404,  412,  413,  418 

Lindegren,  G.,  327,  336 

Lii  der,  P.,  271,  299 

Lindgren,  R.  M.,  35,  44 

Ling,  L.,  361,  366,  370 

Lingane,  J.  J.,  157,  169 

Link,  K.  P.,  390,  391,  397,  398 

Lintzel,  W.,  267,  299 

Lipmann,  F.,  61,  63,  199,  205 

Little,  J.  E.,  257,  264,  390,  397 

Livingstone,  B.  E.,  89,  114 


Lockhead,  A.  C,  199,  205 

Lockwood,  L.  B.,  70,  84,  271,  281,  282, 

299,  318,  326,  336 
Lohrmann,  W.,  70,  84 
Longevity  of  spores,  368 
Lophodermium  pinastri,  190,  195,  196 
Lowther,  C.  V.,  368,  370 
Lundeg&rdh,  H.,  159,  169 
Lycogola  epidendrnm,  367 
Lycoperdon  echinatum,  367 

nigrescens,  367 

perlatum,  350,  351 

pratense,  367 

pyriforme,  367 

umbrinum,  367 
Lyxose,  123 


M 


Ma,  R.,  180,  194,  200,  201,  206,  234,  242, 

329,  337,  400,  418 
McCall,  M.,  258,  263 
McCalla,  T.  M.,  166,  169,  251,  264 
McCallan,  S.  E.  A.,  245,  247,  248,  252- 

255,  261,  264,  265,  319,  320,  336 
McCrea,  A.,  368,  370 
McElroy,  W.  D.,  226,  227,  242,  414,  418 
McGowan,  J.  C.,  261,  264 
McHargue,  J.  S.,  77,  78,  85 
Mcllwain,  H.,  226,  236,  242 
MacLeod,  R.  A.,  74,  85 
McNew,  G.  L.,  261,  264 
Macow,  J.,  229,  243 

Macrae,  R.,  402,  403,  410,  418,  438,  439 
Macrosporium  commune,  98 

sarcinaeforme,  99,  250,  253,  255,  262 

tomato,  316 
McVeigh,  I.,  104,  112,  238,  242,  412,  418 
Mader,  E.  O.,  318,  336 
Magnesium,  68-70,  218 
Maillard,  L.  C.,  16,  22 
Mains,  E.  B.,  381,  397 
Malachite  green,  258 
Maltose  formula,  130 
Mandels,  G.  R.,  91,  113,  114 
Manganese,  77-79,  80 
Mann,  T.,  95,  114 
Mannitol  formula,  124 
Mannose  formula,  119 
Marasmius,  80,  100 

alliaceus,  81,  100 


I 

I 
I 


INDEX 


453 


Marasviius,  androeceus,  100 

chordalis,  95,  100,  132 

epiphyllHs,  80,  81,  100 

foetidis,  100 

fulvobulbillosus,  99 

graminum,  100,  159 

perforrnis,  100 

personatus,  100 

putillus,  100 

rainealis,  100 

rotula,  100 

scorodoniiis,  100 
Margolin,  A.  S.,  16,  17,  22,  121-126,  136, 

147 
Marryat,  D.  C,  392,  397 
Marsh,  P.  B.,  77,  85 
Marten,  E.  A.,  248,  249,  264 
Martin,  G.  J.,  235,  242 
Martin,  W.  J.,  401,  418 
Mass,  J.  M.,  291,  299 
Massospora  cicadina,  384 
Mathur,  R.  S.,  160,  169,  186,  205,  309, 

332.  336 
Meacham,  M.  R.,  159,  169 
Media,  autoclaving  of,  16,  17,  422,  423 

basal  semisynthetic,  427 

choice  of,  13 

comparison  of,  20,  21 

concentration  of,  effect  of,  on  growth, 
38 

constituents  of,  12-14,  421-422 

formula  of,  ghicose-asparagine,  210 
glucose-casein  hydrolysate,  211 
sucrose-ammonium   tartrate-ammo- 
nium  nitrate,  211 

kinds  of,  9-11 

liquid,  14 

naming  of,  14 

natural,  9,  11 

pH  of,  422 

preparation  of,  17,  420-421 

removal  from,  of  metallic  ions,  72 
of  vitamins,  210,  432 

semisynthetic,  10,  427 

solid,  14 

specific  metabolites  in,  12 

sterilization  of,  16,  17,  422,  423 

synthetic,  10-11,  20,  421 

units  of  measure,  18,  19 
Meehan,  F.,  379,  397 
Melanconium  betulinnm,  190,  196,  311 


Melander,  L.  W.,  387,  397 
Melanospora,  323,  325,  428,  429 

destruens,  91,  133,  190,  193,  324,  329 
Melibiosc,  132 

Memnoniella  echinata,  193,  195,  222,  329 
Mercury-toxicity  theory,  251 
Merulius  lacrymans,  95,  135,  159 
Metabolic  products,  effect  of,  on  sporula- 

tion,  327-329 
tests  for  presence  of,  219-221 
variation  in,  401-402 
Metabolism,  intermediary,  139-144 
Metabolite  antagonists,  theory  of  action 

of,  226-229 
Metabolites,  12 
specific     (see    Amino    acids;    Growth, 
factors  affecting;  Vitamins) 
Metarrhizium  glutinosum,  103 

{See  also  Myrothecium  verrucaria) 
2-Methoxy-l  ,4-naphthoquinone  formula, 

257 
a-Methyl-D-glucoside  formula,  129 
j3-Methyl-D-glucoside  formula,  129 
Methylpentoses,  124 
Metz,  O.,  294,  299 
Meyer,  B.  S.,  37,  44 
Meyerhof,  O.,  51,  63,  142,  147,  277,  299, 

300 
Michaelis,  L.,  51,  63 
Michener,  H.  D.,  292,  300 
Microbiological  assays,  standard  curve, 

214 
Microsphaera  alni,  358 
Microsporum  audouini,  197,  327 

canis,  238 
Milhngton,  R.  H.,  143,  148 
Mirsky,  A.  E.,  97,  114 
Mitchell,  H.  K.,  182,  199,  202,  205, 

237,  242,  325,  326,  337 
Mix,  A.  J.,  321,  324,  336 
Moisture  requirements  for  growth,  35 
Molliard,  M.,  68,  85 
Molybdenum,  79,  218,  219 
Monascus  purpurea,  27,  172 
Monilia  Candida,  39 
fructigena,  362 
tarnari,  281 
Monilinia  fructicola,  27,   100,   118,   172, 

304,  321,   325,  375,  402,  425,  426, 

428,  429,  431,  436,  437 
(See  also  Sclerotinia  frurficola) 


454 


PHYSIOLOGY  OF  THE  FUNGI 


Monocotyledons,  immunity  of,  to  cotton 

root  rot,  391 
Monosaccharides,  116-126 
Morchella  esculenta,  267 
Morphology,  effect  of  external  factors  on, 
39-42,  89 
of  yeast,  39 
Mortierella  rhizogena,  100 
Morton,  H.  E.,  283,  300 
Moyer,  A.  J.,  280,  282,  300 
Moyer,  D.,  186,  195,  204,  400,  416 
Mrak,  E.  M.,  309,  336 
Mucor,  9,  39,  271 
flavus,  100 
hiemalis,  100,  295 
mucedo,  295 
nodosus,  100 
pusilhis,  70 
pyriformis,  100,  279 
racemosus,  275,  402 
raviannianus,  103,  133,  231,  232 

carbon  nutrition  of,   118,   122,   124, 

125 
vitamin  requirements  of,  188-191 
rouxii,  281 
saturninus,  100 
spinosus,  362 
stolonifer,  100 
stridus,  100 
Mudd,  S.,  231,  242 
Mueller,  J.  H.,  194,  196,  205 
Mulder,  E.  G.    14,  22,  73,  79,  85,  217- 

219,  224,  225,  294,  300 
Mull,  R.  P.,  96,  114,  140,  147,  277,  300 
Muntz,  J.  A.,  68,  85 
Murphy,  H.  E.,  379,  397 
Muscarin,  292 
Mutations,  back,  413 
chemically  induced,  414-415 
natural,  93 

radiation-induced,   93,   282,   401,   405, 
408 
Mutualistic  symbiosis,  example  of,  173, 

384 
Mycobacterium  tuberculosis,  240,  290 
Mycoderma  cerevisiae,  39 
valida,  176,  199 
vini,  186 
Mycogone,  385 

MycosphaereUa  citrullina,  319 
Mycoiorula  lactis,  197 


Myrback,  K.,  50,  63,  135,  147,  275,  300 
M yrophagus  ucrainicus,  384 
Myrothecium  verrucaria,  104,  109,  222 
(See  also  Metarrhizium  glutinosum) 

N 

Naemosphaera,  311 
1,4-Naphthoquinones,  antibiotic  activity 

of,  390 
Nectria  cinnabarina,  328 
coccinia,  190,  328 
coryli,  328 
cucurbitula,  328 
galligena,  328 
Neidig,  R.  E.,  129,  146 
Neisseria  gonorrhoeae,  290 
Nelson,  J.  M.,  52,  63,  77,  85 
Nematospora  gossypii,  193,  195,  196,  219 
Neocosmopara    vasinfecta,    99,    172,    321, 

325,  426,  429,  431,  432 
Neopyrithiamine,  232 
Neuberger,  A.,  10,  22 
Neufeld,  C.  C,  358,  370 
Neurospora,  29,  102,  144,  216,  223,  238, 
401,  408,  415,  438 
amino  acids  and  spore  germination  of, 

364 
back  mutations  of,  413 
genetics  of,  404-407 
mutants  of,    182,    183,   202,   215-217, 

405-408 
nutritional  adaptations  of,  412,  413 
temperature  and  riboflavin  deficiency 

of,  182 
vitamin  deficiencies  of,  193,  197,  199, 
202 
Neurospora  crassa,  26,  36,  144,  167,  196, 
202,  228,  232,  237,  363,  405,  413,  414 
pH,     and     p-aminobenzoic-acid     defi- 
ciency of,  183,  184 
and  growth  of,  161,  162 
production  of  perithecia  by,  325,  326 
use  of,  in  assays,  211,  215 
Neurospora  sitophila,   29,   34,    184,    193, 
215,  248,  317,  405 
tetrasperma,  357 
Newton,  M.,  411,  417 
Nickerson,  W.  J.,  39,  40,  42-44,  70,  85, 
91,  92,  114,  327,  336,  385,  386,  397 
Nicotinic  acid,  196-198 
antagonists  of,  236 


INDEX 


455 


Nicotinic  acid  amide,  196 

Niemann,  C,  228,  237,  242 

Niklas,  H.,  218,  224 

Nitrate  utilization,  effect  of  molybdenum 

on,  79 
Nitrite  as  source  of  nitrogen,  102 
Nitrogen,  97-110 

fixation  of,  by  fungi,  98 
nitrate,  list  of  fungi  utilizing,  99 
sources  of,  effect  of,  on  sporulation,  321 

laboratory  exercise,  428 
utilization  of,  as  basis  of  classification, 
97 
effect  of  organic  acids  on,  103,  104, 

107 
organic,  101,  105 
Noble,  R.  J.,  366,  371 
Nobles,  M.  K.,  134,  147,  410,  418 
Nocardia  gardneri,  287 
Nord,  F.  F.,  96,  102,  108,  114,  115,  123, 
135,  140,  141,  147,  148,  186,  191,  192, 
207,  270,  295,  300,  303 
Norman,  A.  G.,  49,  63,  134,  147 
North,  H.  E.,  Ill,  113 
Novelli,  G.  D.,  199,  205 
Nutrition,  of  host,  effect  of,  on  disease 
development,  392-393 
special  conditions  of,  laboratory  exer- 
cise, 429-430 
Nutritional   requirements,   variation   of, 

within  species,  400-401 
Nydalis,  385 


O 


Oldium  lactis,  271 
Oligosaccharides,  129-133 
Ophiobolus  graminis,  12,  97,  99,  180,  190, 
193 
miyabeanus,  99 
oryzinus,  193 
Ophiostoma,  401 

catonianum,  193,  200 
muUiannulahun ,  42,  93 

{See     also     Ceratostomella     multian- 
nulata) 
ulmi,  31 

{See  also  Ceratostomella  ulmi) 
Organic  acids,  as  carbon  sources,  126-128 
as  factor  in  resistance,  390 
production  of,  277-283 


Orton,  C.  R.,  7,  381,  383,  397 
Osmotic  pressure,  36,  37 

in  host-parasite  relation,  376,  377,  382 
Oswald,  J.  W.,  311,  335 
Owen,  W.  L.,  274,  300 
Owens,  H.  S.,  136,  147 
Oxidases,  51 
Oxybiotin,  194 
Oxygen,  88-92,  359,  360 


Paneolus,  293 
Pantothenic  acid,  198-199 

antagonists  of,  233-234 

moieties  of,  198 
Pantoyltaurine  formula,  234 
Panus  stijpticus,  402,  403,  410,  438 
Parasitella  simplex,  188 
Parasitism,  375-386 

action  in  advance,  376 

by  balanced  parasites,  375,  380-383 

by  destructive  parasites,  375-378 

by  fungi,  of  fungi,  385 
of  insects,  383-385 
of  man,  385 

types  of,  375 

by  wilt-producing  fungi,  378-380 
Parker-Rhodes,  A.  F.,  250,  264 
Pasteur,  L.,  6,  7,  39,  44,  45,  63,  117,  147, 
219,  224,  275,  276,  300 

biographical  note,  6 
Pathogenicity,  inheritance  of,  411 

variation  in,  402 
Patton,  A.  R.,  16,  22 
Payne,  E.  H.,  291,  300 
Pectinase,  47,  50 

production  of,  376-378 

specificity  of,  377-378 
Pehrson,  S.  O.,  34,  44,  160,  169,  380,  397 
Peltier,  G.  L.,  164,  168,  273,  300 
Penetration,  372-375 

direct,  373-375 

into  immune  plants,  374,  391 

through  stomata,  372-373 

through  wounds,  375 
Penicillin,  288-290 

influence  of,  on  cell  morphology,  39 

type  formula  of,  289 
Penicillium,   93,   94,   99,    172,    197,   222, 
260,   271,   272,   274,  275,   283,   286, 
385,  401,  436 


456 


PHYSIOLOGY  OF  THE  FUNGI 


Penicilliurn,  arenarium,  297 

brevicaule,  110,  111 

hrevicom-pactum,  287 

chnjsogenum,  68,  94,  280-282,  402,  408, 
437 
ion  antagonism  in,  75 
penicillin  production  by,  287-289 

crustaceum,  280 

cyclopium,  360 

digitatum,  98,  193,  232,  261,  283,  375 

divaricatum,  279 

ex-pansum,  98,  262,  375,  425,  426,  431 

flavo-cinereum,  212 

glabrum,  280 

glaucum,  35,  69,  70,  117,  219,  275,  279, 
280,  358,  362 

griseofulvum,  287 

islandicum,  294 

italicum,  360,  375 

janczewskii,  287 

javanicum,  271,  272,  318 

luteum-purpiirogenuri} ,  287 

notatxivi,  164,  284,  288,  408 

olivaceum,  279 

oxalicum,  272 

patulum,  287 

phoeniceum,  294 

piscarium,  272 

purpurogenum  var.  rubrisderotiurn,  280 

roquejorti,  121 ,  272 

rubrurn,  294 

sanguifluus,  279 

spiculisporurn,  431 

variable,  159 
Peniophora  allescheri,  410 
Pentoses,  122-124 
Perlman,  D.,  71,  73,  75,  76,  85,  195,  205, 

279,  300 
Permeability,  change  of,  as  factor  in  re- 
sistance, 382 
in  host  cells,  376,  377,  382 
Peronospora,  353 

effusa,  343 

geranii,  343 

halstedii,  343 

parasitica,  343 

pygrnaea,  358 

tabacina,  343 
Pestalotia  guepinia,  311 

stellata,  255 
Peterson.  M.  S..  202,  205 


Peterson,  W.  H.,  139,  147,  202,  205,  216, 
224 

Peziza  badia,  346 
pH,  definition  of,  150,  151 
equation  for,  151 
indicators  for,  157 
method  of  determining,  156-158 
Phacidium  infestans,  34,  160 
Phalloidin,  292 
PheUorina,  307 

Phenolic  compounds,  toxicity  of,  389-391 
PhiUips,  P.  H.,  236,  242 
Pholiota  autumnalis,  292 
Phoma,  99 
apiicola,  99 

betoe,  98,  100,  172,  321,  325,  429,  432 
causarina,  99 
lingam,  377 
terrestris,  402 
urens,  311,  317 
Phomopsis  californica,  311 

citri,  28,  34 
Phosphorus,  94-97 

in  carbohydrate  dissimilation,  96,  142 
Phycomyces,  364,  415 

blakesleeanus,  15,  17,  32,  35,  100,  164, 

172,  173,  178,  181,  187,  191,  212, 

220,  232,  295,  310,  314,  380,  428, 

431,  434,  437 

pH  and  formation  of  zygospores,  318 

stimulants   and   spore   germination, 

364-365 
thiamine    deficiency    of,    174,    177, 

188-190 
use  of,  in  thiamine  assay,  211,  215 
utihzation  of  sources  by,  carbon,  122, 

124-126,  133 
nitrogen,  100,  103,  104,  107,  108 
nitens,  91,  362 
Phyllachora  graminis,  383 
Phylostida  antirrhini,  356,  358 
opuntiae,  311,  317 
soUtaria,  321,  324,  362 
Phymototrichum  omnivorum,  76,  375,  391 
Physarum  cinereum,  367,  368 

polycephalum,  320,  367 
Phytophthora,  33,  42,  188,  190,  258,  402 
boehtneriae,  327 

cadorum,  17,  122,  132,  326,  327 
cinnamomi,  232 


INDEX 


457 


Phytophthora,  colocasiae,  258,  360 
cry  throne  ptica,  16,  17,  285,  327 

carbon  sources  utilized  by,  122,  124, 

125 
thiamine  deficiency  of,  188,  189,  191 
fagopyri,  122,  136 
hydrophila,  258 

infestans,  34,  343,  352,  360,  373,  425, 
436 
parasitism  of,  376,  377 
temperature    and     germination     of, 

356-357 
variation  of,  401 
megasperma,  327 
melongenae,  258 
palmivora,  132,  360 
parasitica,  132,  360 
richardiae,  258 
terrestris,  28,  34 
Pigman,  W.  W.,  14,  22,  117,  119,  128,  133, 

147 
Pigments,  production  of,  by  fungi,  293- 
295 
relation  to,  of  copper,  77 
of  iron,  76 
Pilaira  anomala,  295 

moreaui,  122 
Pilgrim,  F.  J.,  194,  206 
Pilobolus,  338,  342-345,  349,  353,  438 
kleinii,  339,  340 
longipes,  339 
microsporus,  307 
Pimelic  acid,  194 
Pinkard,  J.  A.,  343,  354 
Piptocephalis,  385 
Piricularia  oryzae,  163,  190,  193,  331,  351 

sporulation  of,  308,  309,  317 
Pirschle,  K.,  102,  114 
Pisum,  382 

sativum,  377 
pKa,  153,  154,  167 
pKft,  153,  154 

Plasmodiophora  brassicae,  307 
Plasmopara  viticola,  358,  360 
Piatt,  B.  S.,  269,  300 
Plattner,  P.  A.,  378,  397 
Platz,  G.  A.,  360,  365,  371 
Plenodomus  destruens,  37,  311,  317 

fuscomaculans,  307,  311,  313,  317 
Pleurage,  344,  353,  438 
anserina,  404 


Pleurage,   curvicoUa,    100,   175,    178,   190, 

193,  322,  323,  325,  344,  428 
Pleurotus  corticatus,  102 

ostreatus,  100,  134 
Plumlee,  C.  H.,  94,  114 
Podaxis,  307 

Podospora  curvula,  190,  193 
Pollard,  A.  L.,  94,  114 
Polyporus  abietinus.  134 
adustus,  219 
betulinus,  135 
cinnabarinus,  134 
hirsutus,  351 
pargamenus,  134,  135 
squamosus,  238 
versicolor,  351 
Polysaccharides,  133-136 

as  wilt  inducers,  378 
Porges,  N.,  280,  281,  300 
Porta  vaillantii,  135 
Porter,  C.  L.,  327,  336 
Potassium,  68,  218 
Prasard,  N.,  380,  397 
Pratt,  E.  F.,  199,  206 
Pratt,  R.,  32,  44,  290,  300 
Prescott,  S.  C,  142,  147,  266,  277,  300 
Preuss,  L.  M.,  271,  300 
Prevost,  B.,  246,  264 
Price,  W.  C,  288,  298 
Pritham,  G.  H.,  158,  169 
Proactinomyces  cyaneus,  287 
Proline,  formula  of,  238 
Protectants,  fabric,  tests  for,  221-222 
Proteins,  amino-acid  composition  of,  268 

synthesis  of,  .110 
Protocatechuic  acid,  390 
Psalliota  campestris,  52,  267,  270 
{See  also  Agaricus  campestris) 
Pseudomonas  saccharophila,  324 
Pseudopeziza  ribis,  193 
Puccinia  antirrhini,  254,  255 
coronata,  348 
glumarum,  358 

graminis,  348,  349,  356,  360,  373,  377 
parasitism  of,  382-383 
resistance  to,  392 
graminis  iritici,  9,  382,  392,  395,  402 
411 
inheritance  in,  411 
graminis  tritici-compacti,  392 
podophylli,  348 


458 


PHYSIOLOGY  OF  THE  FUNGI 


Puccinia,  rubigo-vera  tritici,  402 
(See  also  triticina,  below) 

sorghi,  381,  383,  392 

triticina,  373 

{See  a/.so  nihicjo-vera  tritici,  above) 
Pyridoxal  formula,  200 
Pyridoxamine  formula,  200 
Pyridoxine,  199-201 

antagonists  of,  234-235 
Pyrithianiine,  232 
Pyronema  confluens,  100 
Pyruvic  acid,  action  of  carboxylase  on,  52 

effect   of   thiamine    concentration    on, 
191,  192 

transformations  of,  140-143 
Pythiacystis  citrophthora,  28,  34,  377 
Pythiornorpha  gonapodyoides,  77,  78,  100, 

103,  122,  124-126,  188-190,  232 
Pythium,  33,  327,  374,  377,  378 

arrheno  manes,  190 

ascophallon,  103,  122,  136,  190 

butleri,  182,  190 

debaryanum,  100,  132,  249,  374 

intermedium,  100 

irregulare,  80,  100 

oligandrum,  190 


R 


Raaf,  H.,  271,  300 

Rabinovitz-Sereni,  D.,  69,  85 

Raciborski,  M.,  37,  44 

Radiation  effects  on  fungi,  57,  58 

Raffinose  formula,  132 

Rahn,  O.,  24,  25,  32,  44,  165,  166,  169 

Raistrick,  H.,  94,  114,  139,  147,  275,  276, 

294,  296,  299-301 
Rake,  G.,  290,  301 
Ramsey,  G.  B.,  316,  336 
Rands,  R.  D.,  319,  336 
Rankin,  W.  H.,  368,  369 
Rannefelt,  A.  N.,  201,  206 
Raper,  J.  R.,  326,  337 
Raper,  K.  B.,  34,  44,  282,  284,  285,  301, 

385,  398,  401,  418,  424,  439 
Ratajak,  E.  J.,  136,  147 
Raulin,  J.,  11,  22,  66,  67,  71,  76,  85,  94, 
114 
synthetic  medium  of,  1 1 
Raut,  C.,  30,  44,  60,  63,  212,  224 
Ravel,  J.  M.,  234,  242 


Rawlins,  T.  E.,  17,  22 
Reddy,  C.  S.,  388,  389,  397 
Reeves,  M.  D.,  70,  84 
Reilly,  H.  C.,  288,  301 
Rcnaud,  J.,  273,  301 
Resistance,  386-393 

in  colored  onions,  390 

due  to,  acidity,  388-389 
antibiotics,  390 
organic  acids,  390 
phenolic  compounds,  390,  391 

effect  on,  of  environment,  393 
of  host  metabolism,  393 
of  host  nutrition,  392-393 

functional,  388 

mechanical,  387 

physiological,  388-393 

starvation  theory  of,  383,  391,  392 

toxin  theory  of,  382,  392 
Respiration,  89-92 

effect  of  cyanide  on,  55-57 
Respirometer,  91 
Reticularia  lycoperdon,  367 
Reynolds,  E.  S.,  390,  397 
Rhizobium,  98,  192 

trifolii,  75,  164,  193,  208 
Rhizoctonia  solani,  79,  80,  100,  127 
Rhizophlyctis  rosea,  100 

(See  also  Karlingia  rosea) 
Rhizopus,  49,  76,  274,  281,  319 

arrhizus,  281 

autocar  pi,  50 

chinensis,  50,  281 

delemar,  274 

elegans,  281 

japonicus,  281 

niicrosporus,  50 

nigricans,  37,  50,  70,  76,  78,  100,  122, 
160,  262,  375 
carbon  sources  utilized  by,  133,  136, 

137 
production  of  organic  acids  by,  282, 
283 

nodosus,  281 

oryzae,  100,  274,  281,  326 

pseudochinensis,  281 

salebrosa,  281 

shanghaiensis,  281 

stolonifer,  281 

suinus,  17,  122,  133,  186,  196 

tritici,  281 


( 


INDEX 


459 


Rhodoiorula  aurantica,  201,  231 
(jlulinus,  271 
rubra,  188 
sanniei,  184 

Rihes,  395 

Riboflavin,  202 

production  of,  272-273 

Ribose,  97,  123 

Rice,  M.  A.,  381,  382,  397 

Rich,  S.,  2G0,  264 

Richards,  M.  C,  319,  337 

Richards,  O.  W.,  66,  85 

Richardson,  G.  L.,  165,  166,  169 

Rickes,  E.  L.,  82,  85,  273,  301 

Rickettsia,  291 

Riker,  A.  J.,  17,  22,  305,  337,  420,  439 

Riker,  R.  S.,  17,  22,  305,  337 

Rippel,  A.,  68,  69,  85,  103,  114 

Roach,  W.  A.,  253,  264 

Roark,  G.  W.,  Jr.,  129,  146 

Robbins,  W.  J.,  14,  17,  22,  77,  78,  85,  97, 
99,  100,  114,  155,  160,  164,  169,  172, 
180-182,  190,  194,  196,  200,  201,  206, 
212,  224,  232,  234,  238,  242,  285, 
301,  310,  318,  329,  337,  364,  371,  400, 
412,  418,  422,  432,  439 

Roberg,  M.,  76,  85 

Roberts,  C.,  76,  85 

Roberts,  E.  C.,  216,  225 

Roberts,  M.,  274,  301 

Roblin,  R.  O.,  Jr.,  226,  227,  234,  242 

RochUn,  E.  J.,  390,  397 

Rodenhiser,  H.  A.,  410,  417 

Rogosa,  M.,  196,  197,  206,  218,  225 

Roholt,  K.,  231,  242 

Roine,  P.,  109,  114 

Rolfe,  F.  W.,  266,  267,  301 

Rolfe,  R.  T.,  266,  267,  301 

Rose,  W.  C,  106,  114,  267,  301 

Rosellinia  arcuata,  122,  193 

Rosen,  H.  R.,  380,  398 

Rosenberg,  H.  R.,  187,  206 

Rosenblum,  C,  157,  169 

Rubbo,  S.  D.,  201,  206,  230,  242 

Rubin,  S.  H.,  194,  206,  233,  241 

Rubus,  382 

Ruger,  M.  L.,  273,  298 

Russula  emetica,  292 
foetens,  52 
niger,  52 

Ruta,  365 


Ryan,  F.  J.,  28,  29,  36,  44,  161,  162,  169, 
198,  206,  216,  225,  363;  371,  413,  418 
Rydon,  H.  N.,  229,  242 


S 


Sabouraud,  R.,  386,  398 
Saccharase  {see  Sucrase) 
Saccharomyces,  198 

anamensis,  197 

carlsbergensis,  59,  200,  215,  219 

carlsbergensis  var.    mandshuricus,    199, 
219 

cerevisiae,  49,  60,  76,  78,  105,  120,  193, 
194,  229,  251,  327,  400,  402 
effect  of  penicilhn  on,  39,  40 
growth  antagonists  of,  232-239 
induced  vitamin  synthesis  by,  60 
nutritional  adaptations  of,  412 
use  of,  in  assays,  213-215 
vitamin  content  of,  269 
vitamin  deficiencies  of,  176,  197,  198 

chodati,  199 

fragilis,  140,  197 

ladis, 197 

ludwigii,  181,  199 

macedoniensis,  180 

oviformis,  180,  199 

uvarurn,  195 
Saito,  K.,  281,  301 
Salmon,  E.  S.,  391,  398 
Saltation,  411 
Sandholm,  N.  K.,  261,  264 
Sanger,  F.,  10,  22 
Saprolegnia,  349 

jnixta,  92,  307,  320 

parasitica,  93 
Sarcoscypha  protracta,  345 
Sarett,  H.  P.,  232,  243 
Sarver,  L.  A.,  73,  86,  260,  265 
Sawyer,  W.  H.,  343,  354 
Scandium,  81 

Schade,  A.  L.,  120,  127,  148 
Schatz,  A.,  286,  301 
Scheffer,  T.  C,  89,  114 
Schizophyllum  commune,  94,  351,  409 
Schizosaccharomyces,  25 

octosporus,  49 

pombe,  195 
Schleef,  M.  L.,  270,  301 
Schlenk,  F.,  197,  204 


460 


PHYSIOLOGY  OF  THE  FUNGI 


Schmidt,  M.,  390,  398 

Schmitt,  M.  B.,  17,  22,  154,  164,  169,  310, 

318,  337 
Schneider,  G.  G.,  50,  63 
Schoenbach,  E.  B.,  291,  301 
Schomer,  H.  A.,  57,  63 
Schopfer,  W.  H.,  2,  7,  32,  44,  172,  174, 
177,  186,  187,  192,  196,  206,  211,  214, 
215,  219,  220,  225,  236,  243,  295,  301 
Schubert,  W.  J.,  135,  148 
Schultz,  A.  S.,  127,  148,  215,  225 
Sciarini,  L.  J.,  295,  301 
Scleroderma  aurantirion,  367 
Sclerospora  graminicola,  360,  381 
Sclerotinia,  254 

americana,  252,  253,  255,  261 
cameUiae,  34,  190,  193,  196 

temperature  and  inositol  deficiency 

of,  182,  184,  185 
vitamin  deficiencies  of,  178,  179 
frudicola,  77,  247-249,  258,  262,  288, 
358 
(See  also  Monilinia  frudicola) 
Ubertiana,  136 

(See  also  sclerotiorum,  below) 
minor,  100,  190,  400 
sclerotiorum,   100,    172,  374,  376,  377, 

402,  425 
trifoliorum,  402 
Sclerotium  bataticola,  100 
delphinii,  76 
rolfsii,  80,  376 
Scott,  I.  T.,  160,  169 
Sealock,  R.  R.,  231,  243 
Seaver,  F.  J.,  346,  354 
Seifriz,  W.,  37,  44 
Semeniuk,  G.,  96,  114 
Septobasidium,  352,  353,  384 

burtii,  385 
Septoria  nodorum,  100,  172,  314,  321,  425, 

426,  432 
Septosporium  acerinum,  317 
Sevag,  M.  G.,  231,  243 
Shankman,  S.,  216,  225 
Shanor,  L.,  332,  337 
Sharvelle,  E.  J.,  393,  398 
Shive,  W.,  229,  234,  242,  243 
Shoup,  C.  S.,  121,  148 
Shu,  P.  73,  86,  280,  302 
Siminoff,  P.,  286,  298 
Singer,  T.  P.,  226,  243,  251,  263,  264 


Sinnott,  E.  W.,  42,  43 

Siu,  R.  G.  H.,  91,  113,  114 

Skoog,  F.  K.,  120,  148,  412,  418 

Smart,  R.  F.,  357,  361,  362,  367,  371 

Smit,  J.,  218,  225 

Smith,  E.  C.,  316,  337,  368,  371 

Smith,  E.  L.,  273,  302 

Smith,  F.  G.,  389,  398 

Smith,  L.  D.  S.,  376,  398 

Smith,  P.  E.,  357,  370 

Smith,  R.  M.,  291,  302 

Smith,  V.  M.,  95,  114,  132,  148 

Snell,  C.  T.,  156,  157,  169 

Snell,  E.  E.,  74,  85,  200,  201,  206,  211, 

215,  216,  225,  234,  239,  243 
Snell,  F.  D.,  156,  157,  169 
Snell,  N.,  292,  300 
Snyder,  W.  C,  17,  22,  305,  313,  332,  335, 

337,  412,  417 
Solanine,  390 
Solanum,  390 

tuberosum,  377 
Somers,  G.  F.,  47,  51,  64,  70,  79,  86,  96, 

115,  142,  148,  277,  302 
Sorbitol  formula,  124 
Sorbose  formula,  119 
Sordaria,  344,  345,  349,  353,  438 
fimicola,  39,  40,  100,  122,  164,  173,  175, 
178,  190,  191,  193,  208,  220,  328, 
400,  428,  429,  431,  432,  435,  437 
biotin  and  sporulation  of,  330-331 
carbon  sources  utiUzed  by,  122 
depth  of  medium  and  growth  of,  38 
pH,  and  growth  of,  160,  162,  167 

and  sporulation  of,  319 
spore  discharge  by,  344 
vitamin  deficiency  of,  167,  181 
induced  thiamine,  182,  183 
Spergon,  257 
Sphaerobolus,  347,  349,  353 

stellatus,  100,  347,  348 
Sphaerographium  fraxini,  310,  311 
Sphaeronema  pruinostim,  311 
Sphaeropsis  rnalorum,  100,  172,  314,  428, 

429,  431,  432 
Spiegehnan,  S.,  59,  63,  412,  413,  418 
Spies,  T.  D.,  187,  207 
SpineUus,  385 

Spongospora  subterranea,  374 
Spore  discharge,  of  aeciospores,  348-349 
of  ascospores,  343-346 


I 

I 


INDEX 


461 


Spore  discharge,  of  basidiospores,  348-351 
effect  on,  of  humidity,  343 
of  light,  339-341,  344,  346 
of  temperature,  351 
methods  of,  338-351 
of  peridioles,  347-348 
by  rain,  350-351 
of  sporangia,  338-343 
Spore  dissemination,  agents  of,  351,  352 
Spore  germination,  effect  on,  of  moisture, 
357-359 
of  nutrients,  361-364 
of  pH,  360-361 
of  stimulants,  363-367 
of  temperature,  356-357 
laboratory  exercise,  435-436 
requirements  for,  355 
in  slime  molds,  367,  368 
Sporodinia  grandis,  100,  307 
Sporormia  intermedia,  193 
Sporotrichum,  132 

Sporulation,     effect     on,     of    associated 
organisms,  327-328 
of  carbon  sources,  321-325 
of  light,  310-314 
of  method,  of  inoculation,  332 

of  sterilization,  332 
of  nitrogen  source,  321 
of  nutritional  factors,  320-333 
of  pH,  318-319 
of  temperature,  307-310 
of  ultraviolet  light,  314-316 
of  vitamins,  329-332 
methods  of  inducing,  333-334 
need  for  aeration  in,  316-318 
Srb,  A.  M.,  217,  225,  239,  242 
Stachybotrys  atra,  193,  195 
Stakman,  E.  C,  382,  391,  392,  398,  410 
Standard  curve,  214 
Stanier,  R.  Y.,  16,  22 
Staphylococcus,  284 

aureus,  285 
Starch,  135 

scheme  of  utilization  of,  48 
Stark,  W.  H.,  71,  86 
Starkey,  R.  L.,  159,  169,  246,  264,  271, 

302 
Steinberg,  R.  A.,  10,  22,  23,  67-69,  71-74, 
76,  78-82,  86,  92,  97,  105,  114,  115, 
121,  123-127,  139,  140,  148,  326,  414, 
418,  437,  439 


Steinhaus,  E.  A.,  383,  398 
Stemonitis  axifera,  367 

ferruginea,  368 

fusca,  367 
Stephenson,  M.,  46,  63 
Stereum  frustulosum,  188 

gausapatum,  121,  123,  124 
Stern,  K.  G.,  58,  63 
Stevens,  F.  L.,  315,  316,  337 
Stevens,  N.  E.,  120,  148 
Stoddard,  E.  M.,  231,  243 
Stokes,  J.  L.,    105,    115,    184,    193,    195, 

207 
Stomata,    formation    of,    as    result    of 
parasitism,  382 

penetration  through,  372-373 
Stone,  G.  M.,  373,  395 
Stotz,  E.,  143,  148 
Stout,  P.  R.,  21,  23,  73,  86 
Strauss,  B.  S.,  184,  207 
Streptococcus  faecalis,  16,  200,  201,  228 
Streptomyces,  285,  288 

aureofaciens,  291 

griseus,  58,  82,  127,  273,  286,  287,  290, 
291 

venezuelae,  287,  291 
Streptomycin,  290-291 
Sucrase,  49,  133 
Sucrose,  effect  of,  on  sporulation,  324 

formula  of,  131 
Sugar  acids,  125 
Sugar  alcohols,  124 
Sulfanilamide,  227,  229-231 
Sulfur,  as  fungicide,  251-256 

as  metabolite,  92-94 
Sumner,  J.  B.,  47,  51,  52,  64,  70,  79,  86, 

96,  115,  142,  148,  277,  302 
Sure,  B.,  268,  302 
Syncephalastrum  racemosum,  16,  17,  122, 

132 
Syncephalis,  385 
Synchytrium  endobioticum,  253,  381 


Takamine,  J.  274,  302 

Tamiya,  H.,  89,  115,  121,  123-125,  127, 

129,  132,  136,  148,  158,  169 
Tanner,  F.  W.,  Jr.,  272,  273,  302 
Taphrina,  375 
dejonnans,  311 


462 


PHYSIOLOGY  OF  THE  FUNGI 


Tapke,  V.  F.,  388,  398 

Tatum,  E.  L.,  36,  161,  199,  202,  207,  215, 

225 
Taubenhaus,  J.  J.,  388,  390,  395 
Tauber,  H.,  96,  115,  142,  148,  277,  302 
Taylor,  C.  F.,  386,  398 
Temperature,  effect  of,  on  growth,  34 
on  resistance,  393 
on  sporulation,  307-310 

laboratory  exercise,  425 
Temperature  coefficient,  53 
Temperatures,  cardinal,  34 
Tenebris  moHtor,  58 
Ter  Horst,  W.  P.,  235,  243,  257,  265 
Ternetz,  C,  88,  99,  115,  319,  337 
Thatcher,  F.  S.,  37,  44,  376,  377,  382, 

398 
Thatcher,  R.  W.,  66,  86 
Thaysen,  A.  C,  49,  64,  267,  302 
Theis,  E.  H.,  258,  265 
Thiamine,  antagonists  of,  231-233 

formula  of,  187 

fungi,  deficient  for,  190 

mode  of  action  of,  190,  191 

moieties  of,  187-189 

specificity  of,  191,  192 
Thiamine  pyrimidine  formula,  187 
Thiamine  pyrophosphate  formula,  191 
Thiamine  thiazole  formula,  187 
Thielavia  basicola,  122 

sepedonium,  111 
Thielaviopsis  basicola,  80,  190,  387 
Thimann,  K.  V.,  120,  148,  327,  336 
Thorn,  C,  34,  44,  385,  398,  401,  414,  418, 

424,  439 
Thomas,  H.  E.,  374,  387,  398 
Thorn,  R.  S.  W.,  108,  115 
Thornberry,  H.  H.,  262,  265,  291,  302 
Thraustotheca  clavata,  193 
Thren,  R.,  400,  418 
Tilletia,  250 

levis,  410 

tritici,  410 
Timnick,  M.  B.,  312,  316,  324,  337 
Tomatin,  390 

Tonnis,  B.,  19?,  205,  209,  224 
Torula,  37 

cremoris,  194,  197 

lactosa,  197 

sphaerica,  197 

uiilis,  269,  271 


Torulopsis  kefyr,  197 

pulcherrima,  76 

sanguinea,  366,  367 

tdilis,  75,  109 
Toursel,  O.,  218,  224 
Tove,  S.  R.,  99,  115 
Toxins,  as  pathogenic  agents,  378-380 

produced  by  fungi,  292-293,  378-379 

in  relation  to  enzymes,  376,  382 
Trametes  americana,  135 
Transamination,  109 
Trehalose,  131,  132 
Treponema  pallidum,  290 
Trichodernia  koningii,  260 

lignorum,  100,  286,  314,  426,  437 

viride,  222,  287 
Tricholoma,  366 

personatum,  293 
Trichophyton,  259 

discoides,  196,  200 

gypseum,  238,  287 

inter  digitate,  76 

mentagrophytes,  238,  260,  288,  412 

purpureum,  238 

rubrum,  42,  287 
Trichotheciuni  roseuni,  288 

(See  also  Cephalothecium  roseum) 
Troutman,  M.  C,  66,  85 
Tryptophane,  biosynthesis  of,  229 
Tschang,  J.  L.,  184,  204 
Typhida  variabilis,  122 
Tyrosinase,  52 

U 

Ultraviolet  Hght,  effect  of,  on  sporulation, 
314-316 

inducing  mutations  by,  401,  405 
Umbreit,  W.  W.,  90,  115,  154,  169 
Underkofler,  L.  A.,  274,  275,  302 
Units  of  measure,  18 
Uppal,  B.  N.,  257,  265,  360,  371 
Urease,  52 
Urocystis  cepulae,  307,  356 

occulta,  361,  366 

tritici,  366 
Uromyces  caryophyllinus,  247,  248,  253- 
255,  362,  377 

fabae,  377,  393 

pisi,  348 


1 


INDEX 


463 


Ustilago  avenae,  360 

nuda,  358,  400 

striiformis,    118,    120,    172,    362,    380, 
401,  402,  428 

tritici,  388 

violaceae,  410 

zeae,  360,  365 
Utech,  N.  M.,  288,  302 
Utilization,  definition  of,  87 


Valsa  leucostoma,  321 

pini,  190,  196 
Vanadium,  81 

Vandecaveye,  S.  C,  219,  225 
Van  Lanen,  J.  M.,  272,  273,  302 
Van  Niel,  C.  B.,  132,  148 
Van  Rij,  N.  J.  W.,  39,  40,  44 
Van  Slyke,  D.  D.,  58,  64,  161,  162,  170 
Variation,  basis  of,  in  imperfect  fungi, 
411-412 

physiological,  400-403 
Vasudeva,  R.  S.,  376,  398 
Vaughan,  J.  R.,  288,  302 
Venturia  inaequalis,   254-256,   288,  304, 

356,  358 
VerticiUium,  378 

alho-atrum,  100 
Vincent,  J.  M.,  94,  114 
Vitamers,  186 
Vitamin  Bi  {see  Thiamine) 
Vitamin  Be  (see  Pyridoxine) 
Vitamin  B12,  82,  273 
Vitamin  deficiencies,  172-186 

absolute,  181,  182 

conditioned,  181-184 

methods  of  detecting,  174-176 

multiple,  178-181 

partial,  174,  177,  179 

single,  178 

total,  174,  177,  179 
Vitamin  H  {see  Biotin) 
Vitamin  K,  antagonists  of,  235-236 
Vitamin  K2  formula,  235 
Vitamins,  antagonists  of,  229-236 

characteristics  ot,  171 

concentration  of,  laboratory  exercise, 
433-435 

deficiencies    for,    laboratory    exercise, 
432-433 


Vitamins,  effect  of,  on  sporulation,  329- 
332 
fungi  self-sufficient  for,  172 
inhibitory  effects  of,  184-186 
production  of,  by  fungi,  272-273 
synthesis  of,  by  fungi,  171,   172,  272, 
273 

Vitucci,  J.  C,  123,  135,  147 

Volcani,  B.  E.,  239,  243 

Volkonsky,  M.,  93,  115,  132,  135,  148 

Von  Loesecke,  H.  W.,  269,  279,  302 


W 


Waksman,  S.  A.,  76,  84,  86,  159,  169,  246, 

264,  266,  267,  278,  279,  281-283,  285, 

286,  290,  298,  299,  302 
Walker,  J.  C,  356,  371,  390-392,  397,  398 
Walker,  L.  B.,  347,  348,  354 
Walker,  T.  K.,  278,  302 
Wallerstein,  J.  S.,  267,  303 
Wallerstein,  L.,  275,  303 
Walton,  R.  C,  344,  354 
Warburg,  O.,  58,  64 
Ward,  G.  E.,  271,  272,  281,  282,  299,  303, 

326,  336 
Ward,  M.,  392,  398 
Waring,  W.  S.,  73,  86 
Water,  87,  88,  149 
Waters,  C.  W.,  393,  398 
Webb,  R.  A.,  269,  300 
Webb,  R.  W.,  360,  361,  363,  364,  371 
Weedon,  F.  R.,  253,  264 
Weimer,  J.  L.,  50,  63,  344,  354 
Weinhouse,  S.,  143,  148 
Weinstock,  H.  H.,  198,  207 
Weiss,  S.,  295,  303 
Welch,  A.  D.,  226,  243 
Wellensiek,  S.  J.,  383,  393,  398 
Wellman,  F.  L.,  356,  371 
Wellman,  R.  H.,  245,  261,  264,  265 
Wells,  P.  A.,  280,  303 
Werkman,  C.  H.,  73,  86,  128,  136,  138, 

146,  148 
West,  R.,  273,  303 
Westergaard,  M.,  325,  326,  337 
Weston,  W.  H.,  Jr.,  35,  43,  424,  438 
Weswig,  P.  H.,  231,  243 
Wheeler,  H.  E.,  404,  417 
Whiffin,  A.  J.,  288,  303 
White,  A.  G.  C.,  127,  148,  232,  233,  244 


464 


PHYSIOLOGY  OF  THE  FUNOI 


White,  M.  G.,  138,  139,  148 

White,  N.  H.,  12,  23 

White,  W.  L.,  49,  64,  221,  222,  225 

Wieland,  H.,  292,  303 

Wilcoxon,  F.,  247,  248,  252-255,  264,  265 

Wilkins,  W.  H.,  285,  303 

Willainan,  J.  J.,  138,  139,  148,  275,  299 

Williams,  B.,  291,  297 

Williams,  R.  J.,  198,  199,  204,  206,  207 

Williams,  R.  R.,  187,  207 

Wilson,  A.  N.,  232,  233,  243 

Wilson,  P.  W.,  51,  62,  99,  115 

Wilson,  W.  E.,  121,  148 

Wingard,  S.  A.,  393,  398 

Winzler,  R.  J.,  56,  60,  61,  64,  194,  207, 

233,  241 
Wirth,  J.  C,  102,  108,  115,  141,  148,  186, 

191,  192,  207 
Witkop,  B.,  292,  303 
Wohler,  F.,  277,  303 
Wolf,  F.  A.,  35,  44,  99,  115,  293,  303,  347, 

354,  371,  386,  399,  401,  402,  418 
Wolf,  F.  T.,  35,  44,  99,  115,  121,  148,  231, 
243,  293,  303,  347,  354,  371,  386,  399, 
402,  418 
Wolfrom,  M.  L.,  290,  299 
Wolpert,  F.  S.,  159,  160,  170 
Wood,  H.  G.,  136,  148 
Wood,  W.  A.,  201,  207 
Woods,  D.  D.,  230,  243 
Woolley,  D.  W.,  195,  207,  226,  229,  231- 
233,  235,  236,  237, 239, 240, 242-244, 
378,  399 
Wooster,  R.  C,  199,  207 
Worley,  C.  L.,  28,  44 
Wright,  L.  D.,  215,  225,  226,  244 
Wright,  L.  T.,  291,  303 
Wyss,  O.,  167,  170,  183,  184,  202,  207, 
215,  225 


Xylaria  hypoxylon,  190 

rnali,  100 
Xylose  formula,  123 
Xylulose  formula,  123 


Yarwood,  C.  E.,  311,  337,  358,  359,  371, 

393,  399 
Yaw,  K.,  121,  148 
Yeager,  C.  C,  384,  399 
Yeast,  film  formation  by,  39 

heat  production  by,  61 

measuring  growth  of,  30 

morphology  of,  39 

protein  of,  268 

respiration  of,  effect  of  cyanide  on,  56 

use  of,  for  food,  268,  269 

vitamin  content  of,  269 
Yegian,  D.,  240,  244 
Yoe,  J.  H.,  73,  86,  260,  265 
Yoshimura,  F.,  74,  86 
Young,  H.  C.,  79,  86,  100,  115,  253,  265 
Yuill,  J.  L.,  275,  303 


Z  factors,  effect  of,  on  spore  germination, 

364-365 
Zentmeyer,  G.  A.,  260,  265 
Zikes,  H.,  33,  39,  44 
Zinc,  76,  282 

Zygorhynchus  moeUeri,  100 
Zygosaccharomyces,  199,  327 

acidifaciens,  91,  92 

ladis,  197 

marxianus,  215 


4 


4 


J 


4 


■"uaiHi-i.Hi'--