'-=-'.- -------

■1-^'

AN INTRODUCTION TO
BACTERIOLOGICAL CHEMISTRY

First Edition 1038
Second Edition 194G

AN INTRODUCTION TO

BACTERIOLOGICAL
CHEMISTRY

BY

C. G. ANDERSON

Ph.D. (Birm.), Dip.Bact. (Lond.)

formerly lewis cameron teaching fellow,
bacteriology departxment,
university of edinburgh

Second Edition

EDINBURGH

E. & S. LIVINGSTONE, LTD.

16 & 17 TEYIOT PLACE
1946

Muiie and Fiin'.ed in deal Brilain

PREFACE TO THE SECOND EDITION

WHEN iheliryt edition of tJi is hook appeared in 11)38
it was considered that the suhject of Chemotherapy
of bacterial infections was in too primitive a state
to warrant inclusion. The advance in our knowledge
since then has made it possible and desirable to consider
the topic in some detail. The ideas on which the
explanation of chemotherapeutic action is based, and
which form the foundation for plamiing further investiga-
tion, involve the chemistry of metabolic processes and
naturally come within the scope of this book. Similar
considerations lead to the inclusion of a chapter on
Antibiotics and to a fresh presentation of the facts known
about Gro\\i;h Factors. These chapters constitute the
main difference between the first and the present editions,
but new material has been added to a number of other
chapters in order to keep level, as far as possible, with
the changes involved in such a rapidly growing branch of
Biochemistry.

Some criticism has been expressed that bibhographical
references to the original literature were not given in the
first edition. It is the opinion of the author that a
detailed bibliography is out of place in a small textbook
intended primarily for students, but in order to extend
the usefulness of the volume an endeavour has been
made to give, at the end of each chapter, some references,
mainly to monographs and reviews, which will serve as
a guide to the original papers.

Once more it is a pleasure to acknowledge my debt to
the authors whose works have supplied the material
presented. Especially are my thanks due to colleagues
and friends for much helpful criticism and advice.

C. G. ANDERSON.

Wellcome Physiological Research Laboratories

Laxgley Court, Beckenham, Kent

July, 1946

PREFACE TO FIRST EDITION

THIS text-book is the outcome of lectures on Bacterio-
logical Chemistry presented as part of the course for
the University of London Academic Diploma in
Bacteriology, and recently to students taking Bacteriology
as an Honours subject in the University of Edinburgh.
During the period over which these courses have extended
the need has been increasingly felt for a text -book of
reasonable size yet covermg a sufficiently wide range of
topics. Whilst many excellent monographs on various
aspects of the subject exist, there seems to be no single
book giving a survey of the whole field in a form suitable
for students of such courses as those mentioned.

The present volume makes no claim to be encyclo-
paedic, but an endeavour has been made to cover the
requirements of students, and j)erhaps of those research
w^orkers whose interests may not be primarily chemical
but who feel the need for some understanding of the
metabolic behaviour and chemical nature of the organisms
which they are handling. In order to keep its size within
reasonable limits it has been necessary to assume a
knowledge of elementary organic chemistry and of a
certain amount of bacteriology. In view of the rapid
expansion of the subject within the past ten to fifteen
years, and the consequently ever-changing views and
opinions expressed concerning the various reactions in-
volved, the selection of the appropriate material has not
been easy. Without doubt much has been omitted
which should have been included, and certain matters
admitted which the future will show to be of only passing
importance. Only certain aspects of immunochemistry
have been considered and no attempt has been made to
deal with the subjects of disinfection and chemotherapy,

Vlll PREFACE

nor with the cheiiiistiy of cultiiic Jiiedia and staining
reactions. Their treatment here could be little more than
a catalogue of substances and organisms, for although a
large body of empirical data concerning them is available,
we have as yet but little exact chemical knowledge of the
mechanisms involved.

In the hope that the interest of students reading the
following j)ages will have been stimulated, sources of
further information have been indicated at the ends of
the chapters.

Acknowledgment is gratefully made to the authors of
the many monographs, standard works and papers which
have been drawn upon freely for the material collected
here.

It is a pleasure to record my deep indebtedness to
Professor H. Raistrick, F.R.S., of the London School of
Hygiene and Tropical Medicine, who awakened my
interest in this subject, and to Professor T. J. Mackie, to
whose keenness and encouragement the course in this
University owes its inception.

C. G. ANDERSON.

Defabtmexnt op Bacieriology,

University of Edinburgh,

December, 1937.

CONTENTS

PAirr I

GENERAL CONSIDERATIONS

CHAP. I'A^E

I. Introduction ------- 1

II. Hydrogen Ion Concentration and pR ; Oxida-
tion-reduction Potentials - - - - 4

III. Colloids and Adsorption - - - - 26

IV. Enzymes -------- 36

V. The Chemical Composition of Bacteria, Yeasts

AND THE Lower Fungi - - - - 56

PART II

METABOLISM

VI. The Nutrition of the Autotrophic Bacteria 67

VII. The Nutrition of the Heterotrophic Bacteria 80

VIII. Adaptive and Constitutive Enzymes - - 89

IX. Growth Factors ------ 98

X. Chemotherapy ------- 122

XL Antibiotics ------- I£6

XII. Bacterial Respiration ----- 187

XIII. Nitrogen Metabolism ----- 214

XIV. Carbon Metabolism _ . - - - 240

XV. Alcoholic Fermentation . - _ . 262

XVI. The Fermentation Products of the Lower

Fungi -------- 282

XVII. Industrial Fermentations - - - - 310

XVIII. The Proteins of Micro-organisms - - 328

XIX. The Polysaccharides of Micro-organisms - 345

XX. The Lipoids of Micro-organisms - - - 369

XXL The Pigments of Micro-organisms - - 385

61 ^fiii

X CONTENTS

PAIIT 111

SOME ASPECTS Ot^ IMMUNOCHEMISTRY

CHAP. TAOE

XXII. Antigens, Haptens, Antibodies and Com-
plement ------- 398

XXIII. The Mechanism of Antic4en-antibody Keactions 441

APPENDICES

I. The Isolation and Identification of Meta-
bolic Products ------ 462

II. Synonyms ------- 469

INDEX

BACTERIOLOGICAL CHEMISTRY

CHAPTER I
INTRODUCTION

THE subject of Bacteriological Chemistry has been
steadily growing in scope and importance since the
days when Pasteur studied the fermentation reactions,
normal and abnormal, until, during the past one or two
decades, it has expanded with such rapidity that it has
now almost acquired the dignity of a special branch of
Biochemistry, and is even in danger of itself becoming
subdivided with production of such offshoots as Immuno-
chemistry. This rapid growth is in part due to the
ever-increasing utilisation of microbiological methods
and products in industry, and in part to purely academic
investigations into the mechanisms by which the bacteria,
yeasts and fungi gain the energy for their growth and
reproduction and synthesise the multitudinous products
which they build into their cell structures or excrete into
the medium in which they develop. The combination of
utilitarian and academic motives has resulted in the
accumulation of a vast number of facts which are only
just beginning to be shaped into an ordered whole in
which it is possible to see the relationships between
apparently quite different modes of metabolism, and the
))ewildering variety of substances elaborated during such
l^rocesses .

A great deal of the information which we possess is
still only of an empirical nature and we have not yet
found out how to fit these facts into the general picture.
The present stage of development of microbiological

2 BACTERIOLOGICAL CHEMISTRY

clicniistry is .somowJiat liko tliat of a. Jialf-fiiiisjKMl jig-saw
})Ti/zlo. (Sonio areas are nearly eojiiplele ; in otliers only
a few of the pieces have so far been fitted into place.
Quite recently the filling in of the area which includes
chemotherapy has been ^progressing rapidly as a result
of the development of the sulphonamide drugs, and a
reasoned account of their action can now be presented.
The subject of disinfection, however, is still in a nebulous
state ; we know a great deal about the necessary con-
centrations of disinfectants and the conditions for their
action, but very little as to how the observed results are
brought about. It seems probable that the present views
on the mechanism of chemotherapy may Avell be applied
to the action of disinfectants and antiseptics.

The position of Immunochemistry, in this respect,
has improved considerably in the past decade and we are
at last able to understand something of what is really
happening during immunological reactions, although there
is much detail still to be filled in and much expansion of
our knowledge necessary.

Our knowledge of the metabolic processes of micro-
organisms is perhaps the best developed part of the
whole structure and enables us to see the connections
between the modes of life of many different bacterial
types. Even here, though, we still know comparatively
little of the mechanism of the synthetic assimilation
processes.

The scheme of treatment adopted in the following
chapters has been first of all to deal with the general
conditions which influence and determine the behaviour
or micro-organisms. The importance of the hydrogen
ion concentration and of colloidal phenomena is obvious,
as also is the part played by the enzymes on which almost
every stage of the life of the bacteria, yeasts and moulds
depends. Then follows an account of the different ways
in which micro-organisms obtain the energy and starting
materials for their growth and reproduction. The

INTKODUCTION 3

discussion of chemotherapeutic action falls logically into
place here since it is held to depend on interference with
the metabolic activities of the bacteria. The consideration
of the by-products of metabolic and respiratory processes,
the fermentation products in which man is mainly
interested, follows naturally. Then the substances
s}Tithesised by the organisms for their own use are
detailed. Finally a brief outline of the chemistry of
antigens, antibodies and their reactions is presented.

The usual difficulty with regard to the consistent
naming of bacteria has been encountered. In the absence
of any standardised procedure in this country it has been
considered desirable to adopt the nomenclature sanctioned
by the Society of American Bacteriologists as exemplified
in Bergey's " Manual of Determinative Bacteriology "
(Fifth Edition). The common (as opposed to the
scientific) names of certain organisms have been used in
some instances. A list of synonyms covering cases
which may cause confusion has been added as Appendix II.

CHAPTER II

HYDROGEN ION CONCENTRATION AND pR ;
OXIDATION-REDUCTION POTENTIALS

Hydrogen Ion Concentration. — The course of all biological
processes is profoundly influenced by the degree of acidity
or alkalinity of the fluid in which they take place — whether
the fluid ])e tlie cell contents, or a circumambient fluid
like blood, or a cidture medium in which micro-organisms
are growing, or a solution in wliicli enzymes are acting.
As a notable example one may consider the blood. Should
it become very slightly acid death in coma will result, with
the heart muscle relaxed ; on the other hand, if it should
become but slightly alkaline tetany results and the heart
will cease to function with the muscle contracted. The
heart will only function properly when the blood is within
a very narrow range between acidity and alkalinity.
Similar, though as a rule not such dramatic, changes
follow alteration of the normal state of other biologically
concerned fluids. Some bacteria, for instance, thrive in
quite strongly acid solutions but raj)idly die out in
alkaline conditions ; others, like the cholera vibrio,
develop in alkaline but not in acid media. The same
applies to enzymes ; the gastric enzyme, pepsin, is only
active in breaking down j)roteins when in acid solution,
whilst trypsin, in the pancreatic juice, requires an alkaline
medium for its activity.

According to modern views an acid is defined as a
substance which tends to lose a proton (hydrogen nucleus

■4

HYDROGEN ION CONCENTRATION i)

or H+) and a base is a substance which tends to acquire
a proton. This may be expressed b}^ the equation

A ^=^ H+ + B

where A represents an acid and B a base. This means
that every acid must be associated with a corresponding
or " conjugate " base and vice versa. Generally the
conjugate acid or base is the solvent in which the substance
is dissolved. In aqueous solutions water can act as the
conjugate base of an acid or as the conjugate acid of a
base since it is capable of either taking up or giving up a
proton. For an acid in water the ecpiilibrium is : —

HA + H.OH ^=^ H.OH,+ + A-
(acid) (base)

H.OH2+ is what is usually known as the hydrogen ion
for which the symbol H+ is commonly used.
Applying the Law of Mass Action

(a . H.OHo-) (a . A-)
(a. H.OH) (a. HA)

(1]

where K'a is the dissociation constant of the acid and
(a . H.OH0+), (a . A-), (a . H.OH) and (a . HA) are the
" activities " of the hydrogen ion, conjugate l^ase,
un-ionised water and un-ionised acid respectivel}^ The
" activity " of an ion is the product of its concentration
and its " activity coefficient " which is a measure of the
influence of surrounding ions upon it and which accordingly
depends upon the dilution of the solution. Since the
activity of un-ionised water may be regarded as constant,
because its amount is virtually unaffected by the very
small degree of ionisation which it undergoes, equation (1)
can be rewritten with the new constant K a

(a.H.OH,+) (a.A-)
^'-^ = i^TRA) ... (2)

G BACTERIOLOGICAL CHEMISTRY

Using the more familiar symbol, H+, for the hydrogen
ion this becomes

(a.H+) (a.A-)
^^' ^ (a. HA) .... (3)

For a base dissolved in water the equilibrium is
expressed by the equation

B + H.OH ^=i BH+ + OH-

(Base)

The dissociation constant, K'b, for the base is then given
by the expression

(a.BH+) (a.OH-)

K'b

(a.B) (a.HOH^

Since in this case, too, the activity of the A\'ater can be
regarded as constant this becomes, with the new
constant, Kb,

(a.BH+) (a.OH-) '

^^ = OTBJ • • • ^^\

Water is capable of either giving up or accepting a
proton. Therefore in pure water and in all aqueous
solutions the equilibrium

H.OH + H.OH ^=^ H.OH2+ + OH-

niust exist and the dissociation constant is given by

_ (a . H.OH,+) (a . 0H-)
"" ~ (a.H.0H)2

The activity, (a . H.OH), of un-ionised water can be
regarded as constant and therefore

Kw = (a . H.OH2+) (a . 0H-) ... (5)

The product of the activities of the acidic and basic ions
of a solvent is also loiown as the ionic product of the
solvent. Replacing H.OH^+by the familiar symbol for
tlie hydrogen ion, equation (5) becomes

Kw = (a . H+) (a . OH") .... (0)

HYDROGEN ION CONCENTRATION 7

In dilute solutions, that is when the ionic strength is
small, the activity coefficients are very nearly unity and
the activities (a.H+) and (a. OH") of the hydrogen and
hydroxyl ions can be replaced by their concentrations
(C.H+) and (COH") respectively. Equation (6) can
now be written

Ku' = (C.H-) (C.OH-) .... (7)

In an exactly neutral solution, obviously, the concentration
of the acidic (hydrogen) ions must be equal to the
concentration of the basic (hydroxyl) ions. That is
(C.A+) = (C.OH~) and, from equation (7), each must be
equal to VKw.

The ionic product, Kw, can be determined experi-
mentally from conductivity measurements since the
conduction of electricity through a liquid depends on the
number of ions available to carry the current. The
value for pure water at room temperature has been
found to be 10"^*. Consequently in neutral solution the
concentration of both hydrogen and hydroxyl ions must
be V10~^* or 10^' gram ions per litre. If more hydrogen
ions than this amount are present the solution is acid
and if there are less hydrogen ions the solution is alkaline,
but whatever the state of the solution the ionic j^roduct is
constant and equal to 10-^*. In other words a greater
amount of hydrogen ions means a smaller amount of
hydroxyl ions and vice versa. Consequently the strength
of an alkali, as well as that of an acid, can be expressed
in terms of hydrogen ion concentration.

A normal solution of a strong acid will contain about
1 gram ion per litre of hydrogen ions, the exact amount
depending upon the degree of dissociation of the particular
acid and the activity of the ions. A strong solution of an
alkali will contain about 10-^^ gram ions per litre of
hydrogen ions, derived from the ionisation of the water.
The degree of ionisation of a solution can be measured by
the conductivity of the solution, which depends on the

8 BACTERIOLOGICAL CHEMISTRY

number of ions present to carry the current, or by the
depression of the freezing point of the sokition, which
depends on the total number of ions and molecules present.
A tenth normal solution of hydrochloric acid is 91 per
cent, ionised so that the concentration of hydrogen ions
in such a solution will be

0-1 X 91

— — — — = 0-091 = 9-1 X 10~2 grams per litre.

Acetic acid is a weak acid and is only 1-3 per cent, ionised
in a tenth normal solution, and such a solution will,
therefore, contain only

— --- — = 0-0013 = 1-3 X 10~^ grams per litre of hydrogen ions.

This method of expressing the hydrogen ion concentra-
tion and correspondingly the acidity or alkalinity of a
solution is somewhat cumbersome particularly in the
range of values near neutrality, from about 10^® to 10-^
grams of hydrogen ions per litre, in which the great
majority of biological phenomena occur. The adoption
of the exponential or pH method of expression suggested
by S^rensen in 1909, and which is now universally used,
has greatly simplified the statement and comprehension
of such values. S</>rensen defined the pH value of a
solution as the negative logarithm of its hydrogen ion
concentration. That is

pn = - log (C.H+) or (C.H+) = 10-^H

For neutral water containing lO-*^ grams per litre of
hydrogen ions

^H = - log 10-' = log 10' = 7

For the example quoted above of 0-lN-hydrochloric acid
containing 9-1 x 10-^ grams per litre of hydrogen ions,

pU = - log (9-1 X 10-2) = 2 - log 9-1 - 2 - 0-959 = 1-041

and for 0-lN-acetic acid,

2)H = - log (1-3 X 10-3) = 3 - log 1-3 = 3 - 0-114 = 2-880

HYDROGEN ION CONCENTRATION U

Blood, which contains 4-7 x 10"^ grams per litre of
hydrogen ions, will have

pK = - log (4-7 X 10-8) = 8 - log 4-7 = 8 - 0-672 = 7-328

Since, as we have seen, even strongly alkaline solutions
contain some hydrogen ions, and the amount of hydrogen
ions is inversely proportional to the amount of hydroxyl
ions (on which the alkalinity depends), the alkalinity of
a solution can also be expressed in terms of its hydrogen
ion concentration and the pH. scale. Thus an alkaline
solution containing 3-7x10-^ grams of hydrogen ions
per litre has

pK = - log (3-7 X 10-9) = 9 - log 3-7 ^ 9 - 0-568 = 8-432

As the product of hydrogen ion concentration and
hydroxyl ion concentration is constant at 10"^*, it is
obvious that the more hydrogen ions there are
present, that is, the more acid the solution, the loAver
will be the pH, tending to the value p}l = 0, which is the
theoretical limit when there are no hydroxyl ions present.
On the other hand, the more hydroxyl ions there are
present, that is, the more alkaline the solution, the higher
will be the pH value, tending to the hypothetical limit
;:>H =14 when there are no hydrogen ions present. Since
neutrality occurs at pH 7, all values lower than this refer
to acid solutions, whilst higher values than 7 indicate
alkaline solutions.

It must be remembered that the pH scale is logarithmic,
and that accordingly a change in ^^H value of 1 unit
means a tenfold change in acidity or alkalinity. Thus a
solution of pH 5 will be ten times as acid (will contain
ten times as many hydrogen ions) as one at ^^H 6, and a
hundred times as acid as one at pH 7. Similarly a solution
at ^^H 10 will be ten times as alkaline (will contain ten
times less hydrogen ions) than one at pK 9, and be one
hundred times as alkaline as a solution at pH 8.

The titratable acidity of a solution must not be con-
fused with its pH value. The titratable acidity de]-)ends

10 BACTERIOLOGICAL CHEMISTRY

only on the amount of acid present ; thus, for example,
all tenth -normal acid solutions will have the same titratable
acidity whatever the particular acid present may be,
whether it is a strong acid like hydrochloric acid or a
weak acid like acetic acid. But the ^^H values of different
tenth-normal acid solutions will vary widely, depending
on the acid concerned. This variation in pK value of
acid solutions containing the same equivalent weight of
acid results from the different degrees of dissociation
of the acids with the consequent production of different
concentrations of hydrogen ions. Thus we saw that 0-lN-
hydrochloric acid is 91 per cent, dissociated and has a pK
value of approximately 1, whilst O'lN-acetic acid, which
is 1-3 per cent, dissociated, has pH 2*9 ; that is hydro-
chloric acid is nearly one hundred times as strong an
acid as acetic acid, although the two solutions have the
same titratable acidity. In other words, merely knowing
the amount of titratable acid in a solution does not tell
us enough about its properties and probable effect on a
culture of bacteria or on a fermentation reaction. We
must know also the ^^H value of the solution, which will
give us a much better idea of the magnitude of effect
to expect.

The logarithmic or exponential mode of expressiofi is
also used for the statement of dissociation constants,
which are referred to as ^^K values where

pK = — log K

K being the dissociation constant under consideration.
Thus the dissociation constant of acetic acid at 25°C. is
1-8 X 10"^ and its pK. value is, therefore,

2iK = - log (1-8 X 10-5) = 5 ~ log 1-8 = 5 - 0-255 = 4-745

An acid like phosphoric acid, H3PO4, which has three
ionisable hydrogen atoms, has three pK. values, one
corresponding to the dissociation of each liydrogen ion.
The three dissociation constants and pK values of

HYDROGEN TON CONCENTRATION 11

pliuspliuric acid are —

For the first hydrogen ion K — 11 1(>~-. />K ^ I -Hoi)

For the second hj'drogen ion K = 2 < 10~", pK. — 6-699

For the third hydrogen ion K = 36 x IQ-^^, pK = 12-444

Obviously the higher the pis. value, the smaller the
dissociation constant and the weaker the acid.

The Measurement of Hydrogen Ion Concentration. —
In general there are two main methods of measuring
hydrogen ion concentration, electrical and colorimetric.
The electrical methods are by far the more accurate, but
require the use of costly and delicate apparatus not
available in many ordinary laboratories. The colorimetric
methods, while not so accurate, are much cheaper,
quicker and simpler to perform.

The electrical methods depend on balancing the

potential difference set up between the solution of

unknown hydrogen ion concentration and an electrode

immersed in it against the potential of a standard cell.

If a rod of metal is immersed in a solution of one of its

salts, the metal will tend to dissolve to a degree depending

on its " electroljrtic solution pressure," producing positively

charged ions of the metal, and consequently leaving a

negatively charged mass of undissolved metal. On the

other hand, the salt in solution ionises, and the metallic

ions will exert a definite osmotic pressure, depending on

the concentration of the salt and its degree of dissociation.

If the ionic osmotic pressure is less than the " solution

pressure " of the metal, as is the case with zinc, the

latter will dissolve in order to establish equilibrium.

If, however, the " solution pressure " is less than the

ionic osmotic pressure, as is the case with copper or

mercury, ions will be deposited on the metal and the rod

will acquire a positive charge relatively to the solution.

Each metal in contact with a solution of one of its salts

acquires in this way a characteristic " electrode potential,"

the strength of whicli will depend on the concentration

12

BACTERIOLOGICAL CHEMLSTRY

of the Salt solution and the particular metal in t^uestion.
If two such electrodes, for instance, copper in copper
sulphate and zinc in zinc sulphate, are joined together in
an electrical circuit by connecting the two metals with a
wire and the two solutions either with another wire or
with a ri-tube containing a solution (Fig. 1), then a

COO D ocTo ^ D"

4- rH

Cu

sv

Ca S 0 +

In

lYx SO4

Fig. 1

current will flow in the direction of the higher to the
lower potential. An electrode of this type which is very
frequently employed for purposes of measurement is the
standard calomel electrode in which mercury is in contact
with saturated mercurous chloride and N-potassium
chloride, and which has a potential of +0-56 volt at 18^ C.
Another standard electrode is the hydrogen electrode
in which an electrode of platinum black saturated with
hydrogen gas is immersed in an acid solution normal in
respect to the hydrogen ions. For reasons of convenience
this potential is arbitrarily taken as being zero. If the
concentration of hydrogen ions is different from normal
the potential will also be different from the standard, and
will be a measure of the hydrogen ion concentration.

The potential differences between a standard half -cell,
such as the calomel electrode, and an electrode immersed
in the unknown solution are measured by balancing them

HYDROGEN lOX CONCENTRATION

13

on a potentiometer until no ciuTent flows, as 'determined
by a galvanometer (Fig. 2). Hence, loiowing the electrode
potential of the standard cell, that of the iinknowni may
be calculated, and from this the hydrogen ion concentra-
tion and the pH determined. A third simple and
convenient standard electrode is the quinhydrone electrode

]

<i^

Unknown

in^iij

TANDARD

Fig.

in which quinhydrone is dissolved in a standard birffer
solution of known 2^H, the potential of a platinmn
electrode immersed in such a solution depending on the

A system very commonly used in the laboratory for
measurement of ^^H is the glass electrode in which a glass
membrane separates a solution of normal hydrochloric
acid, into which dips a calomel electrode, from the
solution of unknown pH which is in contact with another
calomel electrode immersed in 3*5 normal potassium
chloride solution. The potential which is set up across the
membrane depends on the jjH of the solution and is
measured by determining the null point on a potentio-
meter and electronic valve circuit.

Colorimetric methods depend on the use of various
dyes which show colour changes over characteristic

14

BACTERIOLOGICAL CHEMISTRY

ranges of pH values of the solution into which they are
introduced. The colour change may be due to the
conversion of an almost non-ionised weak acid or a weak
base of one colour to a strongly ionised salt, where the ion
has a characteristic colour. For example, methyl orange
exists as the yellow non-ionised acid in acid solution,
but on addition of alkali the corresponding salt is formed
which ionises to give the red anion. Or the colour
change may be associated with a change in the internal
structure of the molecule producing a coloured quinonoid
form, as is held to be the case with phenolphthalein. In
acid solution the compound exists in the colourless form : —

HO

\-

CH-CH

-cf ^C~OH
CH=CH
0

In alkaline solution this is converted into the sodium
salt of the tautomeric quinonoid form : —

H0<

I CH=CH

c=c<^ \c=o

CH=CH

COONa

which ionises to give the red anion characteristic of
phenolphthalein in alkaline solution. Some of these
indicators, for example, Congo red, phenolphthalein,
litmus, change colour sharply over a narrow range of pH
value, that is, with a comparatively small change in
hydrogen ion concentration, and consequently are of use

Topfer's lleiigeui

(.Diraethylamino-azobe izenef

1

^rcta-c^esoI Puiple

and Colour Chaages it

HYDKOGEX lOX COXCEXTRATION 1o

ill determining the end point in titration of acid and
alkalies. It must be remembered that the end points
which they indicate are not identical but occur at different
pK values, about pii 4 for Congo red, about 7 for litmus
and about 9 for phenolphthalein. Other indicators change
colour gradually over a range of two to three units of jjK
value, the actual colour observed being determined by the
^H of the solution. By matching the colour of a suitable
indicator in a solution of unknown j^H against the range
of colours which it gives in solutions of known ^^H value,
the hydrogen ion concentration of the unknown solution
can be estimated. Indicators covering piL values from
1 to 13 are available, and each indicator is put up in an
appropriate series of standard buffer solutions of known
pK values in sealed tubes against which the comparisons
are made. An even more convenient method is the use
of the Lovibond comparator in Avhich the standard tubes
containing the indicator are replaced by a disc carrying
tinted glasses of colours which match the standards, at
intervals of ^^H of 0-2 unit. Besides convenience and speed
in use these glass colour standards possess the advantage
of not fading, a fault to which the buffer solution indicator
standards are subject, particularly if they are exposed
to light.

A chart showing the colour changes and 2:)H range of
the more useful indicators is given in Fig. 3.

Buffer Solutions. — In a great deal of biological work,
particularly when dealing with enz}- me systems or reactions
of a similar nature, it is important and often even essential
to maintain the ^^H of the solution constant in spite of
the fact that an acid may be produced or used up during
the course of the reaction. If such a reaction were carried
out in simple aqueous solution the ^^H value would
obviously change j^rogressively with time, but by making
use of buffer solutions these changes can be eliminated,
the buffer solution offering a reserve of acid or alkali
capable of taking up or supplying alkali or acid as occasion

16 BACTERIOLOGICAL CHEMISTRY

demands and thus preventing any considerable change of
pH. As the name implies, the buffer sohition acts as a
shock absorber, or like a sponge, to take up excess acid
or alkali.

Buffer solutions are, in general, mixtures of a weak
acid with one of its salts, the particular acid being chosen
to suit the ^^H range desired. A weak acid, such as
acetic acid, in solution is only ionised or dissociated to a
small extent, so that most of it is present as undissociated
neutral acetic acid molecules : —

HAc^=^H+ + Ac-. (1)

The Law of Mass Action applies to this dissociation,
giving the equation

[H+]x[Ac-] = A:[HAc], or [H+] = ^^^ . . . (2)

On the other hand, the salt, sodium acetate, is strongly
dissociated to give sodium and acetate ions. Now in a
mixture of the acid and the salt practically all the acetate
ions will be derived from the highly dissociated salt and
only very few from the acetic acid, and consequently,
since the number of hydrogen ions cannot be greater
than the number of acetate ions derived from acetic acid,
the hydrogen ion concentration will be lowered. As the
acetate ions in such a mixture come almost entirely from
the salt their concentration wall be a x [NaAc], where a
is the fraction of the salt which is ionised. So that we
can write the last equation as

That is, the hydrogen ion concentration in such a mixture
depends on the ratio of free acid to salt ; the higher the
salt content the lower the hydrogen ion concentration.
It will be seen that dilution of such a solution will have
very little effect on the pH value, as both [HAc] and
[NaAc] are altered to the same extent, the ratio being the

BFFFEPx SOLUTIONS 17

same. The only el'Icft ui (.liluliun is to inciease the value
of a slightly as there is slightly more dissociation in dilute
solutions than in concentrated ones ; hence dilution of a
buffer solution will decrease the hydrogen ion concentra-
tion very slightly, that is the pH value will become very
slightly higher.

If a strong acid such as hydrochloric acid is added to
a buffer solution the hydrogen ions to which it gives rise
will immediately combine with acetate ions to give
imdissociated acetic acid, and consequently there will be
but a slight change of hydrogen ion concentration.
Similarly, if an alkali such as sodium hydroxide is added,
the hydroxyl ions will combine with the hydrogen ions
to form water, and once more there will be very little
change in hydrogen ion concentration. This perhaps
becomes more clear in terms of the acid and conjugate
base view. Hydrogen ions resulting from addition of acid
to the systems are " neutralised " by the acetate ions
acting as a weak base : —

H.OH2+ + A- ^=^ HA + H.O

(acetate (acetic
ion) acid)

Whilst hydroxyl ions are " neutralised " by the reaction

OH- + HA ^=^ HoO + A-

(acetic (acetate

acid) ion)

When the buffer is a weak base and one of its salts such as
ammonia and ammonium acetate the corresponding
reactions are

H.OH,+ + NH3 ^=^ XH4+ + HoO

(ammonia) (ammonium

ion)

and

OH- + NH4+ ^=^ H.O + NH3

(ammonium (ammonia)

ion)

The better the buffer the more acid or alkali is required
to alter the pK of the solution by a given amount.

18

BACTERIOLOGICAL CHEMISTRY

Usually tlic inaxiimmi effect is obtained when ec[uivalent
amounts of acid (or base) and its salt are present in
solution, at which point the ^^H is equal to the ^^K of the
acid or ^^OH (which is the exponential expression for
hydro xyl ion concentration exactly analogous to pK) is
equal to the ^^K of the base. Buffer action is usually
restricted to pJi or ^^OH ranges about one unit above and
below the ^^K value. By making mixtures of appropriate
quantities of the acid and salt, solutions having various
pH values within the range can be prepared, according
to the formula

_ concentration of salt

concentration of acid

The buffer capacity will obviously depend on the
concentration of the mixture and must be selected
appropriately for the purpose required. Some of the more
useful buffer mixtures are shown in Table 1 .

Table 1

Acid

Salt '

pK Range

PhthaUc acid

Potassium hydrogen jihthalate

2-2 - 3-8

Phenylacetic acid

Sodium phenylacetate -

3-2 - 4-9

Acetic acid -

Sodium acetate

3-6 - 5-6

Potassium hydrogen

phthalate.
Sodium dihydrogen

phosphate.
Boric acid -

Dipotassium phthalate -
Disodium hydrogen phosj)hate
Sodium borate (borax) -

40 - 6-2
5-9 - 8-0
6-8 - 9-2

Diethylbarbituric acid

(Veronal).
Sodium borate

Sodium diethylbarbiturate
Sodium hydroxide

70 - 9-2
9-2 - 110

Disodium hydrogen
phosphate.

Trisodium phosphate

110 - 120

OXIDATION-REDUCTION POTENTIALS 1 !)

The phosphate buffers make use of the second and
third hydrogen atoms as the acids with the corresponding
salts so that phosphate buffer solutions can be made
covering a wide range of pK values.

So called " Universal buffer mixtures " consist of a
mixture of acids, with pK values covering a wide range,
to which the calculated amounts of alkali are added to
give solutions buffered at the required pK. Such a mixture
covering the range pH 2 to pK 12 comprises phosphoric
acid, citric acid, boric acid and hydrochloric acid. The
same range is also covered by the mixture boric acid,
citric acid, jDotassium di hydrogen phos2)hate and veronal.

Proteins and amino -acids have a considerable buffering
effect since they may function as weak acids or weak
bases in virtue of their carboxyl and amino groups,
according to the conditions.

Oxidation-Reduction Potentials

It will be seen later that oxidation and reduction play an
extremely important part in the respiratory and metabolic
processes of micro-organisms. In fact it is not too much
to say that their whole existence depends on such reactions,
which not only supply the energy for their gro\\i:h and
reproduction but are also involved in the production of
the intermediate compounds or "building stones " out of
which are synthesised all the complex proteins, fats,
carboh3^drates, pigments, and so on, making up the body
of the organism. As a result of the introduction of means
of measuring the intensity of the oxidising or reducing
power of substances in recent years the study of oxidation-
reduction systems in connection with bacterial metabolism,
gro\^i;h and development has increased rapidly and
afforded considerable knowledge of previously obscure
processes .

Our first ideas of oxidation naturally involve the
addition of oxygen to an atom or a compound, a t}^:)ical

20 BACTERIOLOGICAL CHEMISTRY

oxidation being such a reaction as the combustion of
carbon to yield carbon dioxide,

C + O2 > CO.,

or the combustion of methane to form carbon dioxide
and water,

CH4 + 202 >C0o+2H20.

Besides this type of oxidation by direct addition of
oxygen, there is another in which the proportion of
oxygen in a compound is increased by removal of some
other element such as hydrogen, a typical example being
the oxidation of ethyl alcohol to acetaldehyde,

CH3.CH2OH — > CH3CHO + Ho,

or the conversion of hydroquinone to quinone,

OH 0

I 1.1

+ H.,.

I II -- > II II

\/ \/

I II

OH 0

Even this does not go far enough, however, and we now
recognise that certain reactions in which no oxygen at all
is involved are still of the nature of oxidations. This is
particularly true where such metals as copper or iron,
which have more than one valency, are involved. Thus
we regard trivalent ferric salts as more highly oxidised
than divalent ferrous salts, even though they may contain
no oxygen at all, as is the case with the chlorides. The
conversion of ferrous chloride to ferric chloride is an
oxidation : —

FeCl^ + CI ^=^ FeClg.

Now these reactions, like all chemical reactions, involve
the transfer of electrons from one atom to another. When
liydrogen is oxidised to water the sino^le electron whicli

OXIDATIOX-REDUCTION TOTENTIALS 21

each atom carried is handed over to the oxygen atom to
help it complete its stable octet of electrons. Thus the
oxidation of hydrogen means the loss of electrons to
oxygen. It will be seen that oxidation of ferrous to
ferric chloride also involves loss of an electron from the
iron to a chlorine atom. The same thing applies to all
oxidations : in every case the oxidised atom loses one or
more electrons to some other atom or atoms. It has
become evident that every oxidation (or loss of electrons)
must necessarily be associated with a gain of those
electrons by the other partner in the reactions, that is,
every oxidation is accompanied by an equivalent reduction
(which is a gain in electrons), and conversely every
reduction must have its counterpart in a simultaneous
oxidation. The one reaction cannot occur without the
other.

This transfer of electrons, which is the inevitable
accompaniment of all oxidation - reduction reactions,
affords a means whereby the process may be measured by
electrical means, since a transfer of the charged electron
alters the electrical state of the parts of the system or,
in other words, sets up a potential difference between the
reactants. The magnitude of this potential difference
depends on the ease with which the electrons are lost or
gained, the greater the tendency for a movement of
electrons (that is, the greater oxidising or reducing power
of a substance) the greater will be the potential on one or
other side of zero. The more highly oxidised a substance
is (that is, the more ready it is to take up electrons) the
more positive will be its potential, and the more highly
reduced a substance is (that is, the more ready it is to
part with electrons) the more negative will be the potential.

Here, too, as in all other reversible reactions, the
Law of Mass Action applies, and in general for the
reaction : —

Reductant ;==^ Oxidant + ne

(where " e " represents an electron and " n " the number of

22 BACTERIOLOGICAL CHEMISTRY

them concerned in the particular reaction) the equilibrium
will be expressed by : —

[Oxidant] x [e]'^ _ , r -in _ [Reductant] x h

[Reductant] ~ ''' ^^' ^^^ ~ [Oxidant] * ' ^ '

Obviously the direction in which the reaction will proceed
is influenced by the free electrons ; if their number is
increased the system will tend to produce more of the
reductant ; if they become fewer more oxidant will be
formed. Hence if we know the electronic state of the
system, we have a measure of its reducing or oxidising
power. The electronic state manifests itself in the
electrode potentials set up when non-reacting electrodes
are introduced into the system, and these potentials can
be measured by comparison with standard half- cells.
The electrode potentials depend on the transfer of electrons
from the solution, in which they are present in concentra-
tion [e] , to the electrode which can be regarded as having
a constant concentration of [em]. Now the work W
required to move an electron from concentration [e] to the
metal electrode at concentration [cm] is equal to that
required to transfer a charge of 1 f araday ( F) through the
potential difference E concerned. That is,

W = EF = RT log ^ . . . . (5)

where " log " indicates logarithms to the natural base,
R is the gas constant, and T the absolute temperature.
Rewriting the equation we get

^^ = ^ log [e.J - ^ log [e] ... (6)

But since [cmj is a constant this expression becomes

liT
E ^K-^'-f log [.] .... (7)

OXIDATION-REDUCTION POTENTIALS 23

and substituting the value of [e] from equation (4) above
we get

^ ^^ Rl\ , RT , [Reductant]
^ = ^-^ ^"S ^-^ ^"^ [Oxidant]

„ , RT , [Reductant] ,

^ = ^^-^ l^g [Oxidant] . . • • i8)

where A^i is another constant, smce K ^ log /c is a

constant for any given temperature and reaction. The
electrode potential E can only be measured if it forms
one element of a cell of which the other is a standard
electrode, the hydrogen electrode being used as such in
these cases. The potential referred to the hydrogen
electrode as standard is denoted by Eh, and is given by
Eh =E—lc2 where k^ is the potential of the standard
hydrogen electrode.

„ ^ , RT , [Reductant] ,

Hence Eh = A', =- log ^^ ., — — k^,

^ nF ^ [Oxidant] '^

^ RT . [Reductant] ,^.

^^ = ^--n-^^^g [Oxidant] ' ' ' ^^)

where Eq= ki—k2, which is a constant for the system.

It follows from a consideration of this equation that
the observed oxidation-reduction potential, Eh, depends
on Eo, which is a constant for the particular system under
consideration, and on the ratio of the concentrations of
the reduced and oxidised constituents of the system.
The more reduced substance there is present the lower
will be the Eh value, and the greater the proportion
of oxidised substance the higher will be the potential.
When the concentration of the reductant equals that of
the oxidant, that is, when the system is haK oxidised,
. . . . , „ „ . , . [Reductant]

it IS obvious that Eh = Eo, smce the ratio tq -^^ y.fi "^ ■'■

and its logarithm =0. Thus if the potentials of different

24 BACTERIOLOGICAL CHEMISTRY

systems are compared at half complete oxidation or
reduction they can be arranged in order of their oxidising
or reducing intensities. A system having a certain value
of Eo will oxidise (and be reduced by) all systems having
a negative or less positive potential, and in its turn will
reduce (or be oxidised by) all systems having a more
positive or less negative potential. Alternatively, if we
know the value of Eh and of Eo for any system it is possible
to calculate its state of oxidation or reduction, i.e., the
proportion of oxidised and reduced constituents in it.

Besides the direct electrical method of measuring the
electrode potential use is often made of the simpler but
less accurate method of employing oxidation-reduction
indicators. A number of organic dyes are capable of
existence in the oxidised and reduced conditions, the two
states being characterised by different colours. The
obvious example is methylene blue, which in the oxidised
state has a blue colour, but which in the reduced state,
known as leuco -methylene blue, is colourless. A consider-
able number of other dyes, mostly of the indophenol series,
are known which undergo a colour change on conversion
from the oxidised to the reduced form. These dyes, like
other oxidation-reduction systems, have a characteristic
range of Eh from the oxidised to the reduced state, the
particular Eh value depending on the proportion of
oxidised to reduced dye present. The colour, too, will
depend on the ratio of oxidised and reduced components,
so that an observation of the colour of such a dye will
give information of the degree of oxidation or reduction
which it has undergone, and consequently of the Eh
value of the mixture. The addition of a small quantity
of an appropriate dye to an oxidation-reduction system
will serve as an indicator of the Eh value obtaining in the
system. A range of dyes which cover Eh changes from
about +0-4 volts to —0-01 volts is available, so that by
using a suitable dye an estimate of the oxidising or
reducing intensity of a system can be made very readily.

OXIDATION-REDUCTION POTENTIALS 25

It must be remembered that many organic oxidation-
reduction systems, particularly the indicator dye systems,
involve weak bases or acids, and accordingly their
behaviour and the potentials to which they give rise will
depend on the pK of the solutions in which they are
active. As a result, it is necessary to maintain the pK
value of the solutions constant by the use of buffers and
to record the ^^H value at the time of the measurement
in order that the values of Eh may be of significance.

Another important fact which must be borne in mind
when considering oxidation-reduction potentials is that
Eh is purely a measure of intensity of effect and not of
capacity. It gives information as to whether a given
substance will oxidise or reduce another substance but
not as to how much of the second compound can be
oxidised or reduced. In this respect it is analogous to pH,
which describes the intensity of acidity or alkalinity but
gives no indication of how much alkali or acid may be
required to alter the pH by a given amount. Or, again. Eh
is analogous to temperature, which indicates the intensity
of heat or cold but gives no clue as to how much heat
must be added to or subtracted from a given body in
order to alter its temperature to some other value. We
know that a body with a high temperature v/ill lose heat
to one of lower temperature or vice versa, but from a
knowledge of the temperatures alone we cannot predict
what the final temperature of the pair will be.

The application of oxidation-reduction potentials will
be considered in their appropriate places in connection
with bacterial respiration and metabolism.

For further information : —

W. M. Clark, " The Determination of Hydrogen Ions," Third Edition-
London, 1928.

W. M. Clark, " Recent Studies on Reversible Oxidation - Reduction in
Organic Systems." OJmn. Rev., 2 (1925-26), 127.

L. F. Hewitt, " Oxidation-Reduction Potentials in Bacteriology and
Biochemistry," Third Edition, L.C.C, London, 1935,

CHAPTER III
COLLOIDS AND ADSORPTION

AS in all biological happenings, the behaviour of
colloids and the phenomena of adsorption play a
dominant role in the chemistry of micro-organisms.
Not only is this true in the actual life processes occurring
within the cell and in the action of enzymes isolated from
various bacteria, for instance, and bringing about
reactions outside the cell, but the colloidal nature of the
substances taking part in all the numerous reactions
grouped under the heading immunochemistry is obvious.
It suffices to mention that all antigens are colloidal and
that the antibodies which they provoke are, if not them-
selves serum proteins, always carried in the serum and
associated with proteins which are colloids. For this
reason the properties of colloidal solutions, and particu-
larly those of lyophilic colloids, to which class the proteins
and complex carbohydrates belong, are of the greatest
importance.

Colloid systems are composed of at least two phases :
the disperse phase, consisting of very small particles
ranging in size from 10 to 200 m/x (jLt=one micron=one-
thousandth of a millimetre and m/x or milli-micron,
sometimes erroneously written /x^t, equals one-thousandth
of fjL or one-millionth of a millimetre), which are dis-
tributed through the dispersion medium or continuous
phase. If both continuous and disperse phases are liquids
the system is an emulsion ; if a solid is dispersed through-
out a liquid the resulting system is known as a suspension.
These are the important systems from our point of view,

26

COLLOIDS AND ADSORPTION 27

though others, gas in liquid (foams), liquid in gas (fogs),
solid in gas (smokes) and solid in solid are equally import-
ant in other fields of work. The particles which constitute
the disperse phase may be collections of large numbers
of atoms or molecules, as is the case with gold sols and
cadmium sulphide sols, or they may consist of single very
large molecules or of a comparatively few molecules,
which is usually the case with the proteins and poly-
saccharides (of molecular w^eight of the order of 15,000
to 100,000 or higher) w^hose molecules' are so large that
they fall within the colloidal range of sizes . If the particles
of the disperse phase are much larger than 200 m/x they
tend to settle out comparatively rapidly under the
influence of gravity, whilst if they are much smaller than
10 m/x they cease to behave as colloids and show the
properties of crystalloid solutions.

In a stable colloid system the minute particles are
prevented from cohering and coagulating by two pro-
cesses. In the first place they are constantly bombarded
by the molecules of the continuous phase which keeps
them in the ceaseless zig-zag motion loiown as Brownian
movement, and secondly the particles, as a rule, carry
an electric charge which may be positive or negative
according to the system. Since all the particles in any
one system carry a like charge they tend to repel one
another, and so remain uniformly distributed throughout
the continuous phase. The charge carried by the particles
may be due to the adsorption of ions from the solution,
or it may be due to ionisation of the particles themselves
or to a combination of both factors.

Colloids can be divided into two large classes, the
lyophobic (solvent -hating) sols and the lyophiUc (solvent -
loving) sols. The first group contains those systems in
which the disperse phase has little if any attraction for
the continuous phase. To this group belong most
inorganic sols, like gold and the sulphides, and also
emulsions of oil in water. They show no tendency

28 BACTERIOLOGICAL CHEMISTRY

to gelatinise. The lyophilic colloids comprise systems in
which the disperse phase has a considerable attraction
for the continuous phase. The outstanding examples of
this class are the pix)tein and polysaccharide colloids,
which are characterised by a tendency to gelatinise or
form gels under appropriate conditions.

The lyophobic systems are much less stable than the
lyophilic systems ; in other words, lyophobic sols depend
almost entirely on their surface charge for stability. This
surface charge can be measured by observation of the
direction and speed of migration of the particles when
submitted to a known potential gradient. It has been
found for most lyophobe systems that if the charge or
the surface potential of the particles is greater than about
±15 millivolts the sol is stable. If the charge is reduced
by any means below this critical value the particles tend
to aggregate into larger and larger masses until finally
they settle out completely. One of the easiest ways of
altering the surface charge on the particles is to add an
electrolyte to the sol. The electrolyte will dissociate
into positively and negatively charged ions ; a negatively
charged sol like gold or collodion will adsorb the positively
charged metallic ions with a progressive neutralisation
of the charge on the sol particles . If sufficient electroljrte
is added to reduce the surface charge below the critical
value the sol will coagulate. Since the ions of monovalent
metals, like sodium, carry a less charge than those of
divalent metals, like calcium, and these less than the
ions of trivalent metals, such as lanthanum, it is necessary
to add more of a sodium salt than of a calcium salt and
more of a calcium salt than of a lanthanum salt to have
the same coagulating effect on a given sol. For example,
a certain collodion sol was found to be equally effectively
coagulated by N/2-NaCl, by N/lG-CaCla and by N/680-
LaClg. In the case of positively charged sols, like ferric
hydroxide, it is the negatively charged anion which is
adsorbed on to the sol particles and is important in

COLLOIDS AND ADSORPTION

29

bringing about coagulation by the same mechanism of
lowering the surface charge below the critical value.
Again the same valency rule applies, the higher the
valency of the ion the greater is its effect, except in the
cases of hydrogen and hydroxyl ions which for some
reason are much more effective than other ions.

Similar coagulation can be brought about by adding
a positively charged colloid to a negatively charged one.
When the positively charged colloid is present in sufficient
quantity to lower the surface charge on the particles of

CONCENTRATION OF ADDED ELECTROLYTE >-

Fig. 4

the other below the critical value coagulation occurs.
If addition of the positively charged colloid is continued
a time will come when the charge on the particles becomes
greater than the critical value on the positive side, and
the system will no longer coagulate but will remain
dispersed ; that is coagTilation will only occur within a
zone of concentration of the added colloid, the limits of
the zone being between the concentrations of added
colloid necessary to keep the surface charge between the
positive and negative critical values (see Fig. 4). The

30 BACTERIOLOGICAL CHEMISTRY

same effect is also shown when electrolytes are added
to colloids, the coagulation only taking place within a
zone of concentration, the critical value being overshot
by addition of an excess of the electrolyte with conse-
quent adsorption of the oppositely charged ion.

Lyophilic colloids are much less sensitive to the action
of electrolytes than are lyophobic sols. This is due to
the fact that the particles of a lyophilic colloid have a
strong attraction for the continuous phase, usually water
in systems of bacteriological interest, in which they
are suspended. As a result of this attraction the particles
of the colloid become covered by a layer of water which
acts as a buffer between them and hinders their collision
and consequent aggregation even when the surface charge
is well below the critical value for a lyophobe system.
In other words, lyojDhilic colloids have a lower critical
surface potential than the lyophobes. It varies from ±2
to ±8 millivolts, according to the particular colloid
concerned. Accordingly, considerably more electrolyte
(or oppositely charged colloid) is needed to alter
the charge on a lyophilic sol sufficiently to cause
coagulation.

Polar Groups. — The lyophilic character of proteins and
polysaccharides and similar substances is due to the
presence of polar groups. As is well laiown, the com-
bination of atoms to form molecules is brought about
through the agency of the valency electrons. In the
case of such ionisable compounds as sodium chloride the
bond is effected by the complete transfer of an electron
from the sodium atom, which has one more electron
than its stable octet, to the chlorine atom, which has
one electron less than its stable octet, whereby both
atoms attain the stable octet structure. But in the
process the sodium atom acquires a positive charge by
the loss of the negative electron, whilst the chlorine
atom becomes negatively charged by the gain of the

COLLOIDS AND ADSORPTION 31

same electron, and the two ionised atoms are held
together in sodium chloride by their opposite charges.
In the case of non-ionised compounds, such as the majority
of organic compounds, the bond is formed by the sharing
of a pair of electrons between the atoms, one electron
of each pair being supplied by each atom to give a co-
valent bond : —

H

•C- + 4 -H > H:C:H.

By this sharing of electrons the carbon atom acquires,
in effect, the stable octet structure and the hydrogen
atoms have two electrons each, as in the inert gas,
helium. A double bond is formed by the sharing of two
pairs of electrons and a triple bond by the sharing of three
pairs. Since there is no actual transfer of electrons, the
molecules are neutral and uncharged ; but one of the
atoms may have a stronger pull on the electron pair than
its neighbour, and accordingly the electrons will be
displaced to some extent from the equilibrium position
so that the atom to which they are more strongly attracted
will have a relatively greater negative charge than the
atom from which they tend to be pulled away. As a
result, a group of atoms in w^hich this occurs wdll act as if
it were a minute magnet with two poles, e.g.

+ 0:

Such groups are known as polar groups. The most
commonly occurring of such polar groups are those which
involve oxygen, especially doubly linked oxygen, nitrogen
or halogens. The polar strength of the groups varies
considerably from group to group, but is approximately
constant for any one group. As examples of such groups,
in order of their strength, may be mentioned

32 BACTERIOLOGICAL CHEMISTRY

Amide — cf 13,200

\NH2

Peptide

Carboxyl

YQH

— C^— OH
\H

10,600
9,000

Alcohol

7.250

Aldehyde

\H

4,700

Ketone

\H

4,300

Amine

3,500

-Y"

\H

Methyl

1,800

The nitro group, — NO2, is also polar, and is of approxi-
mately the same strength as the hydroxyl group as it
occurs in alcohols. The numerical values given, which
indicate the comparative strength of the groups, represent
the energy (in calories per gram molecule) required
to separate the molecules from one another and are
calculated from the heat of evaporation of compounds
containing them. The more strongly polar the group
the greater is the attraction between molecules con-
taining the group, and in consequence the greater the
energy required to separate the molecules from one
another on conversion from the liquid to the vapour
state.

COLLOIDS AND ADSORPTION 33

Water is a polar molecule, the oxygen atom being
negatively and the hydrogen atoms positively charged.
When a compound containing a polar group, such as a
carboxyl group, is introduced into it, the polar group of
the water will be attracted to that of the compound and
water will associate itself with the compound : —

oJ

/ % carboxyl
+ H 0-

-0 H +

I
+ H water

If the polar group of the compound is strong compared
with the rest of the molecule (the non-polar part) the
substance will be soluble in water, as is acetic acid ; but
if the non-polar " tail " is long the polar " heads " will
be attracted to and held in the water, leaving the tail
projecting out from the surface and making a film
arranged in an orderly manner on the surface of the
water. If a compound has only very weak polar groups,
or no polar groups at all, as is the case with the paraffins,
it will be insoluble in water, and its molecules will lie
higgledy-piggledy on the surface without forming an
orderly film .

The proteins which make up bacterial protoplasm,
or which form the colloidal carriers of enzymes, or which
occur in serum and take part in the various immunological
reactions, possess considerable numbers of strong polar
groups, principally the peptide linkages, — CO.NH — ,
the carboxyl group, — COOH, and the amino groups,
— NHg. In virtue of these strong polar groups they
have a strong affinity for water, and when they are in
solution they are surrounded by films of adsorbed water,
which lowers their critical surface potential and so renders
them much more stable and less liable to be coagulated
than the lyophobic colloids. The polysaccharides are

34 BACTERIOLOGICAL CHEMISTRY

lyopliilic because they contain a high proportion of polar
hydroxyl groups, — OH, and oxygen links, — 0 — .

Adsorption. — It is the attractive forces between these
polar groups in neighbouring molecules which are largely
responsible for holding groups of molecules together, as,
for example, in protein or cellulose fibres. They also
account in very large measure for adsorption phenomena,
and particularly for specific adsorption. Large molecules,
like those of proteins, will have definite distribution
" patterns " of polar groups according to the amino -
acids of which they are built up, and depending on the
order in which the amino -acids are arranged in the
molecule. Other molecules which may have a distribution
of polar groups giving a " pattern " corresponding to that
on the protein molecule will be adsorbed ; if the polar
patterns do not correspond there will be less and less
adsorption or weaker and weaker adsorption as the pat-
terns differ more and more from one another. This offers
a physical explanation for the well-known " lock and
key " simile which Emil Fischer suggested to explain
enzyme specificity and which Ehrlich used of antibody
specificity.

Now if the polar groups on a protein are in close
proximity to those on the second adsorbed molecule,
which may be represented diagrammatically thus : —
+ ^ + + ^) the charges due to the polar groups will
neutralise one another, to a certain extent at least, if
not completely, and the complex of the two molecules
will no longer be polar in effect, and as a result will no
longer have such an attraction for water molecules. If
enough of the polar groups are masked in this way the
system will more and more lose its lyophilic character
and more and more come to resemble a lyophobic colloid.
In this state it will be readily aggregated by metallic
ions. If the protein is an antibody and the adsorbed
molecule an antigen we get a picture of the sort of thing

COLLOIDS AND ADSORPTION 35

that happens in serological reactions such as agglutination
or precipitation.

The strength of these polar forces falls off rapidly with
the distance from the polar groups, so that they are
effectively exerted only if the groups come into close
proximity. Unless the distribution of the groups in
different molecules corresponds closely, then, there will
be little tendency to adsorption. The application of this
idea to account for the sharp specificity of man}^ enzyme
reactions and of the serological reactions will become
obvious when these subjects are developed (Chapters IV
and XXIII).

For further reading : —

N. K. Adam, "The Physics and Chemistry of Surfaces." 3rd Edition,
Oxford University Press. London, 1941.

E. r. Burton, " The Physical Properties of Colloidal Solutions." Mono-
graphs on Physics. Longmans, Green & Co. London, 1916.

H. Freundlich, " Colloid and Capillary Chemistry." Methuen & Co.
London, 1926.

E. Hatschek, " An Introduction to the Physics and Chemistry of Colloids."
Churchill. London, 1916.

CHAPTER IV
ENZYMES

IT is becoming more and more recognised that enzymes
play a predominant part in the life processes of all
organisms, large and small. The results of innumerable
investigations all go to show that enzymes are concerned
in and control the metabolic and respiratory reactions of
all living things . The supply of foodstuffs in appropriate
form, the supply of energy required for their utilisation,
and the synthesis of simple units into the complex com-
pounds characteristic of living entities all fall within the
province of enzyme action. It is to the action of enzymes
that living cells and tissues owe their ability to perform
at low temperatures and with the mildest of reagents
a vast number of complicated reactions which so
far are beyond the powers of the organic chemist
with all the resources of the modern laboratory at his
disposal.

In this chapter we shall consider some of the general
properties of enzymes before studying the particular
effects of certain of them when we come to deal with the
respiratory and metabolic activities which they initiate
and maintain.

Such fermentations as the production of alcohol from
sugar solution, the production of vinegar from alcohol
and the formation of lactic acid from lactose in the sour-
ing of milk, and such putrefactive processes as the break-
down of plant and animal materials are among the oldest
reactions recognised and used by mankind. A long and

36

ENZYMES 37

bitter controversy raged during the late eighteenth and
the nineteenth centuries as to whether or not such fer-
mentations were due to living entities. Berzelius in 1837
put forward the suggestion that the processes were due
to catalysts, whilst Liebig considered that putrefaction
and similar processes were due to vibrations set up by the
disintegration of living cells. It fell to Pasteur, about
1870-75, to show that all these processes were associated
with the vital processes of minute living organisms,
yeasts, bacteria and fungi, and that if the life of the
micro-organism was destroyed (by heating, for instance)
then the fermentations were brought to a standstill.
Pasteur described the responsible micro-organisms as
" organised ferments."

Parallel with this controversy various digestion and
breakdown reactions by plant and animal juices and
extracts were being described. Substances were isolated
from such extracts which could bring about the same
reactions in the test-tube. For instance, Plane he in 1810
had observed that solutions of guaiacum were turned
blue by extracts of certain roots (this is perhaps the first
recorded isolation of an enzyme) ; amygdalin had been
shown to be hydrolysed by an extract of bitter almonds
from which Liebig and Wohler prepared the enzyme
emulsin ; Payen and Persoz showed in 1832 that starch
was hydrolysed by the enzjrme diastase, which they
obtained by precipitation of barley malt extracts with
alcohol ; the protein degrading preparations pepsin and
trypsin had been obtained from the gastric juice and
from the pancreatic juice respectively. The behaviour
of these and other substances like them was recognised
as being similar in many respects (such as ready destruc-
tion by heat) to the action of the " organised ferments,"
and they came to be known as " soluble " or " unorganised
ferments." Then followed a long discussion as to whether
there was any essential difference between the organised

38 BACTERIOLOGICAL CHEMISTRY

and the unorganised ferments. The discussion was
virtually brought to an end when, in 1897, Buchner ground
up yeast with sand, submitted the mixture to high pres-
sure, and obtained a non-living cell-free juice which was
capable of converting glucose into alcohol and carbon
dioxide in just the same way as the living yeast with
which he started. It is as a result of this investigation
that the term enzyme (Greek: en=m; zyme=yeBBt)
has come into being as the general term for all such
substances. Since that time many other similar active
preparations have been obtained from diverse biological
systems, usually by the maceration of the organism or
tissue in water and precipitation of the enzyme as an
amorphous powder by addition of alcohol or acetone. As
a result of this and later work it has become an accepted
fact that although enzymes are only produced by the
living cell, once they have been so produced the cell is no
longer necessary for their action, which can occur quite
independently of the life of the cell that brought them
into being. Both Pasteur, who maintained that
fermentation only occurred if a living organism was
concerned, and Liebig, who argued that fermentation
could occur in the absence of life, were correct up to a
point, but neither went far enough to complete the
story. Life is necessary to bring about the formation
of the enzyme, but the enzyme may remain active
after the death of its parent cell and still cause fermen-
tation.

Enzymes as Catalysts. — Enzymes can be regarded as
biochemical, organic catalysts which are produced by
living organisms. All living cells contain enzymes of one
sort or another, often a large variety of them. A catalyst
is a substance which changes the rate of a chemical
reaction, usually, although not always, accelerating it ;
in some cases a catalyst may act by removing some
inhibiting factor, thus enabling a reaction to proceed

ENZYMES 39

which otherwise would not do so. The catalyst will, as
a rule, influence a reaction between amounts of reagents
many thousand times its own weight, and it can generally
be recovered unchanged in quantity and constitution at
the end of the reaction.

Enzymes are not only catalytic in their action but
are often much more active than inorganic catalysts
which bring about the same reaction. The enzyme
lactase, for example, hydrolyses the disaccharide, lactose,
to glucose and galactose many hundreds of times more
rapidly than does twice normal hydrochloric acid where
the hydrogen ions act as the catalyst. Inorganic catalysts,
such as acid in the instance just described or in the hydro -
l^^sis of esters,

CHgCCOC^Hs + H2O ^==^ CH3COOH + C0H5OH,

do not alter the equilibrium point of the reactions which
they catalyse but only shorten the time required to
attain the equilibrium condition. That is, the catalyst
accelerates both directions of a reversible reaction to an
equal extent. Likewise the amount of catalyst has no
influence on the final equilibrium quantities of the
reagents, but the velocity with which the equilibrium
state is reached is proportional to the amount of catalyst.
Under strictly controlled conditions this is also generally
true of enzymes, although in certain cases enzjnnes may
alter the equilibrium value as well as the velocity of a
reaction. This is in all probability due to their colloidal
nature and a complete irreversible adsorption of a part
of the reactants, with consequent alteration of the active
concentrations on which the equilibrium depends. During
a reaction enzymes often become partially destroyed or
lose some of their activity as a result of side reactions,
with a consequent slowing down of the velocity of the
reaction, although the final equilibrium reached is un-
altered.

40 BACTERIOLOGICAL CHEMISTRY

Since catalysts and enzymes do not, in general, affect
the equilibrium of a reversible reaction they should be
able to bring about the formation of, say, an ester from
the constituent acid and alcohol as well as influence the
hydrolysis of the ester. This is found in practice to be
the case. For instance, lipase hydrolyses esters, such
as ethyl butyrate, with production of the alcohol and
the acid, but if it is allowed to react with a mixture
of ethyl alcohol and butyric acid it will catalyse the
production of the ester. Similarly the disaccharides
maltose and cellobiose have been obtained by the action
of maltase and emulsin respectively on glucose solutions ;
but the yields of disaccharide are very low, as the equi-
librium state is far over on the side of hydrolysis. Poly-
peptides have also been built up by the action of pepsin
on mixtures of peptides.

The Chemical Nature of Enzymes. — The chemical
nature of most enzymes is still a mystery. They are all
regarded as proteins, or as being protein like, although
for a time there was doubt about some of them, such as
invertase, peroxidase and lipase, which were very highly
purified by Willstatter and his colleagues and then failed
to exhibit the biuret, Millon and ninhydrin reactions
typical of proteins, although possessing 500 to 20,000
times the activity of the original crude preparations.
This failure is very probably due to the fact that the
enzyme solutions contained so little of the very highly
active enzyme that positive chemical reactions could not
be elicited. It may be noted that, generally, these highly
purified enzymes are considerably less stable than cruder
preparations, in which the impurities seem to have a
protective effect. All the crystalline enzymes which
have been prepared, listed in Table 2, are either protein
in character or contain a protein fraction combined with
a prosthetic group.

ENZYMES

Table 2

41

Enzyme

Crystallised By

Urease

Smnner

Pepsin

Northrop

Trypsin

Kmiitz and Northrop

Chyniotrypsin

Kimitz and Northrop

Carboxypeptidase

Anson

Ficin

Watti

Papain

Balls, Lineweaver and Thompson

Ribonuelease

Kunitz

Acetaldehyde reductase

Negelein and Wulff

Catalase

Sumner and Dounce

Amylase

Caldwell. Booker and Sherman

Lysozj-me

Abraham and Robinson

" Yellow Enzyme "

Warburg; and Theorell

Peroxidase

Theorell

It has been held by many workers that certain metals,
notably iron and copper, are essential constituents of
enzymes ; some enzymes, such as catalase, peroxidase,
tyrosinase and ascorbic dehydrogenase, certainly contain
these metals, but very active ^^reparations of others have
been obtained which are quite free from them.

Catalase can be split, by treatment with dilute acid,
to give two inactive fragments, a colloidal protein carrier
and the prosthetic group on which the activity of the
intact enzyme depends. The prosthetic group is identical
with the hsem of haemoglobin, since if it is coupled with
globin from the animal from which the enzyme was
derived it yields the haemoglobin characteristic of that
species.

Tyrosinase and ascorbic dehydrogenase contain copper
in the form of haemocyanin, analogous to the iron
porphyrin compound haemoglobin.

The flavo -protein enzymes such as Warburg's " yellow
respiratory enzyme," which occurs in bottom yeast and
in Lactobacillus delhruchii for instance, (/-amino-oxidase,
occurring in liver or kidney, and diaphorase present in

42

BACTERIOLOGICAL CHEMISTRY

animal tissues and micro-organisms, can also be split by
dilute acid to give a specific protein carrier and either
riboflavin-5-phosphoric acid from Warburg's enzyme

/OH

CH, o.p4o

\0H

HOCH

HOCH

I

HOCH

I

CH^

N N

CH.

CH.

CO

N CO
or f la vine adenine dinucleotide,
OH OH
CRoo P-0- P -0-CH2(CHOH)3-CH2-N-

HOCH O
HOCH
HOCH
CH,

0

CH

N

CH CH

I
NH,.C

N N
CH3/\/\^\

CO
NH

N CO

from the others as prosthetic groups,
sliould l)e remembered, is vitamin Bg
factor for many organisms (see p. 111),
enzyme can be reconstituted by allowing synthetic

Riboflavin, it
and a growth
The " yellow "

ENZYMES 43

riboflavin-5-phosplioric acid (or lactoflaviii-5-pliosphoric
acid as it was originally called because of its isolation
from milk) to react in neutral solution with the carrier.
The enzyme carboxylase, occurring in yeast, and
causing the breakdown of pyruvic acid in alcoholic
fermentation (see p. 277) appears to be a complex of
aneurin diphosphate, magnesium and a protein carrier.
Aneurin diphosphate,

N==C.XHo CI I ' /^^ /^^

I I ■* I /(J = C-CH2.CPI2.O-P-O-P-OH

CH3.0 c — CH2 — n; I ii II

II II \CH-S 0 0
2^ CH

also occurs as the prosthetic group in pyruvic oxidases
occurring in Lactobacillus delhrilckii. Streptococcus
hae7nolyticus and gonococci and possibly in the acetic
acid bacteria.

So far prosthetic groups have only been detected in
the endo -enzymes concerned in respiration and not in
the hydrolytic enzymes.

The mechanism by which these enzymes bring about
their specific activities will be discussed in connection
with bacterial respiration in Chapter XII.

To sum up, we may say that although we loiow a
little about the chemical nature of a very few of the multi-
tudinous enz^Tues, of the vast majority we laiow nothing
beyond the effects they have and the conditions under
which those effects are brought about.

Physical Properties of Enzymes. — Enzymes, in general,
are soluble in water and in dilute alcohol, but are pre-
cipitated from solution by ammonium sulphate or by
high concentrations of alcohol or of acetone. Chemically
and physically they are very unstable substances, one or
two being so sensitive that even mechanical shaking is
sufficient to destroy their activity.

44 BACTERIOLOGICAL CHEMISTRY

All enzymes appear to be colloidal when in solution
in so far as they are unable to diffuse through semi-
permeable membranes, and in showing the T3mdall effect
when a beam of light is passed through such a solution.
Like the proteins, which in fact many enzymes may be,
most of them are amphoteric in nature, that is, they
may behave either as weak acids or weak bases depending
on the acidity or alkalinity of the medium in which they
are dissolved. As a result of their colloidal and amphoteric
character enzymes are usually active adsorbing agents,
and also display their maximum activity at an optimum
2?H value.

Separation of Enzymes. — The majority of enzymes
do not diffuse out of intact cells into the surrounding
fluid but are held within the cell structure probably by
adsorption to various cell constituents. Hence, in order
that they may be isolated, the cell system has to be broken
down by mild means in order not to destroy the enzyme
at the same time. This may be effected by a mechanical
process, such as grinding with sand, as did Buchner when
he obtained zjanase from yeast cells, or by such chemical
action as the use of weak alkali or acid ; or the cells may
be disintegrated by using solvents like ether, chloroform,
toluene, or acetone, but these also remove fatty con-
stituents. The treated cells may then be extracted with
water, salt solutions, dilute acids or alkalies, dilute alcohol,
glycerol, or some similar agent, depending on circum-
stances. Having obtained a crude enzyme solution in
this way, it may be purified in a variety of ways, the
appropriate method depending on the particular enzjnne in
question. Thus salts, acids and alkalies may be removed
by dialysis. The enzjrme may be precipitated from
solution, usually along with considerable quantities of
inactive protein, b}^ alcohol, acetone, ammonium sulphate,
or other protein precipitants . A method which has been
particularly valuable in the separation and purification
of enzymes, especially in the hands of Willstiitter and his

ENZYMES 45

school, is that of selective adsorption. It has been found
that enzymes, in virtue of their colloidal and amphoteric
nature, are readily adsorbed by such materials as kaolin,
kieselguhr, charcoal and alumina. By carrying out
these adsorptions, using appropriate adsorbants and
appropriate conditions of acidity or alkalinity, it has been
found possible to adsorb one enzyme and leave others in
solution, and then, by altering the pH. value of the solu-
tion in which the adsorbed complex is suspended, to
wash out or " elute " the enzyme again and so obtain it
free from other enzymes or inactive accompanying sub-
stances. Kaolin is negatively charged and will adsorb
positively charged basic substances, whilst alumina is
positively charged and adsorbs negatively charged acidic
substances. It is obvious that by altering the 2^H of
the solution the charge on the adsorbant, the enzjrme,
and the complex of the two can be altered and the adsorp-
tion or elution of the enzyme controlled. Thus the enzyme
peroxidase, which has basic properties, is adsorbed on
kaolin from dilute acid solution, and can then be eluted
from the adsorbate (the complex of adsorbant and enzyme)
by dilute ammonia. Invertase which, together with
maltase, is adsorbed on alumina from acid solution can
be selectively eluted by a solution of acid phosphate
which does not remove the maltase. Or invertase can
be adsorbed on kaolin in acid solution and eluted with
dilute sodium hydroxide solution.

The methods adopted for the crystallisation of enzymes
usually involve the use of fairly concentrated enzyme
in an appropriately buffered solution of a salt at fairly
low temperatures. For example, pepsin crystallises when
an alkaline solution is brought to 2^H 3 with sulphuric
acid. Pepsinogen can be crystallised, after preliminary
purification, from 0-4 saturated ammonium sulphate
solution at ^^H 6-5. After fractionation of a pancreatic
extract by ammonium sulphate, chymotrjrpsinogen can
be obtained as long needles by crystallisation at pK 5

46 BACTERIOLOGICAL CHEMISTRY

from 0-25 saturated ammonium sulphate ; chymotrypsin
crystallises from 0-01 N sulphuric acid, whilst trypsin
crystallises from solution in 0-5 saturated magnesium
sulphate solution in borate buffer at ^jH 9.

The course of purification procedures can only be
followed by measurement of the activity of samples
of the enzyme at various stages, since no chemical methods
of estimating the enzyme as such are available. The
enzyme activity is expressed in terms of enzyme units,
which define the amount of change which the enzyme
can bring about under standardised conditions of con-
centration of substance acted on, ^^H, temperature, con-
centration of activators, salt concentration and amount
of change. As an example may be quoted Willstatter's
invertase unit, which is the amount of enzyme which will
reduce the rotation (after addition of alkali to bring
about mutarotation of a-glucose) of 4-0 g. of sucrose in
25 ml. of a 1 per cent, solution of NaH2P04 at 15-5° C.
to 0° in one minute. The activity of an enzyme prepara-
tion is usually quoted as the number of units per gram
of dry weight. It is clear that the unit and the necessary
set of conditions will differ for each enzyme, but must
be rigidly adhered to for any one enzyme.

The Effect of Conditions on Enzyme Action. — (a)

Concentration of Enzyme. — In general the rate of reaction
is proportional to the enzyme concentration as long as
other conditions such as piL are maintained constant by
the use of buffer solutions. In cases where deviations
from this rule have been observed they have usually been
traced to the gradual destruction of the enzyme, or the
production of inhibitors or other substances which com-
bine irreversibly with the enzyme and so alter its effective
concentration. It is to be emphasized that the equilibrium
condition of the reaction is not altered by the presence
or by the amount of the enzyme unless the products of
the reaction are involved in side reactions resulting in a

ENZYMES 47

change in their effective concentration. It is only the
velocity of the reaction which is altered.

(b) Substrate Concentration. — The term " substrate "
is applied to the substance on which the enzyme exercises
its cataljrtic properties . Invertase catalyses the conversion
of the substrate sucrose into glucose and fructose.
Hydrogen peroxide is the substrate which under the
action of the enzyme catalase brealis down into water
and oxygen.

With low concentrations of substrate the reaction
velocity is in many cases proportional to the substrate
concentration for a given concentration of enzyme, but
at higher concentrations the rate of reaction rises less
rapidly than the concentration. This is very probably
due to the saturation of the enzyme surface by adsorption
of the substrate to form the hypothetical intermediate
compounds which have been postulated in most theories
of enzyme action. In some cases, too, the enzyme adsorbs
a part of the products of the reaction with a consequent
slowing of the rate of reaction. It seems to be a general
rule that the oxidising and reducing enzymes are saturated
by the substrates at considerably lower concentrations
than are the hydro lytic enzymes.

(c) Heat. — Rise of temperature at first increases the
rate of reaction of enzjrmes in the ordinary way common
to all chemical reactions ; but at comparatively low
temperatures an optimum is reached, and then the activity
faUs off with further increase of temperature until at
about 70° C. the action of most enzymes is stopped,
whilst 100° C. is sufficient to inhibit all known enzyme
action. It is seen that the effect of temperature is a result
of the competition between the acceleration of chemical
reactions by rise of temperature and the gradual destruc-
tion of the enzyme at higher temperatures . The optimum
temperature varies with the particular enzyme concerned,
but the majority of enzymes have temperature optima
falling between 35° and 45° C. Freezing has no permanent

48 BACTERIOLOGICAL CHEMISTRY

effect on most enzymes, their activity being merely
temporarily inhibited or greatly reduced at low tempera-
tures with recovery to the normal rate on warming up
again.

{d) pH. — The acid or alkaline reaction of the medium
in which an enzyme operates has a profound effect on its
activity. There is an optimum pK value for each enzjrme,
and any considerable departure from that value results
in inactivation of the enzyme. Thus pepsin, the proteolytic
enzyme of the stomach, is only active in acid solution
with an optimum at ^H 1-4, whilst the pancreatic enzyme,
trypsin, will only hydrolyse proteins in alkaline solution
with an optimum between pK 8-2 and 8-7. The salivary
amylase hydrolyses starch in slightly acid conditions, its
optimum being at ^^H 6-7 ; it is completely inhibited by
the acid conditions of the stomach in which pepsin is
most active. The variation of the activity of proteolytic
enzymes with change in ^H value in many cases appears
to run parallel with the calculated dissociation curve of the
protein substrate if it is assumed, for instance, that pepsin
reacts with the acid cation, and that trypsin reacts with
the basic anion of the protein. Papain, with an optimum
at ^H 7-0, appears to attack the undissociated protein
molecule. It is for this reason that the pH for optimum
activity of an enzyme varies somewhat from substrate to
substrate, the optimum corresponding with the 2^11
required for maximum ionisation in the right direction.
In these cases the active part of the enzyme appears to
be the un-ionised part.

(e) Activators. — Often an inorganic or organic activator
is necessary before an enzyme can bring about its normal
effect. Thus papain, a proteolytic enzyme found in
melon seeds, must be activated by hydrocyanic acid or
hydrogen sulphide before it will attack peptones, although
more complex proteins are hydrolysed by it in the absence
of the activator. The animal amylase, ptyalin of saliva,
will only hydrolyse starch when chlorine ions are present ;

ENZYMES 49

if all the mineral constituents of the enzyme preparation
and of the starch are removed by dialysis no hydrolysis
will occur when the solutions are mixed, but immediately
a certain amount of sodium chloride is added to the
mixture breakdo^\Ti of the starch commences. Many
oxidation enzymes, or oxidases, require manganese ions
as activator. Trypsin, as it is obtained from the pancreas,
will only hydrolyse the partially degraded proteins,
protamines and peptones ; in order that it may attack
the complex parent proteins it must be activated by
another enzyme, entero kinase, which is to be found in the
intestinal juice . Enterokinase probably owes its activating
effect to the conversion of the pro-enzjmie, tripsinogen,
into trypsin. A number of activators, for example, the
heat stable co -enzyme in yeast juice (about which we
shall say more when we deal with alcoholic fermentation)
function by forming an essential link in a chain of re-
actions (see p. 267).

(/) Inhibition. — Many heavy metals have the power
of inhibiting enzyme activity. Thus mercury salts
" paralyse " the hydrolysis of sucrose by invertase, and
barium salts inhibit the breakdown of urea by urease.
This action is probably the result of the adsorption of
the metal on to the enzyme with a consequent blocldng
of the adsorption of the substrate. Removal of the
metal by dialysis, for instance, or by appropriate chemical
means restores the activity of the enzyme, no permanent
harm having been done to it. Anaesthetics like chloro-
form and urethane inhibit dehydrogenase activity, whilst
cyanides, carbon monoxide and sulphides inhibit oxidases,
a fact which has had an important bearing in the sorting
out of respiratory mechanisms. Certain organic bases,
such as amines, will inhibit the action of invertase, but
their effect can be annulled by the action of aldehydes .

It is possible that a closer study of the inhibiting
action of compounds or chemical groups on different
types of enz^Tnes may throw considerable light on the

50 BACTERIOLOGICAL CHEMISTRY

nature of the groups in enzymes Avliich are responsible
for their activity.

Some substances of known or unloiown constitution
may act as specific inhibitors of certain enzymes, and
such substances have been called anti -enzymes. Thus
normal serum contains an anti-trypsin which prevents its
activity ; this anti-trypsin is possibly a polypeptide
which combines with trjrpsin to the exclusion of its
normal substrate. The intestinal wall contains an anti-
pepsin by which pepsin is prevented from digesting the
tissue proteins. Heparin, a preparation obtained from
liver and used to prevent the clotting of blood, is an anti-
prothrombin which hinders the clotting of blood in the
veins by preventing the interaction of prothrombin and
calcium ions to form thrombin. Anti-enzymes in the
immunological sense are also known ; for instance, if the
enzyme amylase is injected into rabbits it produces an
anti-amylase Avhich specifically inhibits the action of the
enzyme. If malt-amlyase is injected the anti-enzyme
inhibits only malt amylase and not the salivary or pan-
creatic amylases. Similar antibodies specific for urease
and ribonuclease have also been prepared. An interesting
example is afforded by the a-toxin of Clostridium welchii
which has been shown to be essentially a lecithinase
whose action is specifically inhibited by CI. welchii
antitoxin.

The Specificity of Enzymes. — Enzymes differ from the
majority of inorganic catalysts in being highly specific
in their action, and this is particularly true of the hydro -
lytic enzymes. Enzymes which hydrolyse proteins will
not have any effect on fats or on carbohydrates, nor will
carbohyrdate splitting enzymes hydrolyse proteins or
fats. The specificity goes even deeper than this ; maltase,
for instance, will hydrolyse only those sugars which
have the same type of linkage betAveen the glucose
molecules as occurs in maltose, that is, it will only attack
the a-glucose bond. Emulsin, on the other hand, will

ENZYMES 51

only hydrolyse sugars like cellobiose which have p-
linkages, or glucosid.es like amygdalin which also have
p-glucose bonds. Invertase .attaclvs only sucrose or
the trisaccharides, raffinose or gentianose, which contain
the same glucose-fructose unit as sucrose. It has not the
slightest effect on maltose, cellobiose or lactose. The
acids (inorganic catalysts), on the other hand, attack all
these sugars at approximately the same rate, and, more-
over, they also catalyse other hydrolyses such as the
breakdown of protein or of esters and fats with equal
facility. The proteolytic enzymes besides being specific
as a class also show a certain amount of " internal "
specificity, although this is not so sharp. Pepsin and
trypsin, for instance, are capable of hydrolysing a whole
series of proteins, but the peptidases which attack poly-
and lower peptides are much more specific in their action,
generally speaking only hydrolysing compounds which
have common structures or arrangements of amino-acids.
The specificity is often sharp enough to distinguish
between optical isomers, one isomer (usually the naturally
occurring one) being attacked, while the other is not
attacked at all or only very slowly.

The lipases, or fat -splitting enzymes, while completely
specific as a group, that is, capable of hydrolysing only
fats and esters, show a relatively low degree of " internal "
specificity. Thus any lipase will hydrolyse almost any
fat or ester, but there is a certain amount of relative
specificity. For example, liver lipase hydrolyses esters
readily but fats only slowl}^ whilst the pancreatic lipase
behaves conversely, hydrolysing fats readily and esters
slowly.

Enzymes which bring about other types of reaction
than hydrolysis are, in general, less specific than the
hydrolytic enzymes. They usually catalyse the same
type of reaction, oxidation, dehydrogenation, and so on,
for a whole range of substrates, which only need to have

52 BACTERIOLOGICAL CHEMISTRY

in common the possibility of undergoing the change which
the particular enzjrme effects.

The Classification of Enzymes. — Since, for the most
part, enzymes are not well-defined chemical entities
they cannot be named as compounds in accordance with
the normal chemical usage, but they are named by what
they do, their specificity being made the basis of the usual
classification and nomenclature. An enzyme is normally
named by affixing the syllable " -ase " to the root of the
name of the substrate on which it acts or to the type of
reaction which it catalyses. For example, the enzyme
which hydrolyses invert sugar is called invertase ; enzymes
which break up esters are known as esterases, those acting
on proteins are proteases, those on carbohydrates are
carbohydrases. Enzymes influencing oxidation and re-
duction reactions are called oxidases and reductases
respectively. Some names, such as pepsin and trypsin,
given to enzymes in the past and which have become
generally accepted are still retained, although they do
not conform to the general system.

For a number of enzymes, particularly those involved
in respiratory processes, the nature of the prosthetic
group is known, and they are sometimes classified on this
basis. Thus catalase and peroxidases are porphyrin-
protein enzymes because their prosthetic groups contain
iron porphyrin complexes. The pjn^idino -protein enzymes
are those which involve the di- or tri-phosphopyridine
nucleotides, co -enzymes I and II, attached to specific
protein carriers. The flavoprotein enzymes, in which
the prosthetic group is riboflavin (see p. 42) are con-
cerned with the oxidation and reduction cycles of the
CO -enzymes I and II. The copper-protein enzymes,
such as tyrosinase and ascorbic oxidase, contain haemocy-
anin. Carboxylase, which carries aneurin as its prosthetic
group, is a thiamino -protein enzyme.

There are two large groups of enzymes important in
the chemistry of micro-organisms. The Hydrolases

ENZYMES 53

comprise all those enzymes which bring about hydrolytic
reactions of various sorts. They are further divided into
the carbohydrases, the proteases, the lipases (attacking
fats), esterases, amidases, and so on. Their main function
is the breaking down of complex food materials, proteins,
polysaccharides and fats, into simpler units readily
utilisable by the organism for its nutrition. Generally
speaking, their action involves only very small energy
changes. Since their action is on more or less non-
diffusible substrates they would be virtually useless if
they were retained \vithin the cell, so Nature has decreed
that the hydrolases as a class shall be secreted into the
medium outside the cell ; they are extra-cellular or
Exo -enzymes.

The other large group of enzymes comprises those
involved in the processes of respiration and metabolism.
They are kno^Ti as Desmolases. Most of the reactions
with which they are concerned involve considerable
energy changes and, in fact, it is their function to supply
the energy requirements of the cell. To this group belong
the oxidases and reductases, zymase (the system of
enzymes in yeast responsible for alcoholic fermentation),
catalase and other enzymes involved in anaerobic fer-
mentation. Their activity would not benefit the cell if
it were carried on outside its confines, and normally such
enzymes are held within the cell and are not liberated into
the surrounding medium unless the cell becomes damaged.
These enzymes are Endo-enzjrmes.

Theories of Enzyme Action. — At present our ideas as
to the mode of action of enzymes are somewhat nebulous,
but depend, as is to be expected, on our conception of
the mechanism of catalysis in general. Catalysts may act
in two ways, either reacting chemically to give unstable
intermediate compounds which then break down Avith
formation of the end product and setting free the catalyst
again (as is the case with the oxides of nitrogen in the
manufacture of sulphuric acid), or the catalyst may act

54 BACTERIOLOGICAL CHEMISTRY

as a carrier, increasing the active concentration of one
or more of the reactants, which is probably the mechanism
of the catalytic hardening of oils by hydrogenation in
presence of nickel. Most theories of enzyme action
involve the formation of an intermediate complex between
substrate and enzyme, but the type of compound formed
and its mode of formation are as numerous as the theories.
Michaelis, for instance, considers that the enzyme and
substrate are in homogeneous solution, and that the
union between them is an ionic reaction. Bayliss, on
the other hand, believed that the substrate is specifically
adsorbed on to the enzyme surface and that a chemical
reaction then takes place at the surface, resulting in the
conversion of the substrate into the end product. Fodor
and Abderhalden regard the adsorption as being non-
specific, but consider that a specific decomposition of the
adsorbate occurs. Willstatter suggests that the enzyme
has a specifically reactive group, the prosthetic group,
which is stabilised on a colloidal (usually protein) carrier.
More recently Quastel and his co-workers have developed
Wieland's ideas of hydrogen activation as the cause of
oxidation to account for the behaviour of bacterial
oxidation and reduction enzymes. They regard an
enzyme as being an active centre of high energy in a
cell surface caused by the interplay of the affinities of
neighbouring molecules. The active centre is believed
to exert a specific power of adsorption on the substrate
and to activate it by distorting its electronic system,
rendering the adsorbed molecule of substrate unstable
and capable of undergoing the chemical change character-
istic of the enzyme. Thus an enzyme is considered to be
a property of the surface, but to be specific because of
the groupings involved.

For further reading : —

W. M. Bayliss, " The Natvire of Enzyme Action." Monographs on
Biochemistry. Longmans, Green & Co. London, 1914.

ENZYMES 55

D. E. Green, " Mechanisms of Biological Oxidations." The University

Press. Cambridge, 1940.
J. B. S. Haldane, " Enxymes." Monographs on Biochemistry. Longmans,

Green & Co. London, 1930.
J. H. Northrop, " Crystalline Enzymes." Columl)ia University Pre^a.

New York, 1939.
J. B. Sumner and G. F. Somers, " Chemistry and Methods of Enzymes."

Academic Press Inc. New York, 1943.
H. Tauber, " Enzyme Chemistry." John Wile}'- & Sons Inc. New i'ork,

1937.

CHAPTER V

THE CHEMICAL COMPOSITION OF BACTERIA,
YEASTS AND THE LOWER FUNGI

THE problem of the composition of micro-organisms
can be approached in two ways : either by the
purely qualitative method of microscopical examina-
tion after appropriate selective staining or by chemical
methods of isolation, which may be made quantitative
as well as qualitative.

As examples of the microscopical method may be
quoted the use of osmic acid which stains fats ; the
blue colour given by starch aiid the red-brown colour
by glycogen with iodine, and the blue colour given by
cellulose in presence of zinc chloriodide. Certain dyes, too,
are selective in their action ; Sudan III, for instance,
dyes fat globules red but leaves unstained other portions
of the cell ; the nucleoprotein of metachromatic or
volutin granules is stained selectively by such nuclear
stains as polychrome methylene blue. These methods
are of value in showing the distribution of the constituents
in the cell, particularly in the case of the larger cells,
such as those of yeasts, but their use is obviously attended
with great difficulty when they are applied to such minute
cells as those of bacteria in which, generally speaking,
details of internal structure are not easily visible. The
microscopical methods also suffer from the drawback
that they only identify groups of substances, and usually
do not distinguish between the members of such groups.
Thus Sudan III stains all fats alike, and gives no clue
as to the particular sort of fat in a given organism.

The chemical methods afford a means of separating
the various constituents from one another and allow

66

CHEMICAL COMPOSITION OF BACTERIA, ETC. 57

their individual investigation. Since the organisms are
so very small it is essential to grow them in large quantities
in order that appreciable amounts of their constituents
may be obtained.

The usual methods are (1) to wash off the growth from
solid media, or (2) to separate the cells from a liquid
medium by means of a centrifuge or by filtration. Most
species of bacteria or yeasts grow well on the surface of
appropriate nutrient media rendered solid by the addition
of agar, from which the cells may be scraped or washed.
The mycelia of the moulds or lower fungi, however, fre-
quently grow into such solid media and resist removal.
Centrifugalisation is the simplest and quickest means of
separating the growth of yeasts or bacteria from liquid
cultures, and the deposit of cells can easily be washed
free from the soluble constituents of the medium.
Bacteria, particularly, are not easy to recover by filtra-
tion, since they are far too small to be retained by
ordinary filter papers, and if in any quantity soon clog
the pores of a porcelain filter candle. The moulds are
normally easy to obtain by filtration since they form a
compact mass of mycelium.

The cells of the micro-organism, having been obtained
free from extraneous substances derived from the m^edium,
can be submitted to analysis in bulk to determine the
Avater content and the amount and nature of the mineral
constituents. Usually the organisms are submitted to a
fractionation in order to isolate the various types of
substance present. For instance, the bacterial gums may
be dissolved out in water ; fats and waxes may be
extracted mth such solvents as alcohol, ether, chloroform
or acetone ; nucleic acids are extracted with weakly
alkaline buffer solutions ; the ba.sic proteins can be
dissolved in acid buffer solutions .

The residts of sucli analyses of micro organisms vary
very considerably, from organism to orga.nism, with the
conditions of growth and age of the organism and mth

58 BACTERIOLOGICAL CHEMISTRY

the methods used for isolation and estimation of the
particular component under consideration. The values
given, therefore, can be regarded only as indications of
the general make up of the cells.

Water Content. — The water content of micro-organisms
is usually determined by observing the loss in weight
on drying at 100° to 110° C. in the air or at lower tempera-
tures in a vacuum oven. The values for bacteria vary
with the species, ranging from 73-3 per cent, in the case
of Escherichia coli to 98-3 per cent, for Acetobacter aceti,
the organism commonly known as " mother of vinegar."
The majority of values fall between 75 and 85 per cent.
A certain amount of variation is to be expected as the
result of the differing amounts of water adsorbed by
different bacteria ; capsulated and mucilaginous organ-
isms will naturally retain more water than such bacteria
as Esch. coli.

The yeasts also have a varying water content, from
69-2 to 83 per cent., according to various reports. The
average value lies at about 75 per cent. The moulds
seem not to vary so much in the amount of water they
contain, the values recorded falling between 84-3 and
88-7 per cent.

Spores seem to contain very much less water, of the
order of 40 to 50 per cent.

Mineral Constituents. — The ash or mineral content
of micro-organisms is usually estimated by incineration.
The total ash content of bacteria, yeasts and fungi varies
considerably with the species, and for any one species
with the conditions under which it is grown. Different
investigators give values varying from 2 to 30 per cent,
for bacteria, 3-8 to 7-0 per cent, for yeasts, and 6-0 to
12-2 per cent, for fungi.

The chief constituents of the ash are phosphorus,
sodium, potassium, magnesium, calcium, sihcon and
sulphur together with chlorine as chlorides. The out-
standing feature of the ash content is the high proportion

CHEMICAL COMPOSITION OF BACTERIA, ETC. 59

of phosphorus, especially in the acid-fast bacteria. The
ash of most bacteria contains 10 to 45 per cent, of phos-
phorus, that of the acid-fast bacteria 43 to 74 per cent,
and of yeasts 47 to 59-4 per cent. The ash of yeasts
also has a particularly high potassium content, averaging
38 per cent.

Proteins. — Estimation of the protein content of micro-
organisms is usually based on the total nitrogen content
as determined by the Kjeldahl method, the nitrogen
value being multiplied by the factor 6-25. There are
several fallacies in this method. First of all the Kjeldahl
method estimates only about 85 per cent, of the total
nitrogen present, since the nitrogen of certain types of
compound (nitro-, nitroso-, azo- or azoxy-compounds
and such ring compounds as pyrimidines and purines) is
not capable of estimation by this method. Secondly,
it is assumed that all the nitrogen is present as protein,
which is not the case. Nor is it true that all proteins
contain 16 per cent, of nitrogen, the value on which the
conversion factor 6-25 is based ; 16 per cent, is only an
average value.

The values recorded for nitrogen are 2 to 14 23er cent,
for bacteria, 5 to 12 per cent, for yeasts and 2-3 to 8-3
per cent, for moulds, corresponding to approximate
protein contents of 12-5 to 87-5 per cent., 32 to 75 per cent,
and 14 to 52 per cent, respectively. The carbon content
of each of the three groups of organism lies between 45
and 55 per cent., and assuming that all the nitrogen is
present as protein, and deducting the corresponding
amount of carbon, it can be seen that the bacteria are
relatively rich in proteins, whilst the moulds are richer
in the non-protein carbon compounds.

Most of the protein of micro-organisms is, of course,
found in the protojDlasm of the cells. This protoplasm
is not homogeneous, however, as most cells contain
granules of nuclear material ; in fact, some bacterial
protoplasm seems to be entirely composed of nuclein,

60 BACTERIOLOGICAL CHEMISTRY

as judged by its ability to stain with those aiiihne dyes
which stain cliromatin in the higher plants. The nitro-
genous constituents of protoplasm fall into three groups :
(a) simple proteins, (b) amino -acids derived by the break-
down of proteins and (c) the nucleoproteins.

The bacteria and yeasts appear to contain proteins
of the globulin and albumin types if solubility in buffer
solutions of laiown pH. and the concentrations of
ammonium sulphate required to precipitate the fractions
are used as criteria. Globulin is precipitated from
solution by the addition of ammonium sulphate to 50
per cent, of saturation. Albumin is soluble in this
concentration of ammonium sulphate, but is precipitated
when solutions are saturated with the salt. Globulin
can be separated into euglobulin, which is insoluble in
distilled water, and pseu do -globulin which is soluble.
Partially degraded proteins, protamines, peptones and
polypeptides are also present, as well as such conjugated
proteins as glycoproteins, phosphoproteins, lecitho-
proteins and nucleoproteins.

The amino -acids obtained on hydrolysis of bacterial
and yeast proteins include all the common ones found in
proteins from other sources. The amino-acids proline,
phenylalanine and tyrosine appear to be absent from
the proteins of moulds.

The nucleoproteins, which are all soluble in dilute
alkali and which are precipitated from such solution by
acid, constitute about 2 to 3 per cent, of the dry weight
of bacteria. On hydrolysis those of the tubercle bacillus
give a mixture of the pyrimidine and purine types of
nucleic acid, but those of most other bacteria give the
purine type only. The yeast nucleoproteins are like those
of the tubercle bacillus in yielding both types of nucleic
acid. The nucleoproteins of the moulds appear to have
escaped examination. The metachromatic granules or
volutin found in many bacteria, yeasts and moulds
(particularly in the diphtheria bacillus) appear to be

CHEMICAL COMPOSITION OF BACTERIA, ETC. 61

nucleic acids as such and not nucleoproteins, since they
give nucleic acid staining reactions and are not digested
hy trj^psin and pepsin.

The proteins and nucleoproteins will be considered in
more detail in Chapter XVIII.

Carbohydrates. — Estimates of the carbohydi\ate con-
tent of micro-organisms are not very accurate and vary
from 12 per cent, in some water bacilli to about 28 per
cent, in the diphtheria and tubercle bacilli. The yeasts
may contain from 27 to 63 per cent, and the moulds 7-8
to 40 per cent.

The cell membranes of bacteria have been claimed by
certain workers to contain cellulose, but this has never
been satisfactorily proved except in the case of Aceto-
bacter xyliniim, which synthesises quite large yields of
cellulose from a variety of sugars. Cellulose appears not
to occur in yeasts or moulds.

Less complex polysaccharides, in the sense of smaller
molecules though not necessarily from a chemical point
of view, are very common constituents of nearly all
micro-organisms. The obvious examples are the " soluble
specific substances " so characteristic of many species of
bacteria, the capsules and gums of other species, the
glycogen of yeast and the polysaccharides of many moulds.
These polysaccharides may be built up from glucose,
galactose, mannose, fructose, pentoses, glycuronic acids
or mixtures of these units, as will be seen when their
study is resumed in Chapter XIX.

The presence of chitin, a polysaccharide built up of
glucosamine units, in bacteria and yeasts is stiU very
doubtful, although its presence in the cell wall of many
moulds seems to be established quite definitely.

Polysaccharides are found combined with protein in
the mucoproteins which constitute the capsules of many
bacterial species .

" Reserve carbohydrates " are found in many bacteria
and yeasts and have been given various names and

62 BACTERIOLOGICAL CHEMISTRY

ascribed various structures. The glycogen of yeast is
the most well known and satisfactorily proved of these ;
it is also said to occur in many bacteria. There is still
controversy as to whether the " granulose " of such
anaerobic bacteria as Clostridium hutyricum (Granulo-
hacter butylicum) is true starch or not, in spite of giving
a blue colour with iodine and being hydrolysed by
amylase. The amount of these polysaccharides (starch,
glycogen, granulose, iogen, etc.) varies with the age of
the cell and the amount of nutrient material available ;
their quantity falls off if the organism is starved and
immediately rises again if the cells are transferred to a
rich medium.

Simpler carbohydrates also occur in micro-organisms.
Thus trehalose, a non-reducing disaccharide composed of
two glucose units, is to be found in yeast and many
moulds, up to 4-5 per cent, of Aspergillus niger consisting
of this substance. Pentoses occur in the nucleoproteins
of micro-organisms.

The hexahydric alcohol, maimitol, is common in
moulds, especially in species of Aspergillus and Peni-
cillium, and in some bacteria.

Lipoids. — The lipoids comprise all those compounds
which are soluble in the so-called " fat solvents," ether,
alcohol, acetone, chloroform and light petroleum. The
fats, waxes and phosphatides and certain of their break-
down products, such as the fatty acids, are thus included
in this group. As mentioned earlier the fat droplets in
micro-organisms can be stained in various ways, black
by osmic acid, red by Sudan III, blue by a- or ^-naphthol
and dimethyl-p-phenylenediamine in weak alkaline solu-
tion, or yellow by dimethyl amido-azobenzene.

The total lipoid content is usually regarded as corre-
sponding to the material extracted by ether, which may
vary from 1-6 per cent, of the dry weight of Coryne-
bacterium diphtherice to 41 per cent, of Mycobacterium
tuberculosis. In yeasts the content varies from 2 to 5

CHEMICAL COMPOSITION OF BACTERIA, ETC. 63

per cent, for young actively growing cultures to 15 per
cent, for old cultures. Certain yeasts, however, particu-
larly Torula lipofeni and Endomyces vernalis can produce
up to 60 per cent, of fat under favourable conditions ;
they have been used as a source of fats when more usual
supx3lies have failed due to w^ar conditions, for instance.
The fungi have lipoid contents from 4 to 41-5 per cent.

As with other substances the lipoid content varies
with the conditions of growi^h of the organism. Media
rich in glycerol or sugar yield organisms with a higher
fat content than do media containing but little sugar.
Aeration usually increases the fat content.

The Fats. — The fats appear to serve the purpose of
reserve material in many species, although the high fat
content of old yeast cells and of partially poisoned cells
of other types has led to the view that fat production may
be a sign of degeneration.

Waxes and Higher Alcohols. — In view of the excej)-
tionally high lipoid content of the tubercle bacillus this
organism has naturally been the most studied as regards
such products. As will be seen later (Chapter XX),
this and other acid-fast organisms contain a variety
of waxes and alcohols. The diphtheria bacillus also
yields waxes, which have not yet been investigated
chemically.

Sterols. — It has been claimed that the unsaponifiable
matter of the fat of some bacteria contains mixtures of
sterols in small amounts. They are quite common in
the yeasts, forming up to 20 per cent, of yeast fat. Yeast
is a commercial source of the sterol, ergosterol, used in
the manufacture of calciferol and synthetic vitamin-D.
Certain fungi also contain sterols, ergosterol being found
free and as the ester with palmitic acid.

Phosphatides. — The phosphatides precipitated from
the " ether extract " by acetone are widely distributed
in micro-organisms, and, in fact, are probably present
to more or less extent in all of them. The non-acid fast

64 BACTERIOLOGICAL CHEMISTRY

organisms contain 0-5 to 2-0 per cent, of phosphatides,
whilst the tubercle iDacillus contains about 6-5 per cent.
Yeasts also contain a high proportion of phosphatides.

Pigments. — Many species of bacteria, yeasts, and fungi
are pigmented ; the fungi nearly all contain pigment in
some form, but the majority of bacteria, on the other
hand, show only minimal pigmentation. Beijerinck
has classified the pigment producing micro-organisms
according to the site of the occurrence of the pigment :
Chromophoric organisms contain the pigment in the
protoplasm, as in the case of bacterio-purpurin, the
pigment of certain sulphur bacteria. Parachromophoric
organisms carry the pigment in some other part of
the cell than the protoplasm, usually in the cell wall
or in the capsule. As examples may be quoted some
blue bacteria, the moulds and species of Torula (coloured
yeasts). Chromoparous organisms excrete the pigment as
such into the medium, as does Serratia marcescens
{B. prodigiosus) ; or a colourless leuco -compound may
be excreted and this may become oxidised in the medium
to the coloured compound, which is the case with the
green pigment of Pseudomonas aeruginosa {B. pyocyaneus).
The mould HelmintJiosporiu7n gramineum produces a
pigment, helminthosporin, which sheathes the mycelium
with crystals.

In comparatively few cases has the constitution of
these pigments been worked out, but in general they
are either carotenoid pigments or melanins.

The pigments will be considered further in Chapter
XXI.

Coloured colonies of certain bacteria are produced for
diagnostic purposes by the use of special media. This
may be illustrated by the production of the pink colonies
of lactose fermenting organisms on MacConkey's medium,
and by the black colonies of the diphtheria bacillus on
McLeod's tellurite medium, in which cases a product of
the metabolism of the organism reacts with a substance

CHEMICAL COMPOSITION OF BACTERIA, ETC. 65

added to the medium for that purpose. Such colour
production is not, of course, pigment formation in the
true sense of the term, but an artificial chemical test
applied hy the bacteriologist.

Growth Substances. — Certain bacteria, yeasts and
moulds produce substances which accelerate the gK)^vth
either of themselves or of other micro-organisms. Of
these " bios " is perhaps the best as well as the oldest
known of such substances. Of recent years many such
" growth substances " or " gro^\i:h factors " have been
discovered. They will be discussed in Chapter IX.

Vitamins. — Yeast is a rich source of the vitamin-B
complex, and it has been claimed that Esch. coli, Bacillus
subtilis, B. mycoides, Torula rosea and Oospora lactis
can synthesise it. Bacteria and yeasts do not appear
to produce vitamins -A, -C or -D to any marked extent,
but certain of the lower fungi of the genera Aspergillus
and Penicillium produce a strongly reducing substance
which gives the chemical reactions of vitamin-C (ascorbic
acid).

Antibiotics. — Many micro-organisms produce sub-
stances which have an inhibitory effect on the gro^Hh
of other micro-organisms. The name antibiotic has
been applied to such substances, among which penicillin
is an outstanding example. They comprise compounds
of a very wide range of constitution. They will be
considered in some detail in Chapter XI.

The composition of bacteria, yeasts and fungi, as
regards their main constituents, is summarised in
Table 3.

So far comparatively little appears to be loiown
about the composition of the viruses. Evidence is accu-
mulating, however, that there is a range of viruses of
increasing complexity from the crystalline proteins of
the tobacco mosaic viruses to those like vaccinia and the
influenza viruses which have a composition similar to
that of bacteria.

66

bacteriological chemistry
Table 3

Bacteria.

Yeasts.

Fungi.

Per Cent.

Per Cent.

Per Cent.

Water

73-3 to 98-3

69-2 to 83-0

84-3 to 88-7

Ash - - - -

2-0 ,, 30-0

3-8 „ 7-0

GO „ 12-2

P (as PgOg in ash) -

10-0 ., 74-0

45-0 „ 59-4

...

Carbon

45-0 ,, 55-0

45-0 „ 55-0

45-0 „ 55-0

Nitrogen - . .

20 „ 14-0

50 „ 12-0

2-3 „ 8-3

Protein

12-r> „ 87-0

32-0 „ 75-0

14-0 „ 52-0

Carbohj^drate -

12-0 „ 28-0

27-0 „ 63-0

7-8 ,. 40-0

Total lipoid

1-6 „ 41-0

2-0 „ 15-0

4-0 „ 41-5

Much detailed information on the composition of the
cell constituents of micro-organisms is collected in : —

R. E. Buchanan and E. I. Fulmer, " Physiology and Biochemistry of
Bacteria," Vol. I., Chapter III. BailJiere, Tindall & Cox. London,
1928.

CHAPER VI

THE NUTRITION OF THE AUTOTROPHIC
BACTERIA

THE autotrophic bacteria are tho.se which thrive on
the simplest of inorganic compounds as sources of
energy, carbon and nitrogen. Like plants they are
independent of other organic matter for their growth-.
They derive their carbon from carbon dioxide and their
nitrogen from ammonia, nitrates or nitrites. The energy
necessary for their groAvth and reproduction is obtained
in one of two ways. The photosynthetic autotrophs
utilise radiant energy from the sun. The chemosynthetic
autotrophs are able to grow in the dark and obtain the
energy required for the assimilation of carbon and
nitrogen by means of certain simple chemical reactions.

The autotrophs and the metabolically closely related
blue -green algae and unicellular green plants may be
grouped as shown in Table 4 (p. 68).

It has been suggested that, as these organisms use
very simple substances for their metabolism, they are
the primitive types of bacteria which were first developed
on the Earth before more complex organic nutrients
were available, and that the bacteria which have more
complicated requirements have been gradually evolved
from them as a result of changing conditions.

The autotrophic bacteria contain protoplasm and other
cell constituents very similar to those found in what
we regard as being ordinary bacteria. Obviously, then,
they must be capable of very complex synthetic reactions
in order to build up such compounds from the simple
raw materials carbon dioxide and ammonia,

67

68

bacteriological chemistry

Table 4

{After Knight)

Energy Source.

Carbon

Source.

Nitrogen
Source.

Photosynthetic —
Blue-green algie -

Unicellular green

plants
Green bacteria -

Anaerobic pui-ple
sulphur bacteria
( Thiorhodacece)

Anaerobic purple
"non-sulphur "
bacteria ( Athio-

rhodacece)

light
OO2 + H2O — > HOHO-t-02

light
OO2 + H2O — ^H0HO+O2

light
CO2+2H2S — > H0HO + 2S

light +H2O
OO2-J-2H2S — ^H0H0+2S

-fH20

OO24- simple organic compounds,
or CO2+H2

CO2

CO2
CO2
CO2

CO2

Atmospheric
nitrogen

Ammonia

Ammonia
Doubtful

Chemosynthetic —
Aerobic sulphur bac-
teria
Aerobic obligate

autotrophs
Facultative autotrophs

Oxidation of H2S or thiosulphate

Oxidation of ammonia, nitrite,
FeCOa, iinCOa

Oxidation of thiosulphate H2,
CO, CH«, FeCOa. Can also
grow on ordinary media.

CO2
CO2
CO2

Ammonia

Ammonia
nitrite

Ammonia
nitrate

The autotrophs live and grow in purely mineral media
which must contain the oxidisable substance characteristic
of the particular organism depending for its existence
on the oxidation of that compound. No organic nutrients
are required ; in fact the obligate autotrophs will not
grow on ordinary media containing organic carbon.
There are comparatively few obligate autotrophs ; the
most important among them are the nitrifying bacteria,
the purple sulphur bacteria and some of the iron bacteria.
The facultative autotrophs, of which there is a greater
variety, may derive their energy and growth require-
ments from the oxidation of inorganic substances, with
a corresponding reduction of carbon dioxide to give
the starting materials for synthesis, or they may grow
on already formed organic substances as a source of
carbon. Some of the sulphur and iron bacteria, together
with the hydrogen, carbon monoxide and methane

NUTRITION OF THE AUTOTROPHIC BACTERIA 69

bacteria, are facultative autotrophs. Evidence is ac-
cumulating that those organisms which are now regarded
as obligate autotrophs may all be capable of growi^h
in the presence of organic matter if the proper conditions
can be discovered.

The autotrophic bacteria show a wide variety of
morphology, ranging from coccal forms and rod forms
to multicellular filamentous forms. The photosynthetic
autotrophs contain pigments, for example the bacterio-
purpurin of the purple sulphur bacteria, which act as
respiratory pigments much like the chlorophyll of green
plants .

We can classify the autotrophic bacteria on the basis
of their metabolic activities, as follows : —

A. Oxidise Nitrogen Compounds.

1. Ammonia to nitrite, e.g. Nitrosomonas, Nitro-

sococcus.

2. Nitrite to nitrate, e.g. Nitrobacter.

B. Oxidise Sulphur or Sulphur Compounds.

1. Simple bacteria, e.g. Thiobacillus.

(a) Obligate autotrophs —
(i) Aerobic

(a) Nearly neutral conditions,

e.g. Th. thioparus.
(jS) Acid conditions, e.g. Th.
thio-oxidans.
(ii) Anaerobic, e.g., Th. denitrificans.
(6) Facultative autotrophs.

2. Higher bacteria (complex morphology).

(a) Colourless, e.g. Beggiatat, T hi othrix, etc.

{b) Red or purple pigmented, e.g. ThiQ-'
cystis, etc.

70 BACTERIOLOGICAL CHEMISTRY

C. Oxidise Ferrous or Manganotjs Compounds
(Iron Bacteria).

1. Simple bacteria —

(a) Long sheathed filaments, e.g. Didymo-

helix [Gallionella) .
(6) Coccoid masses, e.g. Sideromonas.

2. Filamentous bacteria, e.g. Leptothrix,

Grenothrix.

p. Hydrogen Bacteria, Hydrogenamonas.

We will now consider some of these in more detail.

The Nitrifying Organisms. — Winogradsky has contri-
buted much of our knowledge of these organisms. They
are divided into two groups, those which oxidise ammonia
to nitrite (Nitrosomonas) and those which oxidise nitrite
to nitrate {Nitrobacter). A species has been reported
which can oxidise ammonia directly to nitrate and
which can use nitrate as a source of nitrogen. They
are strict autotrophs and must have ammonia or nitrite,
as the case may be, for their continued existence. Nitro-
somoims has an optimum pE. of 8-3 to 8-8 for oxidation
of ammonia to nitrite. Nitrobacter oxidises nitrite to
nitrate between pH 8-3 and 9-3. The presence of organic
substances, when tested in vitro, inhibits their growth
and respiration. They are aerobic, non-sporing cocci
and short rods which occur almost universally in soils.
The nitrifying organisms have a considerable agricul-
tural importance in that between them they are largely
responsible for maintaining the supply of nitrate used
in plant metabolism .

The Sulphur Bacteria. — The sulphur bacteria form a
very heterogeneous group. They include obligate and
facultative autotrophs and may be aerobes, facultative
anaerobes or strict anaerobes. There are two important
aerobic obligate autotrophs. One of these, Th. thioparus,
was isolated from soil as small non-sporing rods by

NUTRITION OF THE AUTOTROPHIC BACTERIA 71

Nathansohn. It grows under nearly neutral conditions
and oxidises thiosulpliate, tetrathionate, or sulphides
with separation of sulphur, which is deposited outside
the cells. For example, thiosulphate is oxidised as
follows : —

2X3.28203 + 02 > 2XaoS04 + 2S.

and tetrathionate probably by the reaction : —

NaaS^Oe+NagCOs + O > 2XaoS04 + C02 +2S.

The second type, Th. thio-oxidans, was found by
Waksman in soils in the neighbourhood of sulphur
deposits, but is not ordinarily found in soils which have
not been treated with sulj)hur. Th. thio-oxidans produces
large quantities of sulphuric acid, and in fact only grows
in somewhat strongly acid conditions at pK 2 to 3,
corresponding to 5 to 10 per cent, of sulphuric acid.
It performs this remarkable function by the direct oxida-
tion of sulphur or thiosulphate : —

2S -f 3O2 4- 2H2O ^2H2S0i

XagSgOa + HoO + 2O2 ^XagSO^ + H2SO4

Sulphur is taken into the cell prior to oxidation by
solution in fat globules situated at the ends of the organ-
isms. The oxidation of sulphur, by which energy is
provided, can take place in the absence of carbon dioxide,
and the latter can be assimilated in the absence of sulphur
oxidation, either aerobically when no free sulphur is
available, or anaerobically when oxidation camiot occur,
provided that the products of the latter reaction are
available in the cell. That is, there must be some
" accumulator " mechanism in the cell by which energy
is stored, to be used in the subsequent metabolism of
carbon dioxide. It has been shown that the oxidation
of sulphur is coupled with the conversion of inorganic
phosphate from the medium into organic phosphate
esters in the cells, and that when carbon dioxide is utilised,

72 BACTERIOLOGICAL CHEMISTRY

the esters break down with liberation of inorganic phos-
phate. As sulphur can be oxidised for long periods in
the absence of carbon dioxide and with only a limited
amount of inorganic phosphate it is probable that the
latter is involved in a cycle of reactions, as in alcoholic
fermentation by yeasts (see Chapter XV). It is con-
sidered that the energy provided by the oxidation of
sulphur is used in the synthesis of a storage carbohydrate
which can subsequently be broken down again, via
phosphorylation, during carbon dioxide utilisation. The
phosphate esters which have been isolated from Th.
thio-oxidans include adenosine-3-triphosphate (not
adenosine-5-triphosphate which occurs in muscle, yeasts
Esch. coli, B. suhtilis, Stajoli. aureus and Ps. fhiorescens),
fructose- l:6-diphosphate, glucose-6-phosphate, glucose-1-
phosphate and co -enzyme I (see Chapter XV). This
suggests that the internal carbon metabolism of the
autotrophs is similar to that of the heterotrophs. The
reactions may be expressed as : —

( 1 ) 2S + 3O2 + 2H2O + inorg. phosphate >2H2S04 + phosphate ester.

(2) CO2 -I- 2H2 -f phosphate ester ^(CHgO) +H2O 4-inorg. phosphate.

The symbol (CHgO) does not necessarily represent
formaldehyde but may be a carbohydrate.

The anaerobic organism, Th. denitrificans , is an
obligate autotroph which oxidises sulphur, hydrogen
sulphide, thiosulphate or tetrathionate to sulphuric acid
at the expense of the oxygen of nitrates. For example
thiosulphate may be oxidised as follows : —

5Na28203 + 8KNO3 + 2NaHC03 > 6Na2804 + 4K2SO4 + iNg + 2CO2 + HgO

It occurs widely in soils, water and mud, from which it
was isolated by Beijerinck.

The morphologically more complex sulphur bacteria,
including the purple pigmented, photosynthetic Thiocystis
and the colourless, chemosynthetic Thiothrix and
Beggiatoa, are characterised by the presence of globides

NUTRITION OF THE AUTOTROPHIC BACTERIA 73

of sulphur within their cells. They are obligate auto-
trophs which require hydrogen sulphide for their growth.
The hydrogen sulphide is oxidised in two stages, first to
sulphur and then to sulphate : —

HgS + 0 > H2O + S

2S + 30, + 2HoO > 2H28O4.

As long as hydrogen sulphide is available the organisms
contain globules of sulphur, but as soon as the supply
fails the sulphur is oxidised, and on its complete dis-
appearance the cell dies. The free access of oxygen and
carbon dioxide is necessary for the gro^\i:h of the colourless
organisms.

The purple l^acteria contain the pigment bacterio-
purpurin, which is a mixture of two components, one
a green chlorophyll -like pigment, bacteriochlorin, and
the other a red carotenoid pigment, bacterioerythrin.
The bacteriochlorin apparently is the active pigment in
the respiration of these organisms, behaving, like the
chlorophyll in plants, as a sort of transformer for radiant
energy. That is, these bacteria require light as well as
carbon dioxide and hydrogen sulphide, but can dispense
with free oxygen. It seems possible that the light energy
is needed in the reduction of carbon dioxide with forma-
tion of, probably, formaldehyde, which is used in the
synthetic reactions, and of oxygen, which oxidises the
hydrogen sulphide : —

CO2 ^ 2H2'^ > HCHO ^ HoO ^ 2S.

These sulphur bacteria are found in fresh and salt water
and the mud of lakes and rivers, but not in soils.

The sulphur oxidising bacteria play an important role
in nature in rendering the sulphur of proteins available
again to plants as sulphate. They may also be of value
in neutralising alkaline soils by the production of sulphuric
acid, and probably also convert insoluble phosphates into
soluble, available salts. On the other hand, it is possible

74 BACTERIOLOGICAL CHEMISTRY

that they are responsible, in part at least, for the decay
of stonework and concrete.

The Iron Bacteria. — The iron bacteria are found
associated with deposits of ferric hydroxide around
mineral springs, mines and similar localities. The deposit
is usually in the form of a sheath round chains of rods
wliich thus acquire a filamentous form, or it may occur
as a sheath around true filamentous forms. Some of
these iron bacteria are obligate autotrophs, for example
Didymohelix ferruginea, and some facultative autotrophs,
such as Leptotlirix crassa. The reaction by which they
derive their energy is probably : —

In certain cases the iron may be replaced by manganese.
The Hydrogen Bacteria. — There are a number of
hydrogen oxidising bacteria occurring in such places as
canal mud or swamps and in soils where large amounts
of hydrogen are produced by anaerobic processes. The
hydrogen bacteria are usually facultative autotrophs
which oxidise hydrogen to water in the presence of
carbon dioxide, but they can also utilise organic
compounds. For example a member of the photo-
synthetic "sulphur free" purple bacteria (Aihiorhoda-
cece) is known which can oxidise simple alcohols in
presence of carbon dioxide with formation of the corres-
ponding ketone and reduction of the carbon dioxide to
give cell substances. Thus isopropanol is oxidised
to acetone : —

2(Jfr3.C'H0H.CH3 -!- CO2 > 2CH3CO.CH3 + (CH2O) + H2O.

The oxidation of hydrogen very probably does not pro-
ceed directly to water but through the intervention of
carbon dioxide with formation of formaldehyde : —

K.COa 1 2K2 5-H('H() I 2H2O.

Part of this formaldehyde is used in the synthetic

NUTRITION OF THE AUTOTROPHIC BACTERIA 75

I'eactions accompanying growth and pait is oxidised to
carbonic acid : —

HCHO + O2 > H/'Og

The mechanism

2H2 + CO2 > HCHO + H2O

has also been suggested.

Some strains will only grow in symbiosis with one
another, for example when one of the pair needs the
pyrimidine moiety of anenrin and the other the thiazole
moiety. (See Chapter IX). Pure cultures of each may
be gro^yn if small amounts of the gro^\i:h factors are
provided.

The Carbon Monoxide and Methane Bacteria.^Strictly
speaking these are not autotrophic bacteria since they
can utilise the carbon of their substrates for their gro^\i;h.
They are best regarded as intermediate t^^es between
the autotrophs and the heterotrophic bacteria. An
organism, Carhoxydomonas oligocarbopkila, which oxidises
carbon monoxide to carbon dioxide was isolated by
Beijerinck from soil. It is a facultative autotroph which
exists as a filamentous actinomyces-like organism when
grown in carbon monoxide, but exhibiting a coccal form
when growTi in the presence of organic compounds.

The methane -oxidising organism, Methanoynonas
methanica, was isolated from the mud of canals and
marshes by Sohngen. It oxidises methane, but not
other hydrocarbons, to carbon dioxide and water : —

CH4 + 2O2 > CO., -r 2H2O.

Other hydrocarbon-utilising organisms are loiown
which, although not autotrophs, may be mentioned liere
as forming part of the transition group between them
and the heterotrophs . They are Metha n omonas a lipJmtica ,
Meth. aliphatica liquefaciens and " Paraffin Bakterien,"
which utilise paraffins, including methane in the case of
the two former, a;S their source of carbon and energy.
They can also grow on ordinary media. It has been

?() BACTERIOLOGICAL CHEMISTRY

claimed that Meth. aliphatica liquefaciens can behave as a
true autotroph and live by the oxidation of hydrogen in
presence of carbon dioxide as well as on paraffins.

Sarcina inetluinica decomposes methanol, CH3OH,
in the presence of carbon dioxide with formation of
methane. If carbon dioxide containing radioactive
carbon is used, the methane is also found to contain
radioactive carbon ; that is the methane is produced by
reduction of the carbon dioxide. Meiliaiiohacterium
omelianski similarly oxidises primary and secondary
alcohols to the corresponding fatty acids with simul-
taneous reduction of carbon dioxide to methane. Neither
formate, methanol, nitrate, sulphate nor atmospheric
oxygen can replace carbon dioxide as the oxidising
agent. It is, therefore, considered that formate and
methanol are not intermediate products. It has been
shown by the use of radioactive carbon dioxide and car-
bon balance sheets that most of the carbon of the cell
constituents is derived from sources other than the
carbon dioxide.

Another group of intermediate organisms comprises
the strictly anaerobic purple " sulphur-free " bacteria,
Athiorhodacece. They are peculiar in that they are
photosynthetic, but differ from the autotrophs in requiring
simple fatty acids for their growth as well as carbon
dioxide. In the absence of the fatty acids, carbon
dioxide is not taken up. It is claimed that when sub-
jected to infra-red radiation they can use hydrogen,
that is, they are true autotrophs. The pigment of these
organisms consists of two components, one carotenoid
and the other chlorophyll-like, photocatalytic, and similar
to that in the purple sulphur bacteria.

We know virtually nothing of the way in which the
reduction of carbon dioxide, which seems to be an
essential factor in the metabolism of the autotrophs,
occurs. It has been suggested that it may be by one
of three routes : —

NUTRITION OF THE AUTOTROPHIC BACTERIA 77

{a) via carbon monoxide,
OH

2H

cIh

(h) via formaldehyde,

OH H

I \

C = 0 + 2H, > C=0 + 2HoO

(c) via formic acid,

OH OH

I I

C = 0 + H, > C = 0 + HoO

I " I

OH H

By analogy with plant metabolism it would be expected
that the second method, via formaldehyde, is the most
probable. Support is lent to this view in that formalde-
hyde can be fixed as an insoluble complex with dimedon
(see Chapter XV) in the cases of Nitrosomonas and an
autotrophic sulphur oxidising organism.

Van Niel has suggested that all photospithetic
reactions in which carbon dioxide is reduced conform to
the general equation : —

light
CO2 + 2H,A > (CHoO) + 2A + H./J.

In the case of green plants HgA is water and oxygen is
set free. In photosynthesis by bacteria H2A may be
one of a variety of inorganic or organic substances
characteristic of the particular organism. It is assumed
that each of the necessary four quanta of light energy
is associated with the activation of a hydrogen atom in
the pigment and that the carbon dioxide is reduced by
the activated pigment which thus becomes re -oxidised.

78 BACTERIOLOGICAL CHEMISTRY

In order that the pigment may again become a hydrogen
donor it must be reduced at the expense of the donor
HgA, with formation of A. The only essentially photo-
synthetic step is the activation of the reduced pigment.
The reduction of carbon dioxide and of the pigment can
occur in the dark. Some sulphur bacteria can reduce
carbon dioxide in the dark in the presence of hydrogen
suggesting that the mechanism of reduction is the same
for photosynthetic and chemosynthetic organisms. In
chemosynthetic bacteria the hydrogen donor is not
a pigment but some other substance whose oxidation
provides the necessary energy so that light activation
is unnecessary. These autotrophic processes are not
restricted to autotrophs since it has been shown that
carbon dioxide may be reduced in the dark and fixed
by heterotrophic organisms, probably by the same
mechanisms as in the chemosynthetic autotrophs, the
energy being provided by dissimilation reactions (see
Chapter VII).

As mentioned on p. 72, the symbol (CHgO) is used
to indicate the reduction product of carbon dioxide
which may or may not be formaldehyde, although the
latter is a probable intermediate.

Euben suggests that the reactions by which Meihano-
bacterium wnelianski reduces carbon dioxide to methane
with simultaneous oxidation of an alcohol to the fatty
acid (see p. 76) are as follows : —

RH + phosphate donor ^ ^ Phospho-RH + donor

Pliospho-RH. + CO2 ^=^ RCOOH + phosphate

O

//

RCOOH + phosphate donor ^ ^ R.C- 0- PO3H2 + donor

0^

//
R.C- 0- PO3H2 + 2H ^=^ R.CHO + phosphate

R.CHO + 6H ?=^ RH + CH4 + HgO .

NUTRITION OF THE AUTOTROPHIC BACTERIA

The energy and hydrogen yielding reactions are : —
2C2H5OH ^

2CH3CHO + 2H3PO4 ^=^ 2CH3C- 0- PO3H, + 4H
O

//
2CH3C- O-PO3H2 + 2 donor ^=^ 2CH3COOH + 2 phosphate— : loner

Organic compounds are not only not used by the strict
autotrophs but have a definite inhibitory effect on their
grov>i:,h, under artificial conditions at least. Thus it was
not until Winogradsky grew the nitrifying organisms on
media containing no carbon source other than carbon
dioxide that he was able to obtain cultures of them.
For solid media he employed silica gel in order to avoid
organic substances. The sulphur bacteria and some of
the iron bacteria are less sensitive to organic matter and
can grow if only low concentrations of carbon compounds
are present, especially if large inocula are used. The iron
bacterium, Leptothrix ochracea, however, is susceptible
to peptone, sucrose and asparagine.

The thermodynamic efficiency of the autotrophs is
not very high, only about 5 to 10 per cent, of the energy
liberated by the oxidation of the inorganic substrate
being utilised in the reduction of carbon dioxide to the
organic compounds used for synthesis.

For further reading : —

H. J. Bunker, " A Review of the Physiology and Biochemistry of the
Sulphur Bacteria," D.S.T.R. Chemistry Research, Special Report
No. 3. H.M. Stationery Office. London, 1936.

B, C. J. G. Knight, " Bacterial Nutrition," Sections B. and C. ^Medical

Research Council Special Report No. 210. H.M. Stationery Office.

London, 1936.
S. Ruben, " Photosynthesis and Phosphorylation." J. Amer. Chem. Soc.

65, (1943) 279.
M. Stephenson, " Bacterial ]\Ietabolism," Chapters IX and X. Longmans,

Green & Co. London, 2nd Edition, 1939.

C. B. van Niel, " The Bacterial Photosyntheses and their Importance for

the General Problem of Photosynthesis." Advances in Enzymology,
1 (1941), 263.
C. B. van Niel, " Biochemical Problems of the Chemo -.Autotrophic Bacteria."
Physiol. Reviews, 23 (1943), 338.

CHAPTER VII

THE NUTRITION OF THE HETEROTROPHIC
BACTERIA

WHEN bacteria grow and reproduce there occurs a
synthesis of all the many cell constituents, the
proteins and nucleoproteins of the protoplasm,
polysaccharides, fats, phosphatides and a number of
other carbon compounds. The elements, mainly carbon,
nitrogen, hydrogen and oxygen, but also phosphorus,
sulphur and certain metals in smaller amount, required
for these syntheses have to be supplied in an available
form by the medium in which the organism is grown.

As would be expected from an analysis of the ash
of bacteria, the inorganic constituents which must be
supplied are mainly phosphorus, sulphur, sodium, potas-
sium, magnesium, calcium, iron and chlorine. It is
probable that the metals needed in only small amount
form part of enzyme systems. Iron is an essential part
of the cytochrome complex ; phosphorus, potassium and
magnesium are also intimately involved in respiratory
mechanisms (see Chapter XII). Corynebacterium diph-
therioe, Clostridium tetani and CI. welchii need small
amounts of iron in order to produce their toxins.

We have seen that the autotrophic bacteria derive
their carbon from carbon dioxide and their nitrogen from
ammonia, nitrites or nitrates ; accordingly, they must
possess a very complete equipment of the enzymes
necessary to carry on these syntheses from such simple
starting materials. It seems probable that the hetero-
trophic bacteria, which, in general, require much more
complicated sources of carbon and nitrogen, have lost some

80

NUTRITION OF HETEROTROPHIC BACTERIA 81

of the synthetic power of the autotrophs and depend in
more or less degree on preformed organic material for their
existence. The degree of dependence varies considerably ;
organisms like Escli. coli can thrive on very simple
synthetic media containing a single carbon source, like
lactate or glucose, and a single nitrogen source, such as
an ammonium salt, together with the appropriate mineral
salts. Synthetic media are those which contain only
constituents of known composition and no proteins, broth
or similar components. Further along the scale are the
organisms like the diphtheria bacillus which will grow
on synthetic media, but which require a more or less
extended number of amino -acids. Some of these amino -
acids, for example, tryptophane and cystine, appear to
be essential, whilst others can be replaced by alternatives.
More exacting still are those organisms like the gono-
coccus and the influenza bacillus which demand the
so-called " enriched " media, containing blood or some
tissue fluid or extract, for their growth. Almost cer-
tainly these enriched media support growi^h because of
the gro^vth factors (see p. 98) which they contain. Finally,
there are the viruses which have so far lost their synthetic
powers that they can only live and grow in the presence
of living tissue, on which, it seems possible, the}^ depend
for their supply of ready-made cell constituents, or at
least for materials which are well on the way to being
the finished product. The less exacting organisms might
be compared with country people who bake their own
bread, and the exacting bacteria with town dwellers who
have lost the art of making bread and who must bu}^ it
ready made.

The simplest heterotrophs, from the nutritional point
of view, are those which depend on an organic carbon
source but which can still use inorganic nitrogen, either
as gaseous nitrogen or as nitrate or as ammonia. The
nitrogen-fixing organisms may be free-living, like the
Azotobacter or symbiotic with plants like the RJiizohium.

82 BACTERIOLOGICAL CHEMISTRY

The denitrifying organisms use nitrate as a source of
nitrogen. At the next stage are those organisms which
cannot fix atmospheric nitrogen but can thrive on
ammonium salts ; of course, they also require organic
compounds as energy source and to supply raw materials
for synthesis. An extremely wide range of substances
may serve as the sole carbon source for many micro-
organisms. Of these carbohydrates and similar com-
pounds are most readily assimilated, whilst hydroxy-
acids, fatty acids and monohydric alcohols are pro-
gressively less easily utilised. Amino -acids can often
serve as both carbon and nitrogen source. Amines
are not very satisfactory as carbon sources. Nearly all
saprophytic organisms belong to this group ; as examples
may be mentioned bacteria which can decompose formic
acid and methyl alcohol, the genus Chromohacterium and
the genus Escherichia. It is of interest to note, in this
connection, that often organisms will not grow in syn-
thetic media if they are sown in only very small numbers,
but if a large inoculum is used, growth proceeds vigor-
ously. This may be due to the introduction of essential
growth factors, bacterial vitamins, which are absent
from the medium but present in sufficient quantity in
large inocula to allow growth to commence ; once the
organism has started it can synthesise sufficient of the
growth factor to allow of continued growth. An alter-
native explanation is that in sjmthetic media there is
not, initially, a sufficiently high concentration of carbon
dioxide to permit growth. It has been shown by several
workers that carbon dioxide is an essential prerequisite for
the growth of many organisms, of which Esch. coli is one ;
again, large inocula carry over sufficient carbon dioxide
to allow growth to start. The truth of this explanation
is borne out by the fact that Esch. coli grows quite
regularly from small inocula in synthetic media under
aerobic conditions where carbon dioxide is produced by
respiration ; but under anaerobic conditions, where

NUTRITION OF HETEROTROPHIC BACTERIA 83

practically no carbon dioxide is produced, large inocula
are necessary to establish growth. For the satisfactory
groA\i:h of Brucella abortus about 10 per cent, of carbon
clioxide is necessary.

It has recently been shown that many heterotrophic
organisms, for example yeasts, Esch. coli and the pro-
pionic acid bacteria, can assimilate carbon dioxide by a
mechanism similar to that of autotrophic bacteria (see
pp. 77, 78 and Chap. XII).

The next step in the loss of synthetic power by
bacteria is probably that of the ability to utilise ammonia
as nitrogen source. The nitrogen must be supplied in
the form of organic compounds, usually as amino -acids.
This would appear to be the step in nitrogen metabolism
analogous to the loss of ability to use carbon dioxide in
the change from autotrophic to heterotrophic bacteria.
Just as there is an intermediate group of organisms
between the autotrophs and the heterotrophs, the faculta-
tive autotrophs, so there is a group of organisms which
can utilise either ammonium salts or amino -acids for their
nitrogen supply. An example of this group is Eberthella
typhosa (B.. typhosus), which is capable of growth on either
source of nitrogen but develops better on amino -acids.
Usually strains of any one species in this intermediate
group vary in their ability to use ammonium salts. For
example, the Salmonella, Proteus, the dysentery and
typhoid bacilli and the Vibrios, as a general rule, comprise
two types of strains : (f?) " exacting " strains which
cannot utilise ammonium salts but need amino -acids,
and (b) " non-exacting " strains which will grow on
ammonium salts as well as on amino -acids. The
" exacting " strains are usually pathogenic. Organisms
of this group may be contrasted with those of the coli
group and Serratia rnarcescens which can grow on
ammonium salts or amino-acids, but of which no
" exacting " strains, using only amino-acids, are known.

84 BACTERIOLOGICAL CHEMISTRY

The amino -acid most often demanded as essential by
the " exacting " strains is tryptophane,

li-CH2.CH.c00H

I II II I

%/\/ NH^

NH

p-indole-a-amino-propionic acid. It is one of the most
complicated of the amino -acids, so that it is, perhaps,
not surprising that it should be the most difficult to
synthesise, and accordingly be one of the earliest to be
required in a preformed condition. In many cases this
lost synthetic power may be restored by " training " the
organism by repeated subculture on media containing
less and less tryptophane and more and more ammonium
salt, until it can grow again in the entire absence of the
amino -acid. This so-called " training " may not be a
true change in the metabolism of the organism, but may
be a concentration, by selection, of a few individual cells
in the " exacting " strain which have not lost their
S3nithetic power. Those cells which have lost the power
will die out under the adverse conditions, until ultimately
only non-exacting organisms are left ; the acquisition
of the ability to use ammonium salts is, according to this
view, only apparent, the power really being present all the
time in a small proportion of the bacteria.

Further along the route to complete loss of synthetic
power are those organisms which, in addition to needing
organic carbon and organic nitrogen in the form of one
or more amino-acids, require the so-called " growth
factors " or bacterial vitamins, as they are sometimes
termed. The best-known organisms in this group are
Staphylococcus aureus, Clostridium sporogenes, CI. botu-
linum and Lactobacillus casei. It seems possible, if not
probable, that the organisms of the other groups can
produce their own vitamins, but that those in the present

NUTRITION OF HETEROTROPHIC BACTERIA 85

group have lost the power of synthesis of both amino -
acids and of the vitamins. Many bacterial growth
factors are recognised now as being identical with the
vitamins which play a large part in animal nutrition.
In fact many of the fundamental metabolic reactions of
bacteria are identical with, or very similar to, those
of animals. They are non-specific in the sense that the
CI. sporogenes factor promotes the gro\\i}h of CI. hotulinum
and of CI. tuelchii, and that they are produced by many
bacteria having simpler nutritional requirements ; Esch.
coli, for example, can synthesise the growth factor
required by SUiph. aureus (see Chapter IX).

As with the other changes of nutritional types, here,
again, occurs a group of intermediately placed organisms
which link those requiring growth factors with those
which do not. These intermediate species either exist
as two sorts of strain, one requiring the factor and the
other not, or they may be trained, with more or less
difficulty, to synthesise their own growth factor instead
of requiring it ready made. This seems to be the case
with the tubercle bacillus and such organisms as Coryne-
bacterium diplitherice, which, when freshly isolated, require
complex media containing gro\\i:h factors, but can be
trained to grow on synthetic media comprising only
known carbon and nitrogen sources.

It is possible that the various growth stimulants
which have been described for certain organisms are
really essential growth factors, but which are produced so
slowly by the organism concerned that their addition
from an outside source causes an increased growth. If
the rate of synthesis of the factor were so slow that its
concentration were negligible, it would be regarded as an
essential gro\\i:h factor which must be supplied in the
medium to enable growth to occur. If, on the other
hand, the organism produced it so fast that adequate
growi^h occurred without the necessity of adding it from
outside, neither its stimulating nor essential character

86 BACTERIOLOGICAL CHEMISTRY

would be recognised. It is probable that most, if not
all, organisms require the various growth factors, and
that they differ only in their ability to synthesise one or
more of them.

Organisms of the Hcemophilus group which require
two growth factors are probably the most highly evolved
of the bacteria from the nutritional standpoint. The
X -factor, derived from the haemoglobin of the blood,
may not be a generally required factor, although there
is some evidence that certain bacteria (for example,
C. xerosis and Esch. coli) may be able to synthesise it
from iron compounds. The F-f actor, which is also
present in blood, can be derived from various bacterial
and vegetable extracts and appears to be of much more
general occurrence ; even bacteria which have very simple
nutritional requirements produce it . The various members
of the influenza group of bacteria have lost the power of
synthesising one or other or both of these factors.
H. canis, for instance, has lost the capacity to produce
the Jl -factor and must be supplied with it, but it can
make its own F-factor. The hsemolj^ic influenza bacilli
have lost the power of synthesising the F-factor, but
can do without added X-factor ; whilst H. influenzce
itself cannot synthesise either the X- or F-factors and
must be supplied with both from an external source.
The pneumococcus, meningococcus and gonococcus also
belong to this highly evolved group.

Parasitic organisms, particularly pathogenic ones, have
much more complex growth requirements than the sapro-
phjrtic organisms. It seems reasonable to assume that
this is because the parasites find in their host a source
of nearly all of their needs in a preformed condition and
in the course of time have lost the necessary synthetic
powers to build up their OAvn requirements. This differ-
ence of demands between saprophytic and parasitic
organisms is well illustrated by the acid-fast bacteria.
The purely saprophytic Mycobacterium phlei can grow

NUTRITION OF HETEROTROPHIC BACTERIA 87

freely on ammonium salts and simple carbon compounds,
whilst the parasitic M. tuberculosis and Johne's bacillus
have complex requirements when freshly isolated, but on
prolonged culture in the laboratory gradually become less
fastidious and able to grow on ordinary or even synthetic
media.

Our knowledge of the general nutritional requirements
of bacteria has been summed up by Knight, who divides
the organisms into four groups with increasing complexity
of demands, corresponding to progressive loss of synthetic
power, as follows : —

1 . Carbon derived from carbon dioxide ; nitrogen

from inorganic sources (elementary nitrogen,
nitrites, nitrates or ammonia) ; energy from
light in the case of photosynthetic autotrophs
and from simple inorganic oxidations in the case
of chemosynthetic autotrophs.

2. Carbon and energy from organic carbon com-

pounds (carbon dioxide is not the main source
of carbon) ; nitrogen from inorganic compounds.

3. Carbon and energy from organic carbon com-

pounds ; nitrogen from amino -acids (some,
tryptophane, for example, are in many cases
essential) ; ammonium salts are not assimilated.

4. Carbon and energy from organic compounds ;

nitrogen from amino-acids, of which a con-
siderable number is usually required. One or
more growth factors are also required.
A fifth group, the viruses, may be added to this
Hst :—

5. Live and reproduce only in living tissues ; that

is, exhibit an almost complete lack of synthetic

powers.
It may be stated that in the last year or two the
discovery of saprophytic filter-passing organisms, which
can thrive independently of living tissue, has been
reported.

88 BACTERIOLOGICAL CHEMISTRY

For further reading : —

R. E. Buchanan and E. I. Fulmer, " Physiology and Biochemistry of
Bacteria," \'ol. I., Chapter V. Balli^rc, Tindall & Cox. London, 1928.

B. C.J. G. Knight, " Bacterial Nutrition," Medical Research Council Special
Report No. 210. H.M. Stationery Office. London, 1936.

M. Stephenson, " Bacterial ]\Ietabolism," Chapter YII. Longmans, Green &
Co. London, 2nd Edition, 1939.

CHAPTER VIII
ADAPTIVE AND CONSTITUTIVE ENZYMES

IN the last chapter the " training " of Eherthella typJiosa
to assimilate ammonium salts instead of amino -acids

and of C. diphtherice and M. tuberculosis to grow on
synthetic media were cited as examples to show that
organisms can be made to grow on an initially unfavour-
able medium. Another example is the training of certain
yeasts to ferment galactose (a power which the majority
of yeasts do not possess), whilst the production of lactose
fermenting variants of Esch. coli, of rhamnose fermenting
strains of Eherth. typhosa and of sucrose fermenting
variants of Shigella dysenterice probabh^ occur by a similar
mechanism. As we have said, it has been suggested by
some workers that this might not be due to a true
" training " or induction of variants but to a selection
of the appropriate strain from a mixed inoculum con-
taining only very small numbers of the " non-exacting "
strain, in the case of variation of nutrient requirements,
or of those organisms fermenting galactose, etc., in the
other cases.

There is evidence, however, that this is not always
the correct explanation, although it may be true in some
instances. Often there is a considerable lag before the
mutant ai)pears, much more than would be expected if
organisms of the " trained " type Ave re already present,
even in very small numbers. Moreover, in the majority
of cases the bacteria concerned were isolated as single
colony cultures and repeatedly subcultured and would
thus be expected tQ be pure strains, although it must

69

90 BACTERIOLOGICAL CHEMISTRY

be remembered that there may be a continuous production
of variants.

An alternative and more likely explanation of these
variations is that the organisms produce a new enzyme
or series of enzymes, under the stimulus of the changed
medium, which enables them to deal with the new
nutrient substances provided. Some of these enzymes
may be merely capable of breaking down a new sugar,
as is the case with, galactose trained yeasts or the
Esch. coll mutants ; or they may be responsible for
synthetic reactions which were lacking in the parent
organism, as occurs when Eherth. typhosa is trained to
grow on ammonium salts in the absence of tryptophane.

Karstrom, as a result of growing certain organisms
on a variety of media, showed that the bacterial enzymes
may be divided into two groups. To the first group
belong the

1. Constitutive Enzymes which are always produced

by a given organism, whatever the medium on
which it grows. These enzymes appear to be
essential members of the " battery " of enzymes
carried by the cell.

The second group comprises the

2. Adaptive Enzymes which appear in a given

organism as the result of growth on a medium
containing the corresponding substrate. These
enzymes only appear when their specific sub-
strate is present, and on that account seem not
to be essential enzymes.

The constitutive enzymes can be further divided into
two sorts : {a) those which always appear in approxi-
mately the same amount even on a medium from which
their particular substrates may be lacking, and (b) those
which, although always produced to some extent, occur
in increased amount when the organism is grown on the
specific substrate,

ADAPTIVE AND CONSTITUTIVE ENZYMES 91

According to Karstrom the formation of adaptive
enzymes is always associated wnth the life processes of
the cells. Dead cells can never give rise to enzymes
which were not present in the living cells. If adaptive
enz\Tnes appear in a culture which exhibits no gro^^i:h
(for instance, where the growth has been prevented by
some particular treatment, as is the case with the
" resting " bacteria of Quastel and his co-workers) it
must be assumed that such cells, although incapa})le of
cell division in the given circumstances, are not dead ])ut
in a state of suspended animation.

We will describe some of Karstrom's experiments with
Aerohacfer aerogenes (B. aerogenes) which he grew on a
lactose medium. He separated the cells by centrifugalisa-
tion, washed them and suspended them in solutions
containing xylose and calcium carbonate (the latter to
prevent development of acidity). Four solutions were
used containing, besides the cells, xylose and calcium
carbonate : —

1. Sodium chloride.

2. Potassium phosphate.

3. Potassium phosphate and yeast water.

4. Potassium phosphate and ammonium sulphate.
The fermentation of xylose by the cells in these solutions
was measured by the amount of carbon dioxide evolved,
with the results show^n in Table 5.

Table 5

Time.

Carbon Dioxide Evolved ironi Solution.

1.

2.

3.

4.

hr,<.

0

I
13
22

c.c.
0
0
0
0

c.c.
0
0
0
0

c.c.

0

5

52

66

c.c.
0
2

or,
76

92

BACTERIOLOGICAL CHEMISTRY

The enzyme fermenting xylose is adaptive, and since
the organism was grown on a lactose medium it did not
contain the xylose enzyme. In solutions 1 and 2 there
was no growth of the cells (owing to lack of a nitrogen
supply), and so xylose was not fermented. In solutions 3
and 4, however, nitrogen was present, growth (or at least
synthetic activity) occurred and a delayed fermentation
of xylose took place after the necessary enzyme had been
elaborated.

In other experiments Karstrom grew the pentose
fermenting lactic acid organism, Leuconostoc mesenter-
oides (Betacoccus), on media containing only one of a
series of sugars, separated and washed the cells and
tested their fermenting ability on other carbohydrates,
with the results show in Table 6.

Table 6

Gi'own on

Fermentation of

r; lucose,
Fructose,
Mannose.

Galactose.

Arabinose.

Sucrose.

Maltose.

Lactose.

Glucose

Galactose

Arabinose

Sucrose

Maltose

Lactose

"N'o sugar

+
+
+
+
+

+

+

+

1 i 1 1 + 1 1

+
+

+
+
+

4-

+

The glucose, fructose, mannose and sucrose fermenting
enzymes are thus seen to be constitutive since they are
produced when the organism is grown on any sugar
medium, and even on media containing no sugar at all.
The other enzymes fermenting galactose, arabinose,
maltose and lactose are adaptive since they are only
developed in the presence of the appropriate substrate

ADAPTIVE AND CONSTITUTIVE ENZYMES

93

(except for the maltose enzyme which also appeared in
the absence of any carbohydrate). That the galactose
enzyme appeared when L. nies enter oides was grown on
lactose is not surprising since lactose is built up of glucose
and galactose units.

Usually the glucose splitting enzymes are constitutive,
but an exception is found in the case of the pentose
fermenting organism, Lactobacillus pentoaceticus , in which
the glucose enzyme is adaptive and the xylose and
arabinose enzj^mes constitutive, as may be seen from
Table 7 (also due to Karstrom) : — -

Table 7

Grown on

Ferments.

Arabinose.

Glucose.

Xj^lose.

Arabinose -
Glucose

-r

+

An example of the second type of constitutive enzyme,
those showing increased production in presence of the
substrate, is given by the formation of the sucrose
splitting enzyme by Esch. coli in considerably greater
amount when the organism is grown on sucrose then
when it is grown on media containing glucose, maltose
or lactose, all of w^hich give rise to the production of some
sucrase, however.

Quastel has studied the production of the enzymes
catalase, urease andfumarase by the organism Micrococcus
lysodeikticus in different media. This organism was
chosen because it possesses the peculiar property of being
very easily lysed by egg-white with consequent liberation
of its endo-enzjrmes into the medium. After liberation

94 BACTERIOLOGICAL CHEMISTRY

in this way the amount of enzyme could be easily deter-
mined. It was found that the presence of glucose in an
agar-peptone medium stimulated the production of urease
but depressed that of catalase. The presence of urea
did not stimulate urease production nor did succinate or
fumarate stimulate fumarase production, which is high
in the presence of glucose. Quastel considers that the
effect of the substrate on enzyme production depends on
a balance of factors : (a) whether the substrate tends to
destroy the enzyme or to protect it from destruction
(j)erhaps by forming a reversible complex with it), and
(b) whether the substrate has an effect on the synthesis
of the enzyme by contributing some necessary molecules
or configuration for that synthesis.

Another type of enzyme which is adaptive is the
hydrogenlyase, produced by the coli group of organisms
when grown in the presence of formate, and which breaks
formic acid down to carbon dioxide and hydrogen : —

HCOOH > H2 + CO2.

It is also produced in presence of glucose or glycerol,
which yield formic acid as a result of their fermentation.
Some specific factor in the medium also appears to be
necessary since, generally speaking, the enzyme is only
produced when the organism is grown on a tryptic digest
of casein. If the bacterium is grown on synthetic media,
even one containing formate, no hydrogenlyase is pro-
duced in spite of good growth. It is almost certain that
in this case the enzyme is not produced as a result of
selection of a mutant strain, since the addition of formate
to a young growing culture on tryptic casein digest
caused the appearance of the enzyme in less than an hour,
during which time the number of organisms had increased
by only 18 per cent. A maximum production of enzyme
occurred in two hours with an increase of the viable
count by 34 per cent. In other words a maximum
production of hydrogenlyase had occurred before the

ADAPTIVE AND CONSTITUTIVE ENZYMES 95

organisms had doubled in numbers, which certainly seems
to rule out selection as the mechanism in this case. In
another experiment washed suspensions of Esch. coli were
added to a tryptic digest plus formate, and the production
of enzjTne (as measured by evolution of hydrogen in a
Barcroft apparatus) was followed. Initially there was no
enzyme, but it began to appear after forty-five minutes
and reached a maximum value in 150 minutes, during
which time there was less than a 5 per cent, increase in
the number of organisms ; the hydrogen production in
the same time increased by more than a thousandfold.

Other examples of adaptive enzymes are hyaluronidase
(the " spreading factor " produced by Clostridium
welchii) which is only formed when its substrate, hyalu-
ronic acid, is present and the enzyme, formed by a soil
organism, which hydrolyses the specific polysaccharide
of Type III pneumococcus.

The results of experiments on adaptive and constitutive
enzymes depend largely on the time for which the
organism is allowed to grow or to remain in the medium.
The amounts of some enzymes change considerably with
time (that is, with the state of growth of the organism),
whilst others remain more or less constant in amount.
Thus an organism examined after twelve hours' gro\\i:h
may have quite a different set of enzymes from that
which it possesses after, say, seventy-two hours. When
grown on a particular substrate an organism usually
tends to maintain a normal concentration of the corres-
ponding enzyme for a considerably longer time than it
ordinarily does, and to maintain that concentration after
the activity of most of the other enzymes has fallen off.
Hence the apparent increase in activity of an enzyme
in presence of its substrate may not always be due to a
true increase in production of the enzjnne but to a
contrast with the low values of the other enzymes. This,
of course, will be particularly marked if old cultures are
examined ; if the cultures are examined during the

96 BACTERIOLOGICAL CHEMISTRY

logarithmic phase of growth all the enzymes may be of
approximately equal intensity. Undoubtedly, though, in
many cases a real stimulation of enzymes does occur
under the influence of the specific substrates. The
production of enzymes by bacteria is also influenced
by substances other than the specific substrate. Thus
calcium is necessary for the formation of gelatinase by
Proteus vulgaris and magnesium for the phosphatases of
propionic acid bacteria. The presence or absence of
growth substances in the medium and the pH. of the
medium are factors which must be taken into account
when considering enzyme synthesis.

Some adaptive enzymes are more easily produced than
others. Karstrom showed that Esch. coli, for instance,
produced enzymes to deal with mannitol in seventy-
five minutes, sucrose in 105 minutes and lactose in 165
minutes. The production of the necessary enzymes in
training " exacting " strains is usually much slower.

The production of the galactose fermenting enzyme
by yeast is a true adaptation, since it has also been shown
to occur in the absence of cell division, which rules out
the selection hypothesis. On the other hand, the pro-
duction of lactose fermenting variants by Esch. coli
mutabile seems to be a true selection, since it has been
shown that such variants are being continuously produced
even in the absence of lactose. In this case the effect of
lactose in the medium is to provide the most favourable
conditions for the growth and identification of the
mutants.

The majority of constitutive enzymes are those which
bring about the respiratory and sjmthetic processes of
the bacteria, whilst most of the adaptive enzymes are
to be found among the hydrolases which break down the
more complex nutrient materials to a form suitable for
the attack of the constitutive enzymes. Exceptions to
this rule are the production of the enzymes which enable
" exacting " organisms to become " non-exacting " and

ADAPTIVE AND CONSTITUTIVE ENZYMES 97

use ammonium salts instead of amino-acids in their
synthetic processes.

For further reading ; —

R. J. Dubos, " The Adaptive Production of Enzymes by Bacteria." Bact.

Bev. 4 (1940), 1.
E. F. Gale, " Factors Influencing the Enzymic Activities of Bacteria."

Bact. Bev. 7 (1943). 139.
B. C. J. G. Knight, " Bacterial Xutrition," Medical Research Council

Special Report No. 210. H.M. Stationery Office. London, 1936.
H. Karstriim, " Formation of Enzymes in Bacteria. I, II." Suomen Kern.,

2 (1929), 63; 3 (1930). 42.

" Enzvmatische Adaptation bei Mikro-organismen." Ergehnisse fur

Enzymforsdumg. 7 (1938), 350.

CHAPTER IX
GROWTH FACTORS

PRIOR to the last two or three decades it was con-
sidered that an adequate supply of protein, fat and
carbohydrates, together with some mineral salts,
was all that was necessary for the normal growth and
development of animals and man. Then knowledge of
the more detailed composition of " foodstuff s led to the
recognition of the part played in nutrition by minor and
formerly unsuspected constituents. The cause of several
of what are now called " deficiency diseases " was shown
to be a lack of certain essential growth factors in the diets
of the afflicted persons. Well-known examples are
scurvy, rickets and beri-beri. As the use of diets con-
taining, as far as possible, only known constituents for
experimental investigation of \atamins became common
more and more such substances were discovered. The
stages in the development of our knowledge have usually
been first, the recognition of a condition as due to a
deficiency of some "essential metabolite," secondly, the
discovery of some crude preparation which would supply
the lacking factor or vitamin and finally the identification
of a chemically defined substance which could replace
the crude preparation. Micro-organisms are similar to
animals in that they, too, require essential growth
factors in addition to the normal sources of carbon,
nitrogen, mineral salts and other elements, necessary
for the supply of energy and raw materials for growth.
These growth factors were sometimes called " bacterial
vitamins " by analogy with the vitamins concerned in

GROWTH FACTORS 99

animal nutrition and health. It is now becoming evident
that many bacterial vitamins are in fact identical with
those involved in animal metabolism. ]\Iany of these
substances, possibly all, are intimately connected with
enzyme or co-enzjrme systems, often constituting the
prosthetic group of such systems, and in other cases
serving as an essential intermediate step in a cycle of
reactions.

Whereas, formerly, vitamins which were first dis-
covered in connection with animal nutrition were later
found to be necessary for the growth of micro-organisms,
the position is now rather the reverse. For instance,
p-aminobenzoic acid, biotin and riboflavin were first
investigated in connection with the metabolism of micro-
organisms and not until then was their activity in other
fields suspected. As the metabolism of the micro-
organisms is considerably easier to study than is that of
the much more complex animals it is very probable that
considerable improvement in our knowledge of general
metabolism will result from investigation along such
lines .

The investigation of bios affords an example of the
way in which knowledge of gro\Hh factors evolves.

In 1901 Wildiers showed that yeast would not grow
on synthetic media if sma.ll inocula were used, but that
the introduction of large inocula was followed by satis-
factory growth. He demonstrated that the addition
of boiled yeast to the synthetic medium permitted the
growth of small inocula. He attributed this phenomenon
to the presence in yeast of an essential gro\vth promoter
which he called bios ; he suggested that small inocula
did not contain enough of it to allow grov,i:h to start,
but large inocula carried sufficient bios into the new
medium for gro^^-th to occur. He showed that it could be
extracted from yeast with water, and that it was soluble
in 80 per cent, alcohol but not in absolute alcohol nor in
ether. It was stable to heat and moderately stable to

100 BACTERIOLOGICAL CHEMISTRY

acids, being destroyed by boiling with 20 per cent,
sulphuric acid but not by 5 per cent. acid. Boiling with
sodium hydroxide solutions stronger than 1 per cent,
destroyed bios. It was dialysable through semipermeable
membranes .

When yeast was shown to be a rich source of the
vitamin-B complex it was thought that bios might be
identical with it. This has been shown to be nearly,
but not entirely, true, the two complexes having many
factors in common. It was soon shown that bios was
not a single substance but a mixture of several factors.
It was first split into bios I and bios II by the action of
barium hydroxide solution, which precipitates bios I
but not bios II. Neither fraction alone is active, but
mixing them restores the potency of the preparation.
Bios I has been shown to be optically inactive meso-
inositol.

Further fractionation has shown that bios contains
the following substances : —

Bios I rne so -Inoaitol
Bios Ila Pantothenic acid
Bios lib Biotin
Bios \ Aneurin

j3- Alanine

/-Leucine

Nicotinic acid

Pyridoxine.

Several other substances of known constitution and
some of unknown constitution are also involved in small
amounts in metabolism. The known substances include
adenine, ^-aminobenzoic acid, hsematin, phosphopyridine
nucleotides, pimelic acid, riboflavin and uracil. Among
the substances of unknown composition are folic acid,
the " sporogenes factor " and a fraction from Myco-
bacterium, phlei which stimulates the growth of Johne's
bacillus.

The properties of these growth factors will be con-
sidered in more detail.

GROWTH FACTORS lOl

Adenine. — ^^ = c.XHa it has been shown that

CH c- NH

II !l \

II 11 CH

II II //

N-C-N

adenine is a necessary constituent of media for the
groA\'th of Clostridium tetani. It can be replaced by
hypoxanthine. Very probably adenine is concerned in
the synthesis of nucleic acids and diphosphopyridine
nucleotide (see p. 202 et seq.). It has recently been shown
that adenine can inhibit the bacteriostatic action of some
of the sulphonamide drugs (see Chapter X).

p-Alanine. — NH2.CH2.CH2.COOH. Saccharomyces
species and Corynebacterium diphtherice need ^-alanine
as a gro^Ai:h factor in s^Tithetic media. Since it forms a
part of the molecule of pantothenic acid it is probably
required for its synthesis. It is effective in promoting
the growth of C. diphtherice in concentrations of the
order of 1 /xg. per millilitre or less. (1 jLtg. =0-001 mg.
Sometimes the sjmibol y is used instead of 1 fig.) ^-
Alanine cannot be replaced by a-alanine. It may be
derived by many organisms from asparagine or aspartic
acid.

2)-Aniinobenzoic aeid.— XHg^ ^COOH. It was

shown by Woods and by Woods and Fildes in 1940
that p-aminobenzoic acid was the substance in yeast
extracts, peptone and other substances which inhibited
the bacteriostatic action of sulphonamide. drugs (see
Chapter X). It has since been shown to be a growth
factor for certain strains of Neurospora crassa, Aceto-
hacter suhoxydans (0-005 /xg/ml.), Clostridium aceto-
hutylicum (lxl0~® />tg./ml. in presence of l-5xl0~^ to
1-5 X 10"^ lig.jmX, of biotin) and 01. butyricum.

102 BACTERIOLOGICAL CHEMISTRY

Aneurin. — ch,

N==C.NH, CI I

I 1 «| ! /C = C.CH.X^HaOH.

CH3.C2 50 CH2— N/Sg* 6,1

II3 4 1! ^CH— S

N CH

Pyriniidirie Thiazole

Vitamin B^, aneurin or thiamine was shown to be a con-
stituent of bios by Williams in 1940. It was shown by
Knight that the substance required by Stajyhylococcus
aureus, in addition to nicotinic acid, and supplied by a
gelatin hydrolysate, meat extract or " marmite " (an auto-
lysed yeast preparation) was aneurin. It is active in con-
centrations of 0-003 jLtg./ml. It can be replaced by a mix-
ture of the corresponding 2-methyl-6-amino-5-amino-

N = C.NH2

methyl pyrimidine, CH3.C C— CH2NH2 and 4-methyl-

N— CH
N - C.CH3
5- p -hydro xyethyl thiazole, |ljj ^l ^^jj qjj fragments but

not by differently substituted fragments nor by differently
substituted aneurins. It was shown that the pyrimidine
moiety attached to an inactive thiazole as in thiochrome
(an oxidation jDroduct of aneurin) can act as a source of
the pyrimidine and that the thiazole moiety attached
to an inactive pyrimidine can serve as a source of thiazole.
Aneurin, or a mixture of the two components, has also
been shown to be essential for the growth of Phyccmiyces
blakesleeanus, of lactic acid bacteria and of propionic
acid bacteria. Some protozoa and some parasitic fungi
need intact aneurin and will not grow if the separate
components are supplied instead. Some organisms
need only one component and can synthesise the other.
Thus Mucor rammanianus can sjnithesise the pyrimidine
but not the thiazole whilst the red yeast, Rhodotorula

GROWTH FACTORS 103

rubra, can synthesise the thiazole but not the pyrimidine
part. The two organisms can he grown together in
symbiosis, each producing the component needed by
the other. Many organisms, of which Escherichia coli,
Proteus vulgaris, Aerohacter aerogenes, Alhaligenes
faecalis, Bacillus mesentericus and Thiohacillus
thio-oxidans are examples, are independent of added
aneurin, being capable of synthesising it themselves.

It is highly probable that the aneurin is active in the
decarboxylation of pyruvic acid, since aneurin pyro-
phosphate is loiown to be co-carboxvlase (see
Chapter XII).

Biotin. — ^'<^ , Kogl isolated biotin in 1935

/"\
/ \

15' 4' I

CH 3CH

I ^ I

CH2 2CH.(CH2)4.COOH

from egg yolk and from yeast. It occurs in these sources
in very minute quantities, 360 tons of yeast or one and a
half million eggs being required to yield Ig. of the crys-
talline substance. It has m.p. 230 — 231°C. and

[a]^2== + 92° (in 0-1 N NaOH). It is probably the most
active biological substance known at present ; a dilution
of one part in 10^*^ is sufficient to promote half the maxi-
mum gro^vth of yeast. It has been synthesised and the
synthetic material is indistinguishable from the natural
substance in biological activity. It is inactivated by
treatment with nitrous acid, oxidising agents or by
benzoylation or acetylation. The methyl ester, however,
can be utilised in place of free biotin, but in some cases
less readily. By reduction in presence of the Raney
nickel catalyst the sulphur atom is eliminated and

8

104 BACTERIOLOGICAL CHEMISTRY

replaced by two hydrogen atoms to give desthiobiotin,
CO

/\

/ \ ,. , ....

NH NH which, very surprisingly, is as active as

CH CH

I • I ■

CHg CH2.(CH2)4.COOH

biotin itself. Observable growth of Saccharomyces
cerevisiae occurs in the presence of a dilution of 1 in
4x10^1. Desthiobiotin will not promote the growth
of Lactobacillus casei. Hydrolysis of desthiobiotin by
acid or by alkali yields a diaminopelargonic acid,

CH — CH which on oxidation gives pimelic acid,

I ! '

CH3 CH2.(CH2^4.COOH

COOH.(CH2)5.COOH. Biotin has been shown to be
identical with vitamin H which is protective against "egg
white injury," and with " co -enzyme R," which is necessary
for the resjDiration of the nitrogen fixing organisms,
Ehizobiu7n. The action of biotin is inhibited by avidin,
the substance in egg white which is responsible for " egg
white injury." Since the activity of desthiobiotin is
also inhibited by avidin it seems probable that it is the
urea grouping of biotin which combines with avidin.

0-hetero biotin, the analogue of biotin in which the
sulphur atom is replaced by an oxygen atom, has about
half the activity of biotin lor L. casei, L. arahinosus and
Sacch. cerevisiae. It, also, is inactivated by avidin.

Added biotin is required for the growth of lacto-
bacilli, the propionic acid bacteria, CI. acetobntylicum,
CI. butylicum, Staphylococcus (0-005 to 0-01 ju,g./ml.).
Brucella and hsemolytic streptococci. At present the
mechanism of its activity is obscure.

GIiitamine.--CH2(CONH2).CH2.CH.NH2.COOH. This

GROWTH FACTORS 105

amino-acid has been shown to be necessary for the initia-
tion of gro\\i:h of many strains of Streptococcus, Lacto-
bacillus, Diplococcus pneumonice, and B. anthracis. The
organisms appear to be able to synthesise adequate
quantities of ghitamine once growth has commenced.
For some organisms it can be replaced by considerably
larger amounts of glutamic acid.

Haematin. — ^It has been established that haematin is
the X-factor required by members of the genus Hcenio-
philus. It can be replaced by certain inorganic iron
compounds which have oxidase or catalase activity. It
is, apparently, only necessary for the aerobic gro^vth of
these organisms, since anaerobically they can grow in
its absence. It is highly probable that haematin, or the
other iron compounds, are necessary for the synthesis
of the cytochrome system which plays an important
part in bacterial respiration (see Chapter XII).

CHOH

CHOH CHOH.
Me^o-Inositol. — I I This substance is

CHOH CHOH

CHOH
essential for the growth of yeasts of the genus
Saccharomyces . It is noteworthy that it is required in
amounts considerably larger than for the majority of
growi^h factors, milligrams rather than micrograms being
needed. Its function is still unknown although Eastcott,
who isolated it from bios, states that it is stored unchanged
in the yeast cell. In animal metabolism it can prevent
the development of " fatty liver " which normally arises
when there is an excess of cholesterol in the diet. Inositol
occurs as the phosphoric ester in the phj^in of wheat
germ, which interferes with the normal metabolism of
calcium in bone formation.

106 BACTERIOLOGICAL CHEMISTRY

CH3

/-Leucine.— ^CH.CHa.cH.NH^.cooH, This

(•H3

amino -acid is required by yeast for growth in synthetic
media and is a constituent of bios. It is classed as one
of the essential amino -acids for the growth of most

bacteria.

CH

CH C.CONHo

Nicotinamide. — 1^ II . Knight found that

CH CH

%/

]sr
nicotinamide was one of the constituents of meat or
yeast extract required for the aerobic growth of Staph,
aureus on gelatin hydrolysate with the addition of
tryptophane, cystine and glucose. It was active at a
dilution of 0-05 /xg./ml. of medium. Nicotinic acid,

CH

CH CCOOH "^^^ found to be someM^hat less active than the
I II ' amide. That the activity is not due to an im-

^ p^ purity is shown by the fact pyridine-3-nitrile,

N
CH

CH C.CN, i^ inactive until it is hydrolysed to the acid.

I II Nicotinamide is also required by Shiga's

^•^/'^ dysentery bacillus and by Hceinophilus

pertussis. The ratio of the activity of the amide to that
of the acid is not constant but differs for various organ-
isms. Thus for C. diphtherice the acid is ten times as
active as the amide, for Proteus vulgaris the two are
equally active, for Staph, aureus tlic amide is five times
as active as the acid, for Shigella dysenterice the ratio is
ten to one, whilst some Pasteur ella strains cannot use the
acid at all.

GROWTH FACTORS 107

Shigella dysenterice requires (M /xg./ml. of nicotin-
amide for growth but higher concentrations have an
inhibiting effect. This is probably analagous to the
inhibition of enzyme action by an excess of one or more
of the products .

Nicotinamide forms part of the molecule of Harden
and Young's cozymase or Co -enzyme I, a di23hosphopy-
ridine nucleotide, and of Co -enzyme II, a triphospho-
pyridine nucleotide, which are concerned in carbohydi^ate
metabolism (see Chapter XII). It seems obvious that
nicotinamide is required by some organisms for the
synthesis of these co-enzymes. Bacteria like Esch. coli,
Eberthella typhosa, and Vibrio comma can synthesise the
whole CO -enzyme, whilst Staphylococcus and the dysentery
bacilli, for example, cannot synthesise the pyridine
moiety. H. influenzce cannot s^aithesise any part of the
CO -enzyme which must be supplied intact and cannot be
replaced even by adenylic acid (adenine-|-d-ribose-|- phos-
phate, see p. 332). Nicotinamide can be replaced by
Co-enzjones I or II.

In America the name niacin has been given to
nicotinamide in order to avoid the unfortunate association
in the public mind with nicotine. This is important as
nicotinamide is of interest in human metabolism, a
deficiency of it giving rise to pellagra.

„ , ^, . .^ ^i5J3\o_(jHOH.CO.XH.CH2.CH<,.rOOH.

Pantothenic acid. — ^^3/

CH2OH

Williams isolated pantothenic acid from a number of
sources such as yeast, rice bran, milk, liver and egg
white by extraction with 80 i)er cent, methanol. It was
shown to be identical with the chick anti-dermatitis
factor of liver and to be an important constituent of
bios. It is of very widespread occurrence in bacteria,
moulds and many plant and animal tissues, a fact which

108 BACTERIOLOGICAL CHEMISTRY

gciAc I'iisc to its name. It is active in very low concentra-
tions, of the order of 0-008 fig./ml., in stimulating
carbohydrate fermenting organisms, but not non-carbo-
hydrate fermenters. Pantotlienic acid loses its activity
towards some organisms, e.g. Str. liceirwlyticus , on cleavage,
by acid hydrolysis, into «-hydroxy- p p-dimethyl-y-

^ Ji sXp CHOH C = 0

butyrolactone, ^^Hs/T ' i' ' and p -alanine,

CH2— 0 1

NH2.CH2.CH2.COOH. This is similar to the failure of
some organisms to utilise glutamic acid instead of
glutamine, and to the readier use of nicotinamide than
of the acid. It appears to be associated with the inability
to form amide linkages other than in the a-position, that
in pantothenic acid being p- and that in glutamine
being y--

The effect of pantothenic acid is usually increased
by relatively large amounts of meso-inositol and by
extremely small amounts of bio tin.

Active pantothenic acid has been sjnithesised from
its inactive component parts.

Among organisms for which pantothenic acid is a
growth factor are the lactic acid bacteria, hsemolytic
streptococci, C. diphtherice gravis and Proteus morganii.

The way in which panthothenic acid enters into the
metaboUsm of micro-organisms is not yet understood.

Phosphopyridine Nucleotides. — Diphosphopyridine
nucleotide, Co-enzyme I, which is constituted as nicotin-
amide -ribose -phosphate -phosphate -ribose -adenine , has
been shown to be identical with the F-factor required
by H. influenzce and which is supplied by extracts of
many bacteria, yeasts, blood and plant and animal
tissues. A mixture of nicotinamide and adenylic acid
cannot replace Co -enzyme I in the metabolism of
H. influenzce. The co -enzyme is a co -dehydrogenase
(see Chapter XII).

GROWTH FACTORS l09

The closely related tiiphosphopyridine nucleotide.
Co -enzyme II, which is concerned in reduction and
phosphorylation reactions, is also required by H. in-
fluenzce and H. parainfluenzoi.

Organisms of the genus Hcemophilus afford another
example of symbiosis. H. canis needs hsematin but
synthesises Co-enzyme I ; H. parainfluenzce needs co-
enzyme I but synthesises haematin. If sown separately
in peptone water neither grows. If they are sown
together good growi}h results, each organism supplying
the other with the lacking factor.

Pimelic acid— COOH.CH2.CH2.CH2.CH2.CH2.COOH.
It was shown by Mueller that pimelic acid was one of the
substances in liver extract which was required for the
gro^vth of C. diphtherice in sjoithetic media. The syn-
thetic acid is equally effective, but other dibasic acids,
such as azelaic acid, were not effective. Pimelic acid has
an observable effect in concentrations of 0-005 ju-g./ml.
and optimum effect at 0-01 [ig./ml.

Although nothing is kno^vn of the way in which
pimelic acid acts it is, perhaps, significant that the
acid arises as a result of the hydrolysis and oxidation of
biotin (see p. 104). It is possible that pimelic acid may
be required in the synthesis of the carbon chain of biotin,
the other two carbon atoms of the chain being derived by
condensation with acetaldehyde.

CH2OH

I
C

Pyridoxine. — CHaOH.t^ * ^j.oh ,2-methyl-3-hydroxy-4:5-

Hc!« 2C.CH3

di-(hydroxymethyl) pyridine. This substance has been
shown to be a rat anti -dermatitis factor, vitamin Bg,

110 BACTERIOLOGICAL CHEMISTRY

which occurs in the vitamin B complex. It maj^ also be
concerned as an anti -pernicious ansemia factor in liver
extracts. It has been isolated from rice bran, liver,
molasses and similar sources. It was shown by Williams
to be necessary for the growth of Saccharomyces cerevisice
and Strej^tobacterium plantarum. It is required by Lacto-
hacilhis casei, L. delhrilckii and L. lactis but not by L.
arabinostis or L. pentosus. Leuconostoc mesenteroides can
grow without it but is stimulated by its presence. The
last three organisms are able to synthesise pyridoxine
but L. mesenteroides only to the extent of about one-
fourth of the production by the other two. The amount
required by L. casei depends on the oxygen tension of
the medium, the lower the oxygen tension the lower is
the amount of pyiidoxine required. It stimulates the
growth of staphylococci.

The 4:5-diacetyl derivative is nearly as active as
pyridoxine itself, but the triacetyl derivative is inactive.
Substitution usually reduces or destroys the activity.

It has been shown that pseudopyridoxine, which is
form.ed by the action of hydrogen peroxide on pyridoxine
or by autoclaving solutions of the latter in presence of
cystine, is considerably more active than pyridoxine in
stimulating the growi:,h of Str. lactis and L. casei, although
it had no greater effect on the growth of several moulds,
N euros fora sitophila, yeasts such as Saccharomyces
carlshergensis and Sacch. oviformis or rats.

Pseudoj)yridoxine is now known to be a mixture of

CHO CH.NH2

HO |/\, CH2OH , HO (^\ CH2OH.

pyridoxal ^^^

CH3 'J pyridoxamine, CH3

N N

These two compounds are reversibly interconvertible and
act as the co-enzyme of transamination (see p. 341).

GROWTH FACTORS 111

Pyridoxal is converted, in presence of adenosine triphos-

CHO

HO /\ CH2OPO3H2
phate, to the phosphate

N

which acts as the co -decarboxylase for the amino acid
decarboxylases for tyrosine, tysine, asparagine, arginine
and ghitamic acid (see p. 228).

Riboflavin. — ^'HsOH Warburg and Cln^stian, in

(HOH l^^2, showed that riboflavin

j was an essential part of the

CHOH " yellow " respiratory enzyme,

('HOH which, together Avith Co-

I enzymes I and II, is concerned

y^2 in the carbohydrate metabolism

CH N N of yeasts and bacteria (see

/\/\^\,_ Chapter XII). It occurs in

^^'f !! y y~^^ yeasts and those bacteria which

CH3.C C C XH do not require it as a gTO^\i:h

^CH^X^cf-0 ^^^^^^' ^^^^ ^^ many animal
~ tissues. Since it was originally
isolated from milk it is sometimes known as lactoflavin.
It is a grov>i:.h factor for most lactic acid bacteria, pro-
pionic acid bacteria, streptococci, Thermobacterium and
Clostridium tetani.

Its function seems, obviously, to be built into the
enzjmie by those organisms which cannot sjoithesise it
for themselves.

Uracil.— XH— co Richardson showed that for the

('o CH. anaerobic growth of Staphylo-

I II coccus aureus on synthetic

XH— CH media it was necessary to add

uracil. It is a component of nucleic acids (see Chapter

XVIII), and is, presumably, required for their synthesis

under anaerobic conditions. Staph, aureus appears to be

112 BACTERIOLOGICAL CHEMISTRY

able to Bynthesise uracil under aerobic conditions only.
It has been shown to be necessary for Lactobacillus
arabinosus, Leuconostoc mesenteroides and Group C
hemolytic streptococci. It can be replaced by orotic
acid (uracil 4-carboxylic acid) for the latter. Uracil
is not necessary as a growth factor for Group A hsemolytic
streptococci if carbon dioxide is present at a partial
pressure above 40 mm. of mercury.

Folic acid. — Williams isolated folic acid from the
leaves of spinach and from yeast, liver and kidney.
It is possibly identical with, or forms a part of liver
" eluate factor." It occurs in most green leaves, including
grass. It contains vitamins Bio ^.nd B^ which are
necessary in the growth of cliicks and for the production
of feathers by chicks respectively. It also increases the
growth of rats.

Folic acid is still of unknown constitution, but is
known to contain nitrogen but no sulphur or phosphorus.
It is said to have a molecular weight of about 500. An
active orange yellow crystalline substance of the com-
position C9H10N3O3, probably identical with vitamin Be,
the chick anti-anaemia factor, has been obtained from it.
The crystalline material induces half the maximum
growth of L. casei at a concentration of 0-00005 /xg./ml.

Folic acid is required as a growth factor by many
lactic acid bacteria. In a concentration of 0-00012
/xg./ml. it gives half its maximum growth effect on Str.
lactis R. It also stimulates the growth of L, casei and
L. delbrilckii. A liver extract stimulating the growth
of L. casei E can be replaced by orotic acid, uracil-4
carboxylic acid, but not by uracil itseK. Orotic acid
may be a constituent of folic acid. There appear to be
at least two factors in folic acid required for bacterial
growth because, although an extract from spinach has
equal activity in supporting the groAvth of Str. lactis
and L. casei, extracts from liver and yeast have different

•GROWTH FACTORS 113

values for the two organisms. The difference is due to a
factor which is active for Str. lactis but not for L. casei.
The factor has been isolated and it was found that 1/xg. of
it was equivalent in activity to 56 fig. of a folic acid
concentrate towards Str. lactis but that 1 /xg. was less
active than 0-004 jag. of the concentrate for L. casei.
This factor is different from the crystalline substance
described above which is more active towards L. casei
than towards Str. lactis.

Folic acid, vitamin Be, the L. casei factor and xanthop-
NH— CO

terin, nh=c C— N=C.0H , are closely related and

I II I

XH— C— N = CH

probably concerned in the synthesis of thymine,

NH— CO

CO C.CH3 . Thymine can replace vitamin Be in the

I II
]N^H— CH

nutrition of L. casei but not in that of Str. lactis R (identical
with Str. fcecalis). L. casei factor can be partially con-
verted into folic acid by incubation with chick liver ;

CH2OH

COOH

HOOC |/^| OH

(JH2OH

A OH,

if

a- or

j3-pyracin,

V"-

or

N

is

added to the mixture

even better

conversion

occurs.

Mycobacterium phiel Extract. — In 1912 Twort and
Ingram showed that Jo line's bacillus, on first isolation,
would only grow in the presence of some substance
occurring in an extract of M. tuberculosis honiinis or,
better, in M. phlei. The organism could be " trained,"

114 BACTERIOLOGICAL CHEMISTRY

with some difficulty, to grow in the absence of the extract,
or, in other words, to produce its own growth factor.
The substance, which is acidic in nature, is very stable
and can be extracted by glycerol, hot water, hot alcohol
or hot acetone. It can be replaced by alcoholic extracts
of a number of vegetable tissues and fungi. It can be
partially replaced by 0-1 /xg./ml. of phthiocol, 3-hydroxy-
2 -methyl- 1 : 4 -naphthoquinone, a constituent of tubercle
bacilli (see p. 392), or of 2 -methyl- 1 : 4 -naphthoquinone,
the anti-haemorrhagic vitamin K.

The Sporogenes Factor. — CI. sporogenes, when inocu-
lated as a spore suspension into synthetic media shows
no growth. When active preparations from yeast or
from urine are added in very small quantities (0-4 /xg./ml.)
good growth ensues. The factor is widely distributed in
animal and vegetable tissues from which it can be ex-
tracted by 75 per cent, alcohol. It can be purified by
conversion into an alcohol soluble barium salt. The
regenerated acid is active in concentrations of 0-02
/Ag./ml. It can be further purified by distillation of its
methyl ester in a high vacuum (boiling point, 80 to
100°C. at 0-001 mm. of mercury). The ester is inactive
but the activity is restored on hydrolysis. The sporogenes
factor is an unsaturated hydroxy fatty acid, C11H14O4 or
C11H12O4, of molecular weight about 200. Its presence
appears to be essential for the growth of CI. botulinum
and CI. welchii as well as of CI. sporogenes. It is produced
by many, probably all, aerobic bacteria, for example.
Salmonella typhhnurium, Eherthella typhosa, and M.
tuberculosis and by the mould Aspergillus versicolor. It
is probable that it is required by all micro-organisms
but that the Clostridia have lost the power of sjnithesising
it for themselves.

It will have been noticed that many of the substances
listed as growth factors are the prosthetic groups or the

GROWTH FACTORS 115

parent substances of prosthetic groups of enzymes or
CO -enzymes which are essential for the metabolism of
micro-organisms. It is this which, probably, accounts
for the fact that most growth factors are needed only in
very small amounts, since they remain in circulation,
as it were, and are used over and over again. If they
are not available the particular metabolic process in
which the corresponding enzjine or co -enzyme is in-
volved is brought to a standstill and the organism fails
to grow. It is almost certainly true that all organisms
make use of these substances in their metabolism. Some
organisms, however, appear to be able to make them all
for themselves, whilst others need one or more of them
to be provided from outside sources. It is only when a
substance has to be supplied ready made to a micro-
organism that it is regarded as being a growth factor for
that organism. An organism, although capable of
synthesising a particular gro^^i;!! factor, may do so
relatively slowly so that it is in a chronic state of deficiency
and accordingly develops only poor gro\\i}h. The addi-
tion of the metabolite then has a stimulating effect on
growth. Fildes has suggested that, in reality, most, if
not all, of these compounds can be regarded as " essential
metabolites."

That bacteria really can synthesise " essential meta-
bolites " themselves is showrti by the fact that TJiio-
bacillus thio-oxidans , for instance, when grown
autotrophically on sulphur containing media with no
organic material initially present contains aneurin,
biotin, nicotinic acid, pantothenic acid, pyridoxine and
riboflavin. The same is true of Esch. coli, Proteus
vulgaris, Aerohacter aerogenes, Alkaligenes fcecalis, B.
antJiracis, B. mesentericus , B. vidgatus, Vibrio comyna
and Serratia marcescens in which aneurin, biotin, nicotinic
acid and other substances have been detected after
their growth on synthetic media containing none of these

116 BACTERIOLOGICAL CHEMISTRY

compounds originally. It is possible that intestinal
bacteria may serve as a source of vitamins for human
and animal nutrition, since it has been observed that
vitamin deficiency symptoms often develop when the
growth of intestinal bacteria is suppressed by the use of
such di-ugs as sulphaguanidine or sulphasuxidine.

The Lactobacilli require a considerable proportion
of the known growth factors for their adequate growth
on synthetic media. This fact is used for the detection
and estimation of the amounts of such substances present
in various extracts or foods. Thus Landy and Dickens
have shown that L. casei will grow well on a sjmthetic
medium containing the appropriate amino -acids and
mineral salts together with aneurin, biotin, nicotinamide,
pantothenic acid, pyridoxine and riboflavin. The amount
of growth can be estimated by titrating the lactic acid
formed. If any one of the growth factors is omitted
from the medium growth does not occur. If the missing
factor is added in amounts less than that required for
maximum growth, the degree of growth is proportional
to the quantity of the factor added. The growth of L.
casei on the medium lacking one of the factors can,
therefore, be used as a test for the presence of that factor
in anji^hing added to the medium ; by comparing the
amount of growth in presence of the addendum with
that occurring on the full medium the test can be made
quantitative. By omitting each growth factor in turn
from the medium an analysis of the factors in an extract
can be made.

Another valuable method of assaying growth factors
and essential amino -acids has recently developed from
the use of mutants of the mould Nenrosi)ora crassa
obtained by the action of X-rays on the asexual spores
and crossing them with the heterothallic strain of opposite
character. A series of mutants has been obtained wliich

GROWTH FACTORS 117

can grow on a complete medium but which fail to grow
on a medium lacking a single constituent specific to
each mutant. This is due to a lack of synthetic ability
to form the particular compound, occasioned by destruc-
tion of the controlling gene. Mutants failing to sjoithesise
p-Sbinmo benzoic acid, aneurin, choline, inositol, nicotinic
acid, pantothenic acid, pyridoxine and the essential
amino-acids arginine, leucine, lysine, methionine, proline,
threonine, tryptophan and valine are known. The gro^vth
of such mutants is proportional to the amount of the
specific substance (up to the amount necessary for
optimum gro^vth) which is added to an otherwise adequate
medium.

These and other mutants are also extremely valuable
for elucidating the mechanism of certain metabolic
reactions (see p. 343).

In contrast with the term " Antibiotics " which has
recently come into use for those substances produced by
micro-organisms which inhibit the gro^vth of other
organisms, it has been suggested that those substances
capable of stimulating metabolism which are produced
by micro-organisms, and which are usually highly specific
in their action, should be called " Biotics " as an alter-
native to growth factors.

A list of some micro-organisms and their requirements
of growth factors is given in Table 8. It must be
realised that different strains of a given species may
vary in their ability to synthesise one or more of the
biotics mentioned so that the lists given cannot apply
rigidly to all strains. Some of the strains may also
require growth factors of still unknown composition in
addition to those listed. Considerable variation of this
sort is found amongst strains of (7. diphtherice, Lacto-
bacillus, Streptococcus and the yeasts.

118

bacteriological chemistry
Table 8

Organism.

Growth Factors.

Amount required
per ml. ot modiinn.

A. siihoxyians

^-Aminobcnzoic acid -

G-005

Brucella

Biotin (or methyl ester)
Nicotinamide
Pantothenic acid -
Pyrimidine (or aneurin)

0-000008 to 0-0001
0-02
0-02
0-02

CI. acetobufylicum

CI. botulinum
CI. butylkum,

Cl. butyricum
CI. sporogenes
Cl. tetani -

Cl. uelchil -

_p-Aminoben7,oic acid

Biotin ....

" Sporogenes Factor " -

Biotin

" Sporogenes Factor "

;j-Aminobenzoic acid -

" Sporogenes Factor " -

Adenine . . . .
Aneurin - - . .

Biotin

Folic acid - - . -

Nicotinic acid

Oleic acid - - -

Pantothenic acid -

Pyridoxine - - - -

Ribollavin - - - -

Tryptopha,ne

Uracil

Oleic acid ... -
Pyridoxine ... -
" Sporogenes factor " -
Uracil

0-000001
0-00015 to 0-000001

0-02

5-0
01
0-001
0-001
1-0
2-5
0-05 to 0-25
1-0
0-1
50-0
2-5

C. diphtheriae

j5-Alanine ... -

Nicotinic acid or

Nicotinamide

Oleic acid - - - .

Pantothenic acid -

Pimclic acid - - - -

0-1
1-0

10-0

0-005 to O-Ol

Gontimied on next page

GROWTH FACTORS

Table 8 (Continued)

119

Organism.

Growth Factors.

Amount required
per ml. of medium.

H. canis

Hsematin . - - .

m

H. ducreyii -

Haematin ....

H. influenzce

Hsematin . . . .
Co-enz\nne I or 11

...

H. parainfluenz(B -

Co-enzyme I or II

...

H . pertussis

Nicotinamide

...

Needed by all fol-

Aneurin ....

0-2

low insr strains oi

Biotin

0-0001 to 0-0004

Lactobacillus

Folic acid . . . .

0 00005 to 0-3

Nicotinamide

01 to 0-2

Pantothenic acid -

003 to 0-2

Riboflavin - - - .

0-04 to 0-2

Tryptophane

100-0

L. arahinosns

Adenine ....

20-0

p-Amino benzoic acid -

0-0002 to 0-6

Inositol ....

20-0

Uracil ... -

50

L. casei

Adenine ....

20-0

;;-Amino benzoic acid -

0-6

Guanine . . . .

200

Inositol . - - .

5-0

Orotic acid - - - -

0-01

Pyridoxine . - - -

0-06 to 0-6

Uracil

20-0

L. delbrUckii

Pj'ridoxine . - - .

...

L. lactis

Pyridoxine ....

L. pentoaceticus -

Pyridoxine ....

L. pentosus -

Adenine ....

L. plantarum

Adenine - - -
Guanine ....
Pyridoxine . - - -

Leuc. mesenteroides

Pyridoxine - - - -

Continued on next page

120 BACTERIOLOGICAL CHEMISTRY

Table 8 (Continued)

Organism.

Growth Factors.

Amount required
per ml. of medium.

Pneumococcus

Biotin

Pantothenic acid

i"o

Proi^ionic acid

bacteria

Aneurin . . . .

Biotin

Riboflavin ... -

0-005 to 0-05

Proteus morganii -
Proteus vulgaris -

Pantothenic acid -
Nicotinamide

Nicotinamide

0-2

Bhizobium -

Biotin

■ ...

; Rhodotorula rubra
Rhodotorulaflava -

Aneurin ....
Aneurin ....

0016
0-016

: Shigella dysenterice

Nicotinamide or Co-enzyme I
or II - - . .

01

\ Staph, aureus

i

Aneurin . . . .

Biotin

Nicotinamide

Pyridoxine - - . .

Uracil (anaerobic only)

0-003

0005 to 0-01

005 to 0-2

0-3 to 1-2

0005

\ Streptobacterium
; plantar am

Biotin

Nicotinic acid
Pantothenic acid -
Pyridoxine ....

0001
0-2
0-2
10

[ Strej)iococcus

\ hcemolyticus

■

Aneurin ....
Biotin ....
Glutamine* - - . -
Nicotinamide
Pantothenic acid -
Pyridoxine - - . .
Ribofla\iii - . . .
Urai-il (for Croup ( ')

♦Synthesised once growth has
commenced.

0001

O-i

0008 to 1-25

2-0

0-004 to 01

10 to 20

Continued on next page

GROWTH FACTORS

Table 8 {Concluded)

121

Organism.

Growth Factors.

Amount required
per ml. of medium.

,S7/-. lactis Pv.

Aneurin . . . .
Adenine . . . .

Biotin

Folic acid . . . .
Guanine . . . .
Nicotinamide
Pantothenic acid -
Pvridoxine - . - -
Riboflavin ....
Thymine ....

[J-g-

0-2
100

0-0004

0-0005
10-0

0-6

0-4

1-2

0-2

0-2

Neil rospora c rassa
N. sito/ihila

;j-Aminobenzoic acid
Pvridoxine ....

0-0025
0-1

Phycomyces
blakesleeanvs

Aneurin ....

0-02

Saccharomyces

cerevisirjE

/3-Alanine ....
/^-Aminoben/.oic acid
Aneurin ....

Biotin

Inositol ....
Z-Leucine ....
Pantothenic acid .
Pvridoxine ....

0-08

0-0001
20-0

o-orji

It has been claimed that ascorbic acid is required
as a growth factor by the trypanosomes Schizotrypanum
cruzi, LeisJirminia tropica and L. donovani and some
Trichomonas species. Cholesterol is also said to be
required by some species of Trichomonas.

For further reading : —

K. Hofmann, " The Chemistry and Biochemistry of Biotin." Advances in

Enzymology. 3 (1943), 289.
B. C. J. G. Knight, " Growth Factors in Microbiology." Vitamins and

Hormones. 3 (1945), 108.
S. A. Koser and F. Saunders, " Accessory Growth Factors for Bacteria and

Related Micro-organisms." Bad. ^Rev. 2 (1938), 99.
A. L^voff, " Les Facteurs de Croissance pour les Micro-organismes." Ann.

Inst. Pasteur. 61 (1938), 580.
W. H. Peterson and M. 8. Peterson, " Tlie Relation of Bacteria to Vitamin.s

and other Growth Factors." Bad. Rev. 9 (1945), 49.
R. J. Williams, " The Chemistry and Biochennstrv of Pantothenic Acid."

Advances in Enzymology. 3 (1943), 253.

CHAPTER X
CHEMOTHERAPY

THE term chemotherapy introduced by Ehrlich is
used to describe the treatment of diseases due to
micro-organisms by m«ans of chemicals of known
composition. It is analagous to serotherapy which is
used for the treatment of such diseases by the use of
antibacterial or antitoxic sera. Rather curiously the
treatment of other conditions, such as the " deficiency "
diseases due to lack of vitamins, or of endocrinological
diseases, with drugs, even when their constitution is
known, is not included in the term chemotherapy.

In recent years great advances have been made in
chemotherapy, particularly in that part of it dealing
with bacterial infections and it is now possible to account
for the mechanism of the processes on a fairly certain
basis .

Chemotherapeutic substances cure by the destruc-
tion of the organism causing the disease. The action
may be directly on the organism or by stimulation of
the defence mechanism of the host or, frequently, by a
combination of both means. The organism may be
weakened, or otherwise rendered susceptible, so that the
tissue defences may be strong enough to overcome the
infection.

The application of chemotherapy is obviously more
difficult in the case of a generalised infection, or in an
infection of a deep tissue, such as the central nervous
system, than when the infection is localised or readily
accessible to the action of the drug. In such conditions

122

CHEMOTHERAPY 123

a substance which is more harmful tu the parasite
than to the surrounding cells and tissues can be applied
directly, as did Lister when he used phenol to combat
sepsis in surgery, or as is done when flav^ines are used
for surface staphylococcal infections, or sulphonamides
are dredged into wounds. When the infection is deep
seated or general the chemotherapeutic agent must
circulate in the blood or lymph in order to reach the
organisms and then the body as a whole is subject to the
actions of the drug which may be toxic. Moreover, the
substance is liable to be excreted or destroyed or in-
activated by fixation in the tissues with lowering of the
effective concentration. The difficulty is even greater
if the organisms are situated in avascular tissue or
necrotic areas or inside cells where direct access of the
drug carried in the blood is not possible.

Although a substance may be highly lethal to bacteria
in vitro it does not follow that it will be a good chemo-
therapeutic agent. Thus Koch, as long ago as 1881,
showed that amounts of mercuric chloride many times
the in vitro lethal dose when injected into guinea-pigs
had no effect on anthrax bacilli, subsequently injected.
Similarly Hata showed that an amount of methylene
blue, five hundred times that required to kill Borrelia
reciirrentis (the causal organism of relapsing fever) in
vitro had no influence on the course of the infection in
mice. The converse may also be true ; prontosil is
without effect on Streptococcus pyogenes when tested on
cultures of the organism but is a most useful drug in the
treatment of streptococcal infections.

In all cases its toxicity to the host is the limiting factor
which determines w^hether or not a given substance can
be used as a therapeutic agent. Almost always a sub-
stance which is harmful to micro-organisms is also
harmful to the cells of the host so that the choice of a
suitable drug depends on the difference in intensity
between the two actions. The ordinary disinfectants

124 BACTERIOLOGICAL CHEMISTRY

like phenol, inercuric chloride, or chlorine compounds
are as toxic to animal cells as to micro-organisms and
obviously cannot be used internally. The more damaging
the drug to the parasite compared with its toxicity to
the host the more useful is it likely to be. The ratio

toxicity to micro-organism

, often called the chemothera-

toxicity to host

peutic index, is frequently used as a measure of the value
of a drug, the higher the ratio the more useful is the
substance likely to be, other things being equal. The
chemotherapeutic index is sometimes expressed as the
ratio between the smallest amount of drug which, when
injected in one dose, will effect cure and the largest
amount tolerated by the host.

The route of injection of a drug may influence its
apparent efficacy, as a result of differences in absorption
or excretion. Thus intramuscular injection of a rela-
tively insoluble drug will produce a depot of the drug
which will maintain a more or less uniform concentration
of the drug in the circulation for a considerable time,
whereas intravenous injection is followed by fairly rapid
excretion. As an example salvarsan, when injected
intramuscularly into fowls, protects them against infec-
tion by spirochsetes for several weeks, although they
become susceptible again within six days of an intra-
venous injection. Penicillin is of little value when taken
orally because it is destroyed by the acid conditions
prevailing in the stomach, and is administered intra-
venously. On the other hand, sulphaguanidine is effective
against intestinal organisms because it is only slowly
absorbed from the gut, whilst the rapidly absorbed
sulphanilamide is almost useless for such infections, but
very effective against the bactersemia type of infection.

A micro-organism may be killed in one host by a drug
but may be resistant to the same drug in another host.

CHEMOTHERAPY 125

This is possibly due to differing reactioiLs of the host
to the absolution or excretion of the di'ug. A further
possibihty may be the possession by different hosts of
different amounts of substances inhibiting the drug, as
illustrated by the fact that rats can be protected against
streptococcal infections by pantoyltaurine, whilst mice,
which normally have a higher concentration of panto-
thenic acid in their blood, are not (see p. 149).

The development of chemotherapy can be regarded
as commencing in 1867 with Lister's use of phenol as an
antiseptic in surgery, although knowledge existed much
earlier of such traditional remedies as mercury and
iodides for sjrphilis, cinchona bark for malaria and
ipecacuanha for amoebic dysentery and although, many
years before the causes of the diseases themselves had
been elucidated, the active principle of cinchona bark
was shown to be quinine, and ipecacuanha was known
to act in virtue of the alkaloid, emetine, which it
contained.

Progress in chemotherapy was greatly hampered in
the early days by lack of in vitro methods of testing
drugs. Such methods could not be developed until
methods of culture of the test organisms were available.
Trypanosome infections in rats and mice were originally
used for testing drugs against such diseases as sleeping
sickness. In untreated animals the parasites progres-
sively multiply in the blood stream and death results in a
few days. An adequate dose of an effective compound
leads to permanent elimination of the parasite from the
blood stream, more or less rapidly depending on the
drug. In 1930 a method was devised for maintaining
tr5rpanosomes alive in vitro for about twenty-four hours
and has been of great use in investigating trypanocidal
drugs. The study of amoebecides was greatly facilitated
by Dobell and Laidlaw's in vitro method of culturing
amoeba such as Entamoeba histolytica which causes
amoebic dysentery.

126 BACTERIOLOGICAL CHEMISTRY

Protozoal Infections.— Very little progress was
achieved until Ehrlich's researches gave a stimulus to
the study of the subject. Ehrlich had observed that
some tissues were selectively stained when certain dyes
were injected into animals, whilst other tissues were
almost unaffected. In 1891 he recorded that the malaria
parasite was stained by methylene blue and could be
differentiated from the tissues of the host in this way.
This suggested that dyestuffs might be found which
would be so easily adsorbed by pathogenic micro-
organisms as to kill them without harming the host.
As a result of these investigations Ehrlich and Shiga, in
1904, showed that trypanosomes were readily stained
by the dye trypan-red,

SCNa

and that the substance cured the ordinarily fatal infection
of mice with Trypanosoma equinum. Unfortunately the
dye was effective only against acute laboratory infections
and not against the natural disease.

In 1905 atoxyl, ^^-amino -phenyl arsonic acid,

NH2<^^ NAsOgNa, was shown by Thomas to be

lethal to Trypanosoma gambiense in infected mice. This
discovery led to the production of a number of arsenical
drugs, many of which were dangerously near the toxic
limits for therapeutic use. Their action was rarely
apparent if administered in the later stages of the disease,
and frequently arsenic resistant strains of trypanosomes
were developed when cure was not effected. These drugs
were not lethal to T, rhodesiense. Further research,
however, led to the discovery in 1920 of Baeyer 205 or
germanin (among other names).

CHEMOTHERAPY

127

NH.CO.

CH.

cn.

SOgNa

I

I
CO

CO.NH

I SOgNa

NH

>C0

NH.CO.NH

SO,Na

SOgNa

which is effective against natural trypanosome infections
by T. gambiense and T. rhodesiense, but still ineffective
in the later stages of the disease. This, but not the other
disadvantages, was overcome by the use of tryparsamide,

OH

CONH2.CH2NH

>As = o , and similar drugs such

ONa

as arsanine and neocryl containing pentavalent arsenic.

Atoxyl was shown to be effective also against the
spirochsete causing syphilis. Even more effective is
salvarsan or arsphenamine,

H0<

'^^ . It was later shown that

\

NH2.HCI NH2.HCI

these substances were not themselves active
in vitro but that they were converted in the body, by
reduction and partial oxidation respectively, to deriva-
tives of phenyl-arsenoxide, <^ />-^sO, which were

highly lethal in vitro to spirochaetes and trjrpanosomes .
The therapeutic use of phenylarsenoxide derivatives in
the ordinary way is not possible since they are excreted
too rapidly to be effective unless given in doses which
would be too toxic to the host. The arsenoxide derivative

mapharsen, or mapharside.

H0<

>As0

I
NH.HCl

is used,

however, by the slow intravenous drip method to give a
short but intensive treatment for syphilis.

128

BACTERIOLOGICAL CHEMISTRY

Many compounds, ])a8ecl on the structure <jf the
effective drug, emetine, have been synthesised and
tested for treatment of amoebic dysentery, but none, so
far, has been found to exceed emetine in therapeutic
value, although some are more toxic to Entamceba
histolytica when tested in vitro.

Similar search for antimalarial drugs primarily based
on the structure of quinine,

CHOH— CH— N CH,

CH2— CH— CH.CH ^CHg

N

has led to the production of plasmoquin,

CH3.CH. (CH2)3.N : {C,H,),

CH,0

and atebrine, or mepacrine,

CH3.CH.(CHj)3.N : (C,H5)2

I
NH

I

C2H5O

CI

N

which are of considerable value. A further extension of
this search in recent years has led to the discovery of
paludrine, Ni-p-chlorophenyl-N5-mpropyl-biguanidine: —

CH,

/
>NH.C— NH— C— NH.CH

II II \

NH NH CHj

CHEMOTHERAPY 120

A theory <jf drug action different from Elu'lich\s
receptor theory led to the discovery of a group of drugs
which have proved of great value in several trypanosome
diseases. The drug synthalin, decamethylene diguanidine,
NH XH

C— NH.(CH2)io. NH— c , had been used in diabetes

NH2 NH2

because it had a similar effect to insulin in lowering the
blood sugar content. Jansco thought that it might be
effective in trypanosomiasis by lowering the blood sugar to
such an extent as to starve out the trypanosomes, in the
same way as he had succeeded in preventing trypanosome
infections in mice by inhibiting their carbohydrate
metabolism with iodoacetic acid. It so happened that
the drug was active against. trypanosomes, although not
for the reasons which led to its trial. Investigation of
drugs of similar constitution brought to light stilbamidine,
diamidino-stilbene,

NH NH

C — < >CH = CH

I ^ ^

NH2

which is the most effective drug yet known for the

treatment of kala-azar. and pentamidine,

NH NH

C

NH2

NH3

for babesia in cattle. Propamidine,
NH

' ^ -0— (CHa),— 0—

NH2

has also proved of value in trjrpansomiasis, kala-azar and

babesiasis.

To Summarise : — Trypanosomes are susceptible to the

130

BACTERIOLOGICAL CHEMISTRY

met alio -organic compounds of arsenic, antimony and
bismuth, to derivatives of quinine, to the triphenyl
methane series of dyes, to acid bis-azo dyes, such as
trypan-red and trypan -blue, to acriflavine and to the
amidines.

Leishmania are susceptible to pentavalent antimony
compounds such as stibenyl, ^^-aminophenyl sodium

ONa

stibonate,

NH2<

>Sb = 0 ,

\
OH

or stibacetin, CH3C0.NH<

ONa

/
>Sb = 0

\
OH

and to the amidines.

Amoebae are susceptible to chinoform, iodohydroxy-

SOsNa

quinoline sulphonate.

and to the pentavalent

arsenicals acetarsol,

OH N

H0<

OH

/
)As = 0

\^ ' and carbasone,
OH

NH.CO.CH,

•\ /

OH

/
As = 0

OH

Plasmodia are susceptible to quinine, plasmoquin,
atebrine, and paludrine.

Spirochsetes are susceptible to the metallo -organic
compounds of arsenic, antimony and bismuth, but are
relatively little affected by purely organic drugs.

CHEMOTHERAPY 131

Bacterial Infections. — Until 1935, apart from the
local application of acridine dyes, such as acriflavine
and proflavine, to wounds and surface lesions, developed
during the 1914-1918 war, the only bacterial disease
Avhich had shown itself susceptible to chemotherapy
was a pneumococcal infection of mice which responded
to treatment with optoquin, ethyl hydrocupreine,

CH

CHa

CH.CH2.CH3

CHj,

I II I \L

N ^^

which is also active against trypanosomes.

In 1935 Domagk subjected to clinical trial the drug
prontosil, sulphonamido-chrysoidin,
NHo

.N=N/ \SO2.NH2, ^^hich had been

discovered by Klarer and Mietzsch in 1932, and found
that it was an effective agent against streptococcal
infections, in spite of the fact that it was inactive in
vitro. This phenomenon was explained by Trefouel,
Trefouel, Nitti and Bovet in 1935, who showed that
prontosil was broken down in the body to /)-aminobenzene
sulphonamide, now known as the drug, sulphanilamide,

which was active both in vitro

and in vivo. This discovery led to a tremendous amount
of research for similar and improved drugs. Some
2,500 derivatives of sulphanilamide have been syn-
thesised, most of them by substitution on the nitrogen
atoms. In nomenclature the sulphonamide group —

132 BACTERIOLOGICAL CHEMISTRY

SO2NH2, is regarded as being at position 1, whilst the
amino group is at position 4 : —

SO,NH,.

/ ' \

\ 4 /
NH2
The following is a summary of the findings : —
N^ derivatives.

(a) Alkyl. The introduction of the methyl or ethyl
group causes little change in effectiveness.
Longer chains cause lowered activity.
(h) Isocyclic. About 180 have been prepared, none

of which have been of much value.
(c) Heterocyclic. About 250 have been synthesised.
They include the most useful members loiown.
Among them are sulphapyridine,

CH

CH CH

-\, I "

NH/ >S02.NH— C CH

N

effective against pneumococcal, streptococcal,
meningococcal, gonococcal and coli infections,
sulphathiazole, N— CH

NH,^ ^SO,.NH— C CH.

s
effective against staphylococci, and sulphadia-

CH

/%

zme, II I , which is

NH,/ \S0,.NH— C CH

^ —

CHEMOTHERAPY 133

less toxic than sulphathiazole and is active
against streptococci, pneumococci, staphylococci
and gonococci.
(d) Acyl. About 35 are known, of which sulpha-

guanidine, NHa^^ Ns02.N=C , is very

NH2
useful for intestinal diseases, including bacillary
dysentery, since it is only slowl}^ absorbed from
the gut, and sulphacetamide,

NH2<^ ^SOa.NH.COCHg,

(albucid, sulamyd) which is of value in gonorrhoea
and urinary infections .
N"^ derivatives.

About 550 have been made. It appears that only
those which can break down in the body to give sul-
phanilamide or an active derivative of it are of chemo-
therapeutic use. Prontosil, prontosil soluble,
OH
CH3CO.XH 1^^ —'N^^/ ^SOj.NHj,

proseptasine, <^ \CH2.isrH<^ \SO2.XH2,

and soluseptasine (^ \cH.CH2.CH.NH-<^ \sO2.NH2,

SOgNa SOsNa

are examples. Long chain alkyl or sulphonyl derivatives

are not broken down in this way and are, therefore,

inactive.

N^K* derivatives.

The activities of substances of this type are what
would be expected from considerations of the effect of

134 BACTERIOLOGICAL CHEMISTRY

substituents on the two nitrogen atoms mentioned above.
Succinyl sulphathiazole, sulphasuxidine,

N— CH

COOH.CH2-CH2.CO.NH/' \sO2.NH— C CH,

S

is one of the most useful, especially for intestinal infections
such as dysentery because it is poorly absorbed from the
gut and breaks down slowly with liberation of sulpha-
thiazole. Uleron,

NH2<^ ^S02NH.<^ ^SOa.N(CH3)2,

is active against staphylococci.

Tuberculous infections in guinea-pigs have been
successfully treated with promin, di-aminodiphenyl-
sulphone di glucose sulphonate,

NnH. CH. (CHOH)4.CH20H
SOjNa
.NH. CH.(CHOH)4.CHaOH
SOgNa

which is also active against tubercle bacilli in vitro.
Unfortunately it is fairly toxic to man, producing
hsemolytic anaemia. The related drug diasone, diamino-
diphenylsulphone disodium formaldehyde sulphoxylate

/\ /

NH.CHa.S02Na

S02<^ -^HaO, is less toxic and has

. ^NH.CHg.SOaNa

similar curative effect on guinea-pig tuberculosis. It will
also cure hsemolytic streptococcal infections and pneumo-
coccal infections in mice.

The Mode of Action of Chemotherapeutic Substances. —
The earliest theory of chemotherapeutic action is that

CHEMOTHERAPY 135

due to Ehrlich who suggested that the drugs were taken
up by specific chemoreceptors attached to susceptible
organisms, but lacking in cells not affected by the drug.
The subsequent action of the drugs was not described
except that they were considered not to kill the micro-
organism but to prevent multiplication, enabling the
host to deal effectively by its normal processes with the
consequently milder infection.

Later, when the importance of enzymatic processes
in metabolism began to be realised, suggestions were put
forward that drugs in general might act by inhibiting
enzyme systems and so upsetting the normal course of
events in the animal body. A number of such effects
had been observed in vitro ; cyanides inhibit oxidases,
atoxyl and quinine inhibit lipases, cocaine, atropine and
pilocarpine inhibit yeast invertase, eserine inhibits the
break down of acetylcholine by esterase, acriflavine
inhibits a hydrogen transportase system in trypanosomes,
which is not affected by cyanide.

Another type of enzyme inhibition is that due to the
presence of excess of the products of the reaction or of
substances having a constitution similar to the substrate
or to the products of breakdowTi. For example, the
breakdown of lactic acid, CH3CHOH.COOH, to pyruvic
acid, CH3CO.COOH is partially inhibited by a-hy-
droxybutyric acid, C2H5CHOH.COOH, glyceric acid,

CH2OH.CHOH.COOH, mandelic acid, (^ ^CHOH.COOH,

glyoxylic acid, HCO.COOH, or oxalic acid7H0.C0.C00H.
The action of succinic dehydrogenase in converting
succinic acid, COOH.CH2.CH2.COOH, to fumaric acid,
COOH.CH=CH.COOH, is inhibited by the presence
of malonic acid, COOH.CHg.COOH or glutaric acid,
COOH.CH2.CH2.CH2.COOH, containing the -CHgCOOH
group (see p. 190). The heavy metals such as mercury
and barium also have an inliibitory effect on many

136 BACTERIOLOGICAL CHEMISTRY

enzymes. The effect of mercuric chloride as an anti-
septic has been regarded for a considerable time as
being due to the combination of the mercury with the
-SH groups of proteins. Its inhibition of some enzymes
such as papain or invertase is held to be due to a similar
reaction.

In 1923 Voegtlin suggested that the phenyl -arsenoxides
were lethal to trypanosomes and spirochsetes because
they reacted with the sulphydryl groups of glutathone : —

SG
SHG /

R.AsO + > R.As + HaO,

SHG \

SG

and so interfered with the respiratory mechanism of the
organisms. The addition, in sufficient amount, of com-
pounds containing SH groups was capable of reversing
the inhibition of enzymes or respiration by the arsenoxides
or mercury by themselves combining with the inhibitors.
Accordingly when it was discovered that sulphanila-
mide was effective against micro-organisms in vitro only
when small inocula were used or when the medium did
not contain peptone, the suggestion was soon forth-
coming that some substance or substances contained in
peptone or large inocula were inhibiting the action of
the drug. Confirmation of this view was afforded by
Stamp in 1939 who showed that addition of killed
streptococci to a medium containing sulphanilamide
permitted even small inocula of living organisms to
survive and flourish. He succeeded in extracting the
inhibitor from streptococci with dilute ammonia solution
and obtained it as an alcohol soluble substance stable to
acid and heat ; it contained an amino group. He
regarded it as possibly a complex amino acid required
for growth or as an essential part of an enzyme system.
Similar results liave been reported for Brucella abortus
and other bacteria.

CHEMOTHERAPY 137

Woods found a similar effect with yeast extracts
and brought forward strong evidence that the responsible

substance was ^^ -amino benzoic acid, NH2<[^ NcoOH,

which he showed to have a powerful inhibitory effect on
sulphanilamide. He suggested that p-amino benzoic
acid is essential for the growth of the organism, and is
normally synthesised in adequate amounts. Sulphanila-
mide, which has a structure very similar to that of
2)-aminobenzoic acid, competes with the enzyme involved
in its further utilisation and so prevents groA\'th. Addition
of p-aminobenzoic acid overcomes the competition for the
enzymes and inhibits the action of sulphanilamide.
Very small amounts of p-aminobenzoic acid, about one
five -thousandth of the inhibitory concentration of
sulphanilamide, may suffice to reverse the effect of the
latter. He suggested that the varying sensitivit}^ to
sulphanilamide of different organisms was due to
differences in their power of synthesising 2^-aminobenzoic
acid.

The explanation that sulphanilamide intervenes at
the stage of synthesis of ^^-aminobenzoic acid by com-
bining with the synthesising enzyme seems to be wrong,
because if it were true it would be expected that the
amount of 2:>-aminobenzoic necessary to reverse the
effect would be independent of the amount of sul-
phanilamide present. This is not so, the ratio of
2:)-aminobenzoic acid to sulphanilamide being constant.

Other theories of the action of sulphanilamide have
been put forward but have been displaced by the
" essential metabolite " theory. It was, for instance,
suggested that in presence of sulphanilamide, streptococci
lost their power to produce a capsule and that, as a result,
they were much more susceptible to phagocytosis. This
hypotliesis does not account for the fact that many
normally non-capsulated organisms are susceptil)le to
sulphanilamide, nor for the in vitro bacteriostatic effect

138 BACTERIOLOGICAL CHEMISTRY

of the drug. Another explanation was that sulphanilamide
was oxidised to the hydroxyl-amino compound,

NHOH.<^ ^SOg-NHg, which was known to inhibit the

power of catalase to destroy hydrogen peroxide ;
in presence of the drug, therefore, organisms such as
streptococci and pneumococci, which are sensitive to
hydrogen peroxide, are killed by its accumulation.

2)-Aminobenzoic acid inhibits the bacteriostatic
effects not only of sulphanilamide, but also of the
other sulphonamide drugs, such as sulphapyridine and
sulphathiazole, which have the common grouping

— NH./ \sO2.NH2. Neither ortho- and meto-amino

benzoic acids, ^^-aminophenyl acetic acid,

NHj. / y CH2.COOH, nor 2)-aminophenylglycine di-

hydrochloride, HCI.NH2. ^ ^CH.COOH, ^^^^ replace

NH2.HCI
/)-aminobenzoic acid in inhibiting the sulphonamides .
On the other hand ^^-aminobenzoic acid can reverse the
effect of other drugs which have groupings similar to
sulphanilamide, such as ^^-aminobenzamide,

NH2<^ ^CCNH,, or atoxyl, NHa<^ ^AsOaH,

which slows down the respiration of Esch. coli. However,

it has no effect on the drug marf anil, NH2CH2./ "^SOa.NHa,

in vitro although it reverses its effect in vivo.

It has been found that over a wide range of concen-
trations the molar concentration of /j-aminobenzoic acid
necessary to reverse the bacteriostatic action of sul-
phanilamide is proportional to the molar concentration of

,, 1 .. ^, .. concentration of dm tr ,. , . ,

the latter. The I'atio : — rr: which lust

concentration ot /j-AB *'

CHEMOTHERAPY

139

results in bacteriostasis is known as the antibacterial
index, or bacteriostatic, constant. It varies, with any
particular drug, from organism to organism, and with a
particular organism from dnig to drug. These variations
in value are proportional to the potency of a drug against
the organism, as measured by the minimum bacterio-
static concentration in vitro. These findings are illustrated
in the following tables, from the work of Wyss, Grubaugh
and Schmelkes (Proc. Soc. Exp. Biol Med. 49 (1942) 618).

Table 9

Concentration of sulphonamides permitting 50 per cent, of the maximum
growth in 16 hours.

Medium 1

Medium 2

Sta2)h. aureus

E. coli

E. coli

Sulphanilamide -
Sulphaguanidine -
Sulphapyridine -
Sulphacetamide -
Sulphadiazine
Sulphathiazole

mg. per 100 ml.
140
120

30

20

0-6

0-3

mg. per 100 ml.
150
16-2

2-9

2-8

0-65

0-65

mg. per 100 ml.
5-8
5-8
0-61
0-57
006
006

Table 10

Effectiveness of drugs in overcoming ^-aminobenzoic acid.

^

taph. aureu.

s

JEsch. coli

Mol.

Mol.

Amount

p-Amino-

ratio

p- Amino-

ratio

of drug

benzoic

Drug/

Efficiency

benzoic

Drug/

Efficiency

acid

p-A£

acid

p-AB

Mg. per

Mg. per

Mg per

100 ml.

100 ml.

100 ml.

Sulphanilamide -

50

0-0086

4660

1

0-012

3330

1

Sulphaguanidine

50

0-0070

4510

1

0-0081

3960

0-8

Sulphapyridine -

5

0-0066

1416

11

0-0061

450

7

Sulphacetamide -

5

0-0060

934

9

0-0060

634

Sulphadiaaine -

5

0-030

92

51

0-064

43

78

Sulphathiazole -

5

0-050

53

88

0-065

41

81

140 bacteriological chemistry

Table 11

Neutralisation of /j-aminobeiizoic acid in prt-sence of various bacteria.

Molecular Ratio

Efficiency

Sulphanilamide/

Sulphathiazo]e/

Sulphathiazole/

7)-aminobenzoic

2?-aminobenzoic

Sulphanilamide

acid

acid

E. coli

2000

27

74

A. aerogenes

3220

45

72

Staph, aureus

4660

53

88

Ps. ceruginosa

13350

184

73

Sal. typhimurium-

6650

92

72

L. acidophilus

8000

133

60

Prot. vulgaris

4000

55

73

Table 1 1 shows that the relative efficiency of the two
drugs in inhibiting the use of p-aminobenzoic acid by
various organisms is constant, which means that the
drugs act in the same way in preventing the growth of
all the organisms. This confirms the suggestion by
Woods that different sulphonamides are more effective
against some bacteria than against others, because the
microbes have different abilities to synthesise ^j-amino-
benzoic acid.

It is clear from Table 10 that the sulphonamide
drugs are widely different in their ability to compete
with 2)-aminobenzoic acid in the enzyme system involving
the latter, sulphathiazole being some 80 times as effective
as sulphanilamide. This has been explained by Bell and
Roblin as due to the closeness of resemblance of the
drug to the ^^-aminobenzoic acid ion in molecular structure
and distribution of electric charges. That the effective-
ness of the various sulphonamide drugs is closely related
to their degree of ionisation is shown by the following
values taken from C. L. Fox and H. M. Rose {Proc. Soc.
Exp. Biol. Med. 50 (1942) 142) :—

themotherapy
Table 12

141

Mill.

Min. amt.

effective

Acid

Concn. of

of p-amino-

Ratio

concentra-

dissocia-

% ionised

ionised

benzoic

Ratio

ionised

tion of

tion

at pB. 7-0

drug at

acid

drug/p-AB

drug/p-AB

drug

constant
pKa.

pH7-0

required
to inhibit

Mx 10-«

M X 10-«

M X 10-«

Sulphanilamide -

2500

10-5

0-03

0-71

0-5

5000

1-4

Sulphapyridine -

20

8-5

3-4

0-68

0-5

40

1-4

Sulphathiazole -

4

C-8

61-6

2-46

0-5

8

4-9

Sulphadiazine -

4

6-4

80-0

3-2

U-5

8

G-4

It will be seen that the effective drug concentration
is inversely proportional to the degree of ionisation and
that the amount of ^^-aminobenzoic acid required for
inhibition of the drug is very nearly proportional to the
amount of drug in the ionised state. ^j-Aminobenzoic
acid is completely ionised at ]:>H 7. Its ion may be
represented as at A in Fig. 5.

H

H

'

H

H

I

H

H

\ /

\ /

\ /

N

N

N

A

A

1

1

/\

1 1

6-'

o

6-

A

' H

\/

\/

\ \/

c

T^\^

^\

R ./ %

R /^

0

u-

[

0

0

0

0

2-3 A 2-4 A

A B

Fig. 5

The electrons belonging to the sulphur atom of a sul-
phonamide (C in Fig. 5) are attracted by the oxygen

142 BACTERIOLOGICAL CHEMISTRY

atoms and the pull is transmitted to the electrons on
the amide nitrogen atom, which, consequently, exerts
less attraction for the hydrogen atom which, accordingly,
becomes capable of ionisation. It behaves as a very weak
acid with a dissociation constant of Ka = 3-7 X IQ-^^ in
the case of sulphanilamide (B in Fig. 5). In the ionised
state the electron pair which formed the covalent bond
with the hydrogen atom is available to increase the
electronegative character of the — SO2 group, although
the effect is not very great because the degree of ionisation
is low. The combined effect is approximately equivalent
to the ionised — CO.O" of the carboxyl group. Substitu-
tion of a group R on the amide group has two opposing
effects : (1) a competition with the — SOg group for the
electron pair, which decreases the resemblance to the
charge distribution on the carboxyl ion and so reduces
the activity of the drug ; this effect is considerable when
both hydrogen atoms in the amide group are substituted
so that ionisation is impossible ; the second effect (2) is
observed when the hydrogen atom is present, since then
ionisation is increased because the extra competition for
the ions by the substituent group reduces still more the
attraction of the nitrogen atom for the hydrogen atom.
As a result the more electronegative the substituent
group, the greater is the acid strength of the derivative,
with corresponding increase in activity. There is,
however, an optimum degree of electronegativity since
increase beyond a certain value involves too great com-
petition for electrons, to the extent that they are with-
drawn from the — SO2 group to the R group with loss of
similarity to the carboxyl ion ; that is, the activity of
the drug is lowered. It is possible, therefore, to predict
the activity of a new derivative from a knowledge of
the electronegative character of the substituent group R.
These effects are illustrated by Table 13 and Fig. 6
which are taken from Bell and Roblin's paper (J.A.C.S.
64, (1942) 2905).

chemotherapy
Table 13

Relation between Acidity and Activity of Sulphonamides :-

,R

NH2<

143

COMPOQND

/j-Aminobenzoic acid -
Sulphanilamide -
N^-Methylsulphanilamide
N^-Phenylsulphanilamide

Sulphapyridine
Sulphathiazole

iSulphadiazine
Sulphathiadiazole

Sulphacetamide -
N^-Chloracetylsulph-

anilamide
N^-Ethylsulphonyl-

sulphanilamide

R

H-
CH3-

~N

I r

N N

CH3CO-

Ka

2-1 X 10-5
3-7 X 10-11
1-7 X 10-11
2-5 X 10-10

3-7 X 10-9
7-G X 10-8

3-3 X 10-'

1-7 xlO-5

4-2 X 10-8
1-6 X 10-4
7-9 X 10-4

Per-
cent-

Minimum
molecular

pKa

lonisa-

tion

at

pKl

concen-
tration for
bacterio-

stasis
Ch X 10^

4-68

99-0

10-43

0-03

20-0

10-77

0-01

30-0

9-6

0-25

30

8-43

3-5

0-6

712

430

008

6-48

77-0

0-08

4-77

99-0

0-6

5-38

98-0

0-7

3-79

100-0

100

310

100-0

1000-0

Since the curve shows a maximum, which corresponds
very nearly with the compounds sulphadiazine and

144

BACTERIOLOGICAL CHEMISTRY

O

sulpliathiazole (see |)p. 132, J 33) it in probable that the
most active of the siilphonamide drugs are already known.
Possibilities of more therapeutically useful sulphonamides
lie in eliminating the objectionable side effects, toxicity,
nausea and so on, and in appropriate modification of
solubility and rates of absorption. Thus, although

o

3 •

J--f

,

/\

a J

/ ^

V

\

^

J

ii Sulphacetamide
1) Sulphadiazino

\

c Sulphathiazole
(I Sulphapyridinn

I

c i^iilphamlamiilo

__- 1

1 L.

— 1 1 _ —

2 4 6 8 10 12

pKa
Fig. 6
sulphadiazine is at the peak of activity, it has a low
solubility, especially in acid solution, and tends to
crystallize out in the urinary tract when the urine is
acid or of small volume, as in hot countries. Sulpha-
methazine, 2-sulphanilamido-4:6-dimethylpyridine, is

CHEMOTHERAPY 145

about ten times as soluble as sulpliacliazine at j.>H 7 and

37 °C. and, although it has about twice the toxicity, it

would probably be of greater use in the tropics. To take

another example sulphamerazine, sulphamethyldiazine,

N— CH

/ \ ^ %

NHg/ ^SOa.NH— G CH, IS as active as

X = C.CH,
sulphadiazine, but is less readily eliminated from the
body so that an adequate concentration in the blood
could be obtained by less frequent administration.

It has been claimed that the bacteriostatic power of
sulphonamides can be reversed by adenine, and by
methionine. Adenine sulphate, when administered to
mice infected with &ir. j)yogenes in a dose of 0-8 mg.
per gram prevents the chemotherapeutic effect of 2 mg.
per gram of sulphanilamide or of 4 mg. per gram of
sulphadiazine, sulphapyridine or sulphathiazole, being
more effective than the same amount of ^-aminobenzoic
acid. Guanine and uracil had no such anti-sulphonamide
action. Adenine is an essential metabolite for strepto-
cocci, forming part of the codehydrogenase and co-
phosphorylase systems, and it is considered that the
sulphonamides may interfere with these enzjone systems.
Methionine, CH3.S.CH2.CH2.CH.NH2.COOH, inhibits the
effect of sulphadiazine on E. coli in synthetic medium.
This property of methionine is eliminated by urea which
also reverses the effect of p-aminobenzoic acid on sul-
phanilamide, and increases the potency of sulphadiazine
and sulphanilamide, possiblv bv increasing the penetration

of the drugs into the tissues. Guanidine, ^C=NH,

and thiourea, ^C=S, are even more active than

urea.

14G BACTERIOLOGICAL CHEMISTRY

Following the lead given l)y the discovery of the
^-aminobenzoic acid-sulphonamide inhibition mechanism
of drug action a number of other systems have been
investigated with analagous results. In some cases it
has been possible to devise a substance which should
have antibacterial properties in virtue of its close chemical
relationship with a compound participating in the
metabolism of bacteria. As examples may be quoted

the effect of pyridine-3-sulphonic acid, I ^ ' ^^^

/\

N

its amide, L J , on the growth of Staph, aureus

N

and Proteus vulgaris which require nicotinic acid for
their metabolism. Pyridine- 3 -sulphonamide inhibits the
growth of these organisms in presence of ordinarily
adequate amounts of nicotinamide, and the effect is lost
on increasing the amount of nicotinamide present.
E. coli does not require nicotinamide as a growth factor
and is only little affected by pyridine-3-sulphonamide
at a concentration of 10~^ molar, but the inhibition is
completely reversed by nicotinic acid or the amide.
Pyridine- 3 -sulphonic acid, however, at a concentration
10~2 molar completely inhibits the growth of E. coli
and the effect is not reversed by addition of nicotinic
acid, nicotinamide or co -enzyme.

It is possible that the greater potency of sulphapyridine
compared with that of sulphanilamide is due to its
effect on nicotinic acid metabolism in addition to that
on p-aminobenzoic acid. A similar inhibition by sulpha-
pyridine is observed on the respiration of the dysentery
bacillus stimulated by co -enzyme I (diphosphopyridine

CHEMOTHERAPY 147

nucleotide) or by nicotinamide. Sulphanilamide, sulpha-
thiazole, and sulphapyridine all inhibit the growth of
Sonne's bacillus, and this effect, but not the respiratory
inhibition, is reversed by ^^-aminobenzoic acid. Sulpha-
thiazole is said to have a similar inhibitory effect on
CO -enzyme or nicotinamide stimulated metabolism. The
explanation may be that sulphathiazole and sulpha-
pjrridine are isosteric and accordingly could replace one
another in adsorption on the co -enzyme or that sulpha-
thiazole may act on a different part of the metabolism
chain, involving decarboxylation (see p. 43). The primary
action of the sulphonamides, which is reversed by
2)-aminobenzoic acid, does not affect the respiration of
the cells. The secondary effects due to the pyridine
or thiazole groups, for instance, which are not reversed
by 2>-aminobenzoic acid, are usually concerned with
respiratory processes.

Fildes showed that the growth of Esch. coli and of
Eberthella typJiosa was inhibited by indole -acrylic acid,

/\| -CH =CH.COOH

i{ 11 , but not by other indole

NH
derivatives and that the inhibition was removed by the

addition of traces of tryptophane, ^\ CH2.CH.COOH

II 11 Tshr
NH

^\

but not by indole, II II . It is, therefore, assumed

NH

that the indole -acrylic acid interferes with the synthesis
of tryptophane from indole. By analogy of the equiva-
lence of a-naphthyl-acetic acid and indole-acetic acid

148 BACTERIOLOGICAL CHEMISTRY

aa plant hormones it was expected that p-naphthyl-

/ Y^l CH =CH.COOH,
acrylic acid, would inhibit

the growth of E. coli and that the effect would be reversed
by tryptophane. Inhibition is brought about by a
concentration of 0-0002 M p-naphthyl-1 -acrylic acid and
is reversed by 0-0002 to 0-00004 M tryptophane. Styryl-

acetic acid, ^ ^CH^CH.CHg.CGOH, and cinnamic

acid, <^ ^CH=CH.COOH, have similar but much

weaker effects, whilst dihydro -cinnamic acid, benzoic
acid and fumaric acid, COOH.CH=CH.COOH, have no
such effect.

Pantothenic acid is an essential metabolite for many
organisms and it has been shown that the growth of such
organisms is inhibited by the addition of pantoyltaurine,

CHaX

CH3— C.CHOH.CG.NH.CHa.CHa.SOaH, the Sulphonic

CH2OH /

acid analogue of pantothenic acid. For Str. 'pyogenes
about 500 times as much pantoyltaurine as there is
pantothenic acid present is required to cause inhibition.
The addition of pantothenic acid reverses the effect.

Pantoyltauramide, CHgX

•^ CH3— C.CHOH.CO.NH.CHj.CHa.SOa.NHj,

CH2OH /

has a similar effect. Pantoyltaurine can also inhibit the
growth of yeast. Mixtures of taurine, NHg.CHg.CHa.
SO3H, and ay-dihydroxy- p p-dimethyl-butyrolactone,

CH3\

CH3--C.CHOH.CO,

CHj/ or the compounds alone had no

inhibitory effect on L. arahinosus whicli is inhibited by
pantoyltaurine. N-])antoyl- p-aminoetliyl tliiol,

CHEMOTHERAPY

149

CH3-C

CHjOH /

his (pantoyl- p-aminoethyl) disulphide,

CH3—C.CHOH.CO.NH.CH2.CH2I2SJ,
CH^QH / j

are about equal in their activity against L. arahinosus
and Str. jnjogenes, in vitro and in vivo respectively.

It has been found possible to protect rats against
many thousand lethal doses of Str. pyogenes by frequent
subcutaneous injections of panto yltaurine in amount

Table 14

Ceganism

A

DDENDA

Growth

Inhibitor

Metabolite

Sir. pyogenes
E. coli

0

Sulphanilamide
3 X 10-* M

0
0
/(-Aminobenzoate,

10-^M
Pantothenate,

10-' to 10-*M
Nicotinamide,

10-' to 10-*M

0

-f
0
0

Staph, aureus '

0
Pyridine -3-
sulphonamide
10-2M

0
0

jj-Aminobenzoate,

lO-'M

Pantothenate,

10-' to 10-*M

Nicotinamide,

lO-s.M

0

0
0

4-

Str. pyogenes
Dip. pneumonice -
C. diphtherice

0
Pantoyltaurine

0
0

jy-Aminobenzoate,

lO-'M

Pantothenate,

10- «M

Nicotinamide,

10-' to 10-*M

+
0

0

+
0

150 BACTERIOLOGICAL CHEMISTRY

sufficient to counteract the pantothenic acid present in
the blood. Mice, which normally have a considerably
higher content of pantothenic acid in the blood, are not
protected by such treatment because enough pantoyl-
taurine cannot be administered. Since human blood
has somewhat less pantothenic acid than rat blood it
should be possible to protect man against streptococci and
C. diphtherice, which is also sensitive to pantoyltaurine.

The specific effects of metabolities on certain inhibitors
is illustrated in Table 14, due to Mcllwain, which shows
that inhibition is reversed only by the corresponding
metabolite.

By testing the effect of a number of drugs, such as
sulphathiazole, containing the thiazole ring, it has been
shown that they can interfere with the decarboyxlation
of pyruvic acid (see p. 277), by Staph, aureus, E. coli,
yeast and by a carboxylase preparation from yeast.
Sulphathiazole is most effective against Staph, aureus
and E. coli, whilst sulphanilamide and sulphapyridine
have very little effect. The most active compound was
sulphanilamido-5-ethyl-4-thiazolone.

NH2

It will be remembered that co -carboxylase is aneurin
diphosphate (p. 43),
N=CH

CH,

.C

C— CHo-

-NH-

-C.CH3

OH

1

1

il

II

/

N:

= C.NH2

CH

C-

/

-CH2

.CH2

.O.P = 0

\

OH,

0—

P = 0

OH
which contains a thiazole ring. This may, in part,
explain the greater potency of sulphathiazole, as com-
pared with sulphanilamide, against many organisms.

CHEMOTHERAPY 151

Those Species of bacteria, yeasts and fungi which
require aneurin for their maximum growth are inhibited
^y pyi'ithiamine, the pyridine analogue of aneurin,

jT QTT CH3 CHj.CHaOH

II II J '-^

CH3.G C— CH2— Nf ^

I I I ^=='^

N=C.NH2 Br

whilst other organisms are not inhibited. The inhibition
is overcome by the addition of aneurin. The more
exacting a species is in its requirement for aneurin the
more readily is it inhibited by pyrithiamine. The ratio
of pyrithiamine to aneurin is about 700 for Staph, aureus
and 20,000 for Esch. coli. Species requiring intact
aneurin are much more sensitive than those requiring
only the pyrimidine moiety or those requiring both the
pyrimidine and thiazole portions of the aneurin molecule.
The organisms which are not affected by pyrithiamine
do not synthesise increased amounts of aneurin in its
presence in the same way that sulphonamide resistant
organisms synthesise greater quantities of ^^-aminobenzoic
acid. By growing the yeast, Endomyces vernalis, in the
presence of small amounts of pyrithiamine a strain
resistant to 25 times the normally inhibitory concentra-
tion has been developed. It still required aneurin, or its
p^Timidine moiety, as a gro\^i;h factor but, in their
absence, was capable of converting pyi-ithiamine into the
P3T:'imidine part of the anenrin molecule.

The respiration of Plasmodia species causing malaria
is stopped by the inhibitory action of quinine, plasmoquin,
or atebrine on the hydrogenase and cytochrome oxidase
systems involved.

Drug Resistant Strains. — During the investigation of
chemotherapy it very soon became apparent that micro-
organisms developed resistance to drugs. In fact most
organisms which have survived treatment by a drug
became resistant to its action. Thus Ehrlich showed

n

162 BACTERIOLOGICAL CHEMISTRY

that trypanosomes became resistant to dyes and to
arsenic compounds. He showed that such organisms
no longer took up the arsenical drug, or were not stained
by the dye as were susceptible organisms. He explained
this as being due to loss of affinity of the specific receptor
in the organism for the drug. Trypanosomes which were
resistant to atoxyl were also resistant to (and unstained
by) dyes of the acridine, oxazine, and thiazine series but
not to those of the trypan-blue type nor to those of the
triphenylmethane series. Ehrlich noted that although
trypanosomes might be resistant to atoxyl or tryparsamide
they were not resistant to arsenophenyl glycine. It has
since been shown that tryparsamide resistant organisms
are not resistant to phenylarsenoxide or derivatives of it
containing carboxyl groups and that they take up the
compounds in the same way as non-resistant strains.
It is considered that the active compounds are either
readily water soluble or lipoid soluble and therefore
easily penetrate the parasite. Trypanosomes may con-
tain up to 60 per cent, of lipoid substances. The lethal
arsenic atom can then come into contact with the suscep-
tible groups in the organism and cause its death. The
arsenicals or dyes to which the organism is resistant
fail to act because they are not taken into the organism
or get held up on some non-vital structure. Arsenoxides
react very readily with sulphydryl groups and may kill
the organism by inhibiting essential enzymes which
contain SH groups, in the same way as mercury does.

Strains of pneumococcus become resistant to sulpha-
pyridine if the dosage of the drug used for treatment
has been inadequate. They can also be produced in
vitro by growing the organism in media containing
gradually increasing amounts of sulphapyridine. Such
resistant organisms have the same morphology, virulence
and immunological properties as the parent strains.
Usually a bacterium which has become resistant to one
of the sulphonamide drugs is also resistant to other

CHEMOTHERAPY 153

sulphoiiamide drugs, but not to drugs of different types
such as the dyestuffs of the acridine series or the
propamidines. It has been found that sulphonamide
resistant organisms have acquired the property of in-
creased production of ^^-aminobenzoic acid so that their
groAvth is no longer inhibited by the drug.

Strains of hsemolytic streptococci and of C. diphtherice
have been produced which are resistant to the action of
pantoyltaurine. Some such strains also occur naturally.
These strains are sensitive to the sulphonamides, and
sulphonamide resistant strains are sensitive to pantoyl-
taurine. The varying resistance of naturally resistant
strains of C. diphtherice to pantoyltaurine is associated
with their ability to convert p -alanine into pantothenic
acid, instead of having to be supplied with the latter,
which reverses the effect of pantoyltaurine. Mcllwain
has shown that this mechanism cannot apply to Sir.
pyogenes since many naturally resistant strains and all
experimentally produced resistant strains still need to
be supplied with pantothenic acid and cannot utilise
[3-alanine instead. Resistant strains of Sir. pyogenes and
strains of Proteus morganii, Leuconostoc inesenteroides,
Lactobacillus and propionic acid bacteria become sus-
ceptible to pantoyltaurine when salicylate is added. This
is explained on the assumption that salicylate acts on
the same groups as does pantoyltaurine, that is on
enzymes involved in pantothenic acid metabolism. It
should be pointed out that pantothenic acid antagonises
the action of salicylate. If this is true, pantoyltaurine
resistant strains of Str. pyogenes differ from susceptible
strains in having alternative processes for utilising
pantothenic acid which are not blocked by pantoyltaurine,
although they may still be inhibited by salicA^late. Re-
sistant strains of a normalh^ susceptible organism are
found occurring naturally and must have arisen by a
means other than " training " in tlio presence of tlio
drug. A clue to the mechanism by which this can happen

154 BACTERIOLOGICAL CHEMISTRY

is found in the fact that pantoyltaurine resistant strains
of G. diphtherice can utilise 3 -alanine instead of panto-
thenic acid. If susceptible strains were grown in the
presence of large amounts of p-alanine, or were gradually
trained to do without pantothenic acid, it would be
expected that they would become resistant to pantoyl-
taurine and in fact this has been shown to take place.
The converse of this can also happen ; if 0. diphtherice
is repeatedly subcultured in media rich in pantothenic
acid it becomes progressively more exacting in its need
for pantothenic acid and at the same time more sus-
ceptible to pantoyltaurine. This behaviour is very
similar to that postulated by Knight to account for the
more complex demands of the parasitic organisms and
viruses as compared with those of saprophytic and
autotrophic bacteria (see Chapters VI and VII). In
general drug resistance, once acquired, is stable through
many generations of subculture on ordinary media. The
development of drug resistance by " training " appears
to take place in two stages. In the early stages the
resistance is easily reversed, does not survive continued
sub-culture on ordinary media, and is specific to the drug
used, related dnigs being active. The resistance is
probably due to the stimulation of a reserve, less efficient,
growth mechanism which is present in the organism
but normally plays only a minor role in metabolism.
In the early resistant phase it serves to tide over the
organism, until it has elaborated the final alternative
mechanism which confers permanent resistance on the
organism. This permanent mechanism may be a new
way of by-passing the mechanism normally inhibited by
the drug, or it may involve the development of an enzyme
system which can synthesise enough of an antagonist
to the drug to overcome its effects.

The ease with which it is possible to produce resistance
to different drugs by " training " varies considerably.
Streptococci and staphylococci become readily resistant

CHEMOTHERAPY 155

to .sulplioiiaiuides and to penicillin, l)ut. resistance to
the acridine dyes and propamidine is harder to induce.

are also resistant to propamidine,

NH NH

'2

and vice versa.

Staphylococci resistant to proflavine, I I I I

NH.X/X /X/NHj

N

CHAPTER XI
ANTIBIOTICS

THE term antibiotics is used of those substances
produced by micro-organisms which have an an-
. tagonistic effect, usually specific, on other organisms.
Antibiosis results, therefore, from the growth of an
organism evolving an antibiotic, in presence of another,
susceptible, organism, in contrast to symbiosis which
occurs when two micro-organisms grow together with
mutual benefit (see pp. 102, 109).

The antagonistic effect of some micro-organisms on
others has been known for many years. The anthrax
bacillus, for example, was shown by Pasteur to be in-
liibited by aerial contaminants ; lactic acid bacilli will
overgrow CI. hutyricum in the butyl alcohol/acetone
fermentation, because the large amount of lactic acid
which they form produces conditions under which CI.
butyricum cannot survive (see p. 315). Substances like
lactic acid which act in a non-specific way by altering
the physical condition of the environment are not, as a
rule, called antibiotics ; the expression antibiotic is
reserved for substances which act specifically on a few
species of organisms and which are usually active in
very small amounts. That is they have an action which
is almost the reverse of that of growth factors, probably
by interfering with enzyme systems involved in meta-
bolism. Antagonism also results from other causes such
as the " swamping " of a slow growing organism by a
fast growing one which competes successfully for the
available nutrients, or by the production by one organism
of conditions of oxidation-reduction potential unfavour-

ANTIBIOTICS 157

aljle to another which then dies out. We are concerned
here only with bacteriostatic or bactericidal substances
produced by micro-organisms. vSoil is a rich source of
micro-organisms which have antibiotic properties.

A considerable number of antibiotic substances is
now known. The more important of them will be dealt
with in turn.

Actinomycetin. — Many species of Actinomycetes pro-
duce substances which are lytic to living and dead
bacteria. Thus Actinomyces alhus yields the water
soluble, thermolabile protein-like material actinomj^cetin
which will lyse living or dead Gram -positive organisms
and dead Gram -negative organisms. It can be precipi-
tated by alcohol and appears to have the properties of a
proteolytic enzyme. A similar substance has been
isolated from A. violaceus, which, although heat stable,
otherwise resembles the enzyme lysozyme w^hich occurs
in egg white and tears and lyses most non-pathogenic
bacteria and also streptococci and staphylococci. Its
substrate is a mucopolysaccharide which it breaks down
to an acetylated amino -he xose and a ketohexose.

Actinomycin. — The brown pigmented soil organism
A. antihioticus is very active against almost all bacteria,
and fungi, especially Gram -positive bacteria. An active
substance, actinomycin, was isolated from it by extraction
with ether and fractionated into actinomycin A, soluble
in alcohol and in petrol and giving a clear aqueous
solution, and actinomycin B which is soluble in petrol,
difficultly soluble in alcohol and gives a turbid suspensino
in water.

Actinomycin A is a bright red crystalline polycyclic
nitrogen compound, C, 59-0 per cent. ; H, 6-68 per
cent. ; N, 13-35 per cent. ; m.p. 250°C. (with decomposi-
tion), [a]u^ — 320° It has a molecular weight about 80 0.
It hgis the properties of a reversible oxidation-reduction
indicator and is probably of quinoid structure. It is

158 BACTERIOLOGICAL CHEMISTRY

tlieriiiosiable. It is very strongly bacteriostatic to iriaiiy
Gram -positive organisms even at a dilution of 1 in 10^.
Gram -negative bacteria are usually less sensitive (dilu-
tions of 1 in 5000 to 1 in 10^ being necessary for bacterio-
stasis) but there is no clear dividing line. It is only
slowly bactericidal. Actinomycin A inhibits the fibrino-
\ytic activity of cultures or filtrates of hsemolytic strepto-
cocci and the coagulase activity of staphylococci. It is
highly toxic to animals, when injected intraperitoneally,
intramuscularly or intravenously.

Actinomycin B is a colourless compound which is
only slightly bacteriostatic but highly bactericidal to
Gram -positive organisms at concentrations of 1 mg. in
100 ml. Gram-negative bacteria are more resistant. It
is also highly toxic to animals.

Actinomycin B predominates in young cultures of
A. antibioticus and Actinomycin A in old cultures.

Aspergillic Acid. — Aspergillus f lavas, when grown as
a surface culture on a peptone medium gives yields of
250 to 400 mg. of crystalline aspergillic acid per litre of
medium. Aspergillic acid is a monobasic, amphoteric
acid, C12H20N2O2, m.p. 93°C., [a]D+14°. It can be dis-
tilled in steam or in vacuo without loss of activity and
is stable to acid and to alkali. When grown on a peptone
medium containing 2 per cent, of brown sugar A. flavus
gives a closely related substance, C12H20O3N2, containing
one oxygen atom more than aspergillic acid, and having
m.p. 149°C, [ajo + 42° and about one tenth the activity
of aspergillic acid. The substance isolated by Glister
from a species of Aspergillus related to, but not identical
with A. flavus, active against Esch. coli, Eherthella
typhosa. Salmonella paratyphi, Sal. schottmillleri, Shigella
dysentericB and Vibrio comma, as well as against Gram-
positive organisms, at a dilution of one in 200,000 is
aspergillic acid. It is bacteriostatic to Gram-positive
and Gram-negative bacteria, e.g., streptococci, staphylo-
cocci, pneumococci, Esch. coli, Aerobacter aerogenes, in

ANTIBIOTICS 150

concent latiuiis of 1 in 100,000 to I in 400,000. It is
bactericidal at dilutions of 1 in 25,000 to 1 in 50,000.
It is relatively highly toxic to animals. It will not
protect mice against infection with lisemolytic streptt)-
cocci or pneumococci but prevents the lethal action on
mice of gonococci suspended in mucin solution and
saves guinea-pigs from the action of gas gangrene
organisms. Its antibacterial activity can be measured
by its inhibition of the luminescence of Photobacterhmi
fischeri. A. flavus when groA\TL as a submerged culture
in agitated Czapek-Dox medium does not produce
aspergillic acid but a substance w^hich is very similar
to or identical with penicillin in chemical and biological
properties (see p. 171 et seq.).

Citrinin. — PenicilUum citrinum when grown on Czapek-
Dox or Raulin medium produces citrinin,

C C.OH luuf.. ,oD - - '

I I

CH,.(! (J

^9 ^C.CUOH

I I

CHg.CH 0

which can be precipitated from the medium, in yields of
about 2 g. per litre, as a yellow microcrystalline substance
])y the addition of hydrochloric acid. It is also formed
by Aspergillus candidus. It has m.p. 168° (decomp.).
Its sodiimi salt is soluble in w^ater. The culture filtrate
(containing about 2 g. /lit re of citrinin) is inhibitory to
Staph, aureus in dilutions of 1 in 160 to 1 in 320. Citrinin
itself is bacteriostatic to Gram -positive and Gram-
negative organisms in concentrations of 1 in 9000 to 1 in
30,000, the Gram-positive organisms, in general, being
the more sensitive.

160 BACTERIOLOGICAL CHEMISTRY

Clavacin. — Aspergillvs ciavafus when gn)\Mi on
Czapek-Dox medium gives antibacterial filtrates from
which clavacin can be isolated by extraction with ether
or chloroform or by adsorption on charcoal followed by elu-
tion with ether. It is a relatively stable substance even
in strongly acid solution. It is bactericidal in concentra-
tions of 1 in 10^, to Gram -positive and Gram -negative
organisms. It has also been called clavatin.

At about the same time Penicillimn claviforme was
shown to yield an optically inactive, colourless, crystalline
substance, m.p. 110°C. which could be extracted from
culture filtrates with chloroform. It was stable to boiling
dilute acid, but not to alkali, nor to boiling in neutral
solution. It was bactericidal to pathogenic Gram-positive
and Gram-negative bacteria and killed leucocytes at a
dilution of 1 in 800,000. It is lethal to mice (0-25 mg.
intravenously, 2 mg. subcutaneously or 2-5 mg. per os).
It was given the name claviformin.

Later Aspergillus gigmiteus, when grown on a medium
containing 4 per cent, of glucose, 0-1 per cent, of sodium
nitrate and 0-1 per cent, of potassium dihydrogen phos-
phate, was also shown to yield claviformin. The substance
is also a prduct of the growth of a species of Gymnoascus.

A fifth mould, Penicillium patidmn, produced a
colourless, crystalline substance, given the name patulin,
when grown on Raulin-Thom medium, which could
be extracted by ether, or ethyl acetate. It was found
to be inhibitory to Gram -positive and Gram -negative
organisms at concentrations of 1 in 30,000 to 1 in 80,000.
It was shown to be anhydro-3-hydroxymethylene-
tetrahydro- Y-p3a'one-2-carboxylic acid,
o
II

CHa C =CH.

I I >o .

CHa CH.CO ^

ANTIBIOTICS 161

wliiuli uii Ircatinciit with dilute alkali or on boiling in

water gives the acid, ^^'H, C=CHOH. Patulin is toxic

I I

CHa CH.COOH

\o/

to mice and rabbits in doses of the oixier of 0-25 to 0-5
mg. per 20 g. body weight. In concentrations of 1 in
2000 it is inhibitory to phagocytosis. Conflicting reports
have been published about its efficacy in curing the com-
mon cold. The result probably depends on the particular
organisms concerned in the secondary stages of the
cold, some being susceptible to patulin and some not.
The primary virus stage of the cold appears not to be
affected by patulin.

Since clavacin is inactivated by excess of 8H com-
pounds it is possible that it exerts its action by inhibiting
SH-containing essential metabolites or bacterial enzymes.

It has been shown by chemical, biological and X-ray
evidence that clavacin, claviformm and patulin are
identical compounds.

Clavatih. — This substance is identical with clavacin.

Claviformin. — This substance is identical with clavacin.

Flavacidin. — This substance, produced by A. flavus
in deep, agitated, aerated cultures in a modified Czapek-
Dox medium, is very probably identical with penicillin.

Flavicin. — Aspergillus flavus when grown on a modi-
fied Czapek-Dox medium containing corn steep liquor
gives rise to a bacteriostatic substance which can be
extracted with i^opropyl ether. The purified material is
active against the Gram -positive organisms, staphylo-
cocci, streptococci, C. diphthericE, smd B. aiithracis at con-
centrations of 0-006 to 0-008 mg./ml. Gram-negative
organisms are much less sensitive, E. typhosa, Shigella

162 BACTERIOLOGICAL CHEMISTRY

dysenteriw and Vibrio cotunba l>eing inhibited by 0-8
mg./ml. Flavicin is bactericidal as well as bacteriostatic.
When injected in small doess, 50 mg./kg. of body weight,
it protects mice against infection by Type I pneumococci.

Flavicin resembles penicillin in being a soluble acid
which is unstable in acid solution. It differs from
penicillin in being more active against C. diphtherice,
B. anthracis and Brucella abortus.

Fumigacin. — Aspergillus fumigatus produces, in syn-
thetic media, in the first few days of growth the antibiotic
fumigacin, which is readily soluble in chloroform or
ethanol and to a limited extent in ether or water. It is
precipitated from alcoholic solution on cooling as colour-
less long, slender, needle-shaped crystals. Fumigacin
has m.p.220°C. and [a]D= -132° (c=0-4 in chloroform),
contains neither nitrogen nor sulphur, is weakly acidic,
and gives a methyl ester corresponding to the formula
C29 H38_4o07 for the original substance. The silver salt,
liowever, corresponds to C32H4403Ag. Fumigacin contains
a lactone group in addition to the carboxyl group and on
treatment with alkali gives a crystalline, inactive sodium
salt. Its general properties agree with those of helvolic
acid (see p. 167) which has a formula C32H44O8, corres-
ponding to that of the silver salt mentioned above.
The specific rotation of helvolic acid is given as
[ajo = —49-4° in chloroform, a difference which may,
possibly, be due to complete or partial conversion of the
lactone to the free acid. It is active against Gram -positive
organisms at dilutions of 1 in 200,000 to 1 in 600,000 and
only slightly active against Gram -negative bacteria
(dilution of 1 in 1000). In concentrations of 200 /Mg.
per ml. it is bactericidal. It is thermostable, and not
very toxic to animals but doses near the toxic limit are
necessary to protect mice against infection with
Streptococcus pyogenes.

ANTIBIOTICS

Fumigatin. — Aspergillus fumigatus, in addition to
fumigacin, synthesises the pigment fumigatin which has
been shown to be 3-hydroxy-4-methoxy-2 : 5-tohiquinone,
0

CH3 / \ OH

, which inhibits the errowth of Gram-
" " OCH,

O
positive bacteria at dilutions of 1 in 33,000 to 1 in 50,000,
and Gram-negative organisms at 1 in 12,000. It is par-
ticularly active against V. comma which is inhibited at
a concentration of 1 in 100,000. A number of mono-,
di- and tri-methoxy derivatives of toluquinone and of
benzoquinone have been tested for activity against
Staph, aureus. 4-Methoxy-, 4 : 6-dimethoxy-, 3:4:6-
trimethoxy- (spinulosin dimethyl ether), and 6-hydroxy-
4-mefchoxy- toluquinones and 2 : 6-dimethoxybenzo-
quinone have considerably greater activity than fumigatin.
Introduction of the OCH3 grol^p increases the activity
whilst an OH group reduces it. The most active is

0
II

4:6-dimethoxy-toluquinone, jj || , of which

CH3O V ^ 0CH3,

o

10/xg. per ml. is inhibitory. The active structure
seems to be =C=C.C0.C(0CH3) = C.CO which is also
present in penicillic acid (see p. 170). Replacement of
the terminal — CO group by oxygen (as in the methyl

, . ^. r 1 .. .. rH,0H.C.=CH.C0.C(0CH..)=CH— O.

derivative 01 kojic acid "1 " I

(see p. 294) causes a great loss in activity.

Gigantic Acid. — The growtli of A. giganteus on a
2 per cent, malt extract containing 1 per cent, of peptone

164 BACTERIOLOGICAL CHEMISTRY

and 5 per cent, of tri-ethanolamine buffer at ^^H 8-2
yields a product which is very similar to penicillin in
its properties.

Gliotoxin. — GUodadiuyn fimbriatum, when grown in an
agitated culture medium containing sucrose and peptone
at 2^H 3 to 3-5, yields gliotoxin which can be extracted
with chloroform. It is also formed along with fumigacin
by A. fumigatus in surface or submerged cultures and
from an unidentified strain of Penicillium. The material
can be recrystallised from methanol and has m.p. 221°C.
(decomp). It is optically active, having [aj^f — 290°
in ethanol, —270° in pyridine, and —255° in chloroform.
In ethanol containing sodium hydroxide it undergoes
mutarotation from [aj^-f 111° to -|- 80° after 48 hours
and to 0° after 5 days. Gliotoxin has an ultra-
violet light absorption curve similar to those of
trjrptophane and indole. It is a neutral substance with
the composition C13H14N2O4S2. It appears to be an
indole derivative having a third 6-membered ring at
positions 1 and 2. The third ring contains the second
nitrogen atom, which carries a methyl group, and is
bridged by a clisulphide group. The oxygen is present in
the form of hydroxyl groups. On heating with hydriodic
acid or with dilute alkali, gliotoxin loses its sulphur to
give an a-pyrazindole derivative, C13H12O2N2, m.p.
122°C., which is biologically inactive. The carbon and
nitrogen skeleton of gliotoxin is prol)ably 2 : 3-dimethyl-
1:2:3: 4-tetrahydropyrazino-[l : 2]-indole : —

In a concentration of 10/tg. j)oi- ml. gliotoxin preveiits the

ANTIBIOTICS 165

gro^\'th of all pathogenic organisms tested ; 0-2 to 0-3
fig. per ml. was adequate to inhibit hsemol^^ic streptococci
and Type III pneumococci. It is also inhibitory to A.
niger, P. italicum and Bhizopus. ^ It is lethal to mice
and rabbits in doses of 45 to 65 mg. per kg. of body
Aveight. Less than the lethal dose causes kidney lesions
and hsematuria.

Gramicidin. — In 1939 Dubos isolated from soil a spore
bearing bacillus, which has since been named B. hrevis.
Autolysates of the organism contained a soluble substance,
tyrothricin, capable of lysing living Gram -positive cocci.
The substance is non- volatile, does not dialyse through
coUodion and is heat stable. It is very stable to alkali
but not to acid. 0-02 /xg. per ml. inhibits the growi;h of
pneumococci, whilst staphylococci and streptococci are
not quite so sensitive. Gram-negative organisms are not
affected even by large amounts of the substance. The
intraperitoneal injection of 2 mg. of the extract protected
mice against infection with pneumococcci. Tyrothricin
is not to be confused with the enzyme, produced adap-
tively by some soil bacilli, which destroys the capsule of
Type III pneumococci only, rendering them susceptible
to phagocytosis. Tyrothricin inhibits the fibrinol}i}ic
action of [B-haemolytic streptococcal filtrates and the
coagulase activity of staphylococal filtrates. Tyrothricin
has been shown to be a mixture. By extraction of the
crude material with equal parts of acetone and ether,
evaporation to dryness and extraction with warm acetone
a crystalline substance, gramicidin, is obtained on cooling
the acetone solution. The residue insoluble in the
acetone-ether mixture is taken up in boiling ethanol and
acidified with hydrochloric acid when a crystalline
material, tyi'ocidine hydrochloride, separates out (see
p. 182). Tyrothricin contains about 10 to 20 per cent,
of gramicidin and 40 to 60 per cent, of tyrocidine.
Gramicidin is a neutral substance which is specifically

16(3 BACTERIOLOGICAL CHEMISTRY

bacteriostatic to Gram -positive bacteria. Gramicidin
has the property of depressing the surface tension of
aqueous sohitions ; the property is not lost on heating,
although the bacteriostatic and hsemolytic effects are
destroyed by heat. Gramicidin does not lose its bacterio-
static properties in the presence of serum. It will protect
mice from infection by Gram -positive organisms when
it is injected in contact with the organisms but not
otherwise. Gramicidin has been shown to detoxify
tetanus and diphtheria toxins. It is too toxic to be of
use for internal chemotherapy but has been used for
local application. Gramicidin has been shown to be a
closed ring polypeptide containing 24 amino acid residues,
having a molecular weight of 2790, m.p. 230 — 231°C.,
[all? + 2-5°. It contains no free amino or carboxyl
groups. The amino acid residues comprise 6 of leucine,
6 of tryptophane, 5 of valine, 3 of alanine, 2 of glycine
and 2 of a hydroxy-amino compound, possibly ^<so-serine.
The leucine and two or three of the valine units are in
the d-iorm. It is resistant to the action of trypsin,
pepsin and papain. It is possible that the c?-amino acids,
very rarely found in nature, may be responsible for its
bacteriostatic properties.

The action of formaldehyde on gramicidin greatly
reduces its toxicity and hsemolytic activity without
harming its bacteriostatic properties. The modified
gramicidin may possibly be used for internal therapy.

Gramicidin- S. — A substance of a similar character
to gramicidin has been obtained by Russian workers
from a soil bacillus. Gramicidin-S is very stable to heat,
has m.p. 268-270°C. and molecular weight about 1250.
It is of peptide nature but contains free NHg and COOH
groups. It also differs from gramicidin in containijig
proline and ornithine but no tryptophane. It is different
in that it is bacteriostatic to some Gram-negative organ-
JiJ. typhosa at 50/Ag./m]., Sh. dysenterue at

ANTIBIOTICS 167

l'2ixg./ml., V. comma at 25 ju,g./ml. and P. vulgaris at
100 /Ltg./mL, as well as to Gram -positive organisms.

Helvolic Acid. — A mutant of A. fumigatus mut.
Helvola when grown on a glucose-salts medium gives a
substance which can be adsorbed on charcoal at ^^H 4
and eluted with 80 per cent, acetone. It can be purified
by chromatography of a chloroform solution and re-
crystallised from acetone. The yield is 0-4 g. from 100
litres of medium. It is a monobasic acid, C32H44O8,
containing three active hydrogen atoms per molecule,
m.p.205°C, [a]??-49-4° in chloroform. It is almost
insoluble in water, but gives a soluble sodium
salt ; salts with other metals are very sparingly soluble.
Its activity is not affected by heating to 100°C. with
2 N acid for fifteen minutes, or in neutral solution or
with alkali at pH 10. It is bacteriostatic to Gram-
positive but not to Gram -negative organisms, but has
no effect on the respiration of suspensions of staphylococci.
Human tubercle bacilli are inhibited by a 1 in 10,000
dilution of helvolic acid. ]\Iice will tolerate 5 mg. of
sodium helvolate intravenously and 20 mg. orally,
leucocytes are not injured by a dilution of 1 in 1600
nor are tissue cultures by 1 in 2500 sodium helvolate.
It is absorbed from subcutaneous tissue and from the
gut and is excreted in the urine and bile. Although
antibacterial concentratioixs can be maintained in the
blood stream, repeated injection leads to liver damage.
This probably accounts for the fact that the lives of
mice infected with staphylococci or streptococci can be
prolonged but not saved by sodium helvolate.

Helvolic acid is probably the acid corresponding to
the lactone, fumigacin (see p. 162).

lodinin. — Chroinohacteriiim iodinum gives the pigment
iodinin wliich is the di-N-oxide of a dihydroxy j^henazine,

13

168 BACTERIOLOGICAL CHEMISTRY

O

^\/^\/i% 1

lU 3

■^ . The position of the two hydi'oxyl
OH

groups is unknown, but they are probably not in the
2:3 or 2:5 positions. The growth of streptococci is
partially inhibited by 3 X 10"' molar concentrations and
completely inhibited by 1 to 2 X 10~^ molar iodinin.
Phenazine di-N-oxide has a similar but weaker effect.
When groAvn in the presence of sub -inhibitory quantities of
iodinin or phenazine di-N-oxide, organisms multiply and
destroy the compounds, probably by reduction. Extracts
of a wide variety of yeasts, bacteria and plant and
animal tissues have no effect on the inhibition, but those
which are active contain anthraquinones or naphtho-
quinones. Pure hydroxyanthraquinones and 2-methyl-
1 : 4 -naphthoquinone (vitamin K) antagonise 2 x 10~^
molar iodinin at concentrations between 5 X 10~' and
5 X 10~® molar, the iodinin being destroyed. The
destruction does not take place in the absence of the
organisms. It is suggested that the quinones are probably
concerned in hydrogen transportase systems which can
be interfered with by iodinin or phenazine di-N -oxide.

Notatin. — Penicillium notatum, during growth on a
modified Czapek-Dox medium, produces an antibiotic
which is different from penicillin (see p. 171). It has been
called notatin, penicillin-A, penicillin-B, or penatin.
The formation of notatin is repressed by the addition of
corn-steep liquor, yeast extract, brown sugar or malt
extract to the Czapek-Dox medium. It can be separated
by concentration of the culture filtrate and precipitation
with acetone, or by tannic acid. It is a buff coloured

ANTIBIOTICS 16!)

powder, soluble in water but not in organic solvents.
It is decomposed by 70 per cent, aqueous methanol at
30°C. or by aqueous solutions of trichloracetic acid to
give a protein and a prosthetic group, neither of which
is active alone. Its activity is lost below pH 2 and above
j^H 8 or by heating at 60°C. It is not destroyed by
pepsin at pR 3-8, trypsin at ^^H 5- 7, takadiastase or emul-
sin. It is almost completely inactived by activated papain
at pH 3-8. Blocking of the amino groups by formalde-
hyde, nitrous acid, or phenyl-isocyanate does not cause
loss of activity. It has the properties of a " yellow
enzyme," that is a flavoprotein enzyme (see j). 42).
In the presence of glucose and oxygen it has a very
powerful bactericidal effect on all bacteria which are
sensitive to hydrogen peroxide, e.g.. Staph, aureus,
Str. pyogenes, Dipl. pyieumonice, E. typhosa, Sal. paratyphi,
Sal. schottmillleri, Sal. typhi-murium, V. comma, B.
anthracis and Proteus vulgaris. The effect is reversed
by the presence of catalase. Xotatin is a glucose
aerodehydrogenase which, under aerobic conditions,
oxidises glucose to gluconic acid with production of
hydrogen peroxide which is the directly lethal agent.
Glucose can be replaced by galactose or xylose but not
by other sugars.

The xanthine -oxidase in milk is also a flavoprotein
enzyme which catalyses the oxidation of xanthine or
hypoxanthine to uric acid and hydrogen peroxide. In
the presence of its substrate (present in ordinary meat-
broth media but not in peptone water or hydrolysates of
purified proteins) the gro^^-tli of organisms sensitive to
hydrogen peroxide is inhibited. The effect is eliminated
by the presence of catalase, which destroys the hydrogen
peroxide formed, or by the absence of the appropriate
substrate. The gro\^i;h of E. coli, Bad. lactis aerogenes,
Shigella flexneri or C. diphtherice, which are insensitive
to hydrogen peroxide, is not affected by either of these
enzymes.

170 BACTERIOLOGICAL CHEMISTRY

Parasiticin. — A . parasiticus, when grown on a medium
containing peptone or 7 to 8 per cent, of corn -steep liquor
and small amounts of glucose at ^^H 7, gives culture fil-
trates which are active at dilutions of one in 200 to 600
against Gram-positive organisms but not against Gram-
negative organisms. The substance responsible for the
activity can be adsorbed on charcoal and eluted by
aqueous acetone. It has properties very similar to those
of penicillin and to the substances obtained from A . flavus
(see p. 161) and A. giganteus (see p. 163) and may be
identical with them.

Patulin. — This substance is identical with clavacin
(see p. 160).

Penatin. — This substance is identical with notatin

(see p. 168).

Penicillic Acid. — The growth of P. cydopimn on
Raulin's medium (but not on Czapek-Dox medium)
leads to the formation of about 2g. per litre of penicillic acid

CH3 ^^^^

\c.CO.C =CH.COOH. It is inhibitory to staphylococci,

E. coli and Gram -positive and Gram -negative pathogens
at dilutions of 1 in 30,000 to 1 in 100,000 but is only
poorly bactericidal.

Penicillic acid, which is a colourless, crystalline
substance with m.p. 86°C, soluble in water and chloroform,
can also exist in a closed ring form,

CHgO.C^^CH

CH3 I I

ANTIBIOTICS 171

whicli is /:?-metlioxy-y-liydr(.)xy-}/-/6(yj[)ropylicleiic tutnjiiic

HO.C==C.R
acid. Other tetronic acids produced l)y fungi, j^i ^n qq

'\o/ '

such as carlic, carolic, carolinic and carlosic acids (see

("H
p. 290) and dihvdropenicillic acid, \

^ ' ^ ^ >CH.C.OH CO,

CH3 \o /

are not antil)acterial. The somewhat similarly constituted

CO

CH C.OH

koiic acid, II II (see p. 294) is much less

CH,OH.C CH

0

active than penicillic acid.

It is possible that penicillic acid is active because
it combines with such amino -acids as glycine, alanine
and p-aminobenzoic acid, all of which reverse its effect.

Penicillin. — In 1929 Fleming noted that the growth
of Staph, aurevs on a plate contaminated by Penicillium
notatum was inhibited in the neighbourhood of the
mould. He showed that the growth of P. notatum on
a fluid medium gave a culture filtrate which was strongly
bacteriostatic to certain Gram -positive bacteria. The
active substance, penicillin, which is also formed by
P. chrysogenum and A. flavus, can be extracted from
acidified culture filtrates ^vith ether or other organic
solvents in the cold and can be taken back into aqueous

172

BACTEraOT.OGICAL CHEMISTRY

solutiuii l»y washijig i\\v exiiaci witli dihitc alkali, wiiicJi
converts penicillin into the readily soluble sodium salt ;
concentration of the penicillin can be effected at the
same time by use of a small volume of alkali. Further
purification can be achieved ,by chromatography on
alumina or silica gel carrying an alkaline earth carbonate,
and conversion to the barium salt. The barium salt is
fairly stable iDctAveen j^K 5-5 and 7-5, but is easily in-
activated by heating or by acid or alkali. The free acid,
penicillin, is hygroscopic and loses its activity readily,
although in ether or amyl acetate solution it is stable
for some days. It is inactivated by copper or mercury,
primary alcohols, ammonia, amines, hydrazine, hydroxyl-
amine and oxidising agents.

Culture filtrates have been shown to contain one or
more of several closely related compounds first recognised
by somewhat different degrees of activity against various
bacteria. They have the general empirical formula,
C9H11O4SN2.R, and differ in the nature of the radical R,
as is shown : —

British Name

American
Name

R

Penicillin I -

Penicillin F

A^-pentenovl,

Dihydropenicillin I

—

/i-Amyl

Penicillin II -

Penicillin G

Benzoyl

Penicillin III

Penicillin X

/j-Hydroxy benzyl

Penicillin K -

Penicillin K

n-Heptyl

That the penicillins probably have either the ^-lactam
structure, I, or the incipient azlactam structure, 11, is
in harmony with the following findings : — •

ANTIBIOTICS

173

X X

P? ?l

hJ

H 1

X

o

--ifl

X

o

P
^^

X
o-

9. ^'^

-if

i

^'

a
X ^

0

0

+

P2

;i^

3t

X
0

11

M

0

X^

x^

0 -

~n

^"2
0 ^i-

174 BACTERIOLOGICAL CHEMISTRY

They arc strong monobasic acids with ^^K — 2-8. The
probable presence of a masked basic group is shown by
a slow titration with perchloric acid in acetic acid.
Hydrolysis of penicillin with dilute acid at 30°C. gives
the penillic acids, III, which are dicarboxylic acids
containing a basic group but no SH group. Treatment
of the penillic acids with cold aqueous mercuric chloride
causes loss of carbon dioxide and gives rise to penillamines,
IV, which contain an 8H group, whilst treatment with
baryta converts penillic acid to i^o-penillic acid, X.
Hydrolysis of j)enicillin with hot dilute mineral acids
gives carbon dioxide, penicillamine, V (<i- j8 jS-dimethyl
cysteine) and penillo aldehydes, VII ; that from penicillin
I is A 2-hexenoylamino acetaldehyde, whilst penillo-
aldehyde II is plien3dacetylamino acetaldehyde. The
carbon dioxide and the aldehydes are derived from the
corresponding penaldic acids, VI. Treatment of penicillin
with alkali or with penicillinase causes opening of the
lactone ring with formation of dibasic penicilloic acids,
VIII. Treatment of penicillin with methanol gives rise
to the inactive methyl ester, IX.

The potency of penicillin preparations is defined in
terms of an arbitrary international unit, the "Oxford unit,"
which is the amount of activity contained in 0-6 /xg. of a
particular standard crystalline preparation. The measure-
ment is carried out by comparing the effect of the unknown
sample with that of a sample of known potency on cultures
of a suitable organism in vitro. Staphylococci are in-
hibited by a concentration of penicillin equivalent to
0-01 to 0-02 units per ml. The purest solid preparations
of penicillin so far described contain about 1650 units
per milligram, so that dilutions of 1 in 3 X 10' are
sufficient to inhibit staphylococci ; culture filtrates
from surface growth contain about 20 to 80 units per
millilitre, whilst those from submerged growth ma}^
contain up to 250 units per millilitre, depending on the
strain and conditions.

ANTIBIOTICS 175

The majuiity of organisms wJiicli arc seiisitive to
Ijenicillin are Gram -positive, namely, Stajihylococcus,
Streptococcus, pneiimococciis, C. dipJitherice, B. anthracis,
Actinomyces, CI. loelchii, CI. cedenuitiens and other
Clostridia. The sensitive Gram -negative organisms are
the gonococcus, meningococcus and Micrococcus catarrh-
alls. The other Gram -negative pathogens are not
affected by penicillin nor are Gram-negative saproph;>i3es,
yeasts or moulds. The tubercle bacillus is not sensitive
to penicillin. The activity of penicillin is not affected
by the presence of serum, blood or pus nor by the number
of organisms present.

Penicillin is almost completely non-toxic to man or
animals and has no deleterious effect on leucocytes or
tissue cells. It is, therefore, superior to the sulphonamides
in these respects. From the chemotherapeutic point of
view it has the defects that it camiot be given l)y mouth
as it is destroyed by the acid of the stomach, and that it
is very rapidly excreted in the urine after intravenous
or intramuscular injection. In order to maintain an
adequate concentration in the blood it must be ad-
ministered by the continuous intravenous drip method
or by intramuscular injections repeated at least every
three hours for as long as the infection persists. A daily
dosage of about 120,000 Oxford units is necessary.
Penicillin is of particular value in treating staphylococcal
septicaemia, which is often resistant to treatment with
sulphonamides, osteomyelitis, gas gangrene, gonorrhoea
and infections by sulphonamide resistant strains of
pneumococci. Penicillin can be used effectively in the
local treatment of burns, wounds and skin infections
by application in a cream or as a powder (usually mixed
with sulphanilamide). The methyl and ethyl esters of
penicillin are more stable than penicillin to such an
extent that they can be given by mouth. They have
only about one hundredth of the activity of penicillin

176 BACTERIOLOGICAL CHEMISTRY

/"// vilro, pr()l)al)ly duo to slow hydrolysis. Li doses of
about 2 nig. they will protect mice against many
thousand lethal doses of hsemolytic streptococci.

As with other drugs, bacteria develop resistance to
penicillin when subjected to concentrations which are
inadequate for bacteriostasis. Organisms which have
acquired resistance to penicillin are still susceptible to
sulphonamides and vice versa. The mode of action of
penicillin is miknown but it does not interfere with
respiration ; susceptible organisms subjected to less
than the bacteriostatic dose continue to grow but lose
their power of subdivision so that giant forms are
produced.

Penicillin is destroyed by enzymes (penicillinase)
secreted by several species of bacteria including E. coli,
Micrococcus lysodeikticus, Proteus and some Gram-positive
bacilli such as B. subtilis.

Penicillin B. — This substance is identical with notatin.

Proactinomycin. — A species of Proactinomyces, when
grown on a glucose agar medium, forms an alkaloid
like base, soluble in organic solvents and in water at
pH 4, which has bacteriostatic properties similar to,
but weaker than, those of penicillin. It is more stable,
undergoing only small loss of activity on boiling at
pK 2 or 7 for ten minutes ; it is inactivated by boiling
at pH 10. Proactinomycin is toxic to mice in doses of
2 to 5 milligrams.

Puberulic Acid. — Several species of Penicillimn, P.
puberulum, P. aurantio-virens, P. joJiannioli and
Cyclopium viridicatum yield puberulic acid, CgHeOg,

ANTIBIOTICS 177

iiud |)ul)i'iuluiii<- a<;i(I, ( ^sH^O^,. TJic furnici- is a colcjiiilcss,
crystalline, dibasic acid of m.p. 316 to 31 8 (J., whilst
puberulonic acid is a bright yellow, crystalline substance
having m.j). 298°C., which is thought to be the quinonoid
form of the quinol, puberulic acid. Puberulic acid
inhibits Gram -positive organisms at dilutions of 1 in
6000 to 1 in 33,000, whilst pubenilonic acid is less effec-
tive, inhibiting at 1 in 6000 only. They liave little effect
on Gram -negative organisms.

Pyocyanase. — Pseudomonas ceruginosa (B. pyo-
cyaneus), wliich was among the earliest of organisms
shown to produce antibiotic substances, forms pyo-
cyanase, pyocyanin and a-hydroxy-phenazine (hemi-
pyocyanin). Pyocyanase is l3rtic to many bacteria,
such as E. coli, Ehertliella typhosa, C. dijohtherice, V.
comma, streptococci and staphylococci, and also detoxifies
the toxins of CI. tetam and other bacteria very rapidly,
a property also possessed by sodium lauryl sulphate
and zephiran (a sulphonated mixture of the fatty acids,
Cg to Ci7, contained in coconut oil).

Pyocyanase, in spite of its heat stability, was at one
time thought to be an enzyme attacking nucleic acids
but it is now regarded as being of lipoidal nature, the
activity depending largely on the presence of unsaturated
fatty acids ; it is said to contain a phosphatide and
free fat in addition. The facts that it is soluble in ether,
chloroform a;nd l^enzene and that its activity is not much
affected by changes in temperature between 0 and 37 °C.
are not in accord with the view that pyocyanase is an
enzyme. It has been obtained as a colourless oil forming
an ether soluble lead salt, so that it is probably an
unsaturated fatty acid.

Pyocyanin. — The chloroform soluble blue pigment,
pyocyanin, produced by Ps. ceruginosa, has been shown
to be a phenazonium compound : —

178 BACTERIOLOGICAL CHEIMISTRY

CH N CH

^'\ /\ ^-\
^ \/ \^ \ ,
CH C C CH

I II I II

CH C C CH

% /\ ^\ /
%/ \^ \/
CH N CO

II
CO N CH

/\ ,^\ /%,
/ \^ \/ %
CH C C CH

li I II I

CH C C CH

\^ \/ \/
CH N CH

CH3

which breaks down to

o-

CH N CO CH N C

^\ /% /\ ^^\ /%^ /%

^ \/ %/ \ ^ \/ %/ %

CH C C CH CH C C CH

I II I II .=^ 1 II I I

CH C C CH CH C C CH

% /\ /% / % /\ ^\ ^

%/ \/ %/ %/ \^ \^

CH N CH CH N+ CH

I I

CH3 CH3

with a structure not unlike that of iodinin (see p. 168)
which is also a phenazine derivative. It is formed in the
first two or three days of growth of the organism ;
pyocyanase occurs in cultures two to three months old.
Cultures of intermediate age contain a second pigment,
yellow in colour, which is a degradation product of
pyocyanin, known as hemipyocyanin, and having the
structure , a -hydro xy-phenazine ,

ANTIBIOTICS 17U

N OH

%/%^/\/

Pyocyaiiin, which is thermostable, has strong bacterio-
static power against Gram -positive and Gram -negative
bacteria and comparatively little effect against moulds
or yeasts. Hemipyocyanin is less active against bacteria
but considerably more active against yeasts and fungi.
The similar pigment, chlororaphin, formed by Ps. chloro-
raphis (see p. 391), is also somewhat inhibitory to bacteria.

In virtue of its oxonium structure, pyocyanin is
known to act as an oxygen carrier in the oxidation of
a-hydroxyglutaric acid to a-ketoglutaric acid in presence
of a dehydrogenase occurring in animal tissues. It is
possible that the antibiotic effect of pyocyanin is due to
interference in some such process in bacterial metabolism.

Pyocyanin is highly toxic to animals as w^ell as being
bactericidal ; hemipyocyanin is much less toxic and has
about the same bactericidal power as the f la vines. It
has no staining properties.

Spinulosin. — P. sjnmdosmu gives rise to the red
pigment . spinulosin, 3 : 6-dihydroxy-4-methoxy-2 ; 5-

0

il

toluqumone, , which is hydroxy-fumigatin

0

(see p. 163). It is a weaker bacteriostatic agent than
fumigatin, being active against Gram-positive organisms
at concentrations of 1 in 6000 to 1 in 10,000. The loss
in activity as compared with fumigatin is attributed
to the additional hydroxyl group since, in a series of
tolquinone derivatives, those members containing hy-
droxyl groups are less active than those without.

180 BACTERIOLOGICAL CHEMISTRY

Streptomycin. — Streptomyces griseus, one of the soil
organisms, the Actinomycetes, forms the substance
streptomycin which resembles streptothricin in inhibiting
the growth of Gram -negative as well as Gram -positive
organisms, but is more active against Pr. vulgaris and
Ps. ceruginosa among the former and against B. mycoides
and M. tuberculosis among the latter. It is also bacterio-
static to E. typhosa, Sal. schottmillleri, Br. abortus, H.
influenzce, H. pertussis, Serratia ')narcescens , B. subtilis,
Staph, aureus and CI. butylicum. Streptomycin is poorly
absorbed from the intestine but is not destroyed and so
may be useful for controlling intestinal infections. In
man it is rapidly absorbed and excreted in the urine
after parenteral administration, but therapeutic levels
can be maintained in the blood and urine more easily
than with penicillin. It has low toxicity for animals.
Preliminary trials indicate that it may be of therepeutic
use against typhoid fever and human tuberculosis. If
these claims are substantiated streptomycin is likely to
be as valuable as penicillin.

Streptomycin can be adsorbed on charcoal from the
culture fluid resulting from the growth of S. griseus on
a corn steep liquor medium. After elution from the
charcoal with acid ethanol it can be purified and re-
crystallised as the hydrochloride, [a]D-84°, the sulphate,
the helianthate or reineckate, m.p. 162-1 64 °C. It is
an organic base soluble in water but not in most organic
solvents. It is fairly stable to heat, losing about half
its activity in ten minutes at 100°C. It is inactivated by
cysteine, 2-aminoethanethiol and, to a less extent, by
thioglycollic acid. The inactivation is reversed by
iodine. Its activity increases with the alkalinity of the
medium up to ^H 9.

Streptomycin hydrochloride is completely inactivated
by standing for 24 hours with anhydrous methanol
containing N HCl. Addition of ether to the solutioji
precipitates a diguanidine base, streptidine, leaving an

ANTIBIOTICS

181

optically active substance, methyl streptubiosaminicle
dimethylacetal hydrochloride, in solution : —

QlH37-39^70l2.3HCl + 3 CH3OH

(iStreptomycin hjTli-ochloride)

C8Hi8X604-2HCl + H2O +
(Streptidine)
Ci3H2o-22NO,(OCH3)3.Hri
(Streptobiosaminide dimethyl acetal)

On hydrolysis streptidine loses ammonia and carbon
dioxide and forms a new base, streptamine : —

The six nitrogen atoms of streptidine are present as two
mono -substituted guanidine groups which are replaced
by two amino groups in streptamine. The oxygen atoms
are present as hydroxyl groups since acetyl and benzoyl
derivatives can be obtained. Stre^^t amine is veiy pro-
bably a diamino totrahydroxy c^^lohexane with the
amino groujjs at positions 1:3 or 1:4: —

NH.R

I
CH

CHOH CHOH

I I

CHOH CHOH

CH

I
NH.R

Streptamine, R = H.

NH.R

I
CH

CHOH CHOH

I I

CHOH CN.NH.R

\ /

CH

OH

iStreiDtidiue, R = — C

XH

XH,

Methyl streptobiosaminide dimethyl acetal is most
probably derived by the action of methanol on a nitrogen
containing disaccharide, streptobiosamine. The nitrogen
atom is not present as an amino group but probably as
a methylamino group attached at position 2 of one of
the hexose luiits.

182 BACTERIOLOGICAL CHEMISTRY

The structure of the streptidine moiety of strepto-
mycin makes it tempting to speculate that the antibiotic
is active in virtue of its resemblance to the growth factor

inositol,

CHOH

/\
/ \
CHOH CHOH

I I

CHOH CHOH

CHOH

Streptothricin. — A species of Actinomyces, occurring
in the soil, A. lavendulce, gives rise to the antibiotic
streptothricin, which has a selective action on Gram-
negative organisms, for example E. coli and Shigella
dysenterice, but also acts on some Gram-positive bacteria,
for example B. subtilis and Staph, aureus, moulds and
yeasts. Streptothricin is a base and appears to be built
up of amino -acid residues, resembling tyrocidine. It is
only inhibitory in the undissociated state ; factors such
as the presence of salts and ^H values which cause
dissociation lower the activity of streptothricin. Bacteria
subjected to its action increase in size and tend to form
chains. Resistant strains of bacteria can be developed.
It has a low toxicity for animals and could probably be
used for internal chemotherapy.

Tyrocidine. — Tyrothricin, isolated from auto'lysates of
B. hrevis, is a mixture of gramicidin and tyrocidine
(see p. 165). Tyrocidine is the fraction insoluble in
acetone -ether mixtures, but soluble in hot ethanol and
crystallised out as the hydrochloride. Tyrocidine hydro-
chloride has m.p. 237-239°C. (decomposition), and
[a] Jf — 102° (c = 1 in 95 per cent, ethanol). Tyrocidine
is a polypeptide having a molecular weight about 2500
and containing about twenty amino -acid residues in-
cluding tryptoj)hane, tyrosine, alanine, phenylalanine
and aspartic acid, combined in such a way as to leave

ANTIBIOTICS 183

free two basic amino groups, three amide groups and
one carboxyl group or a phenolic OH group. About
twenty per cent, ol the amino -acid residues have the
^-configuration. It is of interest to note that the capsules
of B. anthracis, B. mesentericus and B. suhtilis are made
up of a polypeptide composed of (^-glutamic acid (see
p. 338). The presence of such a large proportion of
^-amino -acids in gramicidin and tjrrocidine probably
accoimts for their resistance to pepsin, trypsin and
papain.

Tyrocidine has marked bactericidal and l^i^ic action
in vitro against Gram -negative as well as Gram -positive
organisms. Fifty to 100 jug. affords definite protection
to mice infected intraperitoneally with pneumococci.
Tyrocidine blocks the oxidative processes of metabolism.
It is antagonistic to certain lower fungi such as Achorion
schcenlandii, Microsporiu7n gypseum, Trichophyton gyp-
seum and Candida albicans.

Tyrocidine has a strong hsemolytic effect on human
and rabbit red blood corpuscles, haemolysis occurring in
the presence of 0-005 /xg. of the substance, which is also
highly toxic to animals. Tyrocidine loses much of its
bactericidal power in the presence of blood, serum or
pus. It appears to prevent the loss of activity which
tetanus toxin undergoes on heating at 55°C.

Both tyrocidine and gramicidin can be used chemo-
therapeutically by local application, to wounds for
example.

Un-named Antibiotics. — When B. mesentericus, in the
smooth phase, is grown in nutrient broth it gives rise to
a substance which has a specific bactericidal effect on
C. diphtherice at a dilution of 1 in 1250. The toxic
effects of C. diphtherice are eliminated when the organism
is injected along with the B. mesentericus filtrate into
guinea-pigs.

Aspergillus candidus produces a thermostable sub-
stance, similar to citrinin, but which is more powerful

15

184 BACTERIOLOGICAL CHEMISTRY

against Staphylococcus and B. mycoides, being bacterio-
static at concentrations of about 1 in 100,000 and
bactericidal at 1 in 8000.

A. flavipes, when grown on tryptone medium or
corn-steep liquor medium, gives an alkaline bacteriostatic
filtrate active at dilutions of 1 in 320 to 1 in 1300 against
Str. pyogenes, pneumococci, and CI. welchii, very slightly
active at 1 in 5 to 1 in 10 against staphylococci and
inactive towards Str. fcecalis, E. coli and Aerobacter
cerogenes. It can be obtained as a gummy precipitate by
concentration of the medium in vacuo and addition of
ten volumes of acetone.

Penicillium resticulosuni, grown on Czapek-Dox
medium, gives a metabolism solution which inhibits
the growth of Staph, aureus at dilutions of 1 in 320 to
1 in 2500. The antibacterial substance can be precipitated
from solution by the addition of acetone after concentra-
tion, the yield being about 1 g. per litre from a filtrate
inhibiting Staph, aureus at a dilution of 1 in 80. It is
readily soluble in water, contains 3-8 per cent, of
nitrogen, and causes complete inhibition at a concentra-
tion of 1 in 160,000. It is very sensitive to acid, being
destroyed by contact with 0-1 N hydrochloric acid at
room temperature.

Of thirty-nine moulds which were tested seventeen,
all of the genus Penicillium, had antibacterial activity.
They could be divided into two groups ; I, active against
Staph, aureus, Str. viridans and C. diphtherice, were
contaminants of laboratory media, and II, which were
active against E. coli and Eberthella typhosa in addition
to the above organisms, were mainly fi-uit contaminants.

Among a large number of species of Fungi imperfecti,
Wilkins and Harris showed that about 40 per cent, of
Aspergillus species and 25 per cent, of Penicilliuyn
species gave substances antagonistic to one or more of
the test bacteria, E. coli, Staph, aureus and Ps. ceruginosa.
Very few representatives of other genera were active, the

ANTIBIOTICS 185

exceptions being Botrytis cinerea and Helminthosporium
avence against Staph, aureus and Fusarium javanicum
against E. coli.

Among the number of antibiotics recently isolated
and about which little is kno^^'n at present the following
may be mentioned. Allicin, from garlic, Allium sativum,
probably having the structure

which is active against Gram -positive and Gram -negative
organisms ; bacitracin, from a B. subtilis like organism,
which is non -toxic to animals but is very active against
Group A h^emolytic streptococci, staphylococci and the
gas gangrene organisms ; mycophenolic acid (see p. 296),
formed by Penicillium hrevi-compactum, which inhibits
staphylococci, streptococci, C. diphtherice, B. subtilis
and many pathogenic fungi but not Gram -negative
organisms ; viridin, from Trichoderma viride, which is
inhibitory to the growth of Botrytis, Fusariiun, Tricho-
thecium and Cephalosporium strains and to a less extent
to that of Penicillium and Aspergillus strains ; a sub-
stance formed by Group N streptococci Avhich inhibits
many Gram -positive organisms, including streptococci.
Bacillus, Clostridium and Lactobacillus species ; subtilin,
from B. suhtilis, active against Gram-positive but not
against Gram-negative organisms ; violacein, the pigment
from Chr, violaceum (see p. 389), which is very inhibitory
to Gram -positive organisms (meningococcus, the only
Gram -negative organism affected, is inhibited by 0-0005
per cent, of violacein), Saccharomyces cerevisice and some
moulds.

It is interesting that the basic proteins, protamine
and histone, which are of comparatively low molecular
weight, are bacteriostatic or bactericidal to E. typhosa,
Shigella jmradysentericE, E. coli, B. suhtilis, Staph, alhus,
Staph, aureus, CI. ivelchii, CI. tetani, CI. histolyticiuti,

186 BACTERIOLOGICAL CHEMISTRY

streptococci and pneumococci in dilutions varying from
1 in 40,000 to 1 in 3000. It is seen that they have
properties similar to the polypeptide antibiotics gramicidin
and tyrocidine, the resemblance being especially close to
the latter. The effects are eliminated by the presence
of phosphatides, with which all the substances combine.
It is possible that they may be active by interference
with the phosphatide metabolism of the organisms.

Antibiosis appears to be a very widespread
phenomenon, existing among micro-organisms of all
sorts. It is probably the chief mechanism by which the
majority of species manage to survive in natural sur-
roundings, particularly in the soil where they must be
subjected to intense competition. Obviously the anti-
biotic substances produced by micro-organisms influence
profoundly the ecology of their surroundings.

Antibiotic substances have found several applications
in the control of plant and animal diseases, the out-
standing example in human treatment being the use of
penicillin. It is possible that the closed plaster-cast
method of treating wounds may depend for its efficacy
on the antagonistic action of saprophytic organisms
towards any pathogens which might be present originally.

Another use of antibiotics has been in the preparation
of selective media for the isolation of bacteria from mixed
cultures. For instance, the inclusion of penicillin in
Bordet-Gengou medium suppresses the growth of most
organisms occurring in the throat and enables Hcemophihis
"pertussis to be isolated with greater ease from cough
plates or swabs.

For further reading : —

A. Waksman, " Antagonistic Relations of Micro-organisms." Bart.

Reviews, 5, (1941), 231.
A. AVaksnian, " Miorof)ial Antagonisms and Antil»iotic .Substances.'"

The Commonwealth Fund. New York, 1945.

CHAPTER XII
BACTERIAL RESPIRATION

THE term respiration has undergone a continuous
expansion in its meaning. Originally "respiration"
in animals signified the exchange of oxygen and
carbon dioxide through the lungs, then it was used to
describe the transfer of oxygen to and the removal of
carbon dioxide away from tissues. Later still the term
connoted the general oxidation processes of cells, and,
finally, now that it is recognised that these processes are
almost always concerned with the energy requirements of
the cells, the expression has come to mean any energy
producing biological reaction, even when the reaction
takes place under anaerobic conditions. It is in this sense
of a chemical reaction producing energy in the cell,
whether aerobically or anaerobically, that we shall
employ it.

Chemists tend to regard bacteria, yeasts and moulds,
merely as useful reagents which can bring about many
reactions, such as the synthesis of acetone, butyric acid,
butyl alcohol and so on, some of which he cannot yet
carry out in the laboratory. The tendency of the biologist,
on the other hand, is to regard these reactions or products
as accidents more or less incidental to the life of the cell,
useful accidents it may be, in that they sometimes pro-
vide a means of identifying or helping to identify the
organism (as is the case with sugar fermentation re-
actions), or in that they provide some product like alcohol
which he values.

187

188 EACTERIOLOaiCAL CHEMISTRY

These reactions should be regarded as the life processes
of the cell, providing both the energy required and the
raw materials for building up new cells.

As far as we knoAv only two sources of energy are
available for living cells, light and chemical energy. Of
these, light can only be utilised by chlorophyll-containing
plants, by the blue-green algae and by a few autotrophic
bacteria ; for all other forms of life the requisite energy
must be derived from chemical reactions . Heat, electricity
and mechanical energy cannot be utilised by organisms,
probably because they lack appropriate " transformers " ;
the only transformers we know are chlorophyll and
similar pigments for light. Heat, or in other words a
rise in temperature, may cause increased growth and
metabolic activity of a cell, but it is only in so far as
the chemical changes (which supply the essential energy)
are speeded up by a rise in temperature. A cell cannot
economise on food by using the heat energy of the medium;
a starving cell, for instance, derives no benefit from a
rise in temperature.

It follows that the energy liberated in one cell is
of no use to any other cells ; neighbouring cells, even
those closely linked as in tissues, have no direct energy
exchange system. Moreover, the chemical energy neces-
sary for growth must be liberated within the cell, since
if it were produced outside the cell it would have to
take the form of heat or electricitj^ which cannot be
utilised by the cell. As a result of this the only foods of
value to the organism are those which can diffuse into
the cell. Thus complex proteins, fats and carbohydrates
like starch and cellulose, are not directly available to the
organism, but first have to be broken down or hydrolysed
to appropriate smaller, soluble, diffusible compounds.
This is the work of the class of exo -cellular enzymes or
hydrolases which are secreted into the medium by the
organism.

BACTERIAL RESPIRATION

189

As we have already seen in Chapter iV the reactions
catalysed by these enzymes involve only relatively small
energy changes, whilst, on the other hand, those reactions
brought about by the endo-enzymes, inside the cell,
where the energy liberated is of real value to the organism,
involve large energy changes. This is illustrated in
Table 15, w^hich shows the energy liberated from 1 gram
of the appropriate substrata by the action of different
enzymes.

Table 15

Exo-enzymes.

Endo-enzymes.

Pepsin, trypsin, rennet -

Lipase - - - -

Invertase

Maltase - - - -

Lactase - - - -

cals.

0

4

9-3
10
23

Lactacidase - - -

Akoholase -

Urease

Vinegar oxidase -

cals.
82

149-5

239
2,530

An exception to the general rule that hydrolytic
reactions involve little energy change is the case of the
breakdown of urea, w^hich is used as energy source by the
urea bacteria. This is only an apparent exception because
the reaction takes place in two stages, first the hydrolysis
of urea to ammonia and carbon dioxide : — •

.XHo

CO

^H.,0

JXHa -fC02+ about 0 Cab

XHo

which involves practically no liberation of energy. The
second step is the formation of ammonium carbonate
from the ammonia and carbon dioxide : —

2XH3 - CO, + H2O > (XH4)2C03 + 12 Cals.

and it is this part of the reaction which supplies the

190 BACTERIOLOGICAL CHEMISTRY

The products formed by the action of the eiido-
enzymes in the cell are usually essentially different from
the substance fermented, and are in the main useless,
often even harmful, to the organism. It is this accumula-
tion of end products which is often responsible for the
cessation of growth after a time ; for instance, in yeast
fermentation when about 8-5 per cent, of alcohol has been
produced, further growth of the yeast cells is inhibited.

The majority of organisms can utilise several kinds of
food and therefore bring about various fermentations in
the course of their metabolism. They may even be able
to utilise the same food in different ways, according to
the conditions. For instance, certain typical sugar fer-
menting bacteria and the yeasts can be grown in the
absence of sugars ; lactic acid bacteria will grow on
peptone. Yeast normally ferments sugar to alcohol, but it
can also oxidise it completely to carbon dioxide and water
if a sufficient supply of oxygen is available. Most moulds
appear to be omnivorous, attacking almost any substrate
with which they may be supplied. At the other extreme
are the autotrophic bacteria, most of which can utilise
only one substrate as a source of energy (see Chapter VI) .

The chemical activity of the micro-organisms is,
generally speaking, vastly greater than that of animals
or plants. For example, it has been calculated that 1
gram of Micrococcus urece can decompose 180 to 1,200
grams of urea per hour ; and that 1 gram of certain
lactose fermenting bacteria can hydrolyse 180 to 15,000
grams of lactose per hour. If man were capable of meta-
bolism on the same scale he would consume several
thousand tons of food per hour. These figures are pro-
bably subject to a certain margin of error, but they are
sufficiently accurate to indicate the enormous difference
between the metabolic activities of the bacteria and
animals. A major cause for this difference in activity
is the much greater area in bacteria available for the
absorption of nutrients. The area to weight ratio in

BACTERIAL RESPIRATION

191

bacteria is uf the order uf 200,000 times that for man.
Moreover, the whole of the bacterial surface allows the
passage into the cell of foodstuffs, whilst in man and
animals the absorption of nutrients takes place through
only a limited part of the total surface.

If an organism can use a given compound as a nutrient,
its value will depend on its calorific value. For instance,
it has been shown that when Aspergillus niger is grown
on various compounds as the source of energy and carbon,
the weight of mycelium grown runs parallel with the heats
of combustion (that is, the calorific values) of the com-
pounds, as may be seen from Table 16.

Table 16

Nutrient,

Heat of Combustion.

Weight of Mycelium.

Tartaric acid -

Citric acid

Glucose - - - -

Glycerol

Olive oil . - -

2,618 cals./l-5 g.
3,711
5,614
6,461
13,972

0-155 g.

0-240

0-278

0-475

0-810

The value of a compound as a food or energy source
also depends on the degree of oxidation which it under-
goes ; the more complete the oxidation the higher the
energy available. Glucose may be taken as an example,
and the energy liberated with var3dng degrees of oxidation
compared : —

(a) Complete aerobic oxidation

CfiHigOg+eOa > GC02 + 6H20+674Cal,s.

Partial aerobic oxidation

(^)

(c)

2C6H12O6+9O2 > 6C2H204 + 6H,0+493Calr^

(oxalic acid)

Anaerobic oxidation

(1) QHi^Oe
(ii) CeHi^Oe
(iii) CgHjoOe

--> 2C3He03 + 22-5Cals.

(lactic acid)
-^ 2C2H5OH + 2CO2 -

(ethyl alcohol)
-^ 3CH3COOH + 15Cals
(acetic acid)

22 Cals

102 BACTERIOLOGICAL CHEMISTRY

it follows from this consideration that the less com-
plete the oxidation the more of a given substance must
be broken down to supply the needs of the organism.
The incomplete oxidation of non-nitrogenous compounds
is what we usually call fermentation, and the incomplete
oxidation of nitrogenous compounds is usually referred to
as putrefaction.

Pasteur considered that the main factor controlling
fermentation was the oxygen supply. He showed that
under anaerobic conditions the growth of yeast cells is
much restricted but that fermentation is very active,
whilst under aerobic conditions with a good oxygen supply
the growth of the cells is rapid but fermentation is
repressed. The aerobic breakdown of 7 grams of sugar
is associated with the formation of 1 gram of yeast cells,
whilst anaerobically the production of 1 gram of cells
needs 70 grams of sugar, a striking illustration that
complete oxidation is much more economical than
incomplete oxidation.

Various theories have been propounded to explain the
fact that such compounds as carbohydrates, amino-acids
and fatty acids are readily oxidised in part or completely
by cells at ordinary temperatures, whilst in the laboratory
they are only oxidised by much more drastic means.
Usually these theories involve the activation of either the
substrate or the hydrogen acceptor (whether it be oxygen
or some other compound) or sometimes both of them.

Wieland has put forward the idea that all oxidations
are due to a hydrogen transfer. He came to this con-
clusion as a result of experiments in which he found that
many substances could be anaerobically oxidised, in
presence of such metallic catalysts as spongy palladium
or platinum, by loss of hydrogen which is taken up by
the catalyst. Examples are the oxidation of hydro -
quinone to quinone : —

BACTERIAL RESPIRATION 193

and the uxidatiun of ethyl alcohol to acetaldehyde : —

/"

C'Hj.C^— OH y r'Hjf'HO ^ If.

The hydrogen atoms shown in bold type are those trans-
ferred. In cases where there is an actual increase of
oxygen in the oxidised molecule, he regards the first step
as being the formation of a hydrate, the second step
being the loss of hydrogen from the hydrate. Thus the
oxidation of acetaldehyde to acetic acid is considered to
follow the equation : —

CH3C— H ^ HoO > CH3C;— H > CH3C— OH -r Ho

\0H

(acetaldehydf) (acetaldehyde (acetic acid)

hydrate)

Lactic acid and glucose were found to be oxidised in a
similar way by these metallic catalysts, especially if a
hydrogen acceptor such as oxygen or some easily reduced
compound was added. Methylene blue proved a very
useful hydrogen acceptor.

Wieland regards biological oxidations as being similar
in character, the difference being that the metallic catalyst
is replaced by an enzyme. The enzymes are variously
known as a reductase when the hydrogen acceptor is a
dye or a nitrate or some similar compound, as a mutase
when the acceptor is a second molecule of the substance
being oxidised (as, for example, when acetaldehyde is
converted into acetic acid and alcohol by the acetic acid
bacteria), or as an oxidase when atmospheric oxygen is
activated as the hydrogen acceptor.

That bacteria are able to activate a large variety of
compounds in this way has been shown by the use of
the " methylene blue technique." In this procedure a
washed suspension of the organism under investigation

104 BACTEKTOLOaTCAT. CHEMISTRY

is incubated with a .sulutiuii of the compound to be
examined and a standard quantity of methylene blue
solution, all buffered at an appropriate 2>H value in an
evacuated system. If the bacterium is capable of acti-
vating the substrate as a donator of hydrogen (that is,
if it is capable of oxidising it) the methylene blue accepts
the hydrogen and becomes reduced to the colourless
leuco -compound, the loss of colour serving as an indicator
that the reaction has occurred. By this means numerous
fatty acids, hydroxy- and amino-acids, polyhydric alcohols
and sugars have been shown to be activated as hydrogen
donators. Some of these substrates are much more
readily activated than others, the most active, glucose,
being some ten thousand times more effective than the
least active of the lower fatty acids. Formic, lactic and
succinic acids are all fairly active. All bacteria do not
activate the same compounds, and there are marked
differences between the activation by plant and animal
tissues and by bacteria.

Thunberg, who originated the methylene blue
technique, considered these activations as being due to a
series of specific enzymes. This is rather hard to believe,
however, since Esch. coli, for instance, would need to
contain over fifty such enzymes, including some for sub-
stances like chlorates which the organism would be
extremely unlikely to meet in the ordinary course of its
existence. Quastel has suggested that one general
mechanism is responsible for all these activations, which,
after all, are alike in that they are all hydrogen transfers.
He considers that the enzyme action depends on two
factors ; first that the substrate is adsorbed on an active
surface in the cell, and secondly that the adsorbed mole-
cule is rendered unstable in such a way that it is liable
to lose hydrogen if it is a donator or to gain hydrogen
if it is an acceptor (like methylene blue). The cell is
pictured as having a network of internal interfaces,
probably composed of protein and lipoid constituents.

BACTERIAL RESPIRATION 195

Certain areas of these interfaces are endowed with activity
as a result of the arrangement of the molecules composing
them ; certain arrangements of polar groups in the
molecules set up local electric fields of varying intensity
depending on the particular molecular arrangement.
Any molecule, especially one containing polar groups,
like — CO OH, =C0, or double bonds, coming into close
contact with such an active surface by adsorption becomes
distorted with a resulting shift of the hydrogen atoms
rendering the molecule unstable or activated. For
instance, a double bond is supposed to be activated
according to the scheme : —

— CH=CH > _c— CHo—

or an aldehyde group in this way : —

— CH = o — > _c_OH.
The presence of a polar group like carboxyl in a molecule
favours a concentration of hydrogen in its direction : —

R— CH=CH— COOH > R— C-

whilst a non-polar group like methyl favours a concentra-
tion of hydrogen away from it : —

\/
R— CH=CH— CH3 > R.CH2— C— C'Hg.

The mechanism, as described so far, suggests how a
compound can be activated to act as a hydrogen donator
or acceptor, but it does not account for specificity. For
instance, it affords no explanation of the fact that Esch,
coli activates glucose whilst Alcaligenes fcecalis {B.fcecalis
alcaligenes) does not, although both organisms strongly
activate lactates. Nor does it explain why succinic acid
is oxidised by both bacteria and muscle tissue, but that
formic acid is a very active hydrogen donator in presence
of bacteria but quite inactive with muscle. This activity
is held to be due to the presence of definite groupings
of molecules in the active centres of the cells which

19C) BACTERIOLOGICAL CHEMISTRY

selectively adsorb different types of compound which
have a corresponding arrangement of their own polar
groups (see p. 31). Thus sugars are adsorbed by one
grouping, succinic and similar acids by another sort of
grouping, lactic and other hydroxy-acids by a third
grouping, and so on. Once this specific adsorption has
occurred the general mechanism of the activation is the
same in each case. As would be expected, substances
having a configuration similar to those activated, but
which are themselves not activated (possibly because
the field of force is not strong enough), will partially
inhibit the activation of the latter, since they can
be adsorbed on to the active centres to the partial
exclusion of the normal substrate. For example, «-hy-
droxybutyric acid, CH3CH2CHOH.COOH, or tartaric
acid, COOH.CHOH.CHOH.COO^, will inhibit the activa-
tion of lactic acid, CH3CHOH.COOH, in virtue of the
common structure — CHOH.COOH which enables them
to compete for places at the active centres ; but they
will not inhibit the activation of succinic acid,
COOH.CH2.CH2.COOH, because it is adsorbed at different
active centres not affected by the hydroxy-acids. Con-
versely malonic acid, COOH.CHg.COOH, or glutaric acid,
COOH.CH2.CH2.CH2.COOH, will inhibit the activation of
succinic acid but not that of lactic acid, because they are
adsorbed on the same centres as succinic acid which has
the common group, — CH2.COOH. This suggestion of
competitive adsorption at enzyme centres has also been
used in the explanation of the mode of action of chemo-
therapeutic substances (see Chapter X).

Warburg suggested that all aerobic oxidations, that is,
those in which atmospheric oxygen is involved, are brought
about through the intervention of iron compounds, such
as haemoglobin, whereby molecular oxygen is transferred
to the substrate in an activated form. He showed that
amino-acids could be oxidised l)y molecular oxygen when
at the surface of charcoal prepared by licating blood, and

BACTERIAL RESPIRATION 197

that the action could be inhibited by low concentrations
of hydrocyanic acid, the concentration necessary being
proportional to the iron content of the catalyst . Narcotics
can also inhibit the reaction, and this he attributed to
their adsorption on the active surface preventing the
access of oxygen.

Kluyver has combined the above ideas into a general
scheme, applicable to all fermentations, which involves
the transfer of hydrogen to oxygen in the case of aerobic
oxidations or to some other acceptor, suitably activated
by an enzyme, in other cases. Where oxygen is con-
cerned, an iron compound such as Keilin's cytochrome
is usually also involved. The enzymes involved in such
hydrogen transfers are called dehydrases, dehydrogenases
or hydrogen transportases.

It is seen that all these processes are coupled oxidation-
reduction reactions, hydrogen being given up by the
substrate, the donator, and transferred to a second
substance, the acceptor. For the purposes of our study
these reactions can be divided into three types : —

Type I. — The hydrogen acceptor is atmospheric
oxygen, that is, direct oxidation occurs, as is the case
with the production of acetic acid in the vinegar fer-
mentations, or the action of many moulds and of the
Mycobacteria on sugars.

Type II. — The hydrogen donator and acceptor are
the same molecule, giving rise to an intra-molecular
fermentation. As an example one may take the conver-
sion of glucose, C6H12O6, into two molecules of lactic
acid, 2C3H6O3. Apparently no hydrogen or oxygen is
required from outside, the new compound resulting
from a rearrangement of the distribution of the hydrogen
and oxygen wdthin the molecule. Actually the process
is not so simple as this, since the final effect is brought
about by a whole series of intermediate reactions. What
it really amounts to is that a single substance is sufficient
for the growth of the organism.

198 BACTERIOLOGICAL CHEMISTRY

Type III. — The hydrogen donator and acceptor are
different compounds, resulting in inter-molecular fermen-
tation. The Type I oxidation is a special case of Type III
in which the acceptor is oxygen. Examples of this type
are the anaerobic fermentations at the expense of the
oxygen of fumarates, nitrates, sulphates, or similar highly
oxidised substances. Ehizobium, for instance, can be
made to grow as much as 1 cm. below the surface of agar
containing a small amount of permanganate, whereas
normally it grows only on the surface of solid media.
Obviously for Type III reactions to occur the energy
liberated by the oxidation (dehydrogenation) of the
donator must be greater than that required to cause the
reduction (hydi'ogenation) of the acceptor (see also
Oxidation-Reduction Potentials, Chapter II). Besides
the factor of thermodynamic possibility the enzymic
activation of the substrates comes into play. This is
particularly well illustrated in the case of the streptococci
which cannot use oxygen, the best of all hydrogen
acceptors from the energy point of view. This is not due
to an actual sensitivity to oxygen since, for example,
Str. cremoris can grow in milk exposed to oxygen, but no
oxygen uptake can be measured, whilst Esch. coli under
the same conditions takes up oxygen freely. Again,
Str. lactis suspended in aerated buffer solution takes
up no oxygen, although the majority of aerobes and
facultative anaerobes take up from 5 to 25.jti1. of oxygen
per hour under the same conditions. Clostridium sporo-
genes under these conditions behaves like the strepto-
cocci, and cannot utilise oxygen since it, too, lacks the
appropriate enzyme system. Oxygen uptake is usually
stimulated by the presence of methylene blue and inhibited
by cyanide.

The growth of organisms in oxygen usually involves
the production of hydrogen peroxide which is toxic to
most micro-organisms. Normally this hydrogen peroxide
is destroyed by the enzyme catalase with formation of

BACTERIAL RESPIRATION 199

water and oxygen, even in the absence of any oxidisable
compound to take up the liberated oxygen. On the
basis of the presence of catalase McLeod has divided
the bacteria into four groups : —

(a) Strict Anaerobes. — These organisms have no cata-
Jase and are very sensitive to the presence of hydrogen
peroxide. Since they produce the latter in presence of
oxygen they are incapable of growth aerobically. They
are not sensitive to cyanide.

{b) Micro-aerophilic Organisms. — Members of this
group have no catalase but produce hydrogen peroxide ;
however, they are only moderately sensitive to it, and.
can therefore survive if the oxygen tension is not too
great. As examples may be quoted the pneumococcus,
most streptococci and the lactic acid bacteria.

(c) Non-peroxide Producers. — These produce neither
catalase nor hydrogen peroxide and can tberefore grow
aerobically in spite of the absence of catalase. Examples
are Shigella dysenterice and Str. fcecalis.

(d) Catalase Producers. — Bacteria of this group pro-
duce hydrogen peroxide, which is immediately broken
down by catalase. The aerobes and most facultative
anaerobes belong to this class. They are sensitive to
cyanide, which inhibits catalase action.

Hydrogen peroxide and organic peroxides can also be
decomposed by the enzyme peroxidase, which differs
from catalase in that an oxidisable substance must be
present to take up the liberated oxygen or to donate
hydrogen, as the case may be . If such a second substance is
absent no decomposition of the peroxide takes place. The
organic peroxides usually arise from the oxidation by atmo-
spheric oxygen of di- or trihydric phenols, such as catechol,

200 BACTERIOLOGICAL CHEMISTRY

under the influence of yet another enzyme, oxidase : —

OH { \—0

+ Go oxidase 2 | + 2H2O

OH " > K }—o

(catechol) (catechol peroxidase)

It is to this action that the spontaneous browning of
apples or potatoes in air is due. The peroxides so formed
are then activated by peroxidase to regenerate the original
catechol, with oxidation of the oxidisable substance : —

-0 ( \OVL

I + AH2 peroxidase + A

-0 > \/^^

If the oxidisable substance, AH2, is a compound which
is coloured in the oxidised form it can be used as a test
for the presence of peroxidase. The most common of
these substances are a-naphthol (giving a lavender colour),
guaiacol (red), benzidine (blue), ^^-phenylenediamine
(greenish), dimethyl ^^-phenylenediamine (purple to black)
and indophenol (purple).

The oxidation enzymes, catalase, oxidase and peroxi-
dase, are all inhibited by the action of cyanide, sulphide
and carbon monoxide. They all contain iron in the form
of hsem (an iron compound of protoporphyrin) which is
the prosthetic group carried by different specific proteins
to give the complete enzyme . The breakdown of hydrogen
peroxide by catalase is accompanied by a reduction
and re -oxidation of the ferric iron in the enzyme : —

4Fe+++ + 2H2O2 > 4Fe++ + 4H+ + 20o

4Fe++ -f 4H+ + O2 > 4Fe+++ + 2H2O

The oxidation enzymes play a further important part
in the respiration of bacteria in that they are involved
in the action of the respiratory pigments. Perhaps the
best loiown of these is the cytochrome complex which
consists of three iron-containing components, a, h and c,

BACTERIAL RESPIRATION 201

related to haematin and distinguished by characteristic
bands in their absorption spectra. Cytochrome appears
to be present in all cells exposed to oxygen (with the
exception of some streptococci, e.g. Str. lactis). Cyto-
chrome can exist in the oxidised and in the reduced
forms containing iron in the ferric and ferrous forms
respectively : —

Oxidised cytochrome v Reduced cytochrome.

The oxidised form can act as a hydrogen acceptor in
presence of dehydrase (or dehydrogenase) enzymes giving
reduced cytochrome at the expense of the hydrogen of
the donator, DHg : — •

+ Dehydrogenase
Oxidised cytochrome ^Redu?ed cytochrome + D.

Reduced cytochrome can be re -oxidised by atmospheric
oxygen under the influence of cytochrome-oxidase with
production of water : —

HgO 4- Oxidised cytochrome ^ —

Oxidase

Reduced cytochrc

SO that in effect the whole system acts as a catalj^st,
bringing about the oxidation of the substrate, DHg, to
D and water : —

DH2

+

HoO + Oxidised cytochrome

Dehydrogenase
=- Reduced cytochrome + D

Oxidase

Oo

The first step, the reduction of oxidised cytochrome by
dehydrogenase, can be inhibited by narcotics like chloro-
form or the urethanes, so that in their presence there is
an accumulation of oxidised cytochrome. The second
stage can be inhibited by the action of cyanide or
sulphide, under whose influence reduced cytochrome
accumulates.

202 BACTERIOLOGICAL CHEMISTRY

The reduction of oxidised cytochrome to reduced
cytochrome can be catalysed by a number of dehydro-
genases specific for the substrate which is the hydrogen
donator. Thus the dehydrogenases for the oxidation of
« -glycerophosphate to glyceraldehyde phosphate, of
succinic acid to fumaric acid, of lactic acid to pyruvic
acid and of formic acid to carbon dioxide, all transfer
hydrogen to cytochrome. Some of them can make use
of acceptors other than cytochrome ; for instance methy-
lene blue or pyocyanine are acceptors for glycerophosphate
dehydrogenase but riboflavin, the f la vo -proteins, gluta-
thione or ascorbic acid cannot serve this purpose ;
methylene blue can also accept hydrogen from succinic,
lactic and formic dehydrogenases.

Aerobic bacteria contain all the cytochrome com-
ponents, the facultative anaerobes one or two of them,
whilst the strict anaerobes contain no cytochrome at
all. The respiratory activity of aerobic organisms is
proportional to the amounts of cytochrome and cyto-
chrome-oxidase which they contain. Almost all aerobic
respiration takes place through the cytochrome system.

In addition to the cytochrome system there are other
systems which have a similar function in acting as
intermediaries in hydrogen transfer reactions. Thus in
alcoholic fermentation acetaldehyde is reduced to ethyl
alcohol and phosphoglyceraldehyde is oxidised to phospho-
glycerate by a pair of coupled reactions in which co-
enzyme I, diphosphopyridine nucleotide, acts as hydrogen
carrier : —

(1) 3-Phosplioglyceraldehyd3 + phosphate + co-enzyme I
Triose phosphoric enzyme

1 : 'i (liphosplmglycerate + dihydroco-enzymo T

(2) Diliydroco-cJizyme J + ('IfyCilO > ( 'ociizyii-c i i (Uy)H.

BACTERIAL RESPIRATION

203

The pyridine ring in co-enzjTue I becomes reduced
to give dihydroco -enzyme I which can be re -oxidised,
in the presence of the specific flavoprotein acting as a
dehydrogenase, to give the original co-enzyme : —

CH

CH

C.CONH,

II I

CH CH

\//

N 4-
HOCH I

N = C.NHa

I I

CH C— No

N — C— N/'
CH-

CH

HOCH

OH

HOCH

HOCH

I
CH-

I I I I

CHj— 0— P— 0— P— 0— CH2

II II

0 0

CH

CH

II
CH

C.CONHa
CHa

N

I
CH-

I
HOCH

HOCH

I
CH-

I

OH

-f 2H
- 2H

N= C.NH2

I I

CH C— N.\

I! II ^

II II /

N— C— N^

I
CH-

HOCH

I
I

CH

O

OH

HOCH

I
CH

CHo— 0— P— 0— P —0 — CH.

II II

0 0

The breakdown of formic acid to hydrogen and

204 BACTERIOT.OOTCAL CITEMTRTRV

carbon dioxide (sec p. 249) similarly takes place with
CO -enzyme I acting as hydrogen acceptor, and then
becoming re-oxidised.

Co -enzyme I plays a similar role in the action of alde-
hyde mutase in producing alcohol and acetic acid from
acetaldehyde : —

+ H2O

(1) CH3CHO + Co-enzynie I > CH3COOH + reduced co-enzyme I

(2) Reduced co-enzyme I + CH3CHO > Co-enzyme I + C2H5OH.

The overall effect is the coupled oxidation and
reduction of acetaldehyde : —

2CH3CHO + HoO > CH3CH2OH + C'HgCOOH.

The dismutation of triose phosphate in yeast fermenta-
tion takes place by the same mechanism (see page 276).

Co -enzyme II, triphosphopyridine nucleotide, con-
stituted similarly to co -enzyme I but containing three
instead of two phosphate groups, behaves in the same
way in the conversion of hexose-monophosphate to
phosphohexonic acid, and in the citric acid cycle.

The f la vo -protein enzymes which participate in the
oxidation of dihydroco -enzymes contain the prosthetic
group, riboflavin adenine di -nucleotide, which acts as
hydrogen acceptor from the dihydroco -enzyme and gives
rise to the colorless dihydro compound (see opposite page) ;

BACTERIAL RESPIRATION

205

CH2— 0— P—
HCOH OH

HCOH

I
HCOH

CH2

I
CH N N

^\/\^\
CH3C c c c = o

I II I I

CHoC C C NH
CH N C=0

-P— 0— CH2

I I

OH HC—

HCOH

0
HCOH

I
CH-

I
/N — C— N

CH

N— C

I ,

NH2.C=:=N

CH

I

+ 2H

Yellow.

0 O

II II

CH,— 0— P— 0— P— 0— CH.

HCOH
HCOH

HCOH

I
CH,

OH

I I
OH HC

HCOH

HCOH

I
CH -

CH N NH
CH3C c c c = o

I II II I

CH3C C C NH

CH NHC = 0

Colourless,

/N— C

CH

-N— C
NHa.C:=

• N
II

Jh

.N

The dihydro- (or leuco-) flavoprotein can be re-
oxidised by atmospheric oxygen to form the original
enzyme and hydrogen peroxide. In some cases the
leuco -flavoprotein may be re -oxidised with the interven-
tion of yet another carrier, probably cytochrome, before

206 BACTERIOLOGICAL CHEMISTRY

the hydrogen is finally handed over to atmospheric
oxygen.

The flavin adenine dinucleotide and its carrier
protein (sometimes called diaphorase) are present in
animal tissues and micro-organisms and constitute the
enzyme necessary to oxidise reduced co -enzyme I and
co-enzyme II. Warburg's " yellow respiratory enzyme,"
consisting of riboflavin phosphate and protein, serves the
same purpose but is not found in animal tissues.

The aerobic oxidation of a substrate such as lactate
can be summarised by the following equations : —

(1) Substrate + co-enzyme I >

Oxidised substrate + dihydroco -enzyme I

(2) Dihydroco-enzyme I + flavoprotein >

Co-enzyme I -f- leucoflavoprotein.

(3) Leucoflavoprotein + oxidised cytochrome ^

Flavoprotein + reduced cytochrome.

(4) Reduced cytochrome + oxygen >

Oxiclit^ed cytochrome + water.

The necessity for the long series of steps between the
initial substrate and oxygen arises because the reactions
must all take place under conditions of pH. and tempera-
ture compatible with living cells, but must provide a
considerable amount of energy. Obviously if the change
took place in one step the reversal necessary to keep up
a supply of the enzyme would involve somewhat drastic
conditions. As a somewhat crude analogy the process
of enzymatic oxidation of a substrate might be likened
to transferring sacks full of some cargo from the deck of
a ship to the hold and returning the sacks to be refilled.
The sacks might be dropped straight into the hold, but
the distance might well be too great to throw the empty
sacks back again. If, however, the cargo were emptied
from the sack at deck level into one at a slightly lower
level, the empty sack could be handed back and refilled
easily, and the process carried on from level to level until

BACTERIAL RESPIRATION 207

tlie bottom of the hold was reached. The cargo would
all be transferred from the deck to the hold but no sack
would have to be moved through more than a small
distance-. The cargo obviously represents hydrogen
atoms, the sacks are the enzymes, the different levels
are the different co -enzymes, the deck is the substrate
and the hold represents atmospheric oxygen.

The autotrophic hydrogen bacteria contain a hydro -
genase by which the reduction of such substrates as
oxygen, nitrate, sulphate or fumarate by molecular
hydrogen is effected, a process very similar to reduction
by hydrogen in presence of platinum black as a catalyst.

It is possible to induce many anaerobes to grow in the
presence of oxygen by causing the medium to have a
reducing potential sufficiently high to overcome the
effect of the oxygen and prevent the formation of hydrogen
peroxide. This can be accomplished by the addition of
strongly reducing substances like cystein or the oxidation-
reduction system glutathione, both of which contain the
sulphydryl group ■ — 8H. Glutathione is a tripeptide
( Y-glutamylcysteylgiycine) comj)osed of glycine, cystein
and glutamic acid : —

COOH.CH.CH2.CH2.CO.

NH.CH.CO. NH.CHo.COOH

I
CH2

L

(glutamic acid) (cystein) (glycine)

Its formula is usually written in brief as GSH.

Two molecules of this reduced form of glutathione can
combine with elimination of hydrogen to give the oxidised
form : —

GSH

GS

+

+

0

^==

-^ 1

+

HoO.

GSH

GS

The oxidised form can be readily reduced again if a

208 BACTERIOLOGICAL CHEMISTRY

suitable hydrogen doiiator (almost invariably a protein)

is present : —

GS G8H

I + DH2 > + + D.

GS GSH

(protein)

Thus glutathione acts as an intermediary in hydrogen
transfer from the donator to oxygen as acceptor in such
a way that hydrogen peroxide is not formed.

It is very probable that glutathione is the prosthetic
group of the enzyme glyoxalase which brings about the
conversion of methylglyoxal to lactic acid (see p. 246)
by an internal dismutation : —

CH3CO.CHO + H2O — > CH3CHOH.COOH .

We have seen that when an organism grows in a
medium containing an organic compound as the source
of energy it usually oxidises that compound which accord-
ingly loses a certain amount of hydrogen. If the process
is aerobic the hydrogen is taken up by oxygen, but under
anaerobic conditions some substance other than oxygen
must act as the hydrogen acceptor. In order that it may
do so it must be activated by the organism concerned.
Aerobically only one compound, the hydrogen donator,
has to be activated, but anaerobically both donator
and acceptor have to be activated. If a medium
contains two compounds which can be activated in this
way (one as donator and one as acceptor) by an organism,
which is normally aerobic, it will support the anaerobic
growth of that organism, but not of an organism which
can activate only one or neither of the compounds. The
following examples illustrate this point. Each of the
organisms Esch. coli, Serratia marcescens, Proteus vulgaris
and Alcaligenes fcecalis has been shown by the methylene
blue technique to activate lactate as a hydrogen donator.
Both Esch. coli and Ser. marcescens can activate fumarate
and nitrate to act as hydrogen acceptors, and in conse-
quence these two organisms can grow anaerobically on a

B ArTF:RlAL RESPIRATION

209

medium containing eitlier kiftate and f umaratc or lactate
and nitrate, but not on one containing only one of these
substances (see Table 17).

Table 17

Activates

Lactate as

Donator.

Activates as
Acceptor.

Anaerobic
Growth,

Fuinarate.

Xitrate.

L^F.

L+NO3.

Esch. coll -
Ser. marcescens -
Proteus vulgaris -
A Icalige nes fceca lis

-r
+

+

-f

+
— ,

a.
+

Proteus vulgaris can activate nitrate as hydrogen
acceptor, but not fumarate, and therefore it is capable of
anaerobic growth on a lact at e+ nitrate medium but not on
lactate + fumarate. Finally, Alcaligenes fcemlis, which
will activate neither fumarate nor nitrate, cannot be
induced to grow anaerobically on either of the media.

Esch. coll contains enzymes which can activate
glucose, glyceraldehyde, glycerol, acetate, butyrate,
Z-glutamate, lactate, malate, pyruvate and even mole-
cular hydrogen as donators of hydrogen to fumarate
as acceptor under anaerobic conditions.

The majority of bacteria can reduce nitrates to nitrites,
and many of these can further reduce nitrites to ammonia ;
one group, the denitrifying bacteria like Pseudomonas
fluorescens, reduces nitrites with production of gaseous
nitrogen. Some organisms while unable to reduce nitrates
are able to reduce nitrites. Among the organisms which
are capable of reducing nitrates to nitrites but not further
are the Vibrios ; some of these, such as V. comma, can
also produce indole from the tryptophane in peptone
and use is made of this property in their diagnosis. On

210

BACTERTOLOGICAL CHEMISTRY

addition oi" SLdpliuric acid to such a culture in peptone
water a red colour develops (the so-called " cholera red
reaction "), due to the nitroso -indole reaction between
nitrite and indole.

In intra-molecular fermentations of Type II (see
p. 197) complete oxidation does not occur as a rule. In
these cases the energy is supplied by a shift of the oxygen
in the molecule, usually towards the ends of the chain.
This may be illustrated by a comparison of glucose and
its fermentation products, alcohol and carbon dioxide,
as the result of yeast fermentation, or laotic acid following
bacterial fermentation : —

CO,

CH.

C'H,

CHOH
CHOH

O CHOH

Yeast

I
CH2OH

+
CO2

(alcohol + carbon dioxide)

Bacteria

COOH
CHOH

CH3

+
CH,

CHOH

CH CHOH

CH2OH COOH

(glucose) (lactic acid)

Compounds in which this accumulation of oxygen at the
ends of the carbon chain has already occurred are not
caj)able of serving as energy sources by intra-molecular
fermentation. If they are fermented at all it is by the
mechanism of Type III, in which some outside hydrogen
acceptor is necessary. Thus the simple alcohols, fatty
acids and dibasic acids, like oxalic, malonic or succinic
acids, require either oxygen or some hydrogen acceptor
like nitrate in order that they may be utilised by the
organism. It is the uniform distribution of oxygen along
the chain of carbohydrate molecules which renders them
so valuable as nutrient materials. Hydroxy-acids, like
tartaric or lactic acids and amino -acids, which yield
liydroxy-acids on hydrolytic deamination (see Chapter
Xiri), can also be fermented anaerobically by Type II
reactions.

BACTERIAL RESPIRATION 211

From a consideration of all these facts it can be
appreciated that it is not possible to define the classes
aerobic, facultative and strict anaerobic organisms at all
rigidly. For instance, it is not strictly accurate to define
aerobes as bacteria which need oxygen as hydrogen
acceptor, since we have seen that they will grow anaerobi-
cally if other suitable hydrogen acceptors are provided.
The converse also holds ; the normally strict anaerobe,
CI. sporogenes, can be made to grow in oxygen if a strongly
reducing substance like cystein is added to the medium.
The role played by the toxicity of oxygen must also be
borne in mind, particularly in considering micro -aerophilic
organisms which will only tolerate a lowered oxygen
tension in spite of the fact that they normally use oxygen
for their respiratory processes. It should be pointed
out that oxygen is not itself toxic, but gives rise to
hydrogen peroxide which is toxic. On the other hand,
the streptococci, which grow in the presence of oxygen
but do not utilise it for their respiration, are not on that
account called anaerobes, although their metabolism is
exactly that of the anaerobes. While oxygen may not be
required for the supply of energy, it is sometimes necessary
for the growth requirements of the organism. The urea
bacteria afford an illustration of this ; they derive their
energy needs by the breakdown of urea to ammonia
and carbon dioxide, but they cannot grow in the absence
of oxygen, because the synthesis of their cell constituents
involves dehydrogenation reactions, needing oxygen as
a hydrogen acceptor, urea itself being unable to serve
as such.

Strictly speaking, then, aerobes and anaerobes are not
clear-cut groups, but may be more or less interchangeable
under appropriate conditions. The differentiation of
bacteria into aerobic, anaerobic and facultative organisms
is, however, a very useful working classification for
general purposes.

A list of the cliief enzymes involved in the respiration

21

BACTERIOLOGICAL CHEMISTRY

of micro-organisms is given in Table 18. The exo-cellular
enzymes of the Hydrolase class are not included since
their function is preparative and they are only indirectly
concerned in respiration, by assisting in the supply of
raw materials.

Table 18

Enzyme

REACTION

Organism

Acetaldehyde reductase -

Acetaldehyde >■ Ethyl alcohol

Bacteria, veasts.

Carboxylase'

Decarboxylation of pyruvic acid -

Yeast.

Catalase - - - -

Decomposition of hydrogen peroxide -

Aerobic and most
facultative anaerobic
bacteria, yeast.

Cytochrome-oxidase -

Oxidation of reduced cytochrome -

All aerobic organisms.

Deaminase

Z-Glutamic acid — >

fl-ketoglutaric acid + Nils

Bacteria, yeast.

Enolase - - . -

2-Phosphoglyceric acid — >

Phosphopyruvic acid.

Bacteria, yeasts.

Flavoprotciiis (Warburg's

Oxidation of reduced Co-enzymes I and II -

Bacteria, yeasts.

" yellow enzyme,"

Diaphorase)

Formic dehydrogenase

Formic acid — >■ Reduced carrier + VO2 -

Esch. coll.

Fumaric hydrogenase

Fumaric acid — >■ Succinic acid -

Esch. coll.
Anaerobic bacteria.

Glucose dehydrogenase or

Oxidation of glucose to gluconic acid -

Penicillia,

Glucose oxidase

Aspergilli,

B. gluconicum.

Glyoxalase . - -

Methyl glyoxal — > Lactic acid -

Lactic acid bacteria,
yeasts.

Hexose monophosphorylase

Glucose-G-phosphate >

G- I'hosphogluconic acid

Esch. coll, yeasts.

Hydrogenase

Reduction of oxygen, nitrate, sulphate,

Aioiobacter,

fvuuarate

Esch. coH,
Autotrophic bacteria.

Hydrogenlyase -

Formic acid — > H2 + CO2

Aerobacter,
Esch. coll.

Isomerase - - - -

Dihydroxyacetone phosphate — >-

Phospho-glycera Id ( 'hyd e

Yeasts.

Lactic dehydrogenase

Lactic acid >■ Pyruvic acid . . .

Esch. coll.

iX. gmiorrhoecB, yeasts.

Lactic acid enzyme -

2 Pyruvic acid — y

Lactic acid bacteria,

Lactic acid + acetic acid + CO2

X. gojiorrhoecB.

Peroxidase

Oxidation of substrates by hydrogen peroxide

A', gonorrhcece,
Acetobacter peroxidans,
Streptococci.

Phosphogluconic acid

G-1'hosphogluconic acid >

Acetobacter, yeasts.

enzyme

6-Phospho-2-ketoglucouic acid

Phosphoglycero-nuitase

3-Phosphoglyceric acid >■

2-Phosphoglvceric acid

Escli. coll, yeasts.

Pyruvic oxidase

Pyruvic acid — > Acetic acid + CO2 -

L. delbrilcHi,
A', gonorrliccce,
8tr. pyogenes.

Succinic dehydrogenase

Succinic acid > Fumaric acid - - -

Esch. coli, etc.

Triose phosi)horylase

Phosphoglyceraldehyde — >■ Phosphoglyceric
acid (with reduction of co-enzyme I)

Esch. coli, yeasts.

Zymohoxase

Fructose-1 : G-diiihosiiluite — >

Esch. cuU, yeasts.

Triosephosj)hates

BACTERIAL RESPIRATION 213

For further reading : —

D, K. Green, " Mechanisms of Biological Oxidations." The University

Press, Cambridge, 1940.
1). Keilin. " Cytochrome and Intracellular Respiratory Enzymes.'

Ergebnisse fur Enzymforschnny, 2 (1933), 239.
J. H. Quastel, " Dehj^drogenations produced by Resting Bacteria. A

Theory of the Mechanism of Oxidations and Reductions in vivo.^'

Biochem.J., 20 (1926), 166.
M. Stephenson, " Bacterial Metabolism," Chapter II. 2nd Edition.

Longmans, Green & Co. London, 1939.
H. Wieland, " On the ^Icchanism of Oxidation." Oxford University

Press. London, 1932.

CHAPTER XIII
NITROGEN METABOLISM

THE cells of bacteria, yeasts and fungi may contain
as much as 87*5 per cent, of nitrogenous constituents,
although about 70 per cent, is a more usual value.
The greater proportion of these substances comprises the
proteins of the protoplasm and nuclear material of the
cells, but no less important in function, though smaller
in amount, are the enzymes, which are all nitrogenous.
In general the bacterial proteins are like those found in
other organisms, being built up of the same amino -acid
units, but individual differences occur from species to
species. Many of the serological distinctions between
bacteria of different species depend on differences between
the proteins contained in them.

Obviously, in order that bacteria may grow and
reproduce, a supply of nitrogen, as well as of other con-
stituents, must be available from which the cells can
synthesise the proteins and enzymes and other nitro-
genous compounds to be incorporated in the newly formed
cells. We will first consider the forms in which nitrogen
is available to the organism, and then the mechanisms
by which it is converted into an integral part of the
structure of the cell.

Nitrogen Requirements

Elementary Nitrogen. — Free nitrogen can l)e utilised
by certain of tlie soil bacteria, notably Azotohacter and
Bhizobium which fix atmospheric nitrogen, probably
witli the intermediate formation of ammonia. There has

214

NITROGEN METABOLISM 215

been some variation of opinion as to the fixation of
nitrogen by yeasts. About the beginning of the century
it was claimed that species of Torula, Saccharomyces,
Oidium and Monilia were able to use gaseous nitrogen
when grown on artificial media containing only carbo-
hydrates and tap -water. Then followed a period during
which fixation of nitrogen was denied, the growth observed
in the earlier experiments being ascribed to traces of
nitrogenous impurities in the sugars and to ammonia and
nitrates in the water. It was shown that very small
quantities of nitrogen, of the order of 0-01 per cent.,
would serve to support growth. Later Fulmer and Nelson
showed that if Sacch. cerevisice is grown for a long period
on sucrose -phosphate solutions freed from ammonia and
oxides of nitrogen, fixation does occur. The gain in
nitrogen is only apparent after about six weeks, an actual
loss being observed during the early stages, probably
due to conversion of some of the nitrogen in the yeast
into undetectable compounds. This time-lag may well
be the cause of the failure of earlier workers to detect
fixation.

Nitrites and Nitrates. — These substances can be used
as nitrogen source by bacteria of the Azotohacter species,
ammonia probably being formed as an intermediate.
The question is still not settled in the case of the
yeasts, but the balance of opinion is that nitrites and
nitrates are not utilised, especially under aerobic con-
ditions. Their presence in a medium stimulates spore
formation.

Ammonium Salts and Amines. — These substances are
usually readily assimilable, ammonium phosphate being
a good source of nitrogen for all micro-organisms except
Vibrio comyna. Soil bacteria of the N itrosomonas group
oxidise ammonia to nitrite. Free ammonia can be used
by many organisms, but if it is present in any but very
low concentrations it retards growth owing to its toxicity.
The same applies to the amines, some of which, especially

15

216 BACTERIOLOGICAL CHEMISTRY

hydrazines, are very toxic. In the presence of glucose,
ammonium salts are more readily utilised than amines.
The presence of ammonium salts stimulates the utilisation
of amides ; some acids, for example malic acid, which
are not normally fermented can be fermented when
present as the ammonium salt.

In the commercial production of protein by yeasts,
using molasses as the substrate, the replacement of up to
50 per cent, of the nitrogen of the molasses by ammonium
salts leads to an increased yield, but a complete replace-
ment restricts growth ; the optimum concentration of
ammonium salts has been shown to be that which causes
least swelling of the proteins.

Amides. — The amides, particularly urea, can act as
nitrogen source for bacteria and yeasts, probably being
utilised via ammonia as an intermediate. Formamide
appears to be more readily utilised than other amides,
probably because of the constant presence of ammonia
in its solutions.

Amino- Acids. — The majority of amino -acids are
effective as nitrogen sources, although some, for example
tryptophane and tyrosine, may be toxic if present in any
great quantity owing to the end products, indole and
phenol respectively, formed under appropriate conditions.
The open chain amino -acids are more easily attacked than
those containing ring systems (tyrosine, tryptophane,
histidine). Some organisms, for example C. diphthericB,
H. influenzce and Lactobacilli, require complex mixtures
of amino-acids for their growth, although most will
grow on a single simple amino -acid. For the exacting
organisms some amino acids are essential, but others are
not. Thus for the growth of L. arabinosus on a synthetic
medium arginine, cystine, glutamic acid, ^soleucine,
leucine, methionine, phenylalanine, tryptophane, tyrosine
and valine are essential ; in addition other amino-acids
are also required but the need can be supplied by one of
several, whilst omission of any one of those listed causes

NITROGEN METABOLISM

217

failure to grow. Esch. coli and " trained " Eberthella
typhosa can synthesise tryptophane from ammonia and
carbohydrates ; B. anthracis and Staphylococcus aureus
need amino -acids, whilst untrained E. typhosa, C.
diphtherice, CI. sporogenes, CI. hotulinum, etc., need
tryptophane preformed. For the growth of E. typhosa
or C. diphtherice indole can replace tryptophane, but
derivatives such as indole -acrylic acid, indole-acetic
acid or indole -propionic acid cannot. Staphylococcus
aureus cannot convert indole to tryptophane.

Some amino-acids may be toxic unless an adequate
concentration of others is present ; for example, glycine,
3-alanine, serine and threonine are toxic to Str. lactis
in the presence of very small amounts of « -alanine, but
not if larger amounts are present.

The toxic effect of leucine,

CH.NH,.COOH,

I

CH

CH, CH,

threonine.

or a-aminobutyric acid.

CH-NHa-COOH,

I
CH

/ \
OH CH3

[n

H CH3

on B. anthracis can l^e eliminated by valine,

CH.NH2.c00H,

CH

CH,

CH,

218 BACTERIOLOGICAL CHEMISTRY

and vice versa. The toxicity of isoleiic'me, CH.NH2.COOH,

CH

CH3 CHgCHg

norleucine, CH.NHa.COOH,

CH

H CHo.CIlq

or serine, CH.NH2.COOH,

CH

H OH

is removed by a mixture of valine and leucine but not
by either alone. The toxicity of serine can be removed
by threonine. Definite quantitative relationships between
the amounts of the amino-acids exist but they are not
necessarily equimolecular.

The amino-acids are less readily utilised by
yeasts than are ammonium salts, which stimulate
their utilisation. The amino group of asparagine,
COOH.CH.CH2.CONH2, is utilised more readily than

NH2

the amide group.

It is possible that the breakdown of certain amino-
acids serves as the energy source for the anaerobes CI.
sporogenes and CI. hotulinum. The reaction, described
by Stickland, is between two amino-acids, one activated
as hydrogen acceptor and the other as donator ; glycine,
Z-proline and Z-hydroxyproline are acceptors, and Meucine,
c?-alanine, f/- valine, /-phenylalanine, i-aspartic acid and
rf-glutamic acid serve as donators. Tlie reaction is
probably according to the equation : —

KiCH.NHo.COOlI. HiCO.COOH

+ " r H2O — > + 1 2NH3

RCH.NHo.COOH

NITROGEN METABOLISM 210

67. ietcDii does not l)ring about the '' Sticldaml leactioii ''
but breaks down glutamic acid, aspartic acid or serine
to give carbon dioxide, ammonia, acetic acid and butyric
acid, together with some lactic acid and ethyl alcohol
from aspartic acid. The same products are formed in
the dissimilation of pyruvic and fumaric acids, with
malic acid in addition from the latter. Glucose is not
attacked by CI. tetani. CI. tetanomorphum and CI.
cochleariiim can derive energy for growth from the
breakdown of glutamic acid alone.

Ring Compounds. — The purines, pyrimidines and
similar compounds can be used by many bacteria but
not by all. Thus Esch. coll will not utilise uric acid or
hypoxanthine, whereas Aerobacter aerogenes can do so.
Hippuric acid can support the growth of the hsemolytic
streptococci. CI. acidi-urici and CI. cylindrosporum can
utilise uric acid and some other purines.

Proteins, Proteoses and Higher Polypeptides. — Com-
pounds of these types are not utilised directl}' by bacteria
or yeasts, even by the most actively proteolytic organisms.
Purified proteins as a sole source of nitrogen will not
permit the growth of bacteria, most probably because the
cell membranes are not permeable to them. Before
they can be utilised they must be broken down into
simpler substances which can penetrate into the cell,
where they can be acted upon by the true metabolic
enzymes of the cells, the endo-enzymes. This
preliminary breakdow^n of complex proteins to the
simpler nitrogen compounds which can be used is the
function of exo-cellular, hydrolytic enzymes elaborated
by the bacteria. Until enough of such proteoljrtic enzymes
are produced by an inoculum of bacteria proteins cannot
be made available. Usually, of course, there is sufficient
nitrogen available in a simple form in a medium to enable
the culture to start growing and produce enough proteo-
lytic enzymes to allow the organism to utilise any proteins
present. Such organisms as Esch. coll, which do not

220 BACTERTOLOGTCAL CHEMISTRY

produce proteolytic enzymes, can never utilise proteins
or proteoses and their nitrogen must be supplied already
partially broken down to amino -acids or as ammonium
salts. Peptone (which is protein that has been sub-
jected to mild acid hydrolysis) in a medium serves
this purpose.

Nitrogen Fixation
The phenomenon of nitrogen fixation, so important
from an agricultural standpoint, has been known from
time immemorial, but that it is due to micro-organisms
in the soil was only established comparatively recently.
Jodin in 1862 showed that certain micro-organisms (which
he called " mycoderms ") could grow in solutions con-
taining sugar or tartaric acid but no organic nitrogen,
and that if the cultures were allowed to stand in sealed
vessels, nitrogen as well as oxygen was removed from the
enclosed space. Berthelot later showed that soil which
was allowed to stand underwent an increase in nitrogen
content, but that this increase did not occur if the soil
were first sterilised by heat. Later still he isolated organ-
isms which could grow at the expense of atmospheric
nitrogen. Much of our knowledge of nitrogen fixation is,
however, due to Winogradsky, who began to study the
problem in 1893 ; he showed that an anaerobe, CL
pastoriaiium, very closely related to CL butyricum which
also fixes atmospheric nitrogen, was capable of
growth on synthetic media and could derive its
nitrogen from the atmosphere. The fixation of nitrogen
was proportional to the amount of glucose fermented ; for
each 1 gram of glucose destroyed (with formation of
butyric and acetic acids together with carbon dioxide
and hydrogen) approximately 2-5 mg. of nitrogen was
fixed. If other sources of nitrogen, such as ammonium
salts, were present fixation of atmospheric nitrogen
ceased and the nitrogen of the ammonium salts was used
preferentially. Winogradsky suggested that the nitrogen
was reduced with formation of ammonia by nascent

NITROGEN METABOLISM 221

hydrogen evolved during the breakdown of ghicose.

Besides such anaerobic processes of nitrogen fixation,
aerobic organisms are known which effect the same
reaction. Beijerinck isolated organisms from soil and
canal water which, when grown in a nitrogen-free medium
containing an adequate carbon source, actively fixed
nitrogen. The chief organism responsible is Azotobacter
chroococcum, It is usually accompanied in nature by
an organism, Alcaligenes radiobacter, which lives in
symbiosis with it, but is itseK not capable of fixing nitrogen.
A. chroococcum ferments glucose to give mainly carbon
dioxide together with lactic, acetic and formic acids and
some alcohol. Stoklasa showed that when A. chroococcum
is grown anaerobically on a medium containing nitrate it
reduces the latter to nitrite and ammonia, but gives only
feeble growi^h compared with that under aerobic condi-
tions ; a certain amount of nitrogen fixation also occurs
anaerobically. When grown aerobically on nitrate media
good gro\\i;h occurs, the nitrate is reduced mainly to
nitrite, very little ammonia being found since it is probably
used up in the synthetic processes accompanying the in-
creased growth ; fixation of nitrogen, in this case, occurs
only to a very limited extent. Ale. radiobacter seems to
gro'w equally well aerobically or anaerobically and is in
each case an active denitrifier, rapidly converting the
nitrate to free nitrogen, which is lost from the system.
When A. chroococcum and Ale. radiobacter are grown in
symbiosis on nitrate media the nitrogen set free by the
latter is fixed by the former and converted into cell
constituents. When low concentrations of nitrate are
present considerable amounts of atmospheric nitrogen are
fixed, but this does not occur when high concentrations of
nitrate are present in the medium.

It has been shown by other workers that when Azoto-
bacter is grown on synthetic media containing glucose or
mannitol as the carbon source, the organism fixes nitrogen
in four to six daj^s, and that ammonia and amino -nitrogen

222 BACTERIOLOGICAL CHEMISTRY

can then be found in the medium after the organisms have
been removed by centrifugalisation ; nitrites or nitrates
were not detected. These experiments support Winograd-
sky's theory that the nitrogen is fixed by reduction to
ammonia, which is then converted to amino-compounds ;
further evidence is the fact that the fixation is inhibited
by the presence of ammonia and of nitrate, and that
nitrate is reduced to ammonia, as was shown by Stoklasa.

The fixation of nitrogen seems to be intimately bound
up with the growth of the organism since the ratio of
nitrogen fixed to glucose fermented (supplying the energy
required) is highest in the early stages when growth is
rapid, but falls off with the age of the culture until finally
no more nitrogen is fixed, though glucose continues to be
fermented ; nitrogen is no longer fixed at this stage
because there is no further growth, and therefore no
synthetic requirements, but glucose continues to be
fermented by the enzymes which have already been
produced, and liberated by autolysis of the old cells.

The optimum pH for nitrogen fixation by A. agile is
between 7-6 and 7-8. Calcium, which can be replaced
by strontium which is rather less effective, is necessary
for nitrogen fixation by Azotobacter, at a concentration
of 2x 10~* molar. Molybdenum at a concentration of
10~^ molar, vanadium at 2 x 10~^ molar and iron at
10"^ molar stimulate growth and nitrogen fixation of
Azotobacter.

CI. butyricum, after continued cultivation on labora-
tory media, tends to lose its nitrogen fixing power. How-
ever, the power can be restored by cultivation in sterile
soil, a procedure somewhat analogous to animal passage
in the restoration of the virulence of pathogens. Azoto-
bacter shows no tendency to such loss of activity.

Besides the free-living bacteria which fix nitrogen there
is a group which live in symbiosis with the leguminous
plants. The bacteria grow in the nodules on the roots of
the plants which supply the carbon and energy require-

NITROGEN METABOLISM 223

ments of the micro-organisms, which in their turn fix
nitrogen. The legumes, if grown in sterile soil, require
combined nitrogen for their growi^h, do not produce
nodules and do not fix nitrogen. In non-sterile soil
growth can occur in the absence of combined nitrogen,
nodules are produced and atmospheric nitrogen is fixed.
Beijerinck isolated the responsible organism, now know
as Bhizobium leguininosarum and induced it to grow apart
from the plant. The process of fixation of nitrogen by
Bhizobium is considered by Virtanen and Laine to pro-
ceed through formation of hydroxylamine as a first step.
They observed that when peas were grown in sterile
sand inoculated with Bhizobium soluble nitrogen com-
pounds were excreted by roots carrying nodules only.
The nitrogen compounds were identified as the oxime
of oxalacetic acid (1 to 2 per cent.), Z-aspartic acid and
jS-alanine. Ammonia could not be detected. Oxalacetic
acid is also an essential intermediate and is found in
liighest concentration at noon on bright days and lowest
in the dark. The active nodules contain a red pigment
which is identical with, or very similar to, haemoglobin.
Nodules may be formed by non-nitrogen fixing strains
of Bh. leguminosarum but are not pigmented, nor are
those formed by active strains in the absence of oxygen.
If the plants are kept in the dark for several days the
pigment becomes green, due to opening of the porphyrin
ring, and the system can no longer fix nitrogen, even if
the plants are transferred to the light. Virtanen suggests
that the haemoglobin acts as an oxygen transfer system,
becoming oxidised to methaemoglobin with simultaneous
reduction of atmospheric nitrogen to hydroxylamine : —
N2 + Methaemoglobin (Fe^^^) ^=^

NH2OH -f haemoglobin (FeU).
The hydroxylamine condenses with oxalacetic acid,
formed by the plant in the breakdown of sugars, to give
the oxime : —
COOH.CO.CH2.COOH + KH2OH > COOH.C.(NOH).CHoCOOH + H^O

224 BACTERIOLOGICAL CHEMISTRY

The uxiiiie is then rechiced to give aspartic acid,
COOH.CH.NH2.CH2.COOH, which can be decarboxylated
to give p-alanine, CH2.NH2.CH2.COOH. In presence of
the enzyme, transaminase, aspartic acid can transfer the
amino group to pyruvic acid to form a-alanine,
CH3.CH.NH2.COOH, a process by which other amino-
acids may be synthesised.

That nitrogen fixation is intimately connected with
reduction processes involving molecular hydrogen, is
shown by the fact that when nitrogen fixation by
Azotobacter is inhibited by the presence of combined
nitrogen, such as nitrate, the activity of the enzyme
hydro genase is also stopped. The adaptation of the
bacteria to use nitrate more readily is accompanied by
increased inhibition of both nitrogen fixation and
hydro genase activity. This occurs even in the presence
of the substrate, hydrogen. Hydrogenase appears,
therefore, to be an adaptive enzyme whose formation
depends not on the presence of its substrate but on
nitrogen fixation.

Protein and Amino -acid Breakdown

The utilisation of proteins and amino -acids by bacteria
in th^ production of new cells during growth appears to
follow somewhat the same lines as in animal nutrition.
Complex proteins and polypeptides are broken down
outside the cell by the proteolytic enzymes in a manner
analogous to the digestion of the proteins in the stomach
and intestine by the enzymes pepsin and trypsin. The
breakdown products of this hydrolysis, the amino-acids
and the lower peptides, are then taken into the cell, where
they are either directly rebuilt into the proteins character-
istic of the particular bacterial species concerned or are
further broken down into ammonia and simple carbon
compounds, which are then used in the synthetic processes
involved in the growth of the cells.

Protein Degradation. — As indicated above, tlie proteins

NITROGEN METABOLISM 225

are not available to the orgaiiisin as such but must first
be hydrolysed to molecules small enough to penetrate the
cell wall of the bacteria. All the proteolytic organisms
liberate exo -enzymes into the medium where the proteins
are attacked. In other words, it is possible to obtain cell-
free solutions by the filtration of fluid cultures of such
organisms as Proteus vulgaris, B. siibtilis, Serratia ynar-
cescens and most anaerobes, which will hydrolyse proteins,
as shown by the liquefaction of gelatin or the breakdown
of casein. The enzymes responsible are constitutive, that
is, they are produced even when the organism grows on
a synthetic medium containing no protein. In some
cases it has been claimed that they are adaptive enzymes,
only being formed when protein is present in the medium,
although the evidence for this is somewhat doubtful ;
it is very improbable that such proteolytic enzymes
could be adaptive since the protein which would provoke
them cannot penetrate into the cell, where alone it
could influence enzyme synthesis.

Proteoljrtic endo -enzymes also occur in some organisms,
from which they may be extracted after destroying the
cell structure by an appropriate means, such as solution
in bile, or in sodium hypochlorite, or by grinding, or
repeated alternate freezing and thawing. Usually these
endo -enzymes attack peptones and partially degraded
proteins more readily than they do the complex proteins.
For instance, endo-enzymes have been obtained from the
pneumococcus, which will break down peptone to amino -
acids but will not attack gelatin or egg-albumin ; haemo-
Ijrtic streptococci yield an endo -enzyme which destroys
peptone and casein but not serum albumin. Conversely
the proteolytic exo -enzymes do not break down the
proteins completely, but only sufficiently to enable the
fragments to enter the cell, where they undergo further
degradation where the products are of use to the organism.
As we have already mentioned, Esch. coll can grow on
and break down amino -acids but has no effect on proteins,

220 BACTERTOLOGICAL CHEMISTRY

nor even on such sinij^le compounds as the dipeptides,
which are formed by the union of two amino-acids. Even
the active proteolytic bacteria, like Proteus, cannot utilise
pure protein as the sole source of carbon and nitrogen,
since not enough exo-enzjnne is carried over with the
inoculum to break the protein down to diffusible frag-
ments on which the cells must depend for their growth,
and in the absence of growth, of course, no enzymes can
be formed. The addition of small quantities of some
simple nitrogen source is sufficient to allow growth to
start and proteolytic exo -enzymes to be formed, and then
the proteins can be hydrolysed.

Protein Sparing Action. — It is often claimed that the
presence of carbohydrate in media reduces the utilisation
of proteins and the production of proteolytic enzymes.
Kendall states that addition of glucose to gelatin media
delays the formation of the proteolytic enzymes until
all the sugar has been fermented. He explains this by
saying, " When the sugar is exhausted the organism is
forced to derive its energy from the protein constituents,
and the enzyme is then formed to bring about the necessary
changes in the protein to make it assimilable." That this
is not the true explanation is suggested by Berman and
Rettger, who state that the inhibition is due to the acid
produced by the fermentation of the sugar. In the
case of B. suhtilis, which ferments glucose only slowly,
or of Aerobacter cloacce, which yields products which are
not strongly acid, the presence of sugar has little or no
effect on the breakdown of peptone or protein. With
Esch. coli or Proteus, which give much acid, the growth and
chemical activities are quickly brought to a standstill
unless the medium is so heavily buffered that the pH
value never falls low enough to inhibit the breakdown of
protein. When such buffering is employed the breakdown
of protein proceeds as vigorously as in the absence of glucose.

It has also been shown that in the case of Proteus
there is an optimum ^^H at 8-0 for the production of pro-

NITROGEN METABOLISM 227

teolytic enzymes, that good aeration favours production
of the enzymes, and that on synthetic media, even
though good growth may occur in their absence, no
proteolytic enzymes are formed unless small quantities
(0-025 per cent.) of calcium and magnesium are present.

Another possible explanation of apparent protein
sparing action is that the methods of estimating protein
utilisation are fallacious. Normally the extent of protein
degradation is estimated by the amount of ammonia
and amino -nitrogen appearing in the medium as a result
of the breakdown of the protein. In the presence of sugar
less of these breakdown products appear than is the case
when little or no sugar occurs in the medium. Now the
amount of ammonia and amino -nitrogen found in the
medium is that which is left over after the organism has
taken what it needs to build up its protoplasm and other
nitrogenous constituents. The more rapid and profuse
the growth the more of such raw materials will it require
and the less wall be left over in the medium. One of the
chief effects of a plentiful supply of easily assimilated
carbon, such as glucose or other sugar, is to increase the
growth of the organism in such a medium. That is, in
presence of glucose more growth occurs, more amino -
nitrogen is used and less remains in the medium ; in other
words, there appears to be less production of amino -
nitrogen and consequently less breakdown of protein,
whereas in reality there may be just as much protein
breakdow^i, or even more, in presence of sugars.

Amino-acid Degradation. — The amino-acids resulting
from hydrolj^sis of the proteins find their way into the
cell where they are acted upon by the endo -enzymes to
give a variety of products, the nature of which depends
on the amino-acid, the organism and the condition of
the medium. The earliest work on this subject was of
little scientific value since it was done with mixed cultures
of putrefactive organisms on mixtures of proteins, Tliis
period was followed by one in which the action of mixed

228 BACTERIOLOGICAL CHEMISTRY

cultures on pure amino -acids was investigated, but still
it was impossible to connect the action of any one organ-
ism, for instance, with the formation of any particular
type of product. Finally, Harden and Ehrlich instituted
the present stage by studying the effect of pure cultures
of a single organism on single pure amino -acids.

If we take R.CH2.CH.NH2.COOH as representing the
structure of an amino -acid, the various types of break-
down which it may undergo can be summarised as
follows : —

A. Decarboxylation to give the Amine.

R.CH2.CH.NH2.COOH > R.CH2.CH2.NH2 + C()2.

This type of breakdown is apparently only accomplished
by bacteria, and is favoured by anaerobic conditions. As
an example, Ps. fluorescens breaks down glycine to give
methylamine .

B. Deamination.

{a) Reductive to give Saturated Acids,

H2
R.CH2.CH.NH2.COOH — > R.CH2.CH2.COOH + NH3.

The products will be seen to be substituted propionic
acids ; tryptophane gives indole -propionic acid : —

^\ CH2.CH.COOH ^\ CH2.CH2.COOH

I !l II I I II II

[ !l II NH2 > \ II II + NH3

NH NH

The hydrogen accepting amino -acid in the " Stickland
reaction " (see p. 218) undergoes reductive deamination,
at the expense of the second amino -acid which undergoes
oxidative deamination (B.(d), p. 230). Thus glycine gives
acetic acid and ammonia,, ornithine gives a» -amino -
valeric acid and ammonia whilst Z-proline also gives
CO -amino -valeric acid but no free ammonia. Here the
opening of the ring is equivalent to deamination, the
difference being that the freed amino group is held by
the other end of the chain : —

NITROGEN METABOLISM 229

CHa CH, +2H CHa— CH2

I I I I

(Ho CH.COOH " "

NH NH2

(proline) (w-iniino-valeric acid)

Anaerobic conditions are essential, and bacteria are

particularly active in effecting this type of breakdown,

although yeasts and moulds are also capable of it. Almost

all amino-acids are liable to reductive deamination.

(6) Hydrolytic to give a-Hydroxy-acids,

H2O
R.CH2.CH.NH2.COOH ^ R.CH2.CHOH.COOH + NH3.

The products are substituted lactic acids ; indole -lactic

acid from tryptophane, and phenyl-lactic acid from

phenylalanine are examples : —

CH2.CH.XH2.COOH CH2.CHOH.cuoH

I

O'lienylalanine) (phenvl-Iactic acid)

Different bacteria may attack different optical isomers ;
thus Proteus destroys ^Z-tyrosine and leaves the Isevo-
isomer and produces the cZ-hydroxy-acid, whilst B.
syhtilis attacks the Isevo -isomer and leaves (/-tyrosine
unharmed, the product being the Z-hydroxy-acid.
CHo.CH.XH2.COOH
/\"

I I > i I ^ ^^^3

OH OH

(tyrosine) (p-hydroxyphenjl-lactic acid)

(c) Desaturative to give Unsaturated Acids,

R..CH2.CH.NH2.COOH > R.CH=CH.COOH + NH3.

The a- [i linkage is attacked wdth formation of substituted
acrylic acids. This type of breakdown is observed with
the coli-typhoid group of organisms. As examples may
be quoted the formation of fumaric acid from aspartic
acid : —

230 BACTERIOLOGICAL CHEMISTRY

CGOH.CHa.CH.NHs.COOH > COOH.CH=CH.COOH + NH3,

(aspartic acid) (fumaric acid)

and of iminazole-acrylic acid or urocanic acid from

histidine : —

CH =C— CH2.CH.NH2.COOH CH =C— CH =CH.COOH

II II

N NH > N NH + NH3

CH CH

(histidine) (iiiiinazole-acrylic acid or urocanic acid)

(d) Oxidative to give a-Keto-acids,
. R.CH2.CH.NH2.COOH — > R.CH2.C0.C00H + NH3.
Here the products are substituted pyruvic acids. This
type of degradation is not easily detected since the
«-keto -acids are unstable and readily undergo further
breakdown, particularly under the action of yeasts. It
is favoured by aerobic conditions.

The breakdown of Z-glutamic acid and Z-aspartic acid
by H. parainfluenzce to acetic acid, carbon dioxide and
ammonia follows this type of degradation, the steps
being «-ketoglutaric acid, succinic acid, fumaric acid,
malic acid, oxalacetic acid, pyruvic acid, acetaldehyde
and acetic acid from Z-glutamic acid and oxalacetic acid,
pyruvic acid, acetaldehyde and acetic acid from /-aspartic
acid.

The hydrogen donating amino-acid in the " Stickland
reaction " (see p. 218) undergoes this type of breakdown.
Alanine, for instance, yields acetic acid, carbon dioxide
and ammonia : —

H2O
CH3CH.NH2.COOH — > CH3.CO.COOH + NH3 + 2H
CH3CO.COOH — > CH3CH0 + CO2
CH3CH0 — > CH3COOH

All the above types of deamination yield products
which have the same number of carbon atoms as the
original amino-acid. The «-keto -acids, owing to the
ease with Avhich they lose carbon dioxide by the action
of the enzyme carl)Oxylase (see Chapter XV), give rise
to a series of products with fewer carbon atoms than the
parent amino-acid, thus : —

NITROGEN METABOLISM 231

C. With One Carbon Atom Less.
(a) Aldehydes,

R.CHo.CO.COOH > R.CH2.CHO + COo.

The aldehydes are usually not end products since they
undergo further reactions : —

(6) Saturated Acids,

oxidation
R.CHo.CHO > R.CH2.COOH.

The aldehyde is oxidised to give substituted acetic acids,
particularly by bacteria under aerobic conditions. For
example, phenylalanine gives phenyl-acetic acid : —

(c) Saturated Alcohols,

reduction
R.CHo.CHO > R.CH2.CH2.OH.

This type of breakdown of amino -acids is very common
among the yeasts, and is, in fact, the mode of origin of
the higher alcohols which constitute the fusel oil produced
during alcoholic fermentation, as was shown by Ehrlich.
Fungi, Proteus, and the lactic acid bacteria can also
bring about this type of change. As examples may be
mentioned the formation of /50-amyl alcohol from
leucine : —

CH3 CH3

\ \

^CH.CHa.CH.NHo.COOH > ^CH.CH2.CH2.0H.

/ /

CH3 CH3

(Leucine) (jso-amyl alcohol)

Tyrosine gives tyrosol (p-hydroxyphenyl -ethyl alcohol),
h:o<; y('H...{'Hj)]i, and tiyptophane gives tryptophol

IG

232 BACTERIOLOGICAL CHEMISTRY

I II

( p-indole-ethyl alcohol).

NH

{(I) " Hydrocarbons,"

ll.CHa.CH.NHa.COOH > R.CHa.CHa.COOH > K.CHo.CHg + CO2

These products arise as the result of reductive deamination
followed by decarboxylation, that is, by decarboxyla-
tion of the substituted propionic acid. This type of
breakdown occurs by the action of putrefactive bacteria
under anaerobic conditions. Glycine, for example, gives
methane.

D. With Two or Three Carbon Atoms Less.
(a) Acids,

K.CH^.CH.NHa.COOH > R.CH =CH.COOH >

H2O
R.CO.CH2.COOH — > K.COOH + CH3COOH

The substituted formic acids, R.COOH, with two carbon
atoms less than the original amino -acid arise by de-
composition of the unsaturated acid via the p-keto-acid.
For example, Salmonella paratyphi and Sal. schottmillleri
(B. paratyphosus- A and -B) under aerobic conditions
produce j^-hydroxy-benzoic acid from tyrosine and indole-
carboxylic acid from tryptophane : —

CH2.CH.NH2.COOH COOH

OH OH

(tyrosine) Qj-liydroxy-benzoic

acid)

CH,.CH.NHo.COOH ^\ COOH

NH NH

(It-yijlol.lianc) (iinlulc-carbuxylic ;u;ii!)

NITROGEN METABOLISM

233

(6) " Hydrocarbons,"

R.CH2.COOH > RCH3 + CO2 (2 carbon atoms less).

R.COOH > R.H + CO2 (3 carbon atoms less).

These products arise by the decarboxylation of the sub-
stituted acetic and formic acids produced as above. This
type of breakdown appears to occur only with amino-
acids like tyrosine and tryptophane, which contain ring
structures. Tyrosine gives p-cresol (corresponding to
R.CH3) by the action of putrefactive organisms under
anaerobic conditions, and phenol (RH) by the same
organisms under aerobic conditions : —

CH3
/\

I I (p-cresol)

i I

OH

CHa.CH.XHo.COOH

I

OH

OH

(tyrosine)

COOH

(phenol)

OH OH

As much as 0-8 gram of phenol per litre may be formed
by certain organisms isolated from fseces.

Tryptophane gives scatole by the anaerobic action of
Esch. coll and indole by its aerobic action, but only if
sugars are absent from the medium.

_CH2.C00H ,/ \ CH,

NH

CH,.(JH.XH,COOH (indole-acetic acid)

NH

(tryptophane)

COOH

I II II > I II II

^\/\ / ^ /\/'
' NH NH

(^iiiiloli>o;irbo3£:ylir aoiil) (iiulolo)

234 BACTERIOLOGICAL CHEMISTRY

Whether or not an organism decarboxylates the acid
R.COOH to R.H depends on whether or not it possesses
the enzyme carboxylase. Esch. coli does possess it and
can therefore form indole from tryptophane, but Sal.
paratpyhi and Sal. schottmulleri have no carboxylase and
therefore produce no indole but stop at indole -carboxylic
acid.

The following general scheme of amino-acid degrada-
tion (pp. 235, 236) was proposed by Raistrick, as a result
of his observation that histidine under the action of
organisms of the coli-typhoid group gave rise to the
unsaturated acid, urocanic acid (see p. 230). All the
other products of amino-acid degradation are readily
accounted for by reactions of the unsaturated acid, which
is formed as the primary intermediate.

The Factors Influencing the Type of Breakdown.

1. The Organism. — (a) Yeasts. — The yeasts usually
cause hydrolytic deamination followed by decarboxyla-
tion and reduction to give rise to the alcohols, as in the
production of fusel oil.

(b) Moulds. — ^The moulds usually give hydrolytic de-
amination, but do not cause decarboxylation, the product
being the hydroxy-acid. Sometimes the hydroxy-acid
may be completely oxidised.

(c) Bacteria. — The bacteria can bring about any of
the types of reaction described, with formation of a
correspondingly much greater variety of products. De-
carboxylation does not occur so frequently as with yeasts.

2. The Condition of the Medium. — Anaerobic condi-
tions are naturally usually accompanied by reduction
with accumulation of the saturated acids and the " hydro-
carbons." Aerobic conditions favour the production of
other types of product. Generally speaking, the presence
of a readily available source of carbon inhibits the forma-
tion of indole, p()ssil)ly by (vhangino; the course of the
]>]V'dkd(y\vn.

NITROGEN METABOLISM

235

<

X

o
o
o

^ o

o

X

w

o

o

o

w"-

— J

- h

q

X

-,^

o

t-H

fi£

«t

8

1

X

o
o

s

w

1

O-

-^ +

o

x'

o

K

X

9

O

1

+

t-M —

W^

— o ~

^I

o

d

g

o

g

_•

II

X-

o

1 «

.1

p^

Ph

ec

'P

^

r'

s

^

•-S

6

>>

+i

^

i>

eg

^

^

S

^

a

o

— >o

236

BACTERIOLOGICAL CHEMISTRY

o

CD

s

<

o
+ II

o

3g

C>

CJ

X

o

z

g

rS
o

o

"

8"

-> +

.2

o

W

+

z

W

o

i-i»

8

o_

^c

?_
^

o

— ^i

^

o^

c

>y

11

5

s

W

z

o
o

1

z"

«■

9

o~

-^o

P^

ci

ec

M

i

O

^

'1

1 >^

^

o

©

W

z

o

q

9,

■5

z

§

p4

s

a

a:

1-^

a

S
o

O »

o o
o o

->o-

NITROGEN METABOLISM 237

3. The Character of the Group R. — Jiong chaiii amino -
acids are more easily attacked than those containing
ring structures. The ease of attack appears to increase
with the length of the chain ; glycine, the simplest amino -
acid, CH2.NH2.COOH, is very resistant to bacterial
action, but alanine, CH3.CH.NH2.COOH, is much less
resistant. Ring structures containing nitrogen can be
attacked by some organisms but not others. Putre-
factive organisms can open the ring of proline with
formation of w -amino -valeric acid and of n- valeric
acid : —

CH2— — CHo

I I

CH, CH.COOH >

NH

(n-valeric acid) (o-amino-valeric acid)

The iminazole ring of histidine is opened by Esch. coli,
Sal. 2)aratyphi and Sal. schottmidleri and Ps. fluorescens,
but not by Proteus vulgaris. Tryptophane is similarly
attacked. Moulds can break down any type of nitrogen
ring if no other source of nitrogen is available.

The deamination of serine by Esch. coli, which can
occur aerobically or anaerobically, appears to involve the
hydroxyl group since if it is masked by substitution
deamination does not occur. The suggested mechanism
is via the imino acid and pyruvic acid : —

CH2OH.CH.COOH -H2O CH2=C.C00H CH3.C.COOH

I — -> I — > li

NH2 NHo NH

(serine) (imino acid)

NH

+ H2O

> CH3CO.COOH + NH.

The sulphur containing amino-acids like cysteine,
CH2.SH.CH.NH2.COOH, appear to undergo reductive

238 BACTERIOLOGICAL CHEMISTRY

deamiiiatioii accompanied by further breakdown with
liberation of hydrogen sulphide, the products of the
action of Proteus vulgaris on cysteine being hydrogen
sulphide, ammonia, carbon dioxide, hydrogen and acetic
acid.

The course of the breakdown of amino-acids can be
followed by analysing the resulting solutions for the
following fractions : —

1. Total nitrogen (by Kjeldahl's method).

2. Amino -nitrogen (by van Slyke's method).

3. Synthetic nitrogen in proteins and cellular material

(by precipitation with colloidal iron and Kjeldahl
estimation).

4. Ammonia nitrogen (by vacuum distillation).

5. Non-amino-nitrogen, that is nitrogen in the ring,

is given by 1— (2+3+4).

An increase in the amino -nitrogen figure indicates
breakdown of the side chains. A decrease in the non-
amino -nitrogen follows a breakdown of the ring structure.
An increase in the " synthetic nitrogen " value is usually
accompanied by a reduction of the amino-nitrogen and
ammonia nitrogen values since the proteins are built up
at the expense of such compounds. When an alternative
source of carbon, for example glucose or glycerol, is present
there is usually an increased utilisation of amino- and
ammonia nitrogen and a corresponding increase in the
" synthetic nitrogen," which is the probable explanation
of the so-called " protein-sparing " action of carbo-
hydrates (see p. 226).

When organisms act on racemic amino-acids both
isomers are usually attacked, but the naturally occurring
one more readily, so that an optically active mixture
results. Three possibilities arise : (a) The natural com-
ponent is attacked so rapidly compared with the other
one that an almost optically pure amino -acid results ;

NITROGEN METABOLISM 239

(h) liotb enantiomorphs are attacked at the same rate,
giving an inactive residue, and (c) the rates are different
but not very widely so, resulting in an optically active
but impure mixture of the acids. The method can be
used as a means of resolution of the isomers, but has the
disadvantage of giving a maximum possible yield of 50
per cent, and of destroying the naturally occurring isomer.
The problem of protein synthesis by micro-organisms
will be considered in Chapter XVIII,

For further reading : —

E. F. Gale, " Enzymes Concerned in the Primary Utilisation of Amino
Acids by Bacteria." Bad. Reviews, 4 (1940), 135.

P. Hirsch, " Einwirkmig von Mikro-organismen auf die Eiweisskorper."
Die Biochemie in Einzelldarstellmigen. IV.

H. Raistrick, Papers in Biochem. J., 11 (1917), 71; 13 (1919), 446; 15
(1921), 76.

M. Stephenson, " Bacterial Metabolism," Chapter V. 2nd Edition.
Longmans, Green & Co. London, 1939.

(JHAPTER XIV
CARBON METABOLISM

AS we have seen, many organisms are cajDable of
building up all their cell constituents and can
maintain and reproduce themselves, using only a
single organic substance and ammonium salts as the
sources of raw materials and energy. Substances which
can serve in this way to support growth may belong to
almost any type of compound, saturated and unsaturated
fatty acids, hydroxy-acids, keto-acids, di- and tri-basic
acids, alcohols, carbohydrates, amines, amino -acids,
amides, and aromatic compounds among others. Gener-
ally speaking, only a comparatively small proportion of
the compound destroyed ultimately finds its way into the
composition of the cell ; the bulk of the compound is
more or less profoundly altered during the processes by
which energy is obtained. This altered part of the sub-
strate accumulates in the medium as the products of
fermentation which are characteristic of the various
organisms. Some organisms, the yeasts, for instance,
break up sugars with formation of alcohol and carbon
dioxide, others break up sugars with production of such
substances as lactic acid, acetic acid, butyl alcohol, or
acetone. All these fermentation products are, from the
point of view of the organism, waste products, although
they may be very valuable to mankind.

The part of the compound which is converted into
cell material and taken into the composition of the
newly produced cells is said to be assimilated. The
portion which is broken down to provide the energy

240

CARBON METABOLISM 241

requirements is said to ]>e dissimilated. Comparatively
little is known about the processes involved in assimi-
lation, but dissimilation has been more intensively
studied.

Dissimilation, particularly under aerobic conditions,
is often a catalytic oxidation process. The oxidation of
ethyl alcohol to acetic acid by the various vinegar organ-
isms is a case in point ; it is a partial oxidation, for the
organism is not able to carry the reaction further to form
carbon dioxide and water. Wieland, it will be remembered,
proposed that these oxidations were in reality dehydro-
genations, since it was possible to replace free oxygen
by reducible substances like methylene blue or quinone
which can act as hydrogen acceptors . Bacteria and other
organisms are able to bring about these dehydrogenations
by means of the enzymes, dehydrogenases, dehydrases or
hydrogen transport ases, which they produce and which
activate the hydrogen atoms of the substrate, ethyl
alcohol, acetaldehyde, etc., which is to be oxidised. The
vinegar organisms, species of the genus Acetobacter,
usually oxidise substances other than alcohols completely
to carbon dioxide and water ; but one species at least is
loiown, A. suboxydans, which is capable of oxidising
partially many other substances, of which the following
examples will serve as illustrations : —

/OH

1. {a) CHa.C^ H + 0 ^ OHg.cf + H,0

\. • \H "

(ethyl alcohol) (acetaldehyde)

/OH
^O / OH

(b) CHg.cf + HoO > CHg.Cr — OH + 0 > OHg.c/ + HoO

^H \\ ^O

\h

(acetaldehyde hydrate) (acetic acid)

242 BACTERIOLOGICAL CHEMISTRY

2. CH3. M (.'Hg.^

>C< + O ^ >(!-() ■+ HgO

CH3/ \UH CIH3/

isopropyl alcohol) (acetone)

3. (a) — CHOH O /OH OH

I // / /

I C-H C^ — OH C = 0

O (CH0H)3 1 I

+ H2O I ^H +0 I

I > (CH0H)4 > (CH0H)4 > (CH0H)4 + HgO

OH I I I

CH2OH CH2OH CH2OH

CH2OH

(glucose) (glucose hydrate) (gluconic acid)

(6) OH OH

/ /

c=o c=o

(CH0H)3 + O > (CH0H)3 + HoO

HCOH CO

CH2OH CH2OH

(gluconic acid) (5-ketogluconic acid)

A. suhoxydans oxidises glucose to gluconic acid and
5-ketogluconic acid in an acid medium ; if the medium
is kept natural by carrying out the fermentation in
presence of calcium carbonate, 2-ketogluconic acid is
formed.

The point to be noted in each of these examples is
the activation of two hydrogen atoms (shown in bold
type in the formulae) and their transfer to an oxygen
atom. The reactions can all be induced to occur anaerobi-
cally if a suitable hydrogen acceptor like methlyene blue
is provided in place of the oxygen. A. suhoxydans
restricts its activities to the oxidation of the secondary

alcohol group, ^CHOH, to the keto group, ^C = 0, but
A, xylinum, Bertrand's sorbose bacillus, which also
oxidises the same group in the same way, producing
/-sorbose.

CARBON METABOLISM 243

CH2OH CH2OH

I I

CO HOCH

I I

HOCH from sorbitol HOCH

HCOH HCOH

I I

HOCH HOCH

I I

CH2OH CH2OH

and dihydroxyacetone from glycerol,

CH2OH CH2OH

I I

CHOH > C = 0

I I

CH2OH CH2OH,

for example, can, under conditions of vigorous aeration,
carry the oxidation to completion, forming carbon dioxide
and water, probably by a chain of such reactions. A.

xylinum only oxidises the "^CHOH group when the
hydroxyl group is adjacent to a primary alcohol group
and to a second hydroxyl group, that is, alcohols with

i

HCOH

the cis-configuration, | , are oxidised but not

* HCOH

' I

HCOH

those with the ^rari5-configuration, |

HOCH

I

The potential aldehyde group of sugars is oxidised to
a carboxyl group.

Both A. xylinmn and a similar organism called Bad.
ghiconicuiib in addition to gluconic acid and 5-ketogluconic
acid oxidise glucose to 0-aldeliydo-gluconic acid whicli is
identical with /-guluronic acid : —

244

BACTERIOLOGICAL CHEMISTRY

CHO
H.C.OH
HO.C.H

H.C.OH

I
H.C.OH

I
CH2O:

(glucose)

COOH

I
H.C.OH

I
HO.C.H

I
H.C.OH

H.C.OH

I
CH2OH

(gluconic acid)

COOH

i
H.C.OH

I
HO.C.H

H.C.OH

H.C.OH

I
CHO

(«-aIdehydo-
i^-gluconic acid)

CHO

I
HO.C.H

I
HO.C.H

H.C.OH

HO.C.H

I
COOH

(i-gularonic
acid)

Bad. gluconicwm, in neutral solution in presence of cal-
cium carbonate, was also found to oxidise gluconic acid
to 2-ketogluconic acid, which is in contradiction of
Bert rand's rule, since the two hydroxyl groups are in
the tmns--positioii : —

COOH

H.C.OH

I
HO.C.H >

I
H.C.OH

H.C.OH

I
(^HoOH

COOH
CO
HO.C.H
H.C.OH
H.C.OH

Two 2^t>ints should be noticed, however ; first, that a
different organism is involved and second, that a carboxyl
and not a primary alcohol group is adjacent.

Like A. xylinum, Bad. gluconicmn and Bad.
xylinoides oxidise sorbitol to /-sorbose, the yields being
59, 76 and 60 per cent, respectively. A. xylinmn is used
commercially in the production of ascorbic acid to convert
sorbitol, obtained l)y the reduction of glucose, into
/-sorbose, which is oxidised to 2-keto-/-gulonic acid whose
iiKithyl est(uv is readily transformed to ascorbic acid: —

CARBON METABOLISM

245

CO

I

HOCH -

HCOH

I
HOCH

CH2OH

(Z-sorbose)

COOH

Jo

I

HOCH -
HCOH

HOCH

I

(2-keto-Z-gulouic
acid)

CO

I '

HOC

O

CH

I
HC

HOCH

CH2OH

(Z-ascorbic acid)

Acid Fermentation by Bacteria. — The acids most
usually found as a result of bacterial fermentation
are formic, acetic and lactic acids, but propionic,
succinic and butyric acids with some others are also
found with certain organisms and under appropriate
conditions. Various theories to account for their pro-
duction have been proposed, and a number of inter-
mediate compounds suggested. The chief among the
latter are acetaldehyde, pyruvic acid and methyl-
glyoxal.

Acetaldehyde, like aldehydes in general, forms a
water-insoluble compound with sulphites or bisulphites.
If, then, bacterial fermentation is allowed to proceed in
the presence of sulphite (bisulphites are usually poisonous
to bacteria) any acetaldehyde formed as an intermediate
will be " trapped " as the insoluble compound and will
play no further part in the process but will accumulate.
In this way the formation of acetaldehyde has been
detected during the fermentation of glucose, mannitol
and glycerol by members of the Esch. coli group ; among
the products of all organisms giving a positive Voges-
Proskauer reaction ; in acetic acid fermentation ; in
acetone fermentation ; in the fermentation of sucrose by
Aerobacter aerogenes and in the fermentation of pentoses
by B. aceto-ethylicus. Acetaldehyde appears, therefore,
to be a very general, if not universal, intermediate in
ba(;terial fermentations.

246 BACTERIOLOGICAL CHEMISTRY

Pyruvic acid (which is, as we shall see, the normal
precursor of acetaldehyde) has also frequently been
isolated from bacterial fermentation solutions by appro-
priate " trapping " methods, for example, by the use of
[3-naphthylamine with which it forms the insoluble
compound, a-methyl- p-naphthocinchoninic acid : —
CH CH N

CH C C— NHg Q: =C— CH.

I II I ^ I

I II I... ^.. + i COO:!!

CH C CH-

^/\^: HO:-C=CH:-

CH CH I

C.CHg

CH

r*nnTT COOH

+ 2H2O+H2 + CO2

(3-naphthylamine) (2 pyruvic acid) (a-methyl- 3-naphtliociuchomnic acid)

In this way pyi'uvic acid has been identified as an inter-
mediate in the fermentation of lactic and fumaric acids
by Esch. coli, and of glucose, maltose and glycerol by
B. aceto-ethylicus. It has been shown, too, that pyruvic
acid can be utilised by bacteria.

Methylglyoxal is converted into lactic acid by the
action of the enzyme glj^oxalase, which occurs in the
liver, muscle tissues and many bacteria including Esch.
coli, Str. lactis, L. casei, and Acetohacter : —
(HO COOH

I I

CO +H2O > CHOH

1^ I

CH3 CH3

methylglyoxal) (lactic acid)

Methyl glyoxal has been detected following the action
of Esch. coli or of A. xylinurn on magnesium hexose
phosphate in presence of toluene, and in the fermentation
of glycerol by the propionic acid bacteria.

Lusk suggested that glucose is fermented to give
lactic, acetic and formic acids according to the scheme
shown at tlio top of the following page.

CARBON METABOLISM

247

H

CHO

-. I
H:COH

ii
0:CH

M

H;COH

■--: I

H;COH

CHO

I
C.OH

II
CHo

CHO

+ H2O

HCOOH
+
> CH3CHO
(acetaldehyde)

> H2 + CO2

oxidation

> CH3COOH

(acetic acid)

CH,:OH

C.OH +H2O

CH,

(glucose)

COOH

I
HCOH

I

CH3
(lactic acid)

(methylglyoxal)

This is an over simplified expression of the mechanism
of lactic acid fermentation by such organisms as Esch.
coli, but obviously does not account for the fermentation
by the homofermentative lactic acid bacteria which
give almost 100 per cent, lactic acid. Almost certainly
phosphorylation reactions are involved in lactic acid
fermentation by all bacteria (see p. 249).

In alcoholic fermentation by yeast, pyruvic acid is
decarboxylated to give acetaldehyde and carbon dioxide
(see Chapter XV), but this mechanism cannot apply
to the bacterial fermentations since it does not account
for the formic acid and hydrogen found in many such
fermentations. Possibly with some bacteria the pyruvic
acid is broken down in another way with formation of
formic and acetic acids : —

COOH H

CO + OH

HCOOH

CH3

(pyruvic acid)

CH3COOH

(acetic acid)

There is evidence that the reaction occurs through
phosphopyruvic acid and is reversible, for when pyruvic
acid is dissimilat^d by Esch. coli in presence of formic
acid containing " heavy " carbon, C^^, the residual
pyruvic acid contains C^^ in the carboxyl group and the
rate of transfer is accelerated l)y the addition of in-
organic phosphate. Heavy carbon is also found in the

17

248 BACTERIOLOGICAL CHEMISTRY

carboxyl group of pyruvic acid when Esch. coli acts on
pyruvic acid and " labelled " sodium bicarbonate,
NaHCi^Og.

Some bacteria, such as Lactobacillus delbrilckii,
Neisseria gonorrhcece and Streptococcus hcemolyticus , con-
tain pyruvic oxidase which catalyses the conversion of
pyruvic acid to carbon dioxide and acetic acid : —

CH3CO.COOH — > CO2 + CH3COOH.
Probably acetaldehyde is formed as an intermediate
step and its oxidation takes place with f la vine adenine
dinucleotide as co -enzyme or carrier ; the decarboxyla-
tion and oxidation are linked processes and do not occur
independently. Pyruvic oxidase is inhibited by cyanide
or fluoride, but yeast carboxylase is not.

The anaerobic dismutation of pyruvic acid to lactic
and acetic acids and carbon dioxide probably also involves
pyruvic oxidase and a carrier : —

CH3CO.COOH H2 CH3CHOH.COOH

+ +11 — > +

CH3CO.COOH 0 CH3COOH + CO2

Esch. coli yields a certain amount of ethyl alcohol
during fermentation as well as the main acid products.
It is thought that the alcohol arises as the result of a
dismutation (the term applied to the enzymatic equivalent
of the Cannizzaro reaction) of acetaldehyde : —

CH3CHO CH3COOH (acetic acid)

0
+ + II > +

Ha
CH3CHO CH3.CH2OH (ethyl alcohol)

Before the end of the nineteenth century it was shown
by Hoppe-Seyler that the gas production by bacteria was
almost certainly due to the breakdown of formic acid or
of formates. He showed that organisms producing gas
from glucose all fermented formates, whilst those which
were not gas producers did not ferment formates. Harden
showed that Esch. coli and Eberth. typhosa when grown
anaerobically on glucose broke down half the sugar to

CARBON METABOLISM 249

lactic acid and the other half to alcohol, acetic acid and
formic acid. The formic acid Avas further broken down
to hydrogen and carbon dioxide by Esch. coli (to give the
typical " acid and gas " fermentation), but it was not
attacked by Eherth. typhosa (" acid, no gas " fermenta-
tion).

The lactic acid fermentation seems to be more or
less independent of the other acid fermentations, in that
its formation may be stopped without affecting that of the
other products. For instance, Virtanen has shown that
washed suspensions of Esch. coli, which are deprived of
cozymase (see Chapter XV) in this way, no longer
produce lactic acid from glucose, though the other pro-
ducts are formed as usual. Virtanen considers that the
first stages of lactic acid fermentation are identical with
those of alcoholic fermentation by yeast and involve
phosphorylation of the glucose, for which process cozymase
is essential ; methylglyoxal, the j)recursor of lactic acid,
is then formed. That phosphorylation does play a part
in lactic acid fermentation is shown by the fact that
the addition of inorganic phosphates to such a fermenta-
tion brings about an acceleration of the process just as
it does in alcoholic fermentation. Bacterial cozymase
can replace that from yeast in alcoholic fermentation.
Virtanen claims that cozymase does not play a part in
the formation of the other products. It has been shown
that the propionic acid bacteria behave similarly ; washed
suspensions no longer produce propionic acid, but still
give rise to the formation of alcohol, acetic acid, succinic
acid and carbon dioxide.

By grinding Esch. coli Avith powdered glass cell free
extracts can be obtained which contain enzymes which
are capable of converting phosphoglyceric acid to phos-
phopyruvic acid, as occurs in yeast fermentation (see
p. 276). The equilibrium between 3-phosphoglyceric
acid and 2-ph(jsph()glyceric acid also occurs in the presence
of the bacterial enzymes as well as in the yeast and

250 iJACTEIllOLOGlCAL CHEMISTRY

muscle systems. The transfer of phosphate from phos-
phopyriivic acid via adenylic acid to glucose, similarly,
takes place under appropriate conditions. Enzymes
have been obtained from Staph, albus which can bring
about all the reactions of the Embden-Meyerhof scheme
(see p. 275). Lactic acid is only produced under anaerobic
conditions by this organism. There is, therefore, very
considerable evidence that the initial stages of bacterial
fermentation are very similar to, if not identical with,
those of yeast fermentation, and that the variations
producing the additional acids and other substances
arise in the later stages of fermentation.

Kluyver, also, regards methylglyoxal as the inter-
mediate in these acid fermentations : —

: /OH
CHgCChC^OH > CH3CHO + HCOUH . U, + CO,

\H

(methylglyoxal) (methylglyoxal (acetaldehyde)

hydrate)

The acetaldehyde gives acetic acid by direct oxidation
aerobically, or possibly by dismutation, anaerobically.
Kluyver considers that there are three types of fer-
mentation brought about by organisms of the Esch. coli
group : —

1. Succinic acid fermentation, which occurs in the
absence of cozjrmase, and therefore in the absence of
phosphorylation. It takes place by a splitting of the
glucose molecule into a four-carbon atom fragment and
a two -carbon atom fragment giving tartaric dialdehyde
and ethylene glycol respectively, which in turn give
rise to succinic acid and to acetaldehyde, as shown in
the scheme below.

Virtanen also suggested that succinic acid arose by
splitting of hexosos into f()ur-cai'l)on and two-carbon
fragments.

As a result of studies of the fixation of carbon dioxide

CARBON METABOLISM

251

b}' lieterotropliic ]:)a.cteria it has been shown that succinic
acid is formed by tiie condensation of ])yruvic acid with
carbon dioxide followed by reduction via malic and
fumaric acids (see p. 258). It is probable, therefore,
that succinic acid does not arise by the splitting of a
six-carbon molecule into four-carbon and two -carbon
fragments. Succinic acid may also be formed by the
reductive deamination (p. 228) of aspartic acid : —

> COOH.CH,.CH,.COOH + NH,

NHs

2. True alcoholic fermentation (to a small extent).

3. Lactic, acetic and formic acid production, with
or without gas production. The last two types of fer-
mentation need phosphorylation v as a preliminary step.
The three fermentations suggested by Kluyver are sum-
marised by the following scheme : —

CH80H.CH.(OHOH),.CHOK (glucose)

' — o — '

I NajHPO* (as in alcoholic
I fermentation)

2CHs.C0.C^0H >■ CH.CHOH.COOH

OH

(metliylglyoxal (lactic acid)

liydrat*)

4- 4

CHjGH.CHsOH-f-CHO.CHOH.CHOH.CHO
(ethylene glyopl) (tartaric dialdehyde)

CHj.CHO + 11,0

C00H.CH2.CH,.('00H

(sjiccMiic aciil)

CHsCOOH + CHaCHjOH
(acetic (alcohol)

acid)

Non-phosphoiy!atod.

4
CH3CH0

4

H.COOII

I

CH,COOH CH.CH.OH

Pho8phorylat«d.

Hj + COj

The same types of fermentation occur with Eherth.
typhosa, except that the breakdown of formic acid
does not occur. It will be noticed that in this type of
fermentation the ratio of carbon dioxide to hydrogen is
1:1, and it will be remembered that members of the
coli-typhoid group of bacteria have a negative Voges

252

BACTERIOLOGICAL CHEMISTRY

Proskauer reaction, it lias been demonstrated that the
Voges-Proskaiier reaction depends on tlie production of
acetyl-methyl-carbinol, or acetoin, CH3CO.CHOH.CH3,
which, in the presence of potassium hydroxide becomes
oxidised to diacetyl, CH3CO.CO.CH3, which reacts with
some substance in the peptone containing a guanidine
NH.3

residue, C==NH , to give the red-coloured compound.

^NHR
The formation of acetoin by Voges-Proskauer positive
organisms like Aerohacter aerogenes follows the scheme
below, suggested by Kluyver : —

CH20H.CH.(CIiU^i)3CH0H (gluccse)

0

Na^HPO^

(methylglyoxal
hydrate)

1 molecule

i

HCOOH + CH3.CHO

i

H2 + CO2

/OH
-OH
H

CH3.CHOH.COOH

(lactic acid)

1 molecule

2H + CH3.CO.COOH (pyruvic
I acid)

€H3.CH0 + CO2

(acetoin) CH3.CHOH.CO.CH3

\ reduction
CH3.CHOH.CHOH.CH3

(2 : 3-butyleno-glycol)

r i

I CHo.CH.

OH

CARBON METABOLISM 253

Methylglyuxal hydrate, produced its in alcoholic fermenta-
tion (see Chapter XV), is partially converted into lactic
acid (type 3), partially broken down to give pjTuvic acid
and carbon dioxide and partially broken down to give
acetaldehyde and formic acid. The formic acid gives
hydrogen and carbon dioxide, as it does in the case of
Esch. coli. The pyruvic acid is decarboxylated (see
Chapter XV) to give acetaldeh^^de and carbon dioxide.
The molecules of acetaldehyde from this source and
directly from the methylglyoxal condense, under the
influence of the enzyme carboligase, to form acetoin.
The hydrogen evolved when methjdglyoxal hydrate
yields pyruvic acid is partly taken up in reducing some
acetaldehyde to alcohol and partly in reducing some of
the acetoin to 2 : 3-butylene glycol, which is almost
invariably found among the products of the Voges-
Proskauer positive organisms. The 2 : 1 ration of carbon
dioxide to hydrogen which is associated with the Voges-
Proskauer reaction follows from the mechanism suggested,
carbon dioxide arising from two sources and hydrogen
from one. The formation of acetoin is favoured by
conditions, such as aeration or the presence of other
hydrogen acceptors, which restrict the reduction of
acetaldehyde to alcohol.

The formation of propionic acid from glycerol by
the propionic acid bacteria proceeds without any pro-
duction of gas. Propionaldehyde and pyruvic acid have
been detected in fermenting cultures. Wood and Work-
man suggest that the steps in the fermentation are : —

- H3PO4

CH2OH.CHOH.CH2OH — > CH2(O.P03H2).CHOH.CHoOH

(glycerophosphate)

-H3PO4

— 2H

CH2(O.P03H2)CHOH.CHO

(methyl glyoxal) (phosphoglyceraldehyde)

I + 2H

\ — HoO + H2O
CH3.CHOH.CHO -"-> CH3CH2.CHO > CHj.CHo.COOH

(lactic aldehyde) + 2H (propionaldehyde) — 2H (propionic acidj

254 BACTERIOLOGICAL CHEMISTRY

All alternative route is that glycerophosphate is oxidised
to phosphoglyceric acid which gives pyruvic acid which
is reduced to propionic acid.

When glucose is dissimilated by propionic acid
bacteria, phosphoglyceric acid is produced and can be
isolated if toluene and sodium fluoride are present (the
latter inhibiting further breakdown of the phospho-
glyceric acid). Pjrruvic acid can be fixed by using
sodium sulphite, but acetaldehyde cannot be detected.
This suggests that the propionic acid bacteria have no
carboxylase and do not split pyruvic acid to give acetalde-
hyde and carbon dioxide, as do the yeasts. Some strains
also produce lactic acid, but others do not. Lactic acid,
however, is fermented by all strains with formation of
propionic acid. Succinic and acetic acids are also formed,
but undergo further breakdown, the ratio of propionic
acid to acetic acid increasing during the fermentation.
If the culture is buffered by sodium bicarbonate the
ratio of propionic acid to acetic acid remains approxi-
mately constant. Wood, Stone and Werkman suggested
that propionic acid was formed by the following scheme : —

2H.F0,

CeHi.Og — > Hexosephosphate ^ CHoO.PO.Ha.CHOH.CHO

(glucose) _2H20 (Triosc phosphate)

— HsPO^CHaO.POaHo.CHOH.COOH < Noit-reducing

(phosphoglyceric acid) substance

CH3CO.COOH

~2H

+ 2H

CH3CHOH.COOH

(lactic acid)

,or

CH3CO.CHO + H3PO,

(methylglyoxal)

+ 2H
— H3O

CH3CH2COOH

(propionic acid)

CARBON METABOLISM 255

The triose phosphate, derived via phosphorylation of
glucose, gives a non-reducing suljstance of unknown
structure which may give phosphoglyceric acid by
oxidation, or which may give methylglyoxal by loss
of the phosphate group. Phosphoglyceric acid is conveited
to pyruvic acid which is reduced via lactic acid to pro-
pionic acid. Methylglyoxal may be converted to lactic
acid directly (presumably by the action of glyoxalase)
or it may be oxidised to pyruvic acid ; propionic acid is
then formed as in the alternative scheme. These workers
postulated that part of the pyruvic acid might undergo
another series of reactions to give succinic and propionic
acids, pyruvic acid giving the hydrate which would split
to give acetic acid and carbon dioxide ; two molecules
of acetic acid then condeixse to give succinic acid which
in its turn is decarboxylated to form propionic acid and
carbon dioxide : —

/OH — 2H
> CHaCZ—COOH > CH3COOH + CO2

^OH

-2H

CHX'Ho.COOH -r CO,

In view of the recent work on the fixation of carbon
dioxide it seems more probable that the succinic acid
arises by way of condensation of carbon dioxide with
pyruvic acid (see p. 258).

Carbon Dioxide Fixation. — Recently work on the
fixation of carbon dioxide by bacteria has thrown a new
light on the mechanism of acid production by bacteria.
It has been known for a long time that autotrophic
bacteria utilise carbon dioxide, either by photosynthetic
or by chemosynthetic processes, as their sole source of
carbon. It has also been known for many years that
carbon dioxide plays an important role in the metabolism
of some heterotrophic bacteria : thus Brucella aboi'tus,
when first isolated from cattle, will not grow unless the

256 bacterioloCtICal chemistry

atniospheru contains a))out 10 per cent, of carbon
dioxide ; many bacteria, such as Esch, coli, fail to grow
if steps are taken to remove the carbon dioxide from the
medium, by vigorous aeration for example. Until a few
years ago, however, it was not realised that carbon
dioxide was actually assimilated by heterotrophic bacteria,
the reason being that under normal conditions such
micro-organisms form carbon dioxide from carbohydrates
in larger amounts than they use so that the net production
of carbon dioxide masks its assimilation. The first clue
was given by the fact that when the propionic acid bac-
teria ferment glycerol they do not produce carbon dioxide
and it was found that the products of fermentation
contained more carbon than could be accounted for by
that in the medium initially. Since then, by the use of
isotopic " heavy " carbon or radioactive carbon in the
carbon compounds of the medium, it has been shown
that assimilation of carbon dioxide is a general
phenomenon in heterotrophic as well as autotrophic
bacteria. The difference appears to be that autotrophic
bacteria can make use of it in conjunction with inorganic
substances as sources of energy whilst the heterotrophic
bacteria require compounds already containing at least
one carbon atom in organic linkage.

Carbon dioxide has been shown to be fixed by barley
roots, liver, yeast, Esch. coli, the propionic acid bacteria.
Micrococcus lysodeikticus, Aerobacter indologenes, Proteus
vulgaris, Str. paracitrovorus, Staph, candidus, CI. welchii,
CI. acetobutylicum and CI. aceticum.

The mechanism proposed by Van Niel to account for
the photosynthetic reaction : —

CO2 + 2H2A > (CH2O) + H2O + 2A

probably also holds for chemosynthetic reactions. Ruben
has suggested that the fixation of carbon dioxide in the
dark (that is chemosynthetically) takes place with the

CARBON METABOLISM 257

iiiterveiition of a pliosphate-donur complex with a high
energy content, in such a way that an aliphatic compound,
probably an aldehyde, is carboxylated and reduced with
the ultimate formation of carbohydrates. The general
reaction is : —

Phosphate-donor -f RH ?==^ free donor + RH-phosphate

CO 2
RH-phosphate ^ R.COOH + inorganic phosphate

carboxylase

When RH is an aldehyde and if the phosphate donor is
considered to be adenosine triphosphate (which is very
probable, see p. 275) the reactions are : —

0 0 0

i! !! II

C10N5H12O3— 0— P— 0— P— 0— P— OH + R.CHO + HoO

III
OH OH OH

(Adenosine triphosphate) ^1

Enzyme

0 0 0 0

II il ^ II

C10X5H12O3— 0— P— 0— P— OH + R.C 0— P— OH + 2H

1 I I
OH OH OH

(Adenosine diphosphate)

0 0 0 0

^ il Enzyme || 1|

R.r;_0— P— OH + CO2 + 2H ^=^ R_(j_c— OH + H3PO4

ci„

0 o

II II Enzyme

R— C— C— OH + 2H ^=^ R.CHOH.COOH

Enzyme
R.CHOH.COOH -f phosphate -donor ^=^ R.CHOH.CO.PO3H2 + donor

Enzyme
R.CHOH.CO.PO3H2 + 2H F-"=^ R.CHOH.CHO + H3PO4

The aldehyde R.CHOH.CHO, which can be regarded
as a carbohydrate, could give rise to polysaccharides

258 BACTERIOLOGICAL CHEMISTRY

tliroiigJi fiirtlier cyck\^ of pliuy])horylation, carl)Ox\iation
and reduction, or by condensation with similar molecules,
in a manner analogous to the enzymatic synthesis of
starch or glycogen from glucose- 1 -phosphate.

In heterotrophic systems the acceptor of carbon
dioxide is probably a phosphorylated Cg compound, as
has been shown for the propionic acid bacteria. These
organisms, which occur in dairy products (particularly in
Emmenthaler cheese), silage, soil and similar situations,
ferment carbohydrates with production, mainly, of
propionic acid, acetic acid and carbon dioxide, with
smaller amounts of lactic and succinic acids and, some-
times, acetoin. When grown on a glycerol medium
containing " labelled," that is C^^, sodium bicarbonate,
Propionibacterium pentosacemn utilises carbon dioxide
which is found almost entirely in the carboxyl groups of
the succinic acid formed. The succinic acid formed and
the carbon dioxide taken up are very nearly in equi-
molecular proportion ; in the absence of carbonate
practically no succinic acid is formed. The small amounts
of " labelled " carbon found in the other products of
fermentation, propionic and acetic acids and propyl
alcohol, are probably derived by side reactions involving
succinic acid. Wood and Werkman suggested that the
carbon dioxide condensed with pyruvic acid to give
oxalacetic acid : —

CH3CO.COOH + CO2 > COOH.CHo.CO.COOH

Succinic acid then arises via mahc and fumaric acids : —

COOH COOH COOH COOH

CO + 2H CHOH — H2O CH + 2H CHo

I > I > II ^ I

CH2 CH, CH CHo

I I " I I

COOH COOH COOH COOH

(oxalacetic acid) (malic acid) (fumaric acid) (succinic acid)

CARBON METABOLISAL- 259

Evidence that this is so is provided by the fact that
fiimaric and malic acids have been detected in the meta-
bolism solutions, and that oxalacetic, malic and fumaric
acids added to the system become converted to succinic
acid.

An enzyme preparation has been obtained from M.
lysodeikticus which, in presence of magnesium or man-
ganese, and possibly phosphate, catalyses the carboxyla-
tion of p3^ruvic acid to oxalacetic acid. The enzyme only
decarboxylates pyruvic acid if co -carboxylase is also
present. When pyruvic, lactic or oxalosuccinic acids are
decarboxylated by the enzyme in presence of " labelled "
carbon dioxide there is no evidence of exchange of
carbon dioxide. Similarly after glucose or pyruvate
have been treated with Sir. lactis (which can convert
pyruvic acid into lactic and acetic acids and carbon
dioxide) in presence of " labelled " carbon dioxide, the
isolated pyruvic, lactic and acetic acids do not contain
" labelled " carbon.

The mechanism of formation of propionic acid from
glycerol is not yet clear but it may arise via lactic acid : —

CH2OH CH3 CH2 C'Hg

I I -H2O II + 2H I
CHOH > CHOH > CH > CH.

I I I I

CH2OH COOH COOH COOH

or via pyruvic acid : —

CH2OH COOH COOH

I I + 4H I

CHOH > CO > CH2 -f H2O

I I I

CH2OH CH3 CH3

When carbon dioxide is assimilated by other bacteria
it has been found to be distributed as follows : — Esch.
roll : in formic, acetic, lactic and succinic acids ; Aero-
hacter indologenes : in acetic, lactic and succinic acids ;
Proteus vulgaris, Str. paracitrovortis and Staph, aindidus :

2{3() BACTERIOLOGICAL CHEMISTRY

in lactic and succinic acids ; CI. tvelchii : in acetic and
lactic acids ; CI. acetobiitylicum : in lactic acid. When-
ever succinic acid is formed it contains fixed carbon
dioxide, but the other acids do not always contain
assimilated carbon dioxide. The amount and rate of
production of succinic acid formed by Esch. coli from
glucose, galactose or pyruvic acid depends on the quantity
of carbon dioxide available. When carbon dioxide is
removed by aeration the yield of succinic acid is low,
but when the reaction is carried out in presence of carbon
dioxide the yield is high. It has been shown, again by
the use of " labelled " carbon, that succinic acid can be
formed by condensation of two molecules of acetic
acid : —

CH3COOH ('H2.COOH

+ > I + 2H

CH3COOH CH2.COOH

and that the reaction is reversible. Thus acetic acid
containing fixed carbon dioxide could arise by the
breakdown of succinic acid.

Formic acid can be derived by direct reduction of
carbon dioxide in presence of the enzyme hydrogenlyase :

OH OH

I +H2 I

CO2 + H2O > HO— G = 0 > HO— C— OH > HO— C = 0 + HgO

I I

(carbonic acid) H H

(formaldehyde hydrate) (formic acid)

Formaldehyde has been isolated from cultures of pro-
pionic acid bacteria, by fixation with dimedon, con-
firming that carbon dioxide is reduced.

The precursor of lactic acid containing fixed carbon
has been suggested to be a four-carbon dicarboxylic
acid (other than succinic acid) formed by condensation
of three-carbon and one-carbon molecules ; it is
decarboxylated to give lactic acid.

CAKBO^^ METABOLISM 261

For further reading : —

M. M. Barritt, " The Origin of Ac( tvlmethylcarbinol in Bacterial Fermenta-
tion." J. Path, and Bad., 44 (1937)', 679.

E. S. G. Barron, "Mechanisms of Carbohydrate Metabolism. An Essay

on Comparative Biochemistry." Advances in Enzymology, 3 (1943),
149.
A. J. Kluyver, " The Chemical Activities of Micro-organisms." University
of London Press. London, 1931.

F. Lipmann, " Metabolic Generation and Utilisation of Phosphate Bond

Energy." Advances in Enzymology, 1 (1941), 99.
M. Stephenson, " Bacterial ^Metabolism," Chapter I\'. 2nd Edition.

Longmans, Green & Co. London, 1939.
C. H. Werkman and H. G. Wood, " Heterotrophic Assimilation of Carbon

Dioxide." Advances in Enzymology, 2 (1942), 135.

CHAPTER XV
ALCOHOLIC FERMENTATION

ALCOHOLIC fermentation is the most widely studied
of the fermentations, partly because of its wide-
spread industrial importance and partly because of
the ease with which it can be carried out and the condi-
tions modified. About 95 per cent, of the sugar fermented
is normally recovered as e qui -molecular proportions of
alcohol and carbon dioxide. Gay-Lussac long ago ex-
pressed this in the form of the classical equation : —

CeHiaOe > 2CO2 + 2C2H5OH.

The fermentation, however, is not so simple as this
equation suggests ; in reality the end products arise from
a chain of reactions involving a number of intermediate
compounds. Besides the alcohol and carbon dioxide there
is a constant production of some 3 to 4 per cent, of glycerol,
small amounts of fusel oil (derived from the yeast proteins,
see Chapter XIII), and varying small proportions of
hexose mono- and di -phosphates.

The first real advance in our laiowledge of alcoholic
fermentation was the almost simultaneous proof in 1837
by Cagniard-Latour, Schwann and Klitzing that the
fermentation was associated with the living organism,
yeast. If sugar solutions were sterilised by boiling and
only heated air admitted to them no fermentation
occurred.

The next big step forward in the elucidation of the
mechanism of alcoholic fermentation occurred when
Buchner in 1897 succeeded in demonstrating its enzymatic
nature by preparing an active cell -free extract by grinding

262

ALCOHOLIC FERMENTATION 263

yeast with sand and submitting the mixture to con-
siderable pressure. The expressed juice was capable,
in virtue of the enzymes which it contained, of causing
the fermentation of sugar with the production of the same
products as the living yeast. Since then a number of
other yeast preparations have been obtained by various
workers, of which zymin (yeast dried with acetone) and
" maceration juice," obtained by the autolysis of yeast
in water, are the most important.

The Role of Phosphates in Fermentation. — In 1905
Harden and Young established the importance of in-
organic phosphates and phosphoric esters in fermentation
processes. During an investigation of anti -enzymes
Harden had added a serum, prepared by injecting yeast
juice into rabbits, to a fermenting sugar solution and
found that an increase, and not the expected decrease,
of fermentation occurred. Normal serum was found to
have the same property. The addition of boiled yeast
juice to yeast juice and sugar solution had a similar
effect. The cause of the increase in the case of boiled
yeast juice was traced to two factors : (a) the presence of
inorganic phosphates, and (b) to a heat stable co -enzyme
or CO -zymase. In the case of serum the effect was shown
to be due to the phosphates present.

It was found that the addition of inorganic phosphate
to fermentations by yeast preparations caused a con-
siderable increase in both the rate of fermentation and in
the absolute amount of sugar fermented. This increase
did not occur on the addition of phosphate to fermenta-
tions by living yeasts. The rate of fermentation was
measured as the volume of carbon dioxide evolved every
five minutes, and it was found that the extra volume
of carbon dioxide evolved was proportional (molecule for
molecule) to the amount of inorganic phosphate converted
into esters. The normal rate of fermentation by yeast
juice is about 3 ml. of carbon dioxide per five minutes,
represented l)y the straight line (1) in Fig. 7. The rate

18

264

BACTERIOLOGICAL CHEMISTRY

of fermentation is constant, and is controlled by the
supply of inorganic phosphate derived from the hexose
phosphate esters by the action of the enzyme phosphatase.
The inorganic phosphate so formed esterifies more sugar,
which is fermented with liberation of the phosphate again.
There is no accumulation of phosphate esters nor of
inorganic phosphate, and the reaction proceeds at a con-
stant rate until all the sugar has been fermented, at
which stage inorganic phosphate is finally set free and the
sugar of the esters fermented.

If inorganic phosphate is added to such a fermenting
mixture the velocity of fermentation increases temporarily,

20

15 '

10

006M-/»c ^

/^^^^^%r^

(3;

— - — -JIL^^Oj

'/

\0.O6M-K2HPO4

7

^^- — Normal

(1)

. 1 .

,

1 i

10

20 30

Minutes

40

— >-

Fig. 7 {Ajtev Harden)

due to the increased formation of the fermentable ester
which accumulates until all the added inorganic phosphate
has been converted to ester, after which the rate falls
back to the normal value, depending on the rate of
hydrolysis of the ester. Curve (2) in Fig. 7 shows this
effect.

ALCOHOLIC FERMENTATION 265

If the enzyme phosphatase, which hydrolyses the
hexose phosphate esters, is added, an increase in the rate
of fermentation will occur, due to the increased rate at
which the esters are hydrolysed and inorganic phosphate
set free to esterify more sugar. The same effect can be
brought about by the addition of arsenate, which stimu-
lates phosphatase activity (curve (3), Fig. 7). If the
phosphatase is sufficiently active to hydrolyse the esters
as fast as they are formed, a rapid fermentation un-
accompanied by any accumulation of ester will ensue.
The rate of fermentation is about the same as the
maximum obtained on addition of inorganic phosphate
(curve (2) ), but remaiixs constantly high as long as sugar
is available, since in effect the addition of phosphatase
(or its stimulation) is the same as a continuous addition
of inorganic phosphate.

To account for these facts Harden and Young suggested
that sugar was fermented in two stages according to the
equations : —

(1) iK'eHiaOe +2Na2HP04 >CeHio04(Xa2P04)2 + 2H2O +2CO2 +2C2H5OH.

(2) C6Hio04(Na2P04)2 + 2H20 > (:,U,,(\+2:^a,KF0,.

According to this scheme the first stage consists in the
esterification of one molecule of glucose to hexose di-
phosphate at the same time as a second molecule is
broken down to carbon dioxide and alcohol. The second
stage is the hydrolysis of the hexose diphosphate to sugar
and inorganic phosphate, which then go through the cycle
again, with fermentation of half the sugar at each cycle,
until it has all been converted into the end products.
The equation (1) represents the state of affairs occurring
when inorganic phosphate is added to the system (curve
(2)); equation (2) corresponds to the normal rate of
fermentation (curve (1) ). The addition of arsenate
causes tlie liydi'olysis of the esters acconlinu' to e(|uation

266 BACTERIOLOGICAL CHEMISTRY

(2) to proceed at such a rate that inorganic phosphate is
supplied sufficiently rapidly to enable the reactions of
equation (1) to proceed at their maximum velocity.

No such acceleration of fermentation occurs on the
addition of inorganic phosphate or arsenate to living
yeast fermentations, which proceed about twenty to
forty times as fast as those due to yeast preparations.
The addition of inorganic phosphate to the latter increases
the velocity some ten to twenty times. Presumably in
living yeast cells the ratio of the activities of the enzymes
is at an optimum, and the balance is upset during the
extraction of the yeast juice or other treatment. Harden
showed that the velocity of fermentation decreased and
the magnitude of the response to added phosphate
increased, with greater disintegration of the cell structure.

From calculations based on the phosphorus content of
the yeast cell it appears that the phosphate -ester cycle
must be completed once every five or six minutes in
order to maintain the normal rate of fermentation.
For an average yeast preparation the cycle takes
approximately two hours.

Harden and Young showed that the ester which
accumulated after addition of inorganic phosphate was a
hexose diphosphate, later shown to be 1 : 6-fructose
diphosphate, even when the sugar being fermented is
glucose or mannose. Later, Robison isolated a hexose
monophosphate from such solutions, which was ultimately
proved to be a mixture of about 80 per cent, of 6-glucose
monophosphate (loiown as Robison's ester) and 20 per
cent, of 6-fructose monophosphate. More recently
Embden and Myerhof and their co-workers have isolated
a number of triose monophosphates, including a-glycero-
phosphoric acid, 2- and 3-phosphoglyceric acids and
2-phosphopyruvic acid, from alcoholic fermentations.
Tiieir importance will be seen at a later stage.

ALCOHOLIC FERMENTATION 267

Tlie ratio of mono- and di-hexose pliospJiate.s found in
fermentation mixtures is very variable, and depends on
the strain of yeast and on the concentration of the yeast
or yeast preparation used. Living yeast is found not to
ferment added hexose diphosphate, and also leads to the
formation of only a small amount of it during fermenta-
tion. Consequently Harden and Young's equations are
almost certainly not a true representation of the mechan-
ism. It has been suggested by Kluyver and Struyk and
by Meyerhof that the real intermediate is the monophos-
phate, and that the diphosphate is a by-product formed
either by further phosphorylation of the monophosphate
or by condensation of two molecules of a triose mono-
phosphate, which also plays a part in the fermentation
scheme. Hexose diphosphate, however, plays an essential
part in the fermentation reactions.

Co-enzyme. — It was mentioned on p. 263 that the
accelerating effect on fermentation of the addition of
boiled yeast juice was traced to two factors, inorganic
phosphates and the co-enzyme. It was found that if
yeast juice was dialysed or submitted to filtration through
porcelain candles it could be divided into two fractions,
neither of which alone could induce fermentation of
sugar, but which when remixed were once more active.
The residue after dialysis or filtration was heat labile and
destroyed by boiling ; as was to be expected it contained
the undialysable enzjnnes of the yeast juice. The dialysate
or filtrate, on the other hand, was heat stable, and was
termed co -enzyme or co -zymase. Besides the above-
mentioned method for its separation, co-enzyme can also
be obtained by washing acetone dried yeast, zymin, with
water.

Subsequent investigations have shown that the
co-enzjmae, now known as co -enzyme I, consists of two
parts, a magnesium salt and diphosphopyridine nucleo-
tide : —

268

BACTERIOLOGICAL CHEMISTRY

CH

CH

C.CONH2

II I

CH (m

\ //

:C.NH2

I

CH C— No

^CH

/.

N-

-C— N-

I
CH-

I
HOC.H

■ I
HOC.H

I
HC -

I
(JH.,

I OH

I
0 — P — 0

II

o

0-

CH—
HOC.H

HOC.H

I
HC

P _ 0 _ CH,

in which nicotinamide and adenine are joined by two
molecules of ribose-5-phosphate. Its function is to act
as a hydrogen carrier between phosphoglyceraldehyde
and acetaldehyde, being reduced to dihydro -co -enzyme I
by the former and re-oxidised by the specific flavo-
protein enzyme with acetaldehyde as the hydrogen
acceptor (see pp. 203 and 276). Co-enzyme I is very
widely distributed and participates in the respiration of
many bacteria, particularly those producing lactic acid
by fermentation. It is to be found in nearly all animal
tissues, in red blood corpuscles, plants, fungi and bacteria.
It is a growth factor for some members of the genus
Hcemo'philus.

The closely related co -enzyme II, triphosphopyridine
nucleotide containing an extra phosphate group, plays
the same role of hydrogen carrier in animal tissues in
the conversion of glucose to lactic acid.

In view of the discovery of the necessity of magnesium
salts as well as the co -enzyme in yeast fermentation the
following terminology for the enzyme systems concerned
has been proposed : Zymase to indicate the pure enzyme

ALCOHOLIC FERMENTATION 260

free fi'om all activators ; Holozymase (or paiiz^inase) to
designate the complete system of zymase plus the acti-
vators ; and Apozymase, which is the holozymase free
from CO -enzyme but still containing the magnesium salts.
That is,

Holozymase =Zymase-|-Mg+ Co-enzyme
Apozymase = Zymase +Mg.

Zymase is, itself, a complex mixture of several enzymes
(see p. 275 et seq.).

Mechanism of Alcoholic Fermentation. — In elaborating
a mechanism to account for alcoholic fermentation the
following facts have to be considered : —

(1) All the compounds which are produced in a
normal alcoholic fermentation can be derived from
molecules containing three carbon atoms, which may be
obtained by a preliminary split of the glucose molecule
into two such fragments following phosphorylation as an
essential step.

(2) Acetaldehyde has been demonstrated, by fixation
with sulphite or dimedon, as an intermediate.

(3) Pyruvic acid is also regularly formed, as shown
by its fixation by the use of p-naphthylamine.

(4) Neuberg isolated from yeast the enzyme carbo-
xylase, which splits pyruvic acid (and «-keto -acids in
general) into carbon dioxide and acetaldehyde.

(5) Glycerol is always formed to the extent of 3 to
4 per cent, in normal fermentations, and under certain
conditions large quantities of it are formed.

(6) Usually none of the intermediates is left, and
the end products are almost entirely carbon dioxide and
alcohol in equivalent quantities.

A considerable number of theories have been pro-
posed to account for all these facts, of which the three
most important are those due to Kluyver, Neuberg,
and Mj-erhof ; the last of these is the one now generally
accepted.

270

BACTERIOLOGICAL CHEMISTRY

Kliiyver considers that the first step in the fermenta-
tion of glucose is the formation from y-glucose of a
reactive hexose monophosphate which then splits into
two C3 fragments, one of which is still phosphorylated : —

—CHOH

I I
O (CH0H)3

Uh
CH20H

(glucose)

—CHOH

I i
O (CH0H)2

«!_CH

I

CHOH

I
CHjOH

(y-glucose)

CHO
Na2HP04 (CH0H)3

CH.O.POjNa,

(active
monophosphate)

-CHOH

(CH0H)2

I
CHOH

I
— CH

I
CHaO.POaNaj
(Roblson'8 ester)

CHO

CHO
CH.O.POaNaa

CHOH

I
CHjOH CH2OH

(glycetaldehydc) (glyceraldehyde
phosphate)

The triose fragments he considers to be glyceraldehyde
and glyceraldehyde phosphate. Two molecules of the
latter may condense to give hexose diphosphate as a side
product : —

CHO

I

CH.O.PO,Na
CH2OH

CHO
2 + CH.O.POgNaa —
CHoOH

CH.O.POgNa,

CHOH
CH.O.POgNaj

Normally the glyceraldehyde phosphate is hydrolysed
to glyceraldehyde and phosphate. The glyceraldehyde
undergoes a series of hydrogen transfer reactions to give
methylglyoxal hydrate : —

ALCOHOLIC FERMENTATION 271

H t[ H H

CHO C^ — OH C^-— OH C OH C— OH

1 +H2O ', \0H — H,.0 I \ I \0 I \0H

CHOH . CHOH ^>CHOH\o > HOC— I — - CO

I I III I

CH.OH CH2OH CH2 1 CH3 CH3

(glyceral- (plvccraldc- (mcthylglyoxal

dnhyde) liycle hydratv^ liydrate)

The methylglyoxal hydrate loses hydrogen to acetaldehyde
(or some other hydrogen acceptor at the beginning of the
fermentation, before acetaldehyde is formed) to form
pyruvic acid, the acetaldehyde being at the same time
reduced to alcohol.

OH COOH

OH CHO

+ 1
CH3

1 CHoOH

> CO -r 1 "

1 CH3
CH3

(pyruvic
acid)

(ethyl
alcohul)

CO

CH3

The p^Tuvic acid is decarboxylated to give carbon dioxide
and acetaldehyde, which acts as hydrogen acceptor for
the production of further supplies of pyruvic acid from
methylglyoxal hydrate : —

COOH CO2

I +

CO > CHO

I I

CH3 CH3

The only enzymes involved in this scheme are phosphatase,
hydrogen transportase and carboxylase.

The evidence for this scheme is based, in the main,
on analysis of the proportions of the various products
formed and misses many of the steps which are now
known to take place ; it does not take account of the

272 BACTERIOLOGICAL CHEMISTRY

action of co-enzyme, for instance. Moreover, methyl-
glyoxal has only been isolated following the action of an
extract of dried bottom yeast on magnesium hexose
diphosphate, and is not itself fermented.

Neuberg showed that acetaldehyde was an inter-
mediate in alcoholic fermentation as a result of " fixa-
tion " experiments using calcium sulphite or dimedon

CH,

(dimethyl-cyclohexane-dione), | [ , which re-

C(CH3),

move the acetaldehyde from further participation in the
fermentation either as the insoluble acetaldehyde bisul-
phite complex or according to the reaction : —

CH3 CH3

I I
CH, + CHO+ CH CH CH C

/\ ^\ /\ ^\

CO CO HO.C CO > CO CO HO.C CO

II II II II

CHo CH, CH2CH2 CH2CH2 CH2CH2

C(CH3)2 C(CH3)2 C{CH3)2 C(CH3)2

in the case of dimedon fixation, with a consequent loss of
alcohol production. He also showed that yeast contained
the enzyme carboxylase, which could break down «-keto-
acids to the aldehyde and carbon dioxide, and, in
particular, pyruvic acid, which was also shown to be an
intermediate by " fixation " experiments. As a result of
these findings he put forward the following scheme for
the mechanism of alcoholic fermentation ; glucose is
converted into two molecules of methylglyoxal by steps
involving phosphorylation : —

ALCOHOLIC FERMENTATION 273

CgHiaOg ^ 2CH2==('(OH).('H() — ^ L'CHgCO.CHO.

phosphorylation

(eiiolic form) (inethylglyoxal)

These tAvo molecules of inethylglyoxal undergo dismuta-
tion (under the influence of the enzyme mutase) to give
one molecule of glycerol and one molecule of pyruvic
acid : —

CH2=C(0H).CH0 H2 CH2OH.CHOH.CH2OH (glycerol)

+ +H2O+ II >

CH3.CO.CHO 0 CH3.CO.COOH (pyruvic acid).

The pyruvic acid is then decarboxylated by the enzyme
carboxylase to give acetaldehyde and carbon dioxide : —

CH3.CO.COOH — > CH3CH0 + CO2.

In the subsequent dismutation of methylglyoxal, one
molecule of the latter is replaced by acetaldehyde, which
becomes reduced to alcohol, whilst the molecule of
methylglyoxal is oxidised to pyruvic acid : —

CH3.CO.CHO o CH3.CO.COOH

+ +11 — > +

CH3CHO H2 CH3CH2OH

This pyruvic acid in its turn is decarboxylated to give
more acetaldehyde, and so the cycle goes on until all the
sugar is fermented.

This, Neuberg's First Form of Fermentation, accounts
for the small amount of glycerol, always found in a normal
fermentation, as being formed by the reduction of methyl-
glyoxal as hydrogen acceptor before the usual acceptor,
acetaldehyde, is formed. If the acetaldehyde is trapped
and prevented from acting as hydrogen acceptor no
alcohol will be formed, and methylglyoxal will continue
to act as hydrogen acceptor and yield a molecule of
glycerol for every molecule of carbon dioxide formed
and every molecule of aldehyde fixed. This is Neuberg's
Second Form of Fermentation. Neuberg also established
a Third Form of Fermentation, which takes place if the

274 BACTERIOLOGICAL CHEMISTRY

fermenting solution is made alkaline ])y addition of
sodium bicarbonate. These conditions favour a
Cannizzaro reaction of the acetaldehyde with formation
of acetic acid and alcohol : —

CH3CHO O CH3COOH

+ + II — > +
CH3CH0 H2 CH3CH2OH

Since some of the aldehyde is removed in this way it can
no longer act as hydrogen acceptor in the dismutation of
methylglyoxal, and an equivalent amount of the latter
is reduced to glycerol. Since two molecules of acetalde-
hyde, by the Cannizzaro reaction, give one molecule each
of alcohol and acetic acid, and one molecule of glycerol
is formed for every molecule of acetaldehyde diverted
from acting as hydrogen acceptor, it is obvious that for
each molecule of alcohol produced under these conditions
there will be two molecules of glycerol formed : —

^CsHiaOe > 2C02 + 2CH3CHO+2CH20H.CHOH.CH,OH.

2CH3CHO + H2O > CH3COOH+CH3CH2OH.

The chief deficiencies of Neuberg's theories are that
no account is given of the phosphorylation processes,
nor of the influence of co -enzyme, and that the chief
intermediate, methylglyoxal, has not been detected in
normal fermentations, and is not fermented when added.
It has the advantage that it accounts for the constant
small amount of glycerol in normal fermentation and
explains the increased yields in special circumstances.

Meyerhof's theory has points of similarity with both
Kluyver's and Neuberg's schemes but is based on more
complete experimental evidence. Some of the evidence
is derived from the work of Embden, Lohmami and
others on the course of muscle glycolysis, 'many of the
steps in which have been shown also to occur in yeast
fermentation. The main evidence in addition to that
mentioned on p. 269 is that hexose diphosphate, dihydroxy-
acetone phosphate and 3-glyceraldehyde phosphate have

ALCOHOLIC FERMENTATION 275

been isolated from fermentations of glucose by yeast
extract in presence of mono -iodo acetate which inhibits
their further breakdown ; equilibrium is established
between them when the appropriate enzymes are present ;
they are rapidly fermented by yeast extracts. The
breakdown of phosphoglyceric acid to pyruvic acid and
phosphoric acid is inhibited by sodium fluoride so that
phosphoglyceric acid accumulates when fermentations
are conducted in its presence. Both 2- and 3 -phospho-
glyceric acids have been isolated from muscle extracts
and so also has phosphopyruvic acid. Adenylic acid and
adenosine triphosphate have been isolated from yeast
juice and shown to react with fructose- l:6-diphosphate.

According to Meyerhof's scheme glucose is phos-
phorylated by a transfer of phosphate groups from
adenosine triphosphate : —

I I
CH C— N

li II ^.

II II >CH OH OH OH

II II ^ III

N C— X CH— CHOH.CHOH.CH.CH2— 0— P— 0— P— 0— P— OH

I 0 1 11 II 11

u 0 0 0

to give first hexosemonophosphate and then hexose
diphosphate. At the same time the adenosine triphos-
phate is converted through the diphosphate to adenosine
monophosphate or adenylic acid : —

X=-=:C'.NH.,

I I
CH C— X

II II %CH 9^

II II /^^ I

N C— N CH.CHOH.CHOH.CH.CH2.O.P— OH

I 0 I 11

The liexose diphosphate (fructose- I : G -diphosphate),
under the infhience of the enzyme zymohexase, breaks

276 BACTERIOLOGICAL CHEMISTRY

down to give a mixture of dihydroxyacetone mono-
phosphate and 3-phosphoglyceraldehyde : —

CH2O.PO3H2 CH0O.PO3H2

I I

CO CO

I 1

HOCH ^^ CH2OH

I +

HCOH CHO

I I

HCOH HCOH

I I

CH2O.PO3H2 CH2O.PO3H2

The dihydroxyacetone phosphate is converted into
3-glyceraldehyde phosphate by the enzyme isomerase : —

CH2O.PO3H2 CH2O.PO3H2

I I

CO ^=^ CHOH

I • I

CH2OH CHO

The 3-glyceraldehyde phosphate, reacting with inorganic

phosphate and co -enzyme I, is oxidised to 2 : 3-diphospho-

glyceric acid whilst the co -enzyme is reduced (see

p. 203) :—

CHO COOH

I Trio.se- | Dihydro-

CHOH + H3PO4 + Co-enzyme I — > CHO.PO3H2 + co-enzyme

1 phosphorvlase | I
CH2O.PO3H2 ^ CH2O.PO3H2

2 : 3-Diphosphoglyceric acid then loses phosphate to
adenosine diphosphate :—

2 : 3-dipliosphoglyceric acid + adenosine diphosphate >

3-phosphoglyccric acid + adenosine triphosphate

3-Phosphoglyceric acid is converted by the enzyme,
phosphoglyceromutase, to 2-phosphoglyceric acid and
this by the enzyme, enolase, to phosphopyruvic acid : —

COOH COOH COOH

CHOH — =^ (^HO.POgHo .— ^ C.O.PO3H2 + H2O

I I " II

CHaO.POgHa CH2OH CH,

(.■i-|)liosi)lioglyccric uci.l; (:.'-iili(»sphog:Iyct.'ric' m-i.t) (pliosijliopyruvi.- acid)

ALCOHOLIC FERMENTATION 277

Phosphopyruvic acid is now dephosphorylated by adeno-
sine diphosphate to give pyruvic acid and adenosine
triphosphate which hands on its phosphate groups to
fresh hexose molecules. Pyruvic acid is decarboxylated in
presence of the enzyme, carboxylase, to give acetaldehyde
and CO2 : —

CUOH CO.,

I + CHO

CO >

I

The acetaldehyde is reduced by dih^^lro -co -enzyme I
to give ethyl alcohol : —

CHO CH2OH

I + Dihydro- CO -enzyme I > \ + Co-enzyme I

CH3 (-H3

The CO -enzyme I thus becomes available again to oxidise
3-phosphoglyceraldehyde to more 3-phosphoglyceric acid,
and the cycle is maintained.

The reactions outlined above represent the " Stationary
condition," in which acetaldehyde acts as hydrogen
acceptor from dihydro -co -enzyme I. In the " Initial
phase," as Meyerhof calls it, before any acetaldehyde is
available 3-phosphoglyceraldehyde is dismuted with
formation of a -glycerophosphate and 3-phosphoglyceric
acid. The latter is involved in the series of reactions
already described with the ultimate production of CO2
and alcohol whilst the « -glycerophosphate is hydrolysed
to form glycerol and pliosphate. It is in this way that the
2 to 3 per cent, of glycerol jjroduced in a normal alcoholic
fermentation arises. The initial phase can be inhibited
by fluoride which prevents the utilisation of the 3-phospho-
glyceric acid. The effect of fluoride can be overcome
by adding acetaldehyde so that the initial phase is
" by-passed." If acetaldehyde is fixed by addition of
siilpliite or (limedon. or if the stationary (•(Hiditioii is

278 BACTERIOLOGICAL CHEMISTRY

inhibited by mono-iodoacetate which blocks the conver-
sion of glyceraldehyde phosphate to phosphoglyceric
acid, the system continues in the initial phase and
glycerol, via « -glycerophosphate, accumulates at the
expense of alcohol.

This scheme differs from Neuberg's in that glycer-
aldehyde phosphate and not methylglyoxal is the inter-
mediate in the formation of pyruvic acid (compare
Kluyver's scheme). The formation of glycerol under
alkaline conditions is explained in the same way as in
Neuberg's scheme, the acet aldehyde undergoing a
Cannizzaro reaction rather than being reduced to alcohol
at the expense of dihydro -co -enzyme I.

Biological Reduction. — It will have been gathered
from what has gone before that reduction plays an
important part in alcoholic fermentation, particularly
the reduction of acetaldehyde to alcohol and of glycer-
aldehyde to glycerol. If other hydrogen acceptors are
introduced into the system they will compete for the
hydrogen available and themselves become reduced.
For instance, Neuberg showed that if an excess of acet-
aldehyde was introduced into a fermenting mixture it
first of all underwent condensation, as a result of the
action of the enzyme carboligase, present in yeast, to
give acetoin, CH3CO.CHOH.CH3 ; the acetoin could act
as a hydrogen acceptor, and in doing so became reduced
to give 2 : 3-butylene glycol, CH3.CHOH.CHOH.CH3. If
the aeration of the solution was increased, or if other
hydrogen acceptors such as methylene blue or sulphur
were added, the yield of butylene glycol was lessened and
acetoin accumulated, because oxygen or the other sub-
stances diverted a considerable portion of the available
hydrogen .

A great variety of compounds can be reduced in the
presence of actively fermenting yeast. Thus nitro-
benzene gives aniline, benzaldehj^de is converted to
]}onzyl alcohol, and the ketone, methylheptanone, gives

ALCOHOLIC FERMENTATION 279

the alcohol, methylheptanol ; the process is not restricted
to organic compounds, but inorganic substances, too,
can be reduced in this way ; sodium thiosulphate, for
example, yields hydrogen sulphide and sodium sulphite,
and elementary sulphur gives hydrogen sulphide.

The fact that yields of optically active alcohols of
considerably more than 50 per cent, can be obtained
from optically inactive starting material indicates that
true reduction occurs and not a Cannizzaro reaction,

2R.CH0 + H2O > R.CH2OH + R.COOH,

which would give a maximum possible yield of 50 per
cent. For instance, aldol, CH3.CHOH.CH2.CHO, gives
63-5 per cent, of optically active 1 : 3-butylene glycol,
CH3.CHOH.CH2.CH2OH.

The hydrogen used in these redactions is in all
probability that produced during the formation of the
pyruvic acid from the three-carbon intermediate, since in
one case at least, when methylheptanone is reduced to
methylheptanol, an amount of acetaldehyde can be iso-
lated corresponding to the amount of methylheptanol
produced. Moreover, these reductions occur only during
active fermentation and do not take place in the presence
of yeast suspensions alone, which indicates that the reduc-
tions are coupled with the normal fermentation reactions.

These reductions are frequently called ph}i:ochemical
reductions, yeast being regarded as a plant.

Fusel Oil. — Besides the glycerol, alcohol and carbon
dioxide, which are the main products of the action of
yeast on sugar solutions, small quantities of other products
are also formed. The chief among these are the higher
alcohols constituting the high boiling fraction, fusel oil,
which represents 0-1 to 0-7 per cent, of the products. It
was originally thought that the higher alcohols originated
as by-products of the sugar breakdown, but it was proved
]>y Ehrlich that their real source is the amino -acids
derived from the medium or from the autolysis of yeast

]9

280 BACTERIOLOGICAL CHEMISTRY

cells. He showed that leucine, (CH3)2CH.CH2.CHNH2.
CO OH, was converted by yeast in sugar solution to iso-
amyl alcohol, (CHgjgCH.CHg.CHaOH, and that ^5oleucine,
CH3.CH.CH.NH2.COOH, gave rise to cZ-amyl alcohol,

C2H5
CH3.CH.CH2OH, by loss of carbon dioxide and ammonia.

I

C2H5

Other amino -acids were found to undergo the same sort
of conversion giving rise to the corresponding alcohol
(see p. 231).

The production of these alcohols is only brought about
by living cells which are actively growing in a sugar
medium. Yeast juice and yeast preparations, like zymin,
do not lead to the conversion of the amino -acid to the
alcohol, nor does living yeast in the absence of sugar. It
seems that the process is bound up with the life of the
cell, and is the mechanism by which the organism obtains
the nitrogen for its synthetic processes. All the ammonia
produced is assimilated by the cell, and practically none
accumulates in the medium. Yeast juice or zymin,
which do not produce new cells, do not assimilate ammonia
and the reaction does not occur ; living yeast in the
absence of sugar does not grow and reproduce, and again
there is no demand for ammonia for synthesis. This
dependence of fusel oil production on the nitrogen
metabolism is further illustrated by the fact that if a
readily available source of ammonia is present in the
form of salts, the production of the higher alcohols is
very much lowered because the requisite ammonia is
supplied more readily from the salts.

It is very probable that the flavours and bouquets
of fermented drinks depend largely on the proportions
and natures of the various alcohols and esters produced

ALCOHOLIC FERMENTATION 281

in this way, these proportions depending in their turn on
the amino -acids present in the liquors being fermented.

For further reading : —

A. Harden, " Alcoholic Fermentation," Fourth Edition. Monographs on
Biochemistry. Longmans, Green & Co. London, 1932,

A. J. Kluyver, " The Chemical Activities of Micro-organisms," Lecture I.
University of London Press. London, 1931.

0. Meyerhof, " The Intermediate Products and the Last Stages of Carbo-
hydrate Breakdown in the Metabolism of Muscle and in Alcoholic
Fermentation." Nature, September, 1933, 337, 373.

S. C. Prescott and C. G. Duim, " Industrial Microbiology." McGraw Hill
Book Company, Inc. New York, 1940.

R. Robison, "The Significance of Phosphoric Esters in Metabolism." New
York L'niversity Press. New York, 1932.

M. Schoen, " Le Role du Phosphore dans les Processus de Fermentation."
Bull. Soc. Chim. Biol, 11 (1929), 819.

CHAPTER XVI

THE FERMENTATION PRODUCTS OF J
THE LOWER FUNGI I

SINCE Wehmer's classical work on the production of m
oxalic acid by fungi in 1891 a vast bulk of data on |
mould fermentations has accumulated, as a result of
which it emerges that there are three main differences
between bacterial and mould fermentations. The first
of these is that the moulds appear to have considerably
greater and more diverse synthetic powers than the
bacteria. They are able to produce a large variety of
compounds, aromatic as well as aliphatic, besides the
normal cell constituents, proteins and nucleic acids. The
range of these synthetic powers is indicated by the
following examples of some of the types of product
encountered : Benzopyrone, quinones, phenolic acids,
heterocyclic compounds, pyrones, long chain fatty acids,
fats, polysaccharides and sterols. Secondly, and

perhaps most important from a practical point of view,
the fungi produce non-volatile, often polybasic, acids,
of which oxalic, citric and gluconic acids are the most
important. They usually produce about 0-1 per cent.,
and never yield more than 1 per cent., of volatile acids.
Lactic acid has only been reported as a mould product
from Monilia, Mucor rouxii, Rhizopus oryzce and Rhizopus
chinensis. If volatile acids are formed during the meta-
bolism of the fungi they are immediately further broken
down ; indeed, it has been shown that the volatile
acids are even more readily attacked by fungi than are
sugars. Bacteria, on the other hand, produce large
quantities of such volatile acids as acetic, propionic,
butyric and lactic acids. Thirdly, the moulds never
seem to give rise to hydrogen or metliane among their

282

PRODUCTS OF THE LOWER FUNGI 283

products, whilst liydi'ugeii is a very comiuoii product of
bacteria and methane is formed comparatively frequently.

The function that these very varied compounds
perform is still in very large measure unknown. The
carbohydrates and fats probably serve as storage or
reserve materials as also may such acids as gluconic and
citric acids. Some products, such as the yellow pigment,
citrinin, may play the part of oxidation-reduction systems.
Other substances may exert a protective effect by
depressing or inhibiting the growth of other organisms.
For example, many moulds develop such acid conditions
(jjH 1 to 2) that bacterial growth is stopped. An inter-
esting case is the production by Pemcillium notatum and
P. chrysogenum of the antibiotic penicillin which has
marked specific bactericidal powers ; in very high dilution
it entirely suppresses the growth of the pyogenic cocci
and organisms of the diphtheria group, but it is easily
tolerated in higli concentration by other organisms such
as those of the coli -typhoid group, the influenza bacillus and
the enterococci. Many other antibiotic substances are
now loiown (see Chapter XI).

In considering the metabolic products of the lower
fungi it must be remembered that they are essentially
aerobic organisms. If an adequate supply of air is
available they usually completely oxidise, to carbon
dioxide and water, not only the carbohydrate originally
present in the medium but also the metabolic products
which they themselves have formed. Consequently the
incubation periods allowed for formation of the products
must not be too long or low yields result ; the fermenta-
tions are usually stopped just before all the initial nutrient
substance has been used up.

Acid Production by Fungi

Acids of a wide variety of structures are synthesised
by the lower fungi from glucose as the sole source of
carbon. A list of them is provided in Table 19.

284

BACTERIOLOGICAL CHEMISTRY

,

<*

,

^

o

55

••»

>*

m

§1

'r<

•1

if

^'~

ft:;

s

Q

^

^"^

'^

e

y

to

_to

o

o

o 'S

;l.^

g~

'C

"^

£

■|

to

?S to

•S

»

s

4

07iilia
ferum
. niger

■^

-^

^

^

^

f^

^ ■

1^

^

^

^

w

•

■

•

•

•

•

'

«

o

o

o
o

W

GO

w

w

o

o
q

o
o

i

,

,

o

o

II

q

o
o
o

o

o
o

o
o

o
o

§

o
q

q

o

6
q

o

o

d
q

o

o

o

o

o

o

W

W

W

W

W

W

W

d

o

o

o

o

o

o

o

a

W

o

w"

w"

o

o

o

o

o

o

o

o

Q

o

o

o

o

o

o

o

o

.

•

■

'

•

■

■

•

•

>^

•

X

J

3

1

3

1

'3

_o

T

S '

QQ.

>j

Q

^

8

o

_o

i

rO

<5

II

'T3

1

:2

:2

o

1

3

3
o

1

o

1

.2

'S

_o

o

c '3

to o

o

;3

o

"o

>;5 O

-f^

'S,>^

t>.

^

>5

le

§

3

^^

s

JO

O

O

;C

kJ

g

p^

M

S

W

<1

"^

PRODUCTS OF THE LOWER FUNGI

285

i-i

H

>*

&5 V

>

1

w

Si C

g

•CS

it

ii

•2

s

-^

fell

11

1
S

5^

I,

5^

s

o

■S'-p

_:?:

.|.

,o

.l.

^

1

'^■1

■^

§

t'

S

•^
•^

^

^

;:i

"^'

a^

^

^

^*

•

■

M

•

O

'

o

'

»

q

.

W

M

H

q

W

o

X
o
o

O

d

o

HI

o

o
o

9

q

q

o

o

o o

M

W

W

1

o

q iu

II o

o

o

O

o
d

o

5'

d
q

q=^

a"

o
o

q— ;^

O

sT

w

W

Q

W W

o

o

o

o

o

^

^

O Q

o

o

■

'

'

'

'

'

'

O

r-<

c5

=0 •?

,

,

.

o

rs §

.

1

3
<

"i

'i

c3

8

1

pyruvic

o

'2

'S
8

'S

1

C

o

ched C
nethyl

'S

^

s

—

^

5: .3

c5

s

o

§

O

:c

>-

1 ^

1— I

<

28o

BACTERIOLOGICAL CHEMISTRY

' s

eo"'^

o ?:

s ~

• ^ s

o o

s: S

>l

•2^
1 -

■^1

w

Q

0)

w

. ft

^

^

'"'^'i^

p

^ i e

s

5

fi '^

i §

1
1

5
^

*1

^

(^

f^

a^""

ft^

•

■

■ .

o

o —

cq

w

w

w

^ ^ o

o

o

w

w

W W o

m'°

rt

o

O— o— o

7--)

W

H

p

o

1

g-O

o
o

o -

C

CO

(OH).CH
OOH

d

Hi

o

COOH
H.C(OH)

5

8

tl

4

g-

o

q— u

II

q — o

O "

— q

II

>

w'

W

^

._5'

w

o

o

o

o

o

o

CO

n

n

o

w

w

w

^

o

o

o

o

.

K*->

>>

•

X o

O --H

ts •

t- o

'B,

O -tJ

r-1f

c3

■

©

rO O

vLTc

tj ^^ "3

_, ©

o

o

,

3

■MI-

r2 ft

ll

<jH

l3

■ '^

o o

«0 -M

Minioluteic
droxy-jSy
tetradecoi

Spiculisporii
v5-dicarb
lactone)

rs ^

1
.2

o

g

1^

^

PRODUCTS OF THE LOWER FUNGI

fH

25

^

s

u

t3

Q

***

•e»

•»»

O

'S

'^

•?

Ch

<i)

<w

%>

fl^

T^

tf

'k

1J

5

e

.1

e

^

►Si

►ss

o

«

o

«

a^

c^

R^

ft^

'

'

W

'

O

Cl

CT "

w

rt .

'

q

o

W

w'

g

A.

o
o

q

t3

K

o

o

d

^ o

d

^ o

m

o d

d

1 \o

I \o

^ O

q O

s /

5 ^ /

0-'^\

^~*^'\

^ W/

2, W /

\o

^~3'

11 /

W /°

s

X

J-g/

o — q ^

M

o

Q

M

K

o

o

O

o

o

o

__^

_

s .

^^

s .1

>-

ydro for
ityrylj-y
lactone)

1

o ^

^ .in

. 3

6-S o

^•2

^^--

a ^

(J

O u

-^ I

^ d ?

>5

-£3

-»)

^ifi

1l "

Is

irlosic
carbox;
acid)

d

o

6

o

288

BACTERIOLOGICAL CHEMISTRY

g

^

§

1 e*^

>*

■o

•1

M

^

•o*

o

a.

si

^^'

w

ft^

<w

o

•^

u

g

^ ^-r^

S

2^

i

Co"

III

PM

t

a^

ft^

^

^

■

w

.

,

o

o

o

,

,

t£

w"

^

q

II

'

•

w

o

~o

o

d

w

o

•

w

o

d

(>H

w"

„/ %.

9 o

i

q

5

w

o

li V

^c

CO

d

9 o

W-

o

O— o/

II \

q
li

II

d

CO

d W
q

.,.Q

w'

w'^'^w'

o

o o

o

o

_o

1 1^

o O

>J

'g

SI

§^

o

o

^ >.

1-5 '

t-,

a; p^

f^ .S

c^

T.

a 2

g'?^

o

,j3

QQ.§H

ce^?

19

^

^•i

^'>^'i>

Tj

0

1

1

.2

1^^

§1^

Is

II

H

(^

<3l

C<1

PRODUCTS OF THE LOWER FUNGI

289

a

Q

f

P. griseo-fnlvnm
P. flexiiosiun

P. brevi-compactum
P. patidum

A. niger

J-

d

K 5 XX X w
o o o o o o ^

8/ — \ 8/ \ 8/ \S
\ / \ / \ /

^ GO
Q - W M

i

ii

O fc^

Phenolic acids —

O-Methyl-salicylic acid -

Gentisic acid (H-hydroxy-sali-
cylic acid)

Gallic acid (3:4:r)-trihy-
droxy-bcnzoic acid)

290

BACTERIOLOGICAL CHEMISTRY

S

S

Q

§

5S

S

O

o

Q

c

v3

P

s s

S^

S^

o

§ o

I'l

?

s

Uh

5^

?1h e

_ O

8/— \§

o

I

o

^o

^o

o c

O fl

-23

ts

g^

8s

(M r-H

t^^

II

X ©

rQ '^

>.;:;^

f^*?"

-c >>

M >>

'X C

ft fl

ft a

1 Qj

»o ^

'9. ft

CO

CO

^1

O S ;^
CO

PRODUCTS OF THE LOWER FUNGI

291

d

»

CD

w

J^

'3

'3

>>

S

g

e

30

EC

D

3
1

Is

if

g

5

s

•5

t3

p

8

=0

3

2

.1

«;3

1

Ph

'S

•^ o

to -

s

a.

a^

•s

S5

1

•S s

e

.^^

-^

-1

'S^

i

a

'^

ft^

!^^

o;

a.

^'

to

So

a^

"i

-

. ^

;

a

•

«

t3

1

/o-o^

'

8< >

g

•

xn

o

\ / „

W /-=^\ ffi

,

W o W

o o o

8< >

3

as

-J

c

O
g

13

o

o

B 5

00

X^

K

s

2

K

ft-

00

o ^

o

^

^

^

^

o

■

■

.

•

.

■

""H

73

"3

1"

,

,

,

,

.

3.2

o

"S S

OD N

■^

1

o c

^

'

1:5

'

•

■

I— f

£J

>j

"2
'3

1 §

H- 1

'"'

ll-

<

X

c3

.2

If

n

r:i

3

e3

'3

c5

S|

"3

c

0)

8|

1

1
.2

.2
'c
o

O

■g 2

s

•a

1 -

^

'S

3

5

O

■fS>

CO

1

s

5

292 BACTERIOLOGICAL CHEMISTRY

Oxalic Acid. — Oxalic acid was recognised quite early
in the form of crystals of calcium oxalate in many moulds.
Wehmer in 1891 showed that it was a fermentation
product and could be produced by Aspergillus niger
from a variety of substrates, including glucose. He
demonstrated that large yields of calcium oxalate can
be obtained by maintaining the medium more or less
neutral by the addition of calcium carbonate, which
precipitates the oxalic acid as fast as it is formed. Other
species of Aspergillus and some species of Penicillium also
give oxalic acid in good yield.

Citric Acid. — Citric acid was first described as a mould
product by Wehmer in 1893, who claimed that it was pro-
duced only by two species of Penicillium-like organisms
for which he created the genus Citromyces. However, he
himself showed later that citric acid formation is not
so restricted, but is, in fact, a common characteristic of
many species of Penicillium and Aspergillus, particularly
of the black species of the latter. By suitable adjustment
of the conditions of fermentation, namely, by growth in
a medium at pH 1 to 2 instead of the more usual ^H 6 to 7,
it is possible to suppress almost entirely the production
of oxalic acid by A. niger and to obtain citric acid as
almost the sole product. The strong acidity inhibits the
growth of bacteria, yeasts and most other fungi, which
makes the process very useful and easy to control
industrially. It is used in America for the large-scale
production of citric acid.

Gluconic Acid. — Gluconic acid seems to be an almost
constant product of fermentation by A. niger. It was
first reported as such by Molliard in 1922. He showed in
subsequent work that it was possible to produce oxalic,
citric or gluconic acids at will as the main product by
varying the proportion of the mineral constituents of
the medium. Reduction of the amount of phosphate
and nitrogen to a minimum gave greatly increased yields
of gluconic acid, as much as 80 to 90 per cent, being

PRODUCTS OF THE LOWER FUNGI 293

obtained. Lowering the nitrogen supply raises the yield
of citric acid, whilst high nitrogen and phosphate with
low potassium gives rise to good yields of oxalic acid.
The process is worked industrially in the United States,
using P. luteum (variety ruhrisclerotium) .

In view of the high yields obtained from glucose it is
considered that the formation of gluconic acid is a direct
oxidation of glucose by the enzyme glucose -oxidase and
is quite independent of the reactions by which citric acid
is formed. This is further borne out by the facts that
there is no correlation between the times of maximum
formation of the two acids and that there is no formation
of citric acid from gluconic acid. Some, but not all,
other sugars are oxidised in a similar way by Penicillium
luteum to give the corresponding acids ; thus mannose is
oxidised to mannonic acid and xylose to xylonic acid,
but galactose and arabinose are not affected, although
they may be oxidised by other fungi.

Lactic Acid. — Rhizopus oryzce, when grown with urea
as a source of nitrogen in surface culture or as an aerated,
submerged growth in a rotating di'um, may convert
up to 75 per cent, of the glucose utilised into c^-lactic
acid. Other species of Rhizopus produce large quantities
of Z-lactic acid from glucose, together with small quantities
of formic acid, acetic acid, fumaric acid, /-malic acid,
succinic acid and ethyl alcohol.

Fumaric and Malic Acids. — These acids are not very
common mould products but are formed in comparatively
low yields by the action of one or two species of Rhizopus
and Aspergillus from glucose, fructose, galactose or
arabinose. It is possible that they are normally inter-
mediates in the formation of other acids.

Succinic Acid. — Succinic acid is a rare product of
mould metaboHsm, although it is fairly commonly pro-
duced by yeasts and bacteria. It is probably formed by
all three tvpes of organism by the breakdown of the amino -
acid, glutamic acid, COOH.CHg.CHa.CHNHg.COOH,

294 BACTERIOLOGICAL CHEMISTRY

derived from degradation of the cell proteins, by the
mechanism of oxidative deamination followed by de-
carboxylation and oxidation (see p. 230) : —

COOH.CH2.CH2.CH.COOH— >COOH.0H2.Cn2.CO.COOH— >C00H.CH2.CH2.CH0

NH2 I

(glutamic acid) (a-ketoglutaric acid) COOH.CH2.CH2.COOH

(succinic acid)

It is known that A. niger and Ehizopus nigricans fix
carbon dioxide and it is possible that succinic acid may
be formed by this process (see p. 258). The largest
yields of succinic acid are given by species of Fusarimn,
Mucor and Rhizopus.

Kojic Acid. — Kojic acid was first isolated from the
mycelium of Aspergillus oryzoe, an organism used in
Japan to ferment steamed rice, koji, to produce the
alcoholic beverage, sake. It has been since shown to be
produced by a number of other related Aspergilli and by
one species of Penicillium, Some acetic acid bacteria,
for example Bad. xylinoides, produce small yields of
kojic acid from mannitol and fmctose. The closely

CO

related comenic acid, || || > is also produced

HC C.COOH

0

by the gluconic acid bacteria. The constitution of kojic
acid was established by Barger and Yabuta as the y-

CO

., . . HO.C CH

pyrone derivative, n \\

HC C.CH2OH

o

It is produced in 10 to 20 per cent, yields by the fer-
mentation of glucose, fructose, sucrose, galactose, lactose,
xylose, aral)inose, glycerol, mannitol and starch. The
yield of the six-carbon compound, kojic acid, is just as

PRODUCTS OF THE LOWER FUNGI 295

high from pentoses and glycerol as it is from hexoses,
which suggests that it is not formed by a series of oxido-
reductions from glucose, for instance, as its structure
would at first sight indicate : —

M^OlRl CO

HOCiHiiHOiCH ^ HOC CH

|i :.:": M " II II

HCiOHi-HiCCHaOH HC CCH^OH

XoT^ Xo/

(glucose) (kojic acid)

It appears more likely that the sugars first of all undergo
breakdown to a common intermediate which is then
built up into kojic acid. The intermediate appears not
to be acetaldehyde since the formation of kojic acid is
not hindered by fixation of acetaldehyde by dimedon or
sulphite. Aspergillus flavus and A. parasiticus, which
both produce kojic acid from glucose in culture, when
plasmolysed by chloroform or toluene, convert starch,
maltose, sucrose or glucose into glucosone, CHoOH.
(CHOHJaCO.CHO. ^4. flavus converts glucosone into
kojic acid in normal culture, suggesting that glucosone
may be an intermediate between glucose and kojic acid.
A less likely alternative is that all sugars are first con-
verted into a single reserve polysaccharide and that this
is the source of the sugar which is the immediate precursor
of the kojic acid ; thus five hexose units might unite to
give a C30 polysaccharide, and six pentose units or ten
triose units condense to give the same polysaccharide.
The difficulty is that the number of oxygen bridges
would be different in each case. A further objection to
the suggestion that reserve carbohydrates are involved
is that kojic acid is not formed when dry mycelium is
used as carbon source for the mould.

Challenger has suggested that kojic acid may arise by
the condensation of two molecules of dihydroxyacetone
and oxidation : —

296 BACTERIOLOGICAL CHEMISTRY

C = 0
OH H.CHOH /\

/ \ -2H2O / \

HC C.OH > CH C.OH

II + II +iO, II II

CHoOH.C CH CH2OH.C CH

\ / \ /

OH HO

Penicillic and Mycophenolic Acids. — In 1890 Gosio
isolated moulds from diseased maize which gave rise to
products, giving a blue colour with ferric chloride, which
were thought to be toxic substances causing pellagra,
now known to be a deficiency disease due to lack of
nicotinamide. Later P. puherulum giving an acid,
penicillic acid, C8H10O4, and P. stoloniferum giving
mycophenolic acid, Ci7H2oOe, were isolated from mouldy
maize. Penicillic acid has been shown to have the
constitution y-keto- p-methoxy- S-methylene-Aa-hexenoic
acid which exists in both the keto and lactone forms, the
latter being p-methoxy- y-hydroxy- y-zsopropylidene
tetronic acid : —

CH3. CHgv

>.C0.C(0CH3) =CH.COOH ^-=^ \C.C.(0H).C(0CH3) =CH.CO

CH,^ CH.^ I I

'. ^

The constitution of mycophenolic acid has not yet been

fully worked out, but it contains the following nucleus

which represents the demethylated acid : —

C.OH CO

HOOC.C ^

HO.C

C.H

5^^12

Tetronic Acids. — Carolic, carolinic, carlic and carlosic
acids, produced by P. cliarlesii when grown on glucose
as the sole source of carbon, are condensation products

PRODUCTS OF THE LOWER FUNGI 2!J7

of butyrolactone and succinic acid with y-methyltetronic
acid and of butyrolactone and butyric acid with
y-carboxy-methyltetronic acid respecti v^el}^, as shown by
the formulae given in Table 19, (p. 287).

It will be seen that they have stiTictures bearing a
close resemblance to that of ascorbic acid, vitamin G,
which is also said to be produced by some Aspergillus
and Penicillium species, although the evidence rests on
the demonstration of a substance reducing 2 : 6 dichloro-
phenolindophenol and not on isolation of ascorbic acid.
Terrestric and penicillic acids are also shown to be
derivatives of tetronic acid.

Theories of Acid Production by Fungi.— There is still a
considerable amount of controversy as to the mechanism
of the production of acids by fungi, particularly as regards
the commercially important oxalic and citric acids.
Ehrlich, in 1911, suggested that citric acid was formed as
the result of the condensation of three molecules of
acetaldehyde followed by oxidation of a methyl group
and the aldehyde groups to give the carboxyl groups : —
CH2.CHO CH2.COOH

--> CH.CH3 > C(OH).COOH (citric acid)

^ + I I

H.CHa.CHO CH2.CHO CH2.COOH

This cannot be the correct mechanism, however,
since the maximum yield of citric acid from a hexose
would be 71 per cent., whereas in practice yields as high
as 95 per cent, can be obtained readily.

The theory of Chrzaszcz and Tiukow, according to
whom acetic acid, derived by steps like those in alcoholic
fermentation, condenses via succinic, fumaric and malic
acids to give citric acid ; that of Bernhauer, according
to which acetic acid condenses with succinic acid to give
aconitic acid which in turn yields citric acid and that of
Emde which suggests that citric acid arises by oxidation
of (juinic acid are all su]>jeot to the same o])jecti()n that
they do not allow for yields of citric acid as high as those

298

BACTERIOLOGICAL CHEMISTRY

obtained in practice. This limitation obviously applies
to any scheme which involves decarboxylation. That the
same initial steps as occur in alcoholic fermentation
leading to the formation of acetaldehyde (see p. 275)
are not involved in citric acid production is shown by
the fact that formation of the acid is not inhibited by the
presence of mono-iodoacetic acid, but actually accelerated.
Phosphorylation however, is involved since inorganic phos-
phate and glucose disappear from the medium, with forma-
tion of organic phosphate esters, as A. niger produces
gluconic and citric acids. Inorganic phosphate is liberated
again as gluconic and citric acids are further broken
down to oxalic acid in older cultures.

Eaistrick and Clark in 1919 investigated the produc-
tion of oxalic acid from a large number of acids and
showed that it was not formed from any three-carbon
acid, such as propionic, pyruvic or lactic acids, nor from
any mono -basic four-carbon acid, like butyric acid, but
that the dibasic four -carbon acids, malic or succinic acids,
and also acetic acid, gave rise to good yields of oxalic
acid. They considered that the mechanism of its forma-
tion from glucose was via ay-diketo-adipic acid which
split to oxalacetic acid and acetic acid ; the oxalacetic
acid split down further to give oxalic and acetic acids,
whilst acetic acid was oxidised to oxalic acid, according
to the scheme : —
CHO

CHOH

COOH

I
CHOH

I ^

CHOH

CHOH

— 2H2O

CHOH

CHOH

CHOH

CHOH

CHgOH

(gliicosf)

COOH

(saccharic
acid)

COOH

I
CO

CH,

CO OH

•I-- + 1
I H
CH,

COOH

CO OH

•I-- + I
CH, H

COOH

+

COOH

--> COOH
+
CH,

COOH

COOH

COOH

CH3

(JOOH

COOH

(aY-i'ilvct(>- (oxalacetic
a(li|iic ucid) and acetic acids)

COOH

COOH

(oxalic acid)

PRODUCTS OF THE LOWER FUNGI

290

This mechanism accounts for the high yields of oxahc
acid which can be obtained and also for the fact that it is
formed from the four-carbon dibasic acids, which can all
give rise to oxalacetic acid, and from acetic acid, but not
from the three-carbon acids. They consider that citric
acid is formed by the same mechanism as far as the
breakdown into oxalacetic and acetic acids, which then
recondense to give citric acid : —

COOH

1

COOH

CH3

1

CH2

+

1

>

HO.C.COOH

lO.C.COOH

1

II

CH,

CH

1

1

COOH

COOH

(oxalacetic acid

(citric acid)

enolic form)

Challenger has suggested that citric acid may arise via
saccharic acid and [3 y-diketo-adipic acid, which undergoes
a benzilic acid transformation to give citric acid (a process
which occurs in presence of alkali in vitro) : —

COOH

COOH

COOH

CHOH

hn

CH2

COOH

1

CHOH -

-2H2O C.OH

CO OH

CH2

1

^ 1 ^

CO H

> HO.C.COOH

1

CHOH

C.OH

OHOH

CH

CH2

CH2

COOH

COOH

COOH

COOH

iharic acid)

(3y

-diketo-adijiic
acid)

(citric acid)

He regards citric acid as being the source of oxalic acid
via acetic acid, which is formed by decarboxylation of
malonic acid derived from acetonedicarboxylic acid : —

300 BACTERTOLOaiCAL riHEMISTRY

COOH

CH,

CH3COOH

I I +

HO.C.COOH > H.rOOH + CO > COOH CO^

I 11 +

CH, CHo.COOH CH, > CH, > COOH

I "^ III

COOH COOH COOH COOH

(acetonedicarboxylic (maloiiic acid)
acid

The evidence for these reactions is somewhat meagre ;
small amounts of saccharic, acetonedicarboxylic and
malonic acids were isolated, but p y-diketo-adipic acid
could not be detected as an intermediate nor could added
Py-diketo-adipic acid be fermented to citric acid.

Citric acid can be formed from glycerol and pentoses
as well as from glucose, fructose and sucrose. This sug-
gests that the carlwn source is broken down to a common
intermediate, which is then built up into a reserve carbo-
hydrate characteristic of the particular fungus, and that
this in its turn breaks down to give a hexose which is the
immediate precursor of the citric acid. The same explana-
tion may be offered of the formation of gluconic acid from
the pentose, arabinose, of kojic acid from pentoses, of
mannitol from glycerol and pentoses, and of poly-
saccharides composed of hexose units from glycerol and
pentoses .

Assuming the production of a common hexose pre-
cursor the theories of Butkewitsch, of Gudlet and of Ciusa
and Briill become possible. Butkewitsch suggested that
glucose was oxidised to glucuronic acid which underwent
intramolecular aldol condensation to give a f ive-membered
ring compound which was subsequently split and the
terminal C atoms oxidised with formation of citric
acid : — ■

PRODUCTS OF THE LOWER EUXGI

301

HCOH-

-CHOH HCOH-

CHO

CHOH

CHOH

CHO

HCOH
HCOH

(glucose)

CHOH

CHOH

COOH
CH,

CHOH

I
CHOH

/
CHOH

COOH

(Glucuronic
(acid)

COOH

CH2
/

C(OH)
COOH

C(OH)
COOH

(citric acid)

Gudlet suggested that glucose split to give succinic
acid and acet aldehyde. The former was transformed
through fumaric acid to malic acid, whilst the acetalde-
hyde was oxidised to acetic acid which condensed with
the malic acid to give citric acid. Ciusa and Briill found
that increased yields of citric acid were obtained if
glycollic and malic acids were added to A. niger fer-
mentations, and suggested that citric acid was formed
by their condensation : —

COOH
CH,

^HOH.COOH >

I
COOH

COOH

I
CH2

C(OH).COOH

CH2

COOH

This suggestion resembles that of Kluyver and of Virtanen

for the formation of succinic acid (see p. 250) by bacteria.

It is possible that citric acid may be formed by a

series of reactions analogous to part of the tricarboxylic

302 BACTERIOLOGICAL CHEMISTRY

acid cycle involved in muscle metabolism. Oxalacetic
acid can be formed by the condensation of carbon dioxide
and pyruvic acid (see p. 258). Oxalacetic acid condenses
with a further molecule of pyruvic acid to give cis-
aconitic acid which gives rise to citric acid : —

CO2 + CH3CO.COOH > C00H.CH2.C0.C00H^=^C00H.CH = C(0H)C00H

(oxalacetic acid)

— 2H

COOH.CH = C(OH)COOH + CH3.CO.COOH > COOH.C = CH.COOH

I + CO2

CH2.COOH

COOH.C. = CH.COOH "^^^O COOH.C (OH)— CH2.COOH

I ' I

CH2.C00H c:

(cis-aconitiG acid) (citric acid)

Neutral Products of Fungi

Ethyl Alcohol. — Many species of Aspergillus and
Penicillium give more or less yields of alcohol, the most
effective being A. oryzce which is used in Japan for the
production of alcoholic beverages. Many species of
Mucor produce alcohol, whilst most Fusarium species
give large yields, equivalent to those given by yeasts.
Phosphorylation reactions are known to be involved
and it is highly probable that the mechanism is very
similar to that of ordinary alcoholic fermentation by
yeasts.

Acetaldehyde. — Acetaldehyde has been detected by
fixation methods in the case of moulds which produce
alcohol. It is very probable that it is an intermediate in
other fermentations as well.

Glycerol. — Glycerol has been found among the products
of Mucor, Aspergillus and Penicillium species and seems
to be a normal fermentation product. It probably arises
by a mechanism similar to that by which it is formed in
yeast fermentations.

PRODUCTS OF THE LOWER FUNGI

Ethyl Acetate. — P. digitatum, wliich causes the olive -
green rot of citnis fniits, produces ethyl acetate, which
appears to be characteristic of the species.

Mannitol. — Mannitol is produced in yields as high as
50 per cent, of the sugar fermented by several species of
Aspergillus, notably A. nidulans, A. elegans and some
white species, and also by Byssochlamys fidva, Clastero-
sporum, H. geniculatum and P. chrysogemun, from all
sugars except fiiictose. On the other hand, when
mannitol is formed as a bacterial product it is derived
only from fructose and not from any other sugar. The
reaction is, apparently, a direct reduction : —

CH2OH

I
CO

(CH0H)3

CH2OH

(fructose)

I

CHOH

I
(CH0H)3

CH2OH

(mannitol)

When mixtures of fructose and glucose are fermented by
bacteria the former disappears rapidly with formation
of mannitol and acetic acid, whilst the glucose is fermented
more slowly to lactic acid.

In the case of the moulds it has been suggested that
two molecules of glucose, for instance, undergo a
Cannizzaro reaction with formation of gluconic acid and
mannitol : —

CHO + H2O + CHO

I

HCOH
HOCH
HCOH
HCOH
CH2OH

(glucose)

HCOH

HOCH

I
HCOH

I
HCOH

CH2OH

(glucose)

CH2OH

HOCH
HOCH

COOH

HCOH

I
HOCH

HCOH

OH

(mannitol)

ul

OH

CH2OH

(gluconic acid)

304 BACTERIOLOGICAL CHEMISTRY

If this were so it would be expected that the rechiction
product would be sorbitol, CH2OH.HCOH.HOCH.HCOH.
HCOH.CH2OH, corresponding in configuration to glucose,
and not mannitol, which corresponds to mannose.
Sorbitol, however, has never been detected in these
fermentations . It is possible that the conversion does not
involve glucose as such but the phosphorylated sugar,
shown to be fructose diphosphate in the case of yeast
fermentations. It has been shown that yields of 15 to
35 per cent, of mannitol, calculated on the sugar fer-
mented, are formed by the action of a white species of
Aspergillus on mannose, galactose, arabinose and xylose,
but not from fructose. No acid is formed.

According to the Cannizzaro reaction theory of
mannitol and gluconic acid production, .the nature of the
product which accumulates depends on the optimum ^^H
conditions for the organism. Thus the white Aspergilli
thrive in a neutral medium of about pYL 6 to 7 and utilise
the gluconic acid, leaving mannitol in solution. A. niger
on the other hand, favours an aoid medium and utilises
the neutral product mannitol and allows gluconic acid to
accumulate.

The moulds afford a better commercial source of
mannitol than bacteria, since they ferment the cheap sugar
glucose and not the expensive fructose, and do not require
organic nitrogen, in the form of peptone or yeast extract,
as do the bacteria. In addition the moulds produce
mannitol in almost pure solution with very few by-
products. Mannitol is used in the form of its hexa-
nitrate as a detonator.

Products containing Sulphur, Arsenic or Selenium. —

The cases of arsenical poisoning which used to occur as
a result of the action of fungi on wallpapers printed with
arsenic containing substances have long been considered
as due to the formation of volatile substances of the
type trimethyl arsine, (CH3)3As, which has been shown

PRODUCTS OF THE LOWER FUNGI 305

to be a product of P. hrevirtntlc when (^ixjwu on bread
containing inorganic arsenic compounds. Dimethyl
selenium and dimethyl tellurium were similarly formed
from sodium seleiiide and potassium tellurite respectively
by P. brevicaule, P. chrysogenum and P. notatum. Di-
methyl ethyl arsine, (CH3)2.C2H5.As, arises from sodiimi
ethyl arsonate and methyl diethyl arsine, CH3.(C2H5)2.As,
from diethyl arsenic acid ; n-propyl arsonic acid gives
rise to dimethyl n-propyl arsine, (CH3)2.C3H7.As, in the
same way. Similar derivatives of antimony could not be
obtained by the growth of P. brevicaule in potassium
antimony tartrate nor does the mould methylate sulphur.
Challenger has suggested that the methylation is through
the agency of choline or betaine which contains a mobile
methyl group capable of being transferred under biological
conditions, as shown by experiments using choline or
betaine containing methyl groups " marked " with heavy
hydrogen.

The wood rotting fungus, Schizophyllmn commune,
when grown on a synthetic medium containing inorganic
sulphates produces methyl mercaptan, CH3.SH.

A. sydowi, which can also produce volatile arsenic
compounds, is capable of incorporating the inorganic
sulphur of the medium into cyclic choline sulphate,

I I , which IS an anhydride of choline

sulphate. It has been shown that Verticillium albo-atnun
and Botrytis cinereci, when grown on a medium containing
asparagine and ammonium salts as sources of nitrogen,
produce thiourea. The production of biotin (see p. 103)
and of penicillin (see p. 171) obviously also involves the
metabolism of inorganic sulphate with, formation of
organic sulphur compounds, as does, of course, the
synthesis of proteins which all contain sulphur in some of
their amino acids and as SH groups, and which are formed
by all micro-organisms.

306 BACTERIOLOGICAL CHEMISTRY

Chlorine containing Products. — Almost all the inor-
ganic chlorine of Czapek-Dox medium is removed during
the growth of A. terreus and is converted into geodin,
C17H12O7CI2, and erdin, C16H10O7CI2. Although the con-
stitution of the products themselves is not yet known,
dihydroerdin (obtained by catalytic reduction of erdin)
is methylated by diazomethane to give 3^ 5'-dichloro-
4:6:2': 6'-tetramethoxy-4'-methylbenzophenone-2-car-
boxylic acid : —

OCH3 ^ OCH3

I II ^ II r^

CH3O L IICOOH CHgO'l ICH3

CI

Dihydro geodin is the methyl ester of dihydroerdin.
Replacement of the potassium chloride in the medium
by bromide or iodide did not give rise to the corres-
ponding compounds although the organism grew normally.

The yellow crystalline compound sulochrin,

COOCH3 ^ OH

I II ^ !l

HOL lloCHa HO'I icHg

which occurs in the mycelium of Oospora sulphur ea-
ochracea, has a benzophenone structure very like that of
dihydroerdin and dihydrogeodin although it lacks the
chlorine atoms. Sulochrin is also related to the pigment
ravenelin (see p. 396) since treatment of demethylated
sulochrin with concentrated sulphuric acid gives rise to a
xanthone derivative having a methyl group, a hydroxyl
group and the carbonyl group in the same positions as
those in ravenelin.

PRODUCTS OF THE LOWER FUNGI 307

Another chlorine containing metabolic product is the
strongly dextrorotatory, colourless compound griseof ulvin
obtained from the mycelium of P. griseo-fulvum, and
thought to have the structure : —

OCH3 CH

CH30N^^
CI

C CO

I I

C CH,

CH
CH,

Other Products. — A number of other products includ-
ing polysaccharides, pigments, sterols and fats, which
will be described in Chapters XIX, XX, and XXI,
together with the following examples serve to
emphasise the very wide range of the synthetic
abilities of the lower fungi. Some of the substances
which illustrate the diversity of the products

OH

/^jCHaOH

of the moulds are gentisyl alcohol, , which

%/
OH

is produced along with the corresponding gentisic acid

OCH3
^\
by P. patulum ; methyl anisate, , methyl

%/
COOCH3

^\CH=CH.C00CH3,

cinnamate, , and methyl p-methoxy-

%/

^\CH=CH.C00CH3,

cinnamate, , produced by the

CHjO^/

wood rotting fungus Lentinus lepideus ; mellein, which

308

BACTERIOLOGICAL CHEMISTRY

has been shown to be identical with ochracin, produced
by A . melleus and A . ochraceus is a lactone of 6-hydroxy-
2-(a-hydroxypropyl) benzoic acid,

OH

which, on fusion with potassium hydroxide, yields 6-

methyl salicylic acid.

OH

/^COOH

"CH3

, Avhich is a product of

P. griseo-fulvum ; terrein, which is obtained from
A. terreus and has the structure, 4-propenyl-2-hydroxy-
3 : 5-oxidoc2/c?opentane-l-one

CO

HC

CH.CH-

HCOH

-CH

palitantin, C14H22O4, is a colourless, crystalline, unsaturated
dihydroxy aldehyde formed by P. palitans.

It has been noted that in many cases the same products
may be produced from several sugars, including those with
fewer carbon atoms, besides glucose as carbon source.
This suggests, as has been pointed out above, that they
are synthesised from either a common simple intermediate,
possibly acetaldehyde, or else from a common reserve
carbohydrate which is formed irrespective of the sugar
with which the organism is supplied ; this may be
represented schematically in the case of a pentose as : —

0 Pentose > Reserve Carbohydrate — ^-^ 5 Hexose > Other Products.

TRODUCTS OF THE LOWER FUNGI 30!)

In view of the very wide variety of products formed
it seems almost certain that a simple building stone like
acetaldehvde must be involved.

For further reading : —

F. Challenger, " Biological Methylation. ^'III. A Summary of recent
work on Biological Methylation and some Hypotheses regarding
its Mechanism." Chemistry and Industry, 61 (1942), 397, 413, 456.

H. Raistrick, et alia, " Studies in the Biochemistry of Micro-organisms."
Phil. Trans. Roy. Soc. B., 220 (1931), 1. See also numerous papers
in the Biochemical Journal.

H. Raistrick, (a) " Biochemistry of the Lower Fungi." Errjehnisse der
Enzymforschung, 1 (1932), 345.
^b) " The Biochemistry of the Lower Fungi." Ann. Bevieiv
of Biochemistry, 9 (1940), 571.

CHAPTER XVII
INDUSTRIAL FERMENTATIONS

IN recent years there has been a very considerable
expansion of the use of micro-organisms in industrial
processes, largely due to a realisation of the variety of
catalytic properties possessed by their enzyme systems.
Many uses of bacteria and yeasts in industry, it is true,
date back to time immemorial ; it is only necessary to
mention the production of alcoholic beverages of all sorts,
baking, tanning and the retting of flax and hemp in order
that this may be realised.

The Production of Glycerol by Fermentation. — As we
have seen, glycerol is a normal product in alcoholic
fermentation, where it occurs to the extent of about
3 to 4 per cent. During the 1914-1918 war period
Connstein and Liidecke, in Germany, added sulphite and
bisulphite to sugar solutions fermented by Saccharomyces
cerevisice and obtained much increased yields of glycerol,
which ran parallel with the amount of acetaldehyde
fixed by the sulphite (which is probably converted to
bisulphite by the action of carbon dioxide). As the
sulphite solutions are strongly alkaline, and since the
alkalinity increases during fermentation by production
of sodium bicarbonate, the medium soon becomes un-
suitable for continued fermentation, and recovery yields
of about 25 per cent, of glycerol are obtained. This
difficulty was overcome to a certain extent in Great
Britain by the use of a mixture of approximately equi-
molecular proportions of sodium sulphite and sodium
bisulphite which l)uffers the solution at about pTL 7. The
yield is about 30 to 35 per cent.

INDUSTRIAL FERMENTATIONS 311

In America glycerol was produced by maintaining the
medium alkaline with sodium bicarbonate, conditions
under which acetaldehyde tends to undergo dismutation
with formation of acetic acid and alcohol. Glycerol is
formed in amounts corresponding to the quantity of
acetaldehyde diverted from its normal function of acting
as hydrogen acceptor in the formation of pyruvic acid
(see p. 277).

The British method gives the best yields, but the
recovery of glycerol is difficult owing to the interference
of the sulphite. Moreover, there is little or no market
for the acetaldehyde produced as a by-product. The
American method, although giving lower yields, is the
best commercially, since alcohol is also formed and the
proportion of glycerol and alcohol produced can be easily
varied according to demand by adjustment of the degree
of alkalinity.

Molasses or hydrolysed wood pulp or similar vegetable
w^astes serve as an effective sugar source for the growth of
the yeast in glycerol fermentation.

Power Alcohol Production. — The production of alcohol
for fuel purposes and for use in the chemical and other
industries has grown very largely within the past ten to
fifteen years. The fermentation is generally carried out
by the use of an appropriate strain of yeast to ferment
molasses, which provides an excellent source of carbo-
hydrate, nitrogen and mineral salts in immediately avail-
able condition. An inoculum is built up from a pure
culture of the yeast, maintained in the laboratory, by
successive transfers to increasing quantities of molasses
until a " seed " is obtained, about 5 per cent, of the final
volume to be fermented. The molasses is maintained at
pH 5, which allows good gro\vth of the yeast but inhibits
the gro\\i:h of most bacteria which might act as
contaminants. After fermentation for about forty-eight
hours the alcohol, present to the extent of 5 to 10 per
cent., is distilled off from the mash.

21

312 BACTERIOLOGICAL CHEMISTRY

A more recent development is the use of vegetable
wastes as a source of the necessary sugars for fermenta-
tion. Practically all vegetable waste materials, such as
straw, grass, maize cobs and husks, and sawdust contain
insoluble hemicelluloses, built up largely of pentosans.
The pentosans are broken down by a preliminary acid
treatment to give the soluble pentose sugars. The mash
so produced is fermented by yeast.

Ethyl Alcohol-Acetone Fermentation. — Acetone and
ethyl alcohol are produced from maize, potatoes, molasses
and various vegetable waste materials by the action of
B. aceto-ethylicus, which is probably identical with the
starch fermenting organism, B. macerans. The fermenta-
tion proceeds best at a temperature between 40° and
43°C. and in presence of calcium carbonate, to prevent
development of excessive acidity, the mash after pre-
liminary acid treatment being brought to a ^^H value
about 8. The chief products are acetone and ethyl
alcohol corresponding to about 8 to 10 per cent, and
20 to 25 per cent., respectively, of the carbohydrate
fermented ; acetic, lactic and formic acids are also
formed together with hydrogen and carbon dioxide.
The solvents are separated by fractional distillation.

Speakman suggested the following scheme as the
mechanism of fermentation. Glucose is broken down,
probably in a way similar to that in alcoholic fermentation,
to give pyruvic acid : —

CfiHiaOe > 2CH3.CO.COOH + 4H

The pyruvic acid is then transformed in three ways,

(1) to give lactic acid : —

CH3CO.COOH + 2H — > CH3CHOH.COOH

(2) to give carbon dioxide and acetaldehyde, the latter
then giving alcohol : —

+ 2H

--^ CHX'HoOH

INDUSTRIAL FERMENTATIONS 313

(3) to give acetic and formic acids : —

CH3CO.COOH + HO.H > CH3COOH + HCOOH.

The balance between the three modes of fermentation will
depend on the conditions prevailing. At a later stage in
the fermentation acetone and ethyl alcohol are formed
together, probably as a result of the condensation of
acet aldehyde to give aldol,

2 CH3.CHO — > CH3CHOH.CH2.CHO,

which then undergoes dismutation with a further molecule
of acet aldehyde to give p-hydroxybutyric acid and ethyl
alcohol : —

CH3.CHOH.CH2.CHO O CH3.CHOH.CH2.COOH

+ 11 — > +

CH3.CHO H2 CH3.CH2.OH

The p-hydroxybutyric acid undergoes an oxidation-
reduction with acetaldehyde to form acetoacetic acid
and ethyl alcohol : —

CH3.CHOH.CH2.COOH CH3.CO.CH2.COOH

CH3.CHO + CH3.CH2.OH

Finally acetoacetic acid is decarboxylated with formation
of acetone : —

Bakonyi has suggested that aldol undergoes an internal
oxidation and reduction and splitting to give acetic acid
and ethyl alcohol : —

CH3CHOH.CH2.CHO ^ CHs-CHa.OH -r CH3.COOH

Two molecules of acetic acid condense to give acetoacetic
acid which is then decarboxylated to give acetone and
Carlson dioxide : —

(■H3.CUOH + CH^.COOK . ('H3.C0.('H,.(()()H i K.,i)

CH3.CO.CH2.COOH > CH3.CO.CH3 + CO2

314 BACTERIOLOGICAL CHEMISTRY

Although acetoacetic acid may be derived by condensa-
tion of acetic acid it seems unnecessary to postulate that
the acetic acid is formed via aldol when it is known that
acetaldehyde can give rise to it and ethyl alcohol by
dismutation in presence of the enzyme aldehydemutase.

Butyl Alcohol-Acetone Fermentation. — The production
of acetone by fermentation processes was initiated during
the 1914-1918 war period, when it was required in large
quantities for the manufacture of explosives. The fer-
mentation has undergone even greater development since
that time as a result of the greatly increased use of butyl
alcohol and acetone as solvents for lacquers and in the
form of esters as plasticisers in the cellulose paints and
varnishes and in cellophane preparation. When the
process was developed in this country and the United
States in 1915 the butyl alcohol was a by-product for
which no market could be found ; nowadays it is the
more valuable of the two products.

The organisms responsible for the production of these
solvents from starch and sugars are spore -bearing soil
organisms of the type CI. aceto-butylicum, which are
similar to the flax-retting organisms. Two groups of
these organisms are recognised. One group produces the
organic acids, butyric and acetic acids, but cannot ferment
them further to the neutral products butyl and ethyl
alcohols and acetone ; to this group belong the true
butyric acid bacteria such as CI. hiityricum. The second
group comprises the butyl alcohol bacteria which, under
favourable conditions of the medium, reduce the inter-
mediately formed acetic and butyric acids to ethyl alcohol,
acetone and butyl alcohol. Some of these organisms, for
example, CI. butyricum, require the medium to be main-
tained neutral by the addition of calcium carbonate in
order tliat they may bring about fermentation. For
others, like CI. aceto-hutyliciwt,, the addition of calcium
carbonate is unnecessary or even undesirable.

INDUSTRIAL FERMENTATIONS 315

One uf the chief ooutaiiiinants pr(j(hi('iii<z unsalisfactoiy
fermentatioji is a lactic acid organism which can bring
the butyl alcohol fermentation to a complete standstill in
a few hours. A bacteriophage has also been incriminated
as the cause of faulty fermentations. The early method
of sterilisation of the mash in large tanks prior to fer-
mentation had to be abandoned in favour of the use of a
number of small tanks in order to ensure thorough penetra-
tion of the heat. The grain, usually maize, to be fermented
is freed from the embryo (from which an edible oil is
expressed, leaving a cake used as cattle food), sterilised as
a mash in water by steam under pressure at 130° to 140° C,
and transferred through sterile pipe lines to a large fer-
menting tank holding as much as 50,000 gallons. A seed
mash, built up from j)ure laboratory cultures, or withdrawn
from a previous fermentation, is introduced and fermenta-
tion allowed to proceed at 55° C. The production of a
maximum yield of products in the minimum of time
depends largely on the preparation of the seed mash.
This must bear the correct ratio to the total mash to be
fermented (usually about 12 per cent.), and the number
of subcultures used to ])uild the seed up to the requisite
amount is kept as low as possible in order to maintain the
activity of the organism. The seed is withdrawn after
the initial acidity of the fermentation has begun to decline,
since the main contaminant to be expected is the lactic
acid organism, B. volutans, whose presence is indicated
by the maintenance of a high acidity ; hence a sample
taken from a mash whose acidity is declining is more
likely to be pure.

Anaerobic conditions develop automatically ; the
fermentation starts in five to six hours and is complete
in forty-eight hours. In addition to butyl alcohol and
acetone, carbon dioxide, hydrogen and ethyl alcohol are
also formed, together with small amounts of z5o-propyl
alcohol. The chief products occur in the proportions : —

IT)

BACTERIOLOOTCAL

rnEMISTRY

Carbon dioxide -

- GO per cent

y/ -Butyl alcohol -

- ->

Acetone

- 10-5

Ethyl alcohol

- 3

Hydrogen -

- 1-5

The gaseous products are scrubbed to remove ethyl
alcohol and acetone, and the carbon dioxide is removed by
water under pressure, followed by dilute sodium hydroxide.
The residual hydrogen is converted into ammonia by
catalytic combination with nitrogen under pressure. A
recently developed alternative is to remove only three-
quarters of the carbon dioxide and to convert the remain-
ing mixture of carbon dioxide and hydrogen into methyl
alcohol by catalysis. The excess of carbon dioxide is
solidified and used for refrigeration. The solvents are
separated by fractional distillation.

In the early stages of the fermentation acidity due to
the production of acetic and butyric acids develops, but
later falls again as the neutral products appear. If
acetic or butyric acids are added to a fermenting mash,
increased yields of acetone and but 3d alcohol, respectively,
are obtained. It is suggested that the butyl alcohol is
formed by reduction of butyric acid and that the acetone
arises from acetic acid via aceto -acetic acid which is
decarboxylated. The scheme is summarised by the
following equations given by Kluyver : —

/OK

(1) CeHiaOe > 2CH3.CO.C^OH (methylglyoxal hydrate).

\h

Glucose breaks down to give methylglyoxal hydrate,
probably by the same mechanism as in alcoholic fermenta-
tion. The methylglyoxal hydrate then splits to give
acetaldehyde and formic acid, which in turn yields
hydrogen and carbon dioxide : —

TXDTJSTRIAL FER:\rENTATTOXS 317

1 /''^

(2) CHg.CO.C^OH > CH3.CHO + H.COOH

(3) H.COOH > H2 -f CO2

Part of the acetaldehyde is oxidised to acetic acid : —

/OH
(4) CH3CHO -f H,0 — > CHgC^OH — > H2 + CH3COOH,

and part condenses to give aldol, which by internal
rearrangement gives butyric acid : —

OH

-^ /

(5) 2CH3.CHO->Cn3.CHOH.Cnj.CHO->CH,.CH = OH.C — On^CHj.CHz.CFj.COOH.

4 \
H

The butyric acid is reduced by hydrogen formed in the
production of acetaldehyde and acetic acid, to give butyl
alcohol : —

(6) CH3.CH2.CH2.COOH + 4H > CH3.CH2.CH2.CH0OH + H2O.

Two molecules of acetic acid (formed according to
equation 4) condense to give aceto -acetic acid, which is
decarboxylated to give carbon dioxide and acetone : —

OH OH

^ \ /

(7) CHsC + H— C.COOH -> CHjC— GH2.COOH -> CHaCO.CH2.COOH + H2O

\ / \ I

OHH OH CH3COCH3 + CO2

The ethyl alcohol is formed as a side reaction by reduction
of acetaldehyde : —

(8) CH3CHO + 2H > CH3CH0OH,

^18 BACTERIOLOGICAL CHEMISTRY

and similarly the small amyuiit of /6o-propyl alcohol
arises by reduction of acetone :—

(9) CH3.GO.CH3 + 2H — > CH3.CHOH.CH3.

Kluyver's scheme outlined above should be modified
so that the preliminary stages, represented by equations
(1) to (3), agree with modern knowledge of the mechanism
of alcoholic fermentation. In other words the various
phosphorylation reactions leading to the formation of
pyruvic acid (see p. 275) are very probably those really
involved, rather than the formation and breakdown of
methylglyoxal. In the present state of our knowledge
there seems to be no reason to change the suggestions
concerning the later stages.

Lactic Acid Fermentation. — Lactic acid is one of the
commonest of bacterial products and is produced by a
wide variety of bacteria and yeasts. The earlier work was
largely done in connection with the dairy industry and
more particularly cheese manufacture. The first fer-
mentation method for the production of lactic acid was
described in 1841 by Boutron and Fremy before, however,
it was realised that the action was bacterial. When it
was established by Pasteur that this was the case, im-
provements were made, although it was not until 1896
that pure cultures of lactic acid bacteria were used.

The lactic acid bacteria were classified by Orla-Jensen
in 1919 into two main groups : —

1. The Homofermentative or True Lactic Acid Bacteria
which produce almost pure lactic acid from sugars. They
are Gram positive, non-sporing, non-motile rods and cocci
which give no surface growth on liquid media. They
will not grow in the absence of organic nitrogen com-
pounds ; they produce no catalase and do not reduce
nitrates. The true lactic acid bacteria are further sub-
divided into : —

INDUSTRIAL FERMENTATIONS 310

{a) Rods (i) thermophilic, e.g. LactohaciUnshulyaricus,
L. delbrilckii,
(ii) mesophilic, e.g. L, casei.
(h) Cocci, e.g. Streptococcus lactis.

(c) An intermediate group, e.g. Str. cretnoris which
gives some volatile acids (acetic and propionic)
in addition to lactic acid ; Leuconostoc citrovorum
(Str. citrovorus) and Leuc. dextranicum [Str.
paracitrovorus) which give volatile acids and
acetoin from lactic and citric acids.

2. The Hetero fermentative or Pseudo-lactic Acid Bac-
teria which produce volatile acids, carbon dioxide and
hydrogen, as well as lactic acid, which rarely exceeds
half the sugar fermented. They exhibit surface grow^th on
liquid media, reduce nitrates and produce catalase. The
group includes organisms of the coli -aero genes type and a
number of pathogenic organisms.

At the present time thermophilic organisms like
L. delhritckii {B. acidificans longissimus) or L. bulgaricus
are used to ferment the maltose and sucrose in molasses,
or glucose derived from the starch of potatoes, maize or
other cereals (which are first submitted to acid hydrolysis
or to the action of malt diastase), or the lactose in skim
milk or whey. The optimum temperature for fermenta-
tion is about 45° C, at which temperature the growth of
most contaminating bacteria, especially the butyric acid
bacteria, is inhibited. Yields of lactic acid of about 98
per cent., calculated on the sugar fermented, are obtained.
The sterilised mash, containing about 10 to 15 per cent,
of fermentable sugar, is inoculated under aseptic condi-
tions with a pure culture of the appropriate organism,
and fermentation allowed to proceed at the optimum
temperature, which is maintained by steam pipes dis-
tributed in the tanks. Calcium carbonate or lime is
introduced at intervals in order to maintain the solution

320 BACTERIOLOGICAL CHEMISTRY

neutral ov faintly acid ; the fei-nientatioJi is complete in
four to six days. The calcium lactate is recovered from
the mash by filtration, evaporation of the filtrate and
crystallisation. The calcium is removed as calcium
sulphate by addition of sulphuric acid and the lactic acid
solution concentrated to about 50 per cent.

Lactic acid is largely used for " deliming " hides, that
is, for removing the lime employed in the " dehairing "
process preparatory to tanning, and for softening hides.
It is used extensively in the textile industry as a mordant
in dyeing and for the acid dyeing of wool. It is much
used in the form of its ethyl ester as a plasticiser in resins
and lacquers and as a solvent for cellulose finishes. Lactic
acid is finding increasing use also as a flavouring agent
and acidulant in the preparation of " soft " drinks and
food products.

The initial stages of lactic acid fermentation are very
probably the same as those of alcoholic fermentation,
involving phosphorylation and co -enzyme I. If co-
enzyme I is removed from lactic acid bacteria by washing
they no longer produce lactic acid. Lactic acid fermenta-
tion is accelerated by the addition of phosphates in a
manner analogous to that of alcoholic fermentation.
Two mechanisms are possible for lactic acid fermentation,
the first A, going through pyruvic acid and the second,
B, through methylglyoxal.

zymohexase

A. (1) Hexosediphosphate > Dihydroxyacetone phosphate + 3-

phospho -gly ceralde hy de .

isomerase

(2) Dihydroxyacetone phosphate ■ > 3-phosphoglyceraldehyde

(3) 3 -Phosi)hogly ceralde hyde + phosphate + co-enzyme I

triosephosphorylase , , , ,

> 1 : 3-(liphosphoglyceric acid + dihydroco-

enzynie I.

(4) 1 : 3-I)iphosphoglyceric acid + adenosine diphosphate > 3-phos-

phoglyceric acid + adenosine trijjhosphate.

TNDUSTRTAT. FERMENTATIONS 321

(5) 3-t'husphiigly<erir acid — >■ phos])ho|iyruvi<: acid.

aiul phosphoglyccromutase

(6) Phosphopyruvic acid + adenosine diphosphate > pyruvic

acid + adenosine triphosphate.

lactic

(7) Dihydroeo-enzyme I + CH3.CO.COOH > Co-enzyme I -f

(pyruvic acid) enzyme

CH3.CHOH.COOH
(lactic acid)

Alternatively : —

B. (1) Hexose diphosphate > 2-glycerophosphate.

dehydrogenase

(2) Glycerophosphate + 2 cytochrome > glyceraldehyde-

phosphate + 2 reduced cytochrome.

Cytochrome

(3) 2 Reduced cytochrome + 0^— > 2 cytochrome + 2H2O.

oxidase

(4) Glyceraldehyde phosphate > methylglyoxal + phosphate.

(5) CH3CO.CHO + H2O glyoxalase CH3.CHOH.COOH.
(methylglyoxal) > (lactic acid)

Glyoxalgise was first isolated from dog's liver by Dakin
and Dudley, who showed that it converted methylglyoxal
quantitatively to lactic acid. Glyoxalase has also been
shown to be present in yeast and it has been shown that
L. delbrilckii and Aerobacter aerogenes convert methyl-
glyoxal quantitati v^ely to racemic lactic acid, whilst L.
delbrilckii converts hexose -diphosphate into methyl-
glyoxal and the latter to lactic acid.

Some organisms, for example L. pentoaceticus , produce
optically inactive lactic acid, whilst others, for example
Str. lactis, yield the dextro-rotatory acid, and yet others,
for example Leuconostoc mesenter aides, the Isevo -rotatory
isomer. It has been shown that those organisms which
yield inactive lactic acid contain an enzyme, racemiase,
which racemises the d- or /- forms of the acid.

The industrial production of lactic acid by fungi is
possible using species of Rhizopus or Mucor in surface
culture or in aerated submerged culture.

322 BAOTERIOLOGICAL CHEMISTRY

Acetic Acid Fermentation. Vinegar Fermentation.—

The production of vinegar from plant sugars via alcohol
is one of the oldest fermentation industries. It involves
two stages — the conversion of the sugar to alcohol by the
action of yeasts, and secondly, the oxidation of the alcohol
to acetic acid by various bacteria of the genus Acetobacter.
The latter process is strictly aerobic and is accomplished
in practice by trickling the alcohol solution over wood
shavings impregnated with the bacteria. If too little
alcohol is present the acetic acid formed is further oxidised
to carbon dioxide and water and lost. The aeration has
to be adequate or the oxidation of alcohol stops at the
acetaldehyde stage.

In the ordinary vinegar process the acetic acid is
probably formed by direct oxidation of the alcohol via
acetaldeh^^de : —

/OH

CH3CH2OH + 0 > CH3CHO + H2O > CHgC^OH + 0 — ^ CH3COOH + H2O.

(acotaklehydo hydrate) - (acetic acid)

The organisms A. ascendans, A. pasteuria7iimi and
A. xylinum have been shown capable of converting
acetaldehyde anaerobically to acetic acid by dismutation.
The reaction is catalysed by the enzyme aldehyde mutase
with the intervention of co -enzyme I as hydrogen carrier,
the aldehyde acting as both donator and acceptor of
hydrogen : —

CH3CHO+ Co-enzyme I > CH3COOH+ reduced Co-enzyme I

CH3CHO + reduced Co-enzyme I > CH3CH2OH + Co-enzyme I

It is therefore possible that some of the acetic acid in
vinegar fermentation is formed by dismutation of the
acetaldehyde : —

2CH3CHO — > CH3CH2OH + CH3COOH.

The alcohol so formed is then oxidised to acetaldehyde,

INDUSTRIAL FERMENTATIONS 323

which again undergoes dismutation ; the process continues
until all the alcohol has been converted to acetic acid.
Under normal aerobic conditions, however, this reaction
is much slower than the direct oxidation and contributes
but a small proportion of the yield.

The activity of the acetic acid bacteria in bringing
about the oxidation of glucose to gluconic acid and keto-
gluconic acid, of glycerol to dihydroxyacetone, and
of secondary alcohol groups to keto groups has already
been described (see p. 241 et seq.).

Methane and Hydrogen (Power Gas) Fermentation. —
In comparatively recent years the use of methane and
hydrogen produced by the fermentation of cellulose wastes
has developed considerably. The study of cellulose
fermentation has been largely the work of Omelianski.
There are three main types of cellulose degradation : —

1. Anaerobic at 20° to 37° C. with the production of
methane and hydrogen. This fermentation is the result
of the action of tw^o organisms — one CI. fossicularum,
giving the hydrogen, and the other, CI. methanigenes,
forming methane. CI. fossicularum is a long slender
bacillus with terminal spores, which fails to grow on
ordinary media. Its chief products are acetic, lactic and
butyric acids, ethyl alcohol, carbon dioxide and hydrogen.
The methane fermentation also produces much fatty acid
but even more gas than the other, as much as 50 per cent,
of the decomposed cellulose appearing as carbon dioxide
and methane, the remaining 50 per cent, appearing
mainly as acetic acid. The organism concerned is
morphologically very similar to the hydrogen producing
type, but the two can be separated by repeated short
heatings at 75° C, by which the more rapidly developing
methane -bacillus is killed off, leaving the slowly develop-
ing hjTlrogen-producing organism in the resistant spore
stage. Conversely, the hydrogen-bacillus can be elimin-
ated by repeated transfers whilst the methane fermenta-
tion is at its most active, the slow-growing hydrogen-

o24 BACTERIOLOGICAL CHEMISTRY

bacillus being " swamped out." These two organisms
are very widely distributed in soil and mud.

It is possible that Gl. methanigenes does not itself
attack cellulose, but that it forms methane by the
reduction of products formed by CI. fossicularum from
cellulose.

It is claimed that in most herbivorous animals 75 per
cent, of the cellulose which they digest is hydrolysed by
bacteria and not by the digestive fluids.

2. Anaerobic at High Temperature. — A number of
thermophilic organisms decomposing cellulose have been
isolated from soil and rotting plant residues. MacFadyen,
in 1894, isolated several such organisms from rotting
straw and showed that they grew at 60° C. and produced
acetic and butyric acids together with methane and
carbon dioxide. They were not pure cultures. CI.
thermocellum, isolated in America, decomposes cellulose
at 62° to 66° C. with formation of acetic acid and ethyl
alcohol. CI. dissolvens resembles Omelianski's CI. fossicu-
larum morphologically, but grows up to a temperature of
65° C. Cellulose, the only carbohydrate which it will
attack, is broken down to acetic acid, lactic acid, butyric
acid, alcohol, carbon dioxide and hydrogen. CI. cellulo-
lyticum (which may be identical with B. thermocellulyticus)
breaks cellulose down with conversion of 64 per cent, of
it to volatile acids of which 80 per cent, is formic acid
and about 16 per cent, acetic acid, the remainder being
propionic acid.

3. Aerobic Fermentation at 20° to 37° C. — A number of
aerobic organisms decomposing cellulose are known, but
usually they are in mixed cultures and are often symbiotic.
Neither their bacteriology nor chemistry has yet been
worked out satisfactorily. They produce acetic, butyric
and lactic acids, which are further broken down by other
organisms to water and carbon dioxide.

CytopJutga liutcliinsoni converts a])out two-thirds of
the carbon of the cellulose attacked into carbon dioxide.

INDUSTRIAL FERMENTATIONS 3l'5

most of tlie remainder being found in the gimi ^\ hich the
organism produces. Organisms of the genus Cellulomonas
(small motile or non -motile rods which may be pigmented)
also attack cellulose, but may use other organic sub-
stances, though usuaily rather feebly. Some vibrios
attack cellulose ; thus V. agar liquefaciens (Microspora
agar liquefaciens) produces acetic and formic acids from
cellulose and also attacks agar. V . amylocella forms the
same products from cellulose, starch and dextrin. Cellu-
lose is also attacked by the sporing organisms Cellulo-
hacillus rnucosus and C. myxogenes.

The anaerobic fermentations are used as sources of
j)ower gas and solvents from cellulose wastes and from the
decomposition of sewage sludge. Mineral salts are added
to the mash of w^ood pulp, sawdust, corn cobs, maize or
other plant residues, an inoculum of appropriate thermo-
philic organisms introduced, and fermentation allowed
to proceed at about 65° C. The conditions in the mash
soon become anaerobic. Hydrogen, methane and carbon
dioxide are produced together with some alcohol and
acetic acid, which is neutralised by calcium carbonate.
The carbon dioxide is separated from the gases and the
combustible gases used for lighting purposes or for
operating gas engines. The acetic acid is either re-
generated from the calcium acetate or the latter distilled
with formation of acetone. This process is extensively
operated in the " corn belt " of the United States, where
maize stalks and straw are partially fermented by
cellulose- and pectin-destroying organisms Avith production
of power gas and solvents ; the more resistant fibrous
parts of the stalk composing the vascular bundles are not
attacked, and are used for the manufacture of cardboard.

The cellulose in plant wastes may also be partially
decomposed, giving combustible gases and leaving humin
which is used as a fertiliser, and the same applies to the
anaerobic fermentation of sewage, where the gaseous
products are again methane, hydrogen and carbon dioxide.

32(> BACTERIOLOGICAL CHEMISTRY

In this latter case the fatty acids and proteins in the
sewage also contribute their quota to the decomposition
products.

The primary stages in the breakdown of cellulose are
probably hydrolytic with formation, by the action of
the enzyme cellulase, of the disaccharide cellobiose which
in turn is hydrolysed by the enzyme cellobiase to glucose.
These enzymes have been obtained in cell-free filtrates of
Cellulohacillus myxogenes and C. mucosus. Cellobiose
and glucose can usually only be detected as products of
the action of these organisms when the air supply is
restricted (preventing further degradation) or when the
growth of the organisms is prevented by such substances
as toluene. Dextrins may form an intermediate step
between cellulose and cellobiose. Some confirmation of
the intermediary nature of cellobiose and glucose is
afforded by the fact that the action of CytopJuiga
hutchinsoni on cellulose is inhibited by the presence of
0-1 per cent, of glucose or 1 per cent, of cellobiose which
are used preferentially. The acids and other products
of these micro-organisms in all probability arise by the
usual metabolic reactions from the glucose so formed.
It has been claimed that the gums so often produced by
cellulose decomposing organisms may consist of oxycellu-
lose and polyuronides formed by the oxidative break-
down of the cellulose, but the evidence is not at all
sound.

It has been shown that the fatty acids are broken
down by spore -bearing thermophilic bacteria, with the
production of methane and carbon dioxide only in
yields as high as 90 per cent. The process is in effect
an oxidation-reduction process involving water, as shown
by the empirical equation : —

n— 2 n 1-2 :}n— 2 ,^^
CuHanOa + ^-H^O > — — CO^ + ^^^H^,

(fatty acid)

INDUSTRIAL FERMENTATIONS 327

which is borne out by the gas ratios actually found. It is
considered that the methane from cellulose is formed in
this way from fatty acids, w^hich are the first breakdown
products of the cellulose. The amino -acids from proteins
are also subject to the same sort of degradation, the
amino group giving rise to ammonia.

Citric and Gluconic Acids.— The production of citric
and gluconic acids from sugars by fungi has already been
described (see p. 292). The processes are used on a com-
mercial scale in the United States, making use of shallow
pan fermentations, in which a felt of mycelium of A. niger,
P. citromyces or P. luteum develops on the surface of the
medium. Sometimes a continuous process is used in
which fresh sugar solution is fed under the established
mycelial felts as the fermented liquor is drawn off.
Sometimes fermentation in rotating drums, using sub-
merged cultures, has been found more efficacious.

For further reading : —

K. R. Butlin, " The Biochemical Activities of the Acetic Acid Bacteria."

D.S.I.R. Chemistry Research Special Report No. 2. H.M. Stationery

Office. London, 1936.
H. T. Herrick, et alia, "Industrial Fermentations." Ind. Eng. Chem.'

22 (1930), 1148.
A. G. Norman and W. H. Fuller, " Cellulose Decomposition by Micro-

organisms." Advances in Enzymology, 2 (1942), 239.
S. C. Prescott and C. G. Dunn, " Industrial Microbiology." McGraw-Hill

Book Company, Inc. New York, 1940.

CHAPTER XVIII
THE PROTEINS OF MICRO-ORGANISMS

THE proteins are essential constituents of all living
cells and are perhaps the most important as well as
the most complex substances synthesised by micro-
organisms. In spite of this, however, they can be built
up by certain organisms from the very simplest of starting
materials. For instance, B. aminovorans can thrive on
methylamine, in the absence of light, as its sole source of
carbon, nitrogen and energy, producing from it complex
proteins, carbohydrates and fats. The proteins, because
of their complexity, their colloidal nature and the lack
of any criterion of purity, are the most difficult of sub-
stances to study chemically, and we loiow comparatively
little of their internal make-up. We know that they are
built up of amino -acids joined together through the
carboxyl group of one and the amino group of the next
to give peptide linkages : —

R.CH.(NH2).COOH+NH2.CH.R > R.CH.(NH2).C0.NH.CH.R. +H0O.

COOH COOH

The process is repeated, an amino-acid joining on to the
dipeptide first formed to give a tripeptide, and this taking
up another amino-acid and so on until polypeptides,
protamines and, ultimately, proteins are formed. The
proteins have very high molecular weights ; according to
Svedberg's findings they ai'c multiples of about 34,000
or 35,000.

32S

THE PROTEINS OF MICRO-ORGANISMS 329

By hydrolysis proteins can be split up into their
constituent amino -acids and those which go to make up
any particular protein identified, but as to the arrange-
ment of the amino-acids in the protein molecule we are at
present largely ignorant. As a result of the study of the
proj)ortions of various amino-acids in proteins Bergmann
concludes that the total number of amino-acid residues,
Nt, in any protein can be expressed by the equation
Nt=2" X 3"' where 7^ and m are positive whole numbers.
The experimental values suggest that Nt=2^ X 3'^ =288
or whole number multiples of it. Since the numbers of
any particular amino-acid, such as glycine or alanine or
tryptophane, in the protein can also be expressed as
Ni=2'*' X T' it seems plausible to regard the amino-
acids as being arranged in a regular repeating pattern
in the peptide chain, the pattern being characteristic
of the protein. Thus in silk fibroin, for which Nt =
2^ X 3*, half the amino-acid residues consist of glycine,
Ng=2* X 3* ; that is each glycine unit is separated
from the next by another amino-acid. : —

— G — X — G — X — G — X —

The number of alanine residues is found to be Na = 2^ x 3"^
which means that every fourth residue is alanine : —

— G — X — G — A — G — X — G — A — G — X — G — A —

Tyrosine occurs in much smaller quantities, represented by
Nt=2^ X 3^, indicating that every sixteenth amino-acid
is tyrosine : —

— G — A — G — X — G — A — G — X — G — A— G — X — G —

— A — G — T — G — A — G — X — G — A — G— X — G — A —

— G— X— G— A— G— T— G—

8imih\rly for the other amino-acids which fit into thr^ii-
])laces in the ])epti(lo chaiii. which cojistitutos silk fi])i'()iji.

330 BACTERIOLOGICAL CHEMISTRY

in a rhythmic order depending on their proportion in
the molecule. Other proteins, although built up of the
same amino acids, differ from silk fibroin and one another
in having different proportions of the amino -acids and,
therefore, different periodicity or internal structure.

A certain amount of information is being obtained
as a result of studying the action of specific peptidases
which attack only peptide groups between particular
amino -acids. By this means it has been found possible
to determine whether or not certain pairs of amino-acids,
for instance, occur together in a given protein. Obviously,
considering the number of amino-acids available (about
twenty) and the number present in a protein, the number
of possible arrangements and, therefore, of possible
proteins is extremely large. It is to this great variety of
proteins that we owe many of the serological reactions
of bacteria, the reactions of antigens and antibodies
affording a means of detecting the subtle differences in
arrangement of the amino-acids which are at present
beyond the power of chemical methods.

As was mentioned in Chapter V, the proteins of
bacteria and the yeasts are in the main very like those of
plants and animals, containing the same amino-acids in
much the same proportions, and falling into the globulin
or albumin groups, as these are determined by solubility
properties .

An interesting recent development in connection with
the proteins of micro-organisms is the claim put forward
by Stanley that the virus of tobacco mosaic disease is a
crystalline protein. The protein can be isolated only from
diseased plants. Inoculation of healthy plants with as
little as 10~^^ g. of the crystals produces the disease and
gives rise to the production of large quantities of the
protein. It has a molecular weight, determined by
sedimentation in the ultracentrifuge and from its size
according to X-ray analysis, of about 17 million ; the
molecular weight of the normal proteins of the healthy

THE PROTEINS OF MICRO-ORGANISMS 331

plant does not exceed about 30,000. It is inactivated ]>y
treatment with hydrogen peroxide, formaldehyde, nitrous
acid or ultra-violet light, and can then no longer provoke
the disease nor call forth the production of further
protein ; the protein is not denatured, and the molecular
weight and crystalline form are not altered by this treat-
ment, nor is the serological behaviour with antisera
prepared against the active protein or the juice of infected
plants. Denaturation by acid, alkali, heat or oxidation
causes not only loss of activity but also loss of the other
characteristic properties of the protein. Covering of up
to 70 per cent, of the amino groups by acetylation or
conversion to the phenylureido group does not cause loss
of activity a) chough further treatment results in inactiva-
tion of the virus. Twenty to forty per cent, of the
phenolic groups of the tyrosine residues can also be
masked by acetylation without destroying the activity
of the virus. Inoculation of the acetyl- or phenylureido -
virus into tobacco plants gives rise to the disease and to
reproduction of normal virus and not acetylated virus.
A similar crystalline protein having the proj)erties of
the aucuba mosaic viiiis has also been isolated from
the juice of infected plants. It differs from the ordinary
mosaic virus protein in having larger crystals (0-03 mm.
long), an isoelectric point at pK 3-7 instead of 3-3, in
being considerably less soluble and in having a sedi-
mentation constant about 20 per cent, greater. The
virus particles are believed to be thread-like macro -

o o

molecules about 3000 A units long and 150 A units wide,
but their size varies with the treatment used during
isolation, suggesting that they are built up by polymerisa-
tion of smaller molecules of molecular weight about

o o

15,000 and 150 A units long and 15 A wide. It has been
suggested that the virus protein may be formed either by
polymerisation of the normal plant proteins or by direct
synthesis under the autocatalytic influence of the protein
itself.

332 BACTERIOLOGICAL CHEMISTRY

Tobacco mosaic vims has tlie composition of a
nucleoprotein of the yeast type, containing ribose,
guanine, cytosine, adenine and uridylic acid. The
protein portion contains 9-0 per cent, arginine, aspartic
acid, cysteine, glutamic acid, leucine, lysine, 6-7 per
cent, of phenylalanine, 4-7 per cent, of proline, 6-4 per
cent, of serine, 5-3 per cent, of threonine, 4-5 per cent, of
tryptophane and 3-9 per cent, of tyrosine. Alanine,
glycine and histidine appear to be absent. No lipoid
material could be detected.

Besides the proteins, the protoplasm of micro-organ-
isms contains the nucleoproteins and nucleins which
constitute the nuclear material of the cell. This may be
dispersed more or less uniformly throughout the cell
contents, as in most bacteria, or it may be collected in
granules as in the metachromatic or volutin granules of
the diphtheria bacillus, or in true nuclei as in the yeasts.

The nucleoproteins are soluble in dilute alkali, and are
precipitated from such solutions by dilute acids. They
constitue about 2 or 3 per cent, of the dry weight of
bacteria.

The nucleoproteins are complex molecules which break
down on hydrolysis to yield a basic protein — histone or
protamine — and nuclein. Nuclein on hydrolysis breaks
down further to yield another protein and a nucleic
acid. Nucleic acids on hydrolysis with cold alkali are
degraded into nucleotides. There are two main types of
nucleotide (a) those derived from yeasts and plants and
(6) those derived from animals. The yeast nucleic acids
are made up of guanylic acid,

CH C— N

adenylic acid, ! li %

CH, -' II li CH

II II /

N (3— N.R

THE rROTETNS OF :\rTrRO -OKfl ANT.SMS 333

c\*tidylic acid.

=C.NH,

CO CH

I I!

R.N CH

and iiridvlic acid,

XH CO

I I

CO CH

I II

R.N CH

The group R is ribose-3-phosphate,

-CH.CHOH.CH.CH.CHaOH
I OH

I /

0 - P = 0

\

OH

The animal nucleic acids (often called th\Tiionucleic acids
because they were first isolated from the thymus) are
built up of desoxyriboguanylic acid, desoxyriboadenylic
acid, desoxyribocytidylic acid and desoxyribothjTnosine

NH— CO
phosphoric acid, ^'<^ C.CH3 . Tn these nucleic acids

R.N CH

0

the group R is 2-desoxyribose, — cH.CHa.CHOH.CH.CHgOH,

carrying a phosphate group whose position of attachment
is still unknown.

If the nucleic acids are subjected to alkaline hydro-
lysis the phosphate group is split off and the corresponding
nucleosides are formed. The nucleosides are ribosides
or desoxyribosides of guanine, adenine, cytosine, uracil
or thymine (methyl uracil). It is seen, then, that the
nucleic acids consist of phosphoric acid, a pentose, two
purine bases, guanine and adenine, and two pyrimidine

334 BACTERIOLOGICAL CHEMISTRY

bases cytosine and either uracil or thymine. These facts
are summarised in the following scheme : —

Nucleoprotein

I

I
Protein + Nudein

Acid j hydrolysis

I
Protein + Nucleic acid (polytetranucleotides)
Cold alkaline | hydrolysis

I
Tetra-nucleotides
Alkaline | hydrolysis

Phosphoric acid + nucleosides

Acid I hydrolysis

From yeast nucleic acid From animal nucleic acid

Adenine \p^^^-j^p ^^ggg Adenine

Guanine J Guanine

Cytosine "1 ^^ . . ,. , Cytosine

Uracil )Pyrm^idme bases Thymine

Ribose Desoxyribose

The molecular weight of the ribonucleic acids from
various sources corresponds to the presence of from eight
to eighteen tetranucleotide units. Careful deamination
of the nucleic acids does not cause a lowering of the
molecular weight, suggesting that phospho -amide groups
are not involved in the linkage of the nucleotides. Elec-
trometric titration of the poly-tetranucleotides shows
that four acid dissociations are present for each tetra-
nucleotide, three of which correspond to primary phos-
phoric acid dissociations and one to a secondary phos-
phoric acid dissociation. The deaminated nucleic acids
show similar dissociation behaviour. Mild hydrolysis of
the polj^etranucleotides and the deaminated compounds
causes a lowering of the molecular weight with the
liberation of further secondary phosphoric acid groups.
It is suggested, on the basis of these facts, that the

THE PROTEINS OF MICRO-ORGANISMS 335

tetraiiucleotides are constituted as one of the following
three stnictiires : —

(a)

OH

/

Base— ribose— 0— P = 0

\ \

\ OH

0

\

\

Uracil — ribose — 0 — P = 0

\

\ \

0 0

(b)

\

^-^^ OH

\^

^~^-- /

Base— ribose — 0— P = 0

Base— ribose— 0—P = 0

\

\

OH

OH

(a)

or (a

OH

OH

/

/

Base— ribose— 0—P = 0

Base— ribose— 0 — P = 0

\ \

\ \

\ OH

\ OH

0

0

\

\

\

\

Base— ribose— 0—P = 0

Uracil— ribose— 0—P = 0

\

/

\ OH

/

0

0 (b)

\

/ OH

\

/ /

Uracil— ribose— 0—P = 0

Base— ribose— C—P = 0

/

\ \

/

\ OH

0 (b)

0

/ OH

\

/ /

\

Base— ribose— 0—P = 0

Base— ribose— 0—P = 0

\

\

OH

OH

Polymerisation of the tetranucleo tides to give the poly-
tetranucleotides or nucleic acids takes place through the
phosphate groups marked (a) and (b). The order in

38() BACTERIOLOGICAL CHEMISTRY

wliicli the ])ase8 giianiiie, adenine and cytosine are
arranged in the above formulae is still unknown. It is
considered probable that the phosphate groups are
attached at positions 2 and 3 in the ribose molecule in
the ribonucleic acids. In the animal, or desoxy ribonucleic
acids, the phosphate linkages are probably between
positions 3 and 5 of the desoxyribose, and uracil is replaced
by thymine.

The tubercle bacillus gives nucleic acids of the animal,
desoxyribose, type, yielding adenine, guanine, cytosine
and thymine on hydrolysis, whilst M. phlei, the timothy
grass bacillus, contains guanine, cytosine and uracil, but
no thymine, correlating it with the plant nucleic acids.
The nucleic acid of the diphtheria bacillus contains
adenine, guanine, C3^osine, uracil and thymine so that
it is either a mixture of plant and animal nucleic acids
or a new type. Other bacteria yield nucleic acids con-
taining bases of the purine type only ; that from B.
anthracis gives adenine and guanine and that from
Azotobacter chroococciiyn contains guanine, adenine and

NH— CO

I I

CH C ^NH

hypoxanthine, II II \ , which is derived

^^ II II CH '

il II /-

N C N

from adenine by deamination. The cells of streptococci
contain about 80 per cent, of protein and nucleoprotein,
of which nucleic acid constitutes 18 to 24 per cent, in
Smooth organisms and 14 to 17 per cent, in Rough cells.
The nucleic acid is a mixture of 10 to 30 per cent, desoxy-
ribose nucleic acid from nuclear material, and the
remainder ribonucleic acid from the cytoplasm. As was
mentioned on p. 332 the crystalline tobacco mosaic
virus consists of nucleoprotein of the plant or yeast type.
The psittacosis and vaccinia viruses, on the other hand,
are of the animal, desoxyribose type. Rough Type II

THE PROTEINS OF MICRO-ORGANISMS 337

pneiimucocci give lihoiuicleic acud to tiie exteJit of 2 to
6 per cent, of their dry weight. Staphylococci also give
ribonucleic acid but Esch. coli contains desox^a^bonucleic
acid.

It has been shown that the substance responsible for
the Gram -positive staining 7-eaction of CI. perfringens is
the magnesium salt of ribonucleic acid, which can be
removed by extraction with dilute aqueous bile salt
solutions leaving a Gram-negative cytoskeleton. The
extract itself is also Gram -negative but can recombine
with Gram -negative cell residue (but only if the latter is
in a reduced condition) to give a Gram -positive complex.
The ribonucleate will not combine with normally Gram-
negative organisms, nor can desoxyribonucleic acid,
nucleotides or nucleosides replace ribonucleic acid.
Sacch. cerevisice can be similarly extracted to yield
Gram -negative cells and nuclei. The extracted cells of
CI. perfringens and Sacch. cerevisice contain basic proteins
with a high proportion of arginine and it is suggested
that the Gram -positive material is a complex of reduced
basic protein and magnesium ribonucleate. Gram-
positive organisms often become Gram -negative in old
cultures and it has been found that Str. salivarius becomes
Gram -negative in media containing little magnesium or
in media so acid that magnesium ribonucleate cannot
exist.

It has very recently been shown that the substance
responsible for the long known conversion of Rough
pneumococci into vSmooth organisms of another type is a
desoxyribonucleic acid, characteristic of the type of
pneumococcus from which it is derived. Thus an extract
from Smooth Tjrpe III cells will cause conversion of any
Rough pneumococcus to Smooth Type III but not to
any other type ; Smooth Type II extract gives rise to
the corresponding Smooth type only. The distinguishing
feature of Smooth, compared with Rough, pneumococcus
organisms is a non-nitrogenous polysaccharide whose

338 BACTERIOLOGICAL CHEMISTRY

synthesis must be initiated l)y the desoxyribonucleic
acid. The conversion only occurs with actively growing
cultures and not in resting suspensions. The desoxyribo-
nucleic acid appears to be autocatalytic since the new
Smooth organisms, even after repeated subcultures,
contain it in much greater amount than that added to
stimulate the change. This is analogous to the propaga-
tion of the tobacco mosaic virus, when injected into the
plant.

The capsular substance of B. anthracis, apparently
identical with that of B. 7nesentericus, which reacts with
antisera in very high dilution, on hydrolysis with acid
loses it serological activity and forms d{ — ) -glutamic acid.
This is the first recorded natural occurrence of the Isevo-
rotatory isomer of glutamic acid. It is suggested that the
capsular substance has a polypeptide -like structure : —

CO.CH.NH2(CH2)2COOH

COOH NH

L

[.NH.— [CO.CH.(CH2)2COOH] ^.

COOH

or

COOH CO.(CH2)2CH.NH2.COOH.

CH.NH— [CO.(CH2)2.CH.NH.COOHJ ^

I
COOH

Protein Synthesis. — The synthesis of proteins by
micro-organisms has been most extensively studied in the
case of the yeasts. The yeast proteins, constituting
approximately 50 per cent, of the dry weight of the
organism, are valuable as a food since they contain all
the known amino-acids. Beer yeasts have been used as
a cattle fodder and even for human consumption, but

THE PROTEINS OF MICRO-ORGANISMS 339

they have the disadvantage of a bitter taste, due to acid
substances derived from the hops. The taste can be
eliminated by washing with dilute sodium carbonate.
Saccharomyces cerevisice does not grow very satisfactorily
on ammonium salts as the source of nitrogen but requires
expensive organic sources, so that its use for protein
production is not economically sound. During the 1914
to 1918 war the use of Torula utilis, also known as
Mineralhefe or Futterhefe, was developed in Germany.
It grows much more readily on inorganic media than does
Sacchxiromyces, has no bitter taste and is an effective
protein source for man and animals. It is usually grown
on molasses and ammonium salts under conditions of
good aeration. Yields of the organism, as dry weight, up
to 75 per cent, of the sugar consumed can be obtained.
A similar process has been worked out recently in this
country using a thermophilic variant of T . utilis which
gives an almost theoretical conversion of the nitrogen
supplied to protein. The protein content of the dried
yeast is 45 to 50 per cent. The dried product also con-
tains 20/xg. per gram of aneurin, 80 to 85 ju,g. per gram of
riboflavin and 400 to 450 /xg. per gram of nicotinic acid.
A fifteen-fold increase in the inoculum is obtained in
9 hours gro^\i:h on molasses wastes.

It has been shown that the amino-acid, alanine,
CH3CH.NH2.COOH, in yields up to 65 per cent., can be
synthesised from pyruvic acid and ammonium salts. As
a result Oesterlein and Knoop suggested that amino -acids
in general might be synthesised by the following route : —

/OH

R.CO.COOH + NH3 > R.C^COOH

^NH2

x4mmonia condenses with a keto-acid to form the hydroxy-
amino-acid, whicli loses water witli formation of an
imino-acid : —

340 BACTERIOLOGICAL CHEMISTRY

/OH

R.Ce— COOH > R.C.COOH + H,0

^NHa NH

The imino-acid then becomes reduced to give the amino -
acid : —

R.C.COOH + 2H > R.CH.COOH.

II I

NH NH2

The last step is one half of a coupled oxidation-reduction
process, the other half being the dehydrogenation of a
sugar breakdown product, in all probability the formation
of pyruvic acid from glyceraldehyde phosphate or from
methylglyoxal. If the group R is a methyl group the
above scheme illustrates the formation of alanine from
pyruvic acid. Aspartic acid has been shown to be
synthesised by many bacteria from fumaric acid and
ammonium salts : —

COOH.CH=CH.COOH + NH3 > COOH.CH.CHgCOOH.

I
NH2

The amino -acids formed in this way condense with one
another to give peptides : —

Rj Rj

R.CH.COOH + NH2— CH > R.CH.CO.NH.CH + HgO

I III

NH2 COOH NH2 COOH

and the process is repeated with formation of polypeptides
and ultimately of proteins.

The conversion of an a-keto-acid to an amino-acid
may proceed by two other routes instead of via the
imino-acid. The keto-acid might react with hydroxyl-
amine (which is a probable intermediate in nitrogen
fixation, see p. 223) to give the oxime which, on reduction,
would iiive an amino-acid :— -

THE PROTEINS OF 3IICR0 -ORUANISMS 341

-4H

CHaCU.CUUH -r- XHo.UH > CH3.C— COOH > CH3.CH.XH2.COOH + H^O

II
NOH

Alternatively the keto-acid ma}^ take up an amino
group by transamination from another amino-acid : —

CH3

CH3CO.CO.NH.CH.COOH + XHo.CH.COOH — ^

"I

(pyruvyl alanine) (a-amino-phenylacetic acid)

I I

(alanyl alanine)

It has been suggested that the biological synthesis of
peptide chains from non-amino-acid precursors may
result from successive acylation and amination : —

R.CO.COOH-

Reduction of imine

— ^ NH2

Reduction of oxime |
> R.CH.COOH

Transamination

- R'CO.CO.NH.CH.COOH

k

Repetition

R'CO.CO.NH.CH.COOH > R'CH.CO.NH.CH.COOH

I of 1st step I I

R XH2 R

Continued repetition gives polypeptides and finally
proteins .

Probably reactions of this type are only involved in
the synthesis of proteins from inorganic sources such as
in the growth of autotr()'j)liic bacteria or in the fixation

342 BACTERIOLOGICAL CHEMISTRY

of atmospheric nitrogen. When amino -acids are avail-
able, as ordinarily occurs when bacteria grow under normal
conditions in normal nutrient media it is very probable
that proteins are built up by the reversal of the action of
proteolytic enzymes. Bergmann has shown, for instance,
that papain will convert a mixture of benzoylglycine and
aniline into benzoylglycine anilide : —

CsHg.CO.NH.CHa.COOH + CeHs.NHa >

CeHg.CO.NH.CHa.CO.NH.CeHg 4- HaO

although under the same conditions it hydrolyses ben-
zoylglycine amide to benzoylglycine and ammonia : —
C6H5CO.NH.CH2.CONH2 + H2O > CeHs.CO.NH.CHa.COOH + NH3

Papain, in addition to the above reactions, also catalyses
conversion reactions ; for instance it converts a mixture
of benzoylglycine amide and aniline into benzoylglycine
anilide and ammonia : —

C6H5.CO.NH.CH2.CO.NH.CeH4 + NH3

Bergmann pictures the in vivo synthesis of proteins as
consisting of the action of a specific enzyme breaking
down, synthesising and rearranging a number of peptide
fragments until a protein is formed which is stable in
the presence of the enzyme . The particular protein formed
will depend on the fragments available and on the
specificity of the enzyme. In any one organism or cell
several enzymes may be present, resulting in the forma-
tion of a corresponding number of different proteins.

The proteinases are, in all probability, proteins
themselves or at least contain protein constituents.
It must be assumed, therefore, that there are proteins
which are capable of multiplication or autocatalysis.
Apart from the enzymes such autocatalytic proteins
are to be found in the crystalline viruses and in the
desoxyribonucleic acids responsible for the conversion of
Rough pneumococci to Smooth organisms (see p. 337).

'A i-ecently developed method wJiicJi promises to be

I

THE PROTEINS OF MICRO-ORGANISMS

343

of great value in the study of the synthetic mechanisms
of cells involves the use of mutants of the moulds Neuro-
spora crassa or N . sitophila. When asexual spores of the
moulds are treated with X-rays or ultra-violet light,
germinated and crossed with the heterothallic strain of
opposite sex, mutants arise which lack the ability to
bring about certain syntheses which the normal strains
can perform. Besides the strains which can no longer
synthesise some of the growi^h factors, others have been
produced which cannot make argim"ne, lysine, leucine,
valine, methionine, tryptophane, proline or threonine.
By using mutants in which particular stages in a synthesis
are blocked it becomes possible to trace the course of
synthetic processes. For instance, the cycle of formation
of arginine from ornithine via citrulliue : —

NH2

I

(CH2)3

CH.NH,

COOH

CO.NH,

I

NH

+CO2 + NH3

Gene 2
Gene 3

(CH2)3

I
CH.NH2

COOH

XH2

C=NH
+NH3 I
> NH

Gene 1

Prote

(CH,)3
CH.NHo

COOH
(Ornithine) (Citrullinc) (Arginine)

has been worked out using mutants lacking genes 1, 2
or 3 and thus incapable of carrying out the corresponding
stage in the synthesis. Similarly tryptophane has been
shown to arise by the following route : —

^ / ^— COOH ^

Gene 1 | | ^^ Gene 2

(Anthranilie acid)

I I -1- CH2OH.CH.NH2.COOH —

N
(Indole)

(Serine)
CH2CH.NH2.COOH

N
(Tryptojjhane)

23

344 BACTERIOLOGICAL CHEMISTRY

Strains which lack gene 1 will grow only if anthranilic
acid, indole or tryptophane is supplied in the medium.
Strains lacking gene 2 will grow if indole or tryptophane
is supplied. The lack of a particular gene is specifically
responsible for the absence of the corresponding enzyme.

For further information : —

G. W. Beadle, " Genetics and Metabolism in Neurospora.''' Physiol. Rev
25 (1945), 643.

M. Bergmann, " The Structure of Proteins in Relation to Biological Pro-
bkms." Che?n. Rev., 22 (1938), 423.

R. E. Buchanan and E. I. Fulmer, " The Physiology and Biochemistry
of Bacteria," Vol. I, Chapter III. Bailliere, Tindall & Cox. London,
1928.

W. M. Stanley, " The Biochemistry of Viruses." Ann. Rev. Biochemistry,
9 (1940), 545.

CHAPTER XIX

THE POLYSACCHARIDES OF MICRO-
ORGANISMS

THE production of polysaccharides by micro-organisms
is almost as widespread and universal as that of the
proteins. Polysaccharides of a more or less degree
of complexity are to be found in nearly all bacteria,
yeasts and fungi. Of the common polysaccharides,
starch, cellulose and glycogen are found as the result
of the synthetic activities of micro-organisms, and
besides these a considerable number of other poly-
saccharides characteristic of particular organisms is
known.

The chief poh^saccharide produced by yeasts is
glycogen, but others comprising the various " yeast
gums " have been described. Glycogen has also been
isolated from the higher fungi and from certain species of
Aspergillus. It has been claimed that it is also present in
certain bacteria, including Mycobacteriiim tiiherculosis,
CI. hutyricum and Shigella dysenterice.

As was mentioned in Chapter V the presence of cellu-
lose in micro-organisms has been established satisfactorily
only in the case of Acetobacter xyUnum, in which it was
detected as early as 1886 by Brown. It is sjoithesised by
the cell from a variety of sugars, for example glucose,
fructose, sucrose and pentoses, and even more readily
from such polyhydric alcohols, with three to seven carbon
atoms, as glycerol, erythritol, arabitol, dulcitol, sorbitol,

345

:]4G BACTERIOLOGICAL CHEMISTRY

mannitol and a- and j8-glucoheptitols, as Avas shown by
Hibbert and by Khouvine and their co-workers.

Starch, or a polysaccharide giving a blue colour with
iodine, has been recorded as occurring in a number of
bacteria and fungi. Colman in 1862 claimed that it was
to be found in the spore tubes of Ascomycetes , whilst it
is formed by A. niger growing on a synthetic medium.
The spores of P. glaucum contain a " spore-starch "
which gives a blue colour with iodine and yields glucose
on hydrolysis.

Many bacteria and fungi produce large yields of carbo-
hydrate " gums " which may be built up of one or more
of several sugar units. Thus Leuconostoc mesenteroides
forms dextran, a dextro-rotatory polysaccharide giving
glucose on hydrolysis, when grown on glucose and sucrose
but not when other sugars, such as melezitose, raffinose,
fructose, galactose, lactose, maltose, xylose or glycerol,
serve as the source of carbon. The polysaccharide,
[ajo + 180° and containing 0-5 per cent, of nitrogen, is
also formed by the action on sucrose of a cell-free, sterile
extract of the organism. The enzyme has no action on
glucose -1 -phosphate, and no inorganic phosphate is
liberated during its action on sucrose. Potato phos-
phorylase, which converts glucose- 1 -phosphate to starch,
has no effect on sucrose ; it is therefore concluded that
glucose -1 -phosphate is not an intermediate in the forma-
tion of dextran. This dextran has been shown to give
serological reactions with antisera prepared against
pneumococcus, Types II, XII and XX, but not Types I
and III, some Salmonella and Sir. salivarius, as well as
the homologous organism. As a result of the proportions
of di-, tri- and tetra-methyl methylglucosides formed by
hydrolysis of methylated dextran Hil^bert has suggested
that the polysaccharide has the branched chain struc-
ture : —

POLYSACCHARIDES OF MTCRO-ORO ANTS:MS ?A'

o W

/

"O

wo

o -

w'_

o

WO 4

wo

o

w

348 BACTERIOLOGICAL CHEMISTRY

in which the glucose units are united though the 1 : 6
positions.

Some strains of Str. salivarius and of Sir. bovis syn-
thesise a water insoluble dextran, [a]i) + 180°, from
sucrose or raffinose, as does Betabacterium verrniforme
(probably a Lactobcicillus) from sucrose but not from
other sugars. This dextran, too, has [a]o+180° when
dissolved in acid or alkali. It is made up of « -glucose
units linked through positions 1 : 6, that is, it has the
gentiobiose structure.

Phytomonas tumefaciens excretes into the medium a
polysaccharide which has [a]^^-9°, yielding only glu-
cose on hydrolysis. Its molecular weight, about 3600,
indicates that it is built up of 22 glucose units, probably
of pyranose form and joined by ^-linkages.

Leuconostoc dextranicum, when grown on sucrose,
gives a mixture of dextran and a fructosan, but when
grown on glucose it yields mainly the dextran and very
little fructosan.

Azotobacter chroococcum gives rise to a gum which
appears to consist mainly of an araban since it yields
arabinose on hydrolysis. The root nodule bacteria,
Bhizobium, also produce a gum which splits up into
glucuronic acid, CH0H-(CH0H)3-CH-C00H , and

glucose on hydrolysis. ! q ^

It gives cross reactions with antisera to Types III,
VI and XIV pneumococci, probably due to a common
cellobiiironic acid structure. The Rh. radicicolum poly-
saccharide has the structure : —

~ H OH CH2O H OH ~

— O— , / \ H H / 0\

OH H
H

-0— 1\ OH H

H \| 0/ N K H H

OH H OH OH

-0- /I !\ H

/ OH H \,

|\ H /l-O-

POLYSACCHARIDES OF MICRO-ORGANISMS

340

\\liicli vsliould be compared with that giv^en for Type III
pneumococciis on p. 356.

B. rnesentericus and B. suhtilis when grown on sucrose
or raffinose produce levan, a laevo rotatory polysaccharide,
[a]D-45-3°, built up of fructofuranose units linked
together through positions 2 and 6 : —

It is not formed when the organisms are growai on melezi-
tose, maltose, lactose, glucose, xylose or fructose. That
is, a terminal fructofuranose grouping, such as is present
in sucrose,

HCOH

I
0 HOCH

I
HCOH

CH

I
CHaOH

(glucose)

CH2OH

(fructofuranose)

350 BACTERIOLOGICAL CHEMISTRY

or ill raffiiiose,

CH2OH

(galactose)

but not in melezitose,

(glucose)

CH3OH
(fructofuranose)

CiljOH
(glucose)

nor in normal fructose,

(fructos.)

(glucose)

CH2OH— C.OH— HOCH— HCOH— HCOH— CH2

I 0 I

which has a fructopyranose structure, is necessary for
the formation of levan. The conversion of sucrose smd
raffinose into levan can also be accomplished by an
enzyme, levansucrase, which is secreted into the medium
by B. mesenteric us and B. suhtilis ; B. polymyxa and
Aerohacter levanicwn form an endocellular levansucrase
which has the same effect. Levansucrase has no action on
Neuberg's ester, Harden and Young's ester, methyl- y-
fructoside or inulin, all of which contain a terminal
fructofuranose group. Along with levan a reducing

POLYSACCHARIDR8 OF MICRO-ORGANISMS 351

sugar is formed, glucose from sucrose, and melibiose (not
galactose) from raffinose, showing that the latter is not
first hydrolysed to sucrose. No carbohydrate is des-
troyed in the conversion, showing that the process is
independent of respiration. Many strains of Str. saUvarius
synthesise levan from sucrose aii^ raffinose. B. lactis
produces a fructosan from sucrose only. In the sugar-
refining industry contamination with Leuconostoc and
B. mesentericiis causes considerable loss as the result
of the " viscous fermentations " to which they give rise.
B. lactis pituitosi gives rise to the formation of a galactan
which is excreted into the medium.

Mycodextran, [a]D-|-251°, a polyglucose giving no
coloration with iodine, is produced by the mould P.
expansum. A number of other glucose polysaccharides
of varying degrees of molecular complexity have l)een
described giving colours with iodine ranging from no
colour through red to blue and purple. Mycogalactan,
[a]D-f284°, yielding galactose on hydrolysis, is formed
by the growth of A. niger on a glucose medium. This
production of a polysaccharide built up of sugar units
different from those of the sugar on which the organism
grew is not an isolated case ; a number of such conversions
are now known. The formation of cellulose from a variety
of carbon sources by Acetohacter xylinum has already been
described. P. liiteum produces an acid polysaccharide,
luteic acid, when grown on glucose. Luteic acid is
hydrolysed by dilute mineral acid to yield one molecule
of malonic acid and two molecules of glucose. The
malonic acid is linked through one carboxyl group to a
neutral polysaccharide, luteose, from which it can be
removed by treatment with dilute alkali. Besides luteic
acid, however, the mould also produces other polysac-
charides built up of fructose, galactose or mannose, even
when it is grown on glucose as the sole source of carbon.
The converse of this phenomenon also occurs ; when
P. liitemn is grown on galactose, mannose, fructose,

352

BACTERIOLOGICAL CHEMISTRY

pentoses or glycerol it produces the same liiteic acid,
together with the other products, as it does when grown
on glucose. Some moulds produce polysaccharides
which are built up of more than one carbohydrate unit.
Thus P. chadesii gives a polysaccharide containing
glucose and galactose as well as a second polysaccharide,
mannocarolose, [a]587o+66°, which consists of eight or
nine units of f?-mannose linked together through the
1 : 6 230sitions : —

CHaOH

P. varians gives rise to an even more complex poly-
saccharide, varianose, [a]D+15°, which on hydrolysis
yields three sugars, galactose, glucose and either d-idose
or Z-altrose, in the proportions 6:1:1. They are linked
together with the glucose molecule at one end and the
galactose units in the middle : —

CH

I
CHjjOH

(glucose)

l-CHOH
CHOH

(P-galactoae;

((Mdose or /-altrose)

rOLYSACCHARTDES OF MICRO-ORGANISMS 353

8ume bacteria also produce mixed polysaccharides, the
pneumococcus, for instance, giving products containing
ghicose and galactose, the tubercle bacillus glucose,
arabinose and mannose, and the Vibrios glucose, galactose
and arabinose.

The most important polysaccharides from a bacterio-
logical point of view are the so-called " soluble specific
substances " which are responsible for the serological
behaviour of many organisms. They were first described
in the case of the pneumococcus, but since then have been
found to occur in a number of other bacteria of several
genera. It was shown by Dochez and Avery that filtrates
of cultures of pneumococcus contained a substance which
gave specific reactions with antisera prepared against
the same type of pneumococcus but not with antisera
prepared against other types. Later Zinsser and Parker
isolated " residual antigens " from alkaline extracts of
the organisms ; the residual antigens reacted with the
homologous sera, but gav^e no reactions for proteins and
were non-antigenic. Dochez and Avery's " soluble
specific substance " was heat stable, was precipitated
from aqueous solution by acetone or alcohol, was free from
protein, and shown to be of a polysaccharide nature.
It was accompanied by a nucleoprotein which was common
to all the types of pneumococcus, that from any one
type reacting with antisera prepared against any of the
other types ; it was antigenic and antisera prepared
against it gave no reaction with the soluble specific
polysaccharides.

The specific polysaccharides from each type of pneumo-
coccus are not only different in their serological behaviour
but, also, have been shown to be chemically different,
as may be seen from Table 20, where their main properties
are summarised.

354

BACTERIOLOaiCAL CHEMISTRY

Table 20

Acid

Acetyl

[a]D

Equiva-

Total

Amino-

(Amino

Products of

lent

X

Nitrogen

Group)

Hydrolysis

Per

cent.

Per

cent.

Per

ceut.

1 'iK'umococcus —

'J'yi^c I

+ 280°

G50

4-85

2-5

Amiuo-sugar. galac-
turonic acid.

II

+ 55°

950

0-2

Glucose, aldobionic
acid (glucose, glu-
curonic acid).

III

- 33°

340

0-1

Glucuronic acid, glu-
cose.

IV

+ 30°

1,500

5-5

0-1

5-5

Acetic acid, amino-
sugar, glucose.

„ VIII

+ 125°

750

0-2

Glucose, aldobionic
acid.

„ XIV

+ 12-5°

2-7

2-7

9-5

Acetic acid, glucos-
amine, galactose.

Carbohydrate-C

+ Gl-3°

1,050

5-9

1-14

3-7

Acetic acid, phos-

(species specific)

phoric acid, amino-
sugar.
Acetic acid, glucos-

Carbohydratc-P

+ 08-9°

5-0

0-99

(Forsinann)

amine, reducing
sugar, lipin, phos-
phoric acid.

Inactive carbo-

+ 10°

4,500

G-0

5-G

Acetic acid, glucos-

hydrate

amine.

Priedlander —

Typo A -

-100°

430

...

Glucose, aldobionic
acid, G5 per cent,
reducing sugar.

„ B -

+ 100°

G80

Glucose, aldobionic
acid, 75 per cent,
reducing sugar.

„ c -

+ 100°

085

Glucose, aldobionic
acid, 70 per cent,
reducing sugar.

POLYSACCHARIDES OF MICRO-ORGANISMS 355

Although the Type I pneumococcus polysaccharide
contained nitrogen it gave no protein reaction and, as it
was first isolated^ was not antigenic. Since then, however,
by avoiding the use of alkpJi in its extraction, it has been
obtained in an acetylated form, containing one acetyl
group for each glucose unit ; the acetyl polysaccharide
is antigenic, and on removal of the acetyl group yields a
non-antigenic polysaccharide identical with that originally
isolated. Some doubt is cast on this finding by later
work by Felton who could find no correlation between
the acetyl content and antigenicity. Half the nitrogen of
the Type I polysaccharide is in the form of amino -
nitrogen since it is eliminated by treatment with nitrous
acid, with production of reducing sugar and loss of
serological activity. The polysaccharides of Types II
and III pneumococci are not affected b}^ treatment with
nitrous acid. Type I polysaccharide is amphoteric,
acting as a strong acid and a weak base ; it has an
iso -electric point at about pH 4.

The Type II polysaccharide is a weak acid ; on acet}^-
ation it yields a serologically inactive product. Removal
of the acetyl groups restores the original activity of the
compound. The Type III polysaccharide is similar to
that of Type II, but is laevorotatory. On hydrolysis it
yields cellobiuronic acid, [a]D+10°, an aldobionic acid
having the structure 4-^-glucuronosidoglucose. In the
polysaccharide these units are linked through the 3-
carbon atom of the uronic acid to the reducing group of
the glucose in the next unit : —

356

BACTERIOLOGICAL CHEMISTRY

c

)

1 m

/\

/ w^

— o

w

o

o^

5-

x-S-

r3

\/

1

s 1

'3
o

o
1

^

s

W '

_o

/\

1

o-

^ o

w

o

w

\ «7

-8

y«

M

o

1 w

/\

w

o/ "«^

^w ^

o

_2

~^\

-h

to

\

O a
O o

POLYSACCHARIDES OF MICRO-ORGANISMS 357

The polysaccharide from Type XIV pneumococcus
resembles that of Type IV in not containing a uronic
acid group. It is constituted of one molecular proportion
of acetyl glucosamine and three molecular proportions
of galactose. It closely resembles the Blood group A
specific polysaccharide which occurs in group A red
blood corpuscles and which can be isolated from saliva,
gastric mucin, commercial pepsin and peptone. The
two substances are nob identical, since the blood group
A substance contains nitrogen in addition to that as
glucosamine, and the blood group A substance does not
give a cross precipitin reaction with antisera to Type XIV
pneumococci prepared in rabbits, although strong cross
reactions are found when horse antisera are used. Horse
antisera to Type XIV pneumococci agglutinate human
red cells of all groups.

Pneumococci resemble Shigella dyseiiterice and Sal-
monella schottmuUeri in containing the Forsmann hapten
which is capable of provoking lysins for sheep's red
blood corpuscles when injected into animals. In the
case of the pneumococci it is a lipo -polysaccharide
complex (carbohydrate F) associated with the bacterial
bodies. It is probably made up of the species specific
carbohydrate-C and a lipin fraction, chemically bound
to it. The lipin, constituting 6-5 per cent, of the poly-
saccharide, is devoid of nitrogen and phosphorus, has
m.p. 39-41 °C. and an acid equivalent 372 ; it is possibly
a C24 compound. Carbohydrate-C contains no lipin.

Enzymes have been found in various soil organisms,
for example Bhodobacillus palustris and a Myxococcus,
which hydrolyse the pneumococcal polysaccharides
specifically. Eh. palustris attacks only Type VIII
polysaccharide and not those of Types I, II or III.
Another soil organism gives an enz^Tne attacking Type III
polysaccharide only.

358 BACTERIOLOGICAL CHEMISTRY

Autolytic enzymes isolated from pneumococci them-
selves have been shown to attack the corresponding
polysaccharides and also those of Str. salivarius, of the
bovine vitreous humour and of the umbilical cord, all
of which contain an acetylglucosamine-glucuronic acid
complex, hyaluronic acid.

The enzyme, hyaluronidase, is also found in many
anaerobes, including CI. perf ring ens and those of the
gas gangrene group ; it is identical with or very closely
related with the so-called " spreading factor " of these
organisms .

Similar specific polysaccharides were isolated from
Friedlander's bacillus, Klebsiella pneumonice, Types A,
B and C. Those from Types B and C, although very
similar chemically (see Table 20), are quite distinct
serologically.

The tubercle bacilli afford very complex mixtures of
polysaccharides which have not yet been thoroughly
worked out. A polysaccharide having a rotation of
[a]D+67° was isolated, which on hydrolysis yielded 30
per cent, of c?-arabinose, together with glucose, galactose,
mannose and a sugar acid. It was also found to be present
in tuberculin. Mycobacterium phlei yields a polysac-
charide, precipitated by basic lead acetate, which gives
rise to arabinose, mannose and inositol on hydrolysis.
The avian tubercle bacillus produces a polysaccharide
which gives two molecules of mannose and one of inositol
on hydrolysis . On a synthetic medium the human tubercle
bacillus forms a polysaccharide, having a specific rotation
[a]i>+32°, which yields 19-7 per cent, of mannose and
10 per cent, of c?-arabinose. The lipoid fractions contain
glycerophosphoric acid, mannose, inositol, arabinose,
glucose, fructose, glucosamine and other unidentified
carbohydrates. The acetone soluble fat fraction of the

POLYSACCHARIDES OF MICRO-ORGANISMS

359

human tubercle bacillus and of 31. jjhlei contains fatty
acids linked to the non-reducing disaccharide, trehalose,

CH-

-CH-

HCOH I
HOCH O

HCOH

I
HC

HCOH

I
HOCH 0

I
HCOH

CH2OH

(glucose)

(glucose)

instead of to glycerol as in ordinary fats. The leprosy
bacillus also contains trehalose together with a dextro-
rotatory polysaccharide yielding pentoses on hydrolysis.
Heidelberger and Menzel have separated the polysac-
charide from the human tubercle bacillus into three
fractions having the properties outlined in Table 21 : —

Table 21

Acid

Products of

Type

Isolation

[ajD

Equiva-
lent

Nitrogen

Phosphorus

Hydrolysis

Per cent.

Per cent.

A

Precipitated by
Ba(0H)2

+ 81°

1,500

0-7

1-8

B

Soluble in 75 per
cent. methyl
alcohol

+ 30°

2,200

0-7-

1-0

61 per cent, reducing
sugar, d-arabinose.

C

Insoluble in 75
per cent, methyl
alcohol

+ 90°

6,700

0-1

0-2

87 per cent, reducing
sugar, arabinose,
mannose, magnesium
palmitate.

The polysaccharide C is common to avian, bovine and
human tubercle bacilli, whilst B, if present at all, occurs
in only very small amount in the avian and bovine types.
The removal of the magnesium palmitate from polysac-
charide C does not affect its serological behaviour (com-
pare the polysaccharide from Salmonella typhimurimn,
p. 361). An acetyl containing polysaccharide has been

360 BACTEE^IOLOGICAL CHEMISTRY

obtained from human tubercle bacilli which gives precipi-
tin reactions with the sera of tuberculous patients and
with antisera to the organism. The complex mixture of
polysaccharides from the bovine tubercle bacillus contains
much more inactive carbohydrate than does that from
human tubercle bacillus. An inactive carbohydrate,
precipitated by 80 per cent, acetic acid, is common to
both human and bovine bacilli ; another, which is soluble
in 96 per cent, acetic acid, occurs in bovine strains only.
The acetic acid soluble carbohydrate from human strains
is serologically active. The inactive carbohydrates from
bovine tubercle bacilli are strongly dextro-rotatory and
contain phosphorus but little or no pentose. The sero-
logically active polysaccharides contain t^-arabinose. The
wax fraction of the human tubercle bacillus (see p. 375)
contains fatty acids esterified by a specific polysaccharide
which gives precipitin reactions with anti -tubercle sera.
On hydrolysis it yields mannose, f?-arabinose and galac-
tose, with small amounts of inositol and glucosamine.
Two polysaccharides have been separated from the
attenuated tubercle organism, Bacille de Calmette-Guerin
(B.C.G.). Polysaccharide A, soluble in water, has
[«]D-f77*4°, and, on hydrolysis, gives 77 per cent, of
reducing sugar containing mannose, arabinose and
inositol, with 3 per cent, of an amino -sugar. The other
polysaccharide, insoluble in water but soluble in acids, is
a complex of about equal weights of a polysaccharide
(giving 95 per cent, of reducing sugars on hydrolysis) and
calcium phosphate.

Sal. typhimurium has been investigated by three
methods which have given almost identical results. If
smooth strains of the organism are extracted with dilute
trichloracetic acid a polysacchride is removed which can
be recovered by dialysis, concentration and precipitation
by acetone. It is toxic to mice, is antigenic and reacts
specifically with the corresponding antisenim. On treat-
ment with hot dilute acetic acid it yields four components

POLYSACCHARIDES OF MICR0-0RGANIS3IS 361

(a) 69 per cent, of a soluble specific polysaccharide,

[b] 16 per cent, of insoluble conjugated protein, (c) 3 to 4
per cent, of a benzene soluble lipin fraction and {d) 8 per
cent, of an alcohol soluble acetyl polysaccharide. The
conjugated protein is toxic but is neither antigenic nor
a hapten. The lipin fraction, insoluble in acetone,
appears to be a phosphatide. The main fraction, which
can be precipitated by alcohol, is a hapten, reacting with
antisera, but is neither antigenic nor toxic. It has
[«]d+103°, contains no nitrogen and on hydrolysis
yields 93 per cent, of reducing sugar, of which 31 per cent,
is glucose, 21-5 per cent, mannose and 19 per cent,
galactose. It contains no ketose, pentose or uronic acid.
The antigenic complex can be dissociated by precipitation
from weakly alkaline solution to give a small amount of an
amphoteric protein and a non-antigenic " undegraded "
polysaccharide which reacts specifically with Sal. typhi-
murkirn antisera. The complete antigen, possibly a
calcium salt of a phosphatide-polj^saccharide-protein
complex, occurs only in the smooth organisms ; the
rough variants contain the " residual antigen," which is
the complete antigen deprived of the phosphatide fraction.
The rough variants apparently contain an enzyme which
breaks down the complete antigen.

The second method of isolation depends on removal of
the protein of the organism by digestion with trypsin
and precipitation of the polysaccharide in the solution
with alcohol. The complete antigen so obtained behaves
in the same way as that extracted by trichloracetic acid.
These are the so-called " F 68 " polysaccharides since
they are precipitated by 68 per cent, of alcohol. The
third process is extraction of the dried organisms with
diethylene glycol, which gives products which may not
be degraded to such an extent as those obtained by the
more drastic methods, although their properties are
essentially the same.

The trichloracetic acid method has been applied to a

362 BACTERIOLOGICAL CHEMISTRY

number of organisms, including Escli. coli, Eherth, typhosa,
Sal. paratyphi, Sal. schottmillleri, Shigella dysenterice,
Proteus, Serratia marcescens, Ps. ceruginosa, B. ayithracis,
Phytomonas tumefaciens and V. comma, with similar
results.

The polysaccharide antigens isolated by these methods
are the somatic 0 -antigens of the smooth organisms.

The polysaccharides of a number of bacteria have
been investigated. Shigella dysenterice in the smooth
form produces a polysaccharide with a specific rotation
[a]D+98°, containing 1-6 per cent, of nitrogen. It has
a molecular weight about 5,100 and acid equivalent
about 9,000. It contains no protein, no pentoses and no
uronic acids. The nitrogen is present as an amino group,
which, however, is masked by acetylation (the acetyl
content is 5 per cent.). The composition and molecular
weight correspond to four hexose units, probably glucose,
and one acetamido -hexose unit, all repeated six times.
This polysaccharide also appears to be responsible for the
heterogenetic reaction between Shigella dysenterice antisera
and sheep red blood cells. The polysaccharide-protein-
phospholipin complex can be dissociated by treatment
with formamide into the non-antigenic phospho-lipin and
a polysaccharide-protein moiety which has the properties
of the somatic antigen of the smooth organisms. Treat-
ment of the complex with trypsin removes the protein
and leaves the feebly antigenic phospholipin-polysac-
charide. The free polysaccharide is a non-antigenic
hapten which gives precipitin reactions with antisera.
The polysaccharide-protein complex can be split by
solution in 90 per cent, phenol and dialysis to give the
polysaccharide hapten and the antigenic protein, which,
however, has lost the somatic specificity of the complex.
The complex can also be degraded by boiling with 1 per
cent, acetic acid, yielding an almost non-antigenic
protein whi(;li can he further dissociated by solution in
phenol, when a prosthetic group is probably removed.

POLYSACCHARIDES OF MICRO -ORC ANISMS 303

The polysaccharide and the protein can ])e recombined
by solution in formamide and precipitation with alcohol.
The conjugated protein gives rise to an antigen having
the properties of tlie original somatic antigen of Shigella
dysenterice, but the simple protein, when coupled with
the polysaccharide, gives a non-antigenic complex,
suggesting that the prosthetic group is essential for
antigenicity.

A similar somatic antigen complex has been extracted
from the 0 901 strain of Eberthella typhosa by the tri-
chloracetic acid, the trypsin and the diethylene glycol
extraction methods. It can be dissociated by boiling
1 per cent, a.cetic acid to give an ether soluble phospho-
lipin, a water soluble polysaccharide and an insoluble
protein. The purified polysaccharide, which has
[a]546i+1^8°, 1-2 per cent, of organic phosphorus and
less than 0-1 per cent, of nitrogen, is non-antigenic and
non-toxic to mice but gives a precipitin reaction with
typhoid 0-antisera. The purified protein, containing
11-5 per cent, of nitrogen, 0-47 per cent, of phosphorus
and having [o^]oier'55'^, i-'^ soluble in alkali but not in
aoid and a]3pears to be identical with that from Shiga's
bacillus. The two proteins can replace one another in
combination with the polysaccharide from either organism
to give antigens having specificity which is determined
by the polysaccharide. A similar protein has also been
obtained from Shigella parady sentence (Flexner 88), it
contains 10-7 per cent, of nitrogen and 1-1 per cent, of
phosphoms and has [a]546i-50°.

The virulent Vi strains of E. typhosa yield a similar
polysaccharide complex which reacts specifically witli
antisera to Vi organisms.

Different polysaccharides corresponding to the rough
(avirulent, Type B) and smooth (virulent, Tyj)e A)
variants of Staph, aureus are known. They are acid to
litmus, give no protein reactions but contain phosphorus

364

BACTERIOLOGICAL CHEMISTRY

and iiitrugen. Tlieir <-hief properties are given' in

Table 22 :—

Table

22

[^]d

Acid
Equivalent

Nitrogen

Phosphorus

Eeducing Sugar

Type A

+ 7°

770

Per cent.
4-1

Per cent.
6-3

25 per cent, glucose,
mannose (?).

„ B

+ 69°

80G

3-8

64

37 per cent, glucose.

The polysaccharide from the smooth organisms is precipi-
tated by barium hydroxide, whilst that from the rough
variant is not.

The streptococci afford both serologically active and
inactive polysaccharides. The former, obtained from
Str. salivarius by Lancefield, is a hapten, reacting with
homologous antisera but being non-antigenic. The
inactive polysaccharide, [a],)-73°, obtained from mucoid
hsemolytic streptococci of Lancefield's Group A, contains
no phosphorus, sulphur or amino -nitrogen. It contains
8-7 per cent, of nitrogen, 11 per cent, of acetyl group and
46 per cent, of uronic anhydrides ; it has an acid equiva-
lent of 380. It is an acetyl-glucosamine -glucuronic acid
complex, apparently identical with or closely related to
the hyaluronic acid in bovine vitreous humour, since it is
hydrolysed by the autolytic enzyme of the pneumococcus
(see p. 358). A group specific polysaccharide has been
extracted from the organisms with formamide and
contains 1-72 per cent, of nitrogen, 0-7 per cent, of
phosphorus and has [a]D-71-5°. It contains glucos-
amine and uronic acid residues. It reacts at a dilution of
1 in 2x10*^ with homologous antisera.

The Vibrios have been divided into six groups by
Linton on the basis of their content of proteins and poly-
saccharides. Those derived from cholera patients, Group
I, form a polysaccharide which on hydrolysis yields

POLYSACCHARIDES OF MICRO -ORCIANISMS 365

galactose and an al(l()1)i()nic acid composed of galactose
and glucuronic acid. The water vibrios produce a poly-
saccharide containing arabinose linked with the same
aldobionic acid. The Inaba variant of V. comma, Group
VI, gives a polysaccharide having [a]n-f58°, and con-
taining 2-6 per cent, of nitrogen, which yields 58 per cent,
of reducing sugar on hydrolysis. The sugar is glucose
only, no galactose or arabinose being present. The
polysaccharides present in rough strains, and which also
occur in the corresponding smooth strains, are stable to
alkali, whilst the polysaccharides peculiar to the smooth
variants are not stable to alkali.

The gonococcus and meningococcus contain non-
antigenic and non-toxic polysaccharides which both react
with antisera to Type III pneumococcus. That from
the meningococcus is probably the sodium salt of an
acid polysaccharide. The gonococcus gives two poly-
saccharides corresponding to two serological types.

Both capsulated and non-caps ulated forms of B.
anthracis yield a non-toxic, non-antigenic polysaccharide,
containing 0-8 per cent, of nitrogen, which gives 60 per
cent, of glucose on hydrolysis together with pentoses and
a uronic acid. Another polysaccharide from virulent and
avirulent B. anthracis has been described. It gives
equimolecular proportions of galactose and of acetylated
fZ-glucosamine on hydrolysis (corresponding to 68 per
cent, of the pure polysaccharide) but no pentose and no
uronic acid. This polysaccharide is antigenic. The
capsules, unlike those of the pneumococcus, are not
polysaccharide in nature but contain peptides (see p. 338).

Proteus yields two polysaccharides, one of which,
stable to hot alkali, appears to be the common antigenic
factor between Proteus XI 9 and Rickettsia.

Micrococcus lysodeikticus yields a polysaccharide of
high molecular weight which is the specific substrate of
the enzyme lysozyme, which splits it into an N-acetyl-
aminohexose and a ketohexose. The polysaccharide can

366 BACTERIOLOGICAL CHEMISTRY

be extracted by cold hypochlorite solution (aiitiformin),
diethylene glycol or hot formamide, and precipitated
with alcohol or acetone.

Hapten polysaccharides have also l)een isolated from
Brucella abortus and other Brucella species, from CI.
perfringens, C. diphthericB and the diphtheroid bacilli,
H. pertussis, H. parapertussis, from the capsules of
H. influenzoe, from a number of Sahnonella, from certain
Pasteur ella and from Leptospira hiflexa.

Carbohydrate Synthesis. — We know practically nothing
of the mechanism of the synthesis of this great variety of
complex polysaccharides. Kluyver has suggested that
their synthesis involves coupled oxidation-reduction pro-
cesses, as do the syntheses of fats and proteins, and
that it may follow similar lines to the resynthesis of
glycogen from lactic acid in muscle during the recovery
period : —

(1) 3CH3CHOH.COOH + 30 — > 3CH3CO.COOH + 3H2O.

/OH

(3) 3ch/jh0 + 3h.0 > 3ch3c^oh

\h

:h\ /O :h: 0

(4) 3 H — ^C— C ^OH + 30 > 3CH2— CHOH + 3H2O

H^ ^H

0

■/ \

(5) 3CH2— CHOH ^=^ CH2OH.CHO (glycolaklehyde).

(G) 3CH2OH.CHO > CgHiaOe > {Ce-R^oO.h (glycogen).

The lactic acid is oxidised to pyruvic acid, which is
decarboxylated to give acetaldeliyde and carbon dioxide.
The acetaldeliyde is oxidised via the hydrate to give

POLYSACCHARIDES OF MICRO-ORGANISMS 367

gl^culaldehyde, which coiideiuses to give glucose aiid
finally glycogen. In summary, three molecules of lactic
acid give one molecide of glucose and three molecules of
carbon dioxide and water : —

3CH3CHOH.COOH + 30^ ^ 3CO2 + 3H2O + CeHioOg.

Or we may regard the process as being the oxidation of
one-third of the lactic acid to provide the energy for the
synthesis of the other two -thirds to glucose and
glycogen :—

CH3CHOH.COOH + 3O2 > 3CO2 + 3H2O (energy).

2CH3CHOH.COOH > CeHiaOg (synthesis).

Hanes showed that starch was synthesised from glu-
cose-1 -phosphate under the influence of an enzyme,
phosphorylase, present in potatoes, inorganic phosphate
being liberated during the process. It has been suggested
as a possible general mechanism that glucose is phos-
phorylated at the expense of adenosine triphosphate, as
in yeast fermentation, and that some of the ester may be
acted on by enzymes other than the normal fermentation
enzymes to give rise to polysaccharides : —

Glucose + adenosine triphosphate

Hexokinase (in yeast, animal

T tissues)

Isomerase | '

Fructose-G-phosphate ^=1:1^=:===^ glucose-G-phosphate + adenosine

30% diphosphate

1j Q-Q, Phosphoglucomutase

\l '° (yeast, animal tissues
Glucose - 1 -phosphate

'jl „m^, Phosphorylase (yeast
I '° potatoes, muscle and
kidney)
Polysaccharide + phosphate

The necessary energy for the process is held to be
derived from the conversion of adenosine triphosphate to
the di -phosphate. The latter is re-esterified to adenosine -
riphosphate by the phosphate set free in the final stage

368 BACTERIOLOGICAL CHEMISTRY

ot" syiitliosis. An alteriiativo sou r(^e of ghicoso-Ophosphatc
may be from phosphopyruvic acid by reversal of the
reaction of alcoholic fermentation according to the
Embden-Meyerhof scheme. Pyruvic acid is also formed
in the fermentation of lactic acid and in most, if not all,
bacterial fermentations . It will be remembered, however,
that glucose- 1 -phosphate is not an intermediate in the
conversion of sucrose to dextran by enzymes of
Leuconostoc mesenteroides, and that potato phosphorylase
is not effective in that synthesis (see p. 346).

The synthesis of carbohydrates and polysaccharides
may also occur by way of the condensation of aldehydes
with the intervention of adenosine phosphates as was
suggested by Ruben for the autotrophic bacteria (see
p. 78).

In the early stages of dissimilation of glucose by
yeasts and by Esch. coli potassium and glucose disappear
from the medium and a fermentable non-reducing poly-
saccharide is formed in equimolecular proportions. The
disappearance of glucose is more rapid than is accounted
for by the products of fermentation formed. It appears
that the potassium is concerned in the synthesis, but the
mechanism is unknown.

For further reading : —

R. E. Buchanan and E. I. Fulmer, " Physiology and Biochemistry of
Bacteria," Vol. I, Chapter III. BalJiere, Tindall & Cox. Londor,
1928.

E. Mikulaszek, " Bakterielle Polysaccharide." Ergebnisse der Hygien e

17 (1935), 415.
R. W. Linton, " Chemistry and Serology of the Vibrios." Bad. Rev., 4

(1940), 261.

CHAPTER XX
THE LIPOIDS OF MICRO-ORGANISMS

THE term " lipoid " (lipide and lipin are sjTionymoiis)
is employed as a general name for all the fat-like
substances which are soluble in the " fat solvents "
ether, alcohol, acetone, chloroform and light petroleum.
Thus the f?«ts, waxes, higher alcohols, sterols and phos-
phatides, together with certain of their degradation
products, such as fatty acids and glycerol, are all lipoids.
Fats. — The fats are all glycerides, that is, esters of
fatty acids with glycerol. When it is remembered that
glycerol, CH2OH.CHOH.CH2OH, contains three hydroxyl
groups capable of esterification and one, two or three
of them may be involved, that the acids attached may
be all the same or all different and may be chosen from
a large number, both saturated and imsaturated, it is
obvious that a great variety of fats is possible. Usually
fats occur together as complex mixtures which are
extremely difficult to separate, and their analysis is
restricted to the identification of the fatty acids present
and the determination of the ratio of saturated to un-
saturated fatty acids. The fats of micro-organisms
usually contain a high proportion of unsaturated fatty
acids and are consequently liquid at ordinary temperatures
or have low melting points. Palmitic, C16H32O2, stearic,
CigHgeOg, and oleic acids, C18H34O2, are the chief acids
found in the fats of micro-organisms as they are in animal
fats, but butyric, C4H8O2, caproic, C6H12O2, lauric,
C12H24O2, dihydroxystearic, C18H36O4, linoleic, CigHggOa,
linolenic, C18H30O2, tuberculostearic, C19H38O2, arachidic,

369

370

BACTERIOLOGICAL CHEMISTRY

020^140^2' (^'t^rotic, C26H52O2, phthioic, ('26^5202, myiistic,
C30H60O2, and isocetinic acids have also been reported
as occurring in the fats of bacteria.

Tlie acid-fast ])acteria contain the following fatty
acids in the acetone soluble fat fraction (Table 23) : —

Table 23
( After R. J . Ayiderson)

Bacillus

Human
Tubercle

Bovine
Tubercle

Timothy
Grass

Butyric acid, C4H8O2 -
Palmitic acid, CigHgaOa
Stearic acid, CigHggOa
Cerotic acid, CagHgaOg
Linoleic acid, CigHgaOa
Linolenic acidjCigHgoOa
Tuberculostearic acid,

CigHggOg

Phthioic acid, Q^^Yi^^O^

Trace.
Large amount
Small amount

Trace
Small amount
Small amount
Large amount

Large amount

Trace
Large amount

None
Small amount
Small amount
Small amount
Large amount

Large amount

Trace
Large amount

None

None
Small amount
Small amount
Large amount

Large amount

Tuberculostearic acid is a liquid, saturated, fatty acid,

CH3
very probably 10-methylstearic acid, |

CH3(CH2)7CH.
(CH2)8COOH. It is optically inactive and has m.p.
10-1 1°C. Phthioic acid is also a liquid saturated fatty
acid with a branched chain having m.p. 20°C. and [a]^D-f-
12-6°. From a study of X-ray diffraction and the areas
of monomolecular films it is likely that phthioic acid is
constituted as

CH3(CH2)j/-
CH3(CH,)2,

\

-C.COOH

THE LIPOIDS OF MICRO-ORGANISMS 371

where x and y are 9 to 12 and z is equal to 0 or 1. The
most probable structure is ethyl-n-decyl-zz-dodecjd acetic
acid.

The fats of the acid-fast bacilli are peculiar in that the
fatty acids are linked to the disaccharide trehalose, in
the case of the human tubercle bacilkis and M. leprce,
and to an unidentified substance in the case of bovine
and avian tubercle bacilli and not to glycerol.

The fat of the diphtheria bacillus consists mainly of
free fatty acids, of which the solid saturated fatty acid,
constituting about one -third, is exclusively palmitic acid ;
the chief liquid unsaturated fatt}^ acid is A^-hexadecenoic
acid, CH3(CH2)5CH=CH(CH2),COOH ; 1 per cent, of the
fatty acids consists of an unsaturated acid, C14H26O2 ;
a higher unsaturated fatty acid, diphtheric acid, CgsHggOs,
m.p. 35°, was also isolated.

Lactobacillus acidophilus yields lauric, myristic, pal-
mitic, stearic and oleic acids, together with a dihydroxy-
stearic acid having an optical rotation [a]D + 7-8°.

In yeasts butyric, caproic, lauric, palmitic, stearic,
oleic, linoleic, linolenic, arachidic, myristic and isocetinic
acids have been found. The yeasts normally contain about
18 per cent, of fat, but End. vernalis and Oospora [Oidium]
lactis can produce up to 50 per cent., whilst Toriila
lipofera has been shown to give as much as 60 per cent,
of its dry weight as fat. The yeast fats usually contain a
high proportion of unsaturated fatty acids, oleic and
linoleic acids, and are accordingly usually liquid, closely
resembling olive oil.

The mould fats contain palmitic, stearic, tetracosoic,
C24H48O2, oleic, linoleic and linolenic acids, in the pro-
portion of approximately one-third saturated acids and
two -thirds unsaturated acids.

Waxes, Sterols and Higher Alcohols. — The waxes are
esters of fatty acids with higher monohydric alcohols
instead of with glycerol as in the fats. The sterols are
unsaturated alcohols having a condensed ring structure

372 BACTERIOLOGICAL CHEMISTRY

of high molecular weight ; cholesterol has the structure :-

CH,

CH,

-CH,

.CH.{CH2)3CH

CH,

CH,

CH3 CH3 CHg

1 I /

CH.CH^CH.CH.CH

"^CH,

/

CH,

\/
and ergosterol,

CH,

HO

In view of their exceptionally high lipoid content
the acid-fast bacteria have naturally been most closely
examined in this respect. As early as 1898 a wax re-
sembling beeswax had been isolated from the tubercle
bacillus, and in 1904 a higher alcohol was isolated. In
1914 Tamura isolated an alcohol, C29H55OH, m.p. 66° C,
which he named mykol. He showed that it stained
Gram positive and had the property of acid-fastness ; he
ascribed these properties of the tubercle bacillus to mykol.
Other alcohols, including phytoglycol, C26H54O2, have
also been isolated from the tubercle bacillus, together with
a wax shown to be an ester of mykol with lauric acid.
R.J. Anderson and his co-workers have shown that most
of the " waxes " of the acid-fast bacteria consist of

THE LIPOIDS OF MICRO-ORGANISMS 373

optically active hydroxy-acids esterified with carbo-
hydrates together with smaller amounts of true waxes
which are esters with higher secondary alcohols. Phthio-
cerol, a crystalline alcohol, m.p. 73°, [a]D-4-8° (in
chloroform), C35H72O3, containing two hydroxyl groups
and one methoxy group, is found in all the wax fractions
of human and bovine tubercle bacilli but not in those of
other acid-fast bacteria. Avian tubercle bacilli, the
Timothy grass bacillus and the leprosy bacillus contain
the secondary alcohols fZ-2-eicosanol, CH3.(CH2)i7.CHOH.
CH3, m.p. 62°, [«]d+4-2°, and f/-2-octadecanol,
CH3.(CH2)i5.CHOH.CH3, m.p. 56°, \a]u-^5-l\ The wax
of the human tubercle bacillus contains the specific
polysaccharide, to which fatty acids are attached ;
avian tubercle bacilli, and the Timothy grass bacillus
contain trehalose, whilst the leprosy bacillus gives only
glycerol as the water soluble product of hydrolysis.

The properties of the various waxes are summarised
in Table 24.

374

BACTERIOLOGICAL CHEMISTRY

^P . CD

P c« O

o

§ g 2 ^ ^ .^

"* "^ " "'§'0

aJ >,

o S o o

W O -23 "s '^H o T3

1^ S?

^O'S -I

"+6 6l|o;|^5?'gfJw:

^ ^, ^ j. -3

^■2 9 2 ^

' tua.2,

CO +i !-( -2 -2 r^

CO O.'C ■^ r^ ^

S-*?!

^ t« e)

^5-"^

'-? ^ ^ 2 'EZ^ ^ ^
^|+6h&p^""

S b .
q do.

i^ -d e?;

a

73 o

sa

b^

:p?s':

o o i;3 -g ;=! .y

o o o<^

ill

„ „ "2 ^ -

"^-^l^^lll 1^

o t>>

o >>

o >>

^a5§e^M

THE LIPOIDS OF MICRO-ORGANISMS 375

Mycolic acid of Imman and bovine tubercle bacilli
m.p. 54-56°, [ajo + 1*8°, C88H176O4, is a saturated acid
containing one hydroxyl group, one methoxy group and
one carboxyl group. On vacuum distillation at 280° C. it
splits to give ?i-hexacosanoic acid, C26H52O2, ni.p. 87-88°
and a colorless non-volatile residue. It is the acid-fast
staining substance of both organisms. Avian tubercle
bacilli give two mycolic acids a- and j3-, both of which are
acid fast. On pyrolysis at 210° a-mycolic acid, m.p.
69-70°C, [ajc + 5-6°, mol. wi:. 500, gives 25-4 per cent,
of a branched chain crystalline pentacosanoic acid,
m.p. 78-79°C. and jS-mycolic acid, m.p. 60-61°C.,
[ajo -f 5-5°, mol. wt. 1300, at 280° C, gives n-tetracosanoic
acid, m.p. 83°C. Phleimycolic acid, from the wax of the
Timothy grass bacillus, is a mixture of a saturated acid
and an unsaturated, dibasic, hydroxy-acid, CToHisgOe,
which has m.p. 56-57° and [ajo -f 6-1°. Its methyl ester,
on vacuum distillation, gives the volatile methyl ester of
a branched chain tetracosanoic acid and a neutral non-
volatile residue. The optically active, dibasic, hydroxy-
acid, leprosinic acid, from the leprosj^ bacillus has m.p.
62-63°, [ajc + 4° and has the formula CgsHi^eOg.

The specific polysaccharide of the human tubercle
bacillus, which occurs in the purified wax and which
gives precipitin reactions with homologous antiserum,
contains nitrogen and phosphorus and, on acid hydrolysis,
yields 2 per cent, of mannose, 36 per cent, of c?-arabinose,
17-5 per cent, of galactose and traces of inosito, and
glucosamine. The carbohydrate from the bovine o rgan-
isms, containing 2-2 per cent, of phosphorus and traces
of nitrogen, gives mannose, inositol and inositol mono-
phosphate on hydrolysis.

The acid-fast bacteria contain lipoids which can only
be removed by extraction after treatment of the cells
with 1 per cent, alcoholic hydrochloric acid. These

376

BACTERIOLOGICAL CHEMISTRY

" firmly bound " lipoids can be precipitated from ether
solution by alcohol or acetone. By filtration through a
Chamberland candle the extracts from human and avian
tubercle bacilli can be separated into filterable and
unfilterable fractions. The extract from the leprosy
bacillus is all filterable. Their composition is shown in
Table 25 :—

Table 25
( After B. J. Anderson)

Unfilterable

Filterable

Human

Avian

Human

Avian

Leprosy

Per cent.

Per cent.

Per cent.

Per cent.

Per cent.

Firmly bound lipoid -

4-7

2-68

7-5

8-16

19-5

Polysaccharide -

50-5

31-3

25-5

150

40-5

Glycerol

none

none

2-0

none

none

Hydroxy-acids -

510

52-8

411

61-0

56-3

Lower fatty acids

40

3-8

28-4

3-7

4-3

Neutral material

none

10-6

0-8

8-2

5-5

The unfilterable fractions contain twice as much
polysaccharide as the filterable fractions. The poly-
saccharides are similar to that of the human purified
wax and give mannose, c?-arabinose and galactose on
hydrolysis (that from the leprosy bacillus contains no
mannose and only 1 per cent, of galactose, being almost
entirely composed of c?-arabinose, with a small amount
of an unidentified pentose). The polysaccharides all give
about 50 per cent, of reducing sugar on hydrolysis.
The non-reducing portion has not yet been identified.
They give precipitin reactions with the homologous
antisera.

Tlie lower fatty acid fraction from the filterable
fraction contained tuberculostearic acid but no phthioic
acid. The hydroxy-acid from both human tubercle

THE LIPOIDS OF MICRO-ORGANISMS 377

fractions is mycolic acid ; but that from the avian
bacilli is different from the wax mycolic acids ; it is
called y-mycolic acid and on vacuum distillation gave
18 per cent, of a branched chain tetracosanoic acid,
C24H48O2. The hydroxy-acid from the leprosy bacillus is
leprosinic acid, which also gives a branched chain tetra-
cosanoic acid on pyrolysis. The unsaponifiable material
consists of r/-2-eicosanol and 6?-2-octadecanol.

The chloroform soluble wax from the attenuated
strain of tubercle bacillus, B.C.G., is a complex mixture
giving a pentose polysaccharide, palmitic and cerotic
acids and an acid-fast Avax, C51H102OH.COOH, on
hydrolysis. It has been claimed that the wax content
of the ether extracts of the various acid-fast bacteria is
characteristic of the different types ; thus the lipoids of
human and bovine tubercle bacillus contain 60 to 70
per cent, of wax, those of avian and cold-blooded tubercle
bacilli, the leprosy bacillus and M. pJilei contain 27 to
30 per cent., whilst the smegma bacillus and dung bacilli
contain only 4 to 10 per cent. Waxes have also been
obtained from the diphtheria bacillus, but have not
been investigated chemically.

The evidence for the presence of sterols in bacteria
is somewhat conflicting ; it has been claimed that
Azotobacter chroococcum contains ergosterol to the extent
of 0-1 per cent, of the fat fraction, and that sterols occur
in human and bovine tubercle bacilli, in the B.C.G. strain,
in 31. jjhlei and in Esch. coli. Cholesterol has been
reportecl as occurring in L. acidophilus. The majority
of reports, however, state that sterols are absent from
bacteria. On the other hand, yeasts contain large amounts
of ergosterol, up to 20 per cent, of the fat being quite
common. Yeast is the chief commercial source of ergo-
sterol, which is used in the mamrfacture of synthetic
vitaniiu-T) or calciferol. Ergosterol also appears to be

378 BACTERIOLOGICAL CHEMISTRY

a common constituent of many moulds, having been found
in P. javanicum, P. puheridum,, P. aurantio-brunneum,
A.fischeri, A. oryzce, A. sydowi, A. niger and Rhizopus
jajyonicus in amounts varying from 0-1 to 0-4 per cent,
of the dry weight of the mycehum. It has been found
as ergosteryl palmitate in P. brevi-co7npactum, P. italicurn
and P. atirantio-griseum in amounts between 0-02 and
0-5 per cent, of the dry mycelium.

The mycelium of Aspergillus sydowi contains fungus

C1BH31.CHOH.CH.CHOH.CH2.CH2.OH

cerebrin, I , m.p.

NH.CO.CHOH.C24H49

143°C., [ajo + 11*9° (in pyridine), which is identical
with that found in yeasts and in mushrooms.

Phosphatides. — The phosphatides yield fatty acids,
glycerol or carbohydrates, phosphoric acid and choline or
other nitrogenous bases on hydrolysis. According to the
ratio of nitrogen to phosphorus in the molecule, they are
classified as monoamino-monophosphatides (1:1), dia-
mino-monophosphatides (2:1) and mono amino -diphospha-
tides (1:2). The phosphatides are soluble in ether and are
precipitated from such solution by acetone, by which
means they can be separated from the fats which are
soluble in acetone. The commonest phosphatide is
lecithin, having a nitrogen to phosphorus ratio of 1:1.
It is very probably built up of glycerol esterified with one
molecule each of stearic, oleic and phosphoric acids with
a molecule of choline linked on to the phosphoric acid
group, as represented in the formula : —

CH2O.CO.C17H36 (stearic acid)
CHO.CO.C17H33 (oleic acid)

CH20.P.(OH).O.C2H,

II I

O N(CH3)3 (chulinc)

I
OH

THE LIPOIDS OF MICRO-ORGANISMS 379

The phosphatides are widely distributed in inicio
organisms and are almost certainly present to more or
less extent in all of them. The non-acid-fast organisms
contain about 0-5 to 2 per cent., whilst the tubercle
bacillus contains about 6-5 per cent, of phosphatide.

The phosphatides of the tubercle bacillus on hydi'olysis
yield palmitic and oleic acids and the two liquid saturated
acids, tuberculostearic acid, CigHggOg, and phthioic acid,
C26H52O2. The leprosy bacillus phosphatide contains an
unsaturated Cie fatty acid. The B.C.G. strain yields also
a phosphorylated polysaccharide, giving mannose on
hydrolysis. The phosphatides of the acid-fast bacteria
contain very little nitrogen (see Table 26) and no choline
or aminoethanol. The nitrogen ai:)pears to be in the
form of ammonia.

Phthioic acid appears to be responsible for the forma-
tion of the tubercles which are a characteristic of tuber-
culosis, since injection of the acid or of various fractions
containing it gives rise to their production in experi-
mental animals. Tuberculostearic acid is irritant but
does not cause tubercle formation.

The lipoids of the acid-fast bacteria contain, besides
the phosphatides and waxes, substances akin to the
cerebrosides which, on hydrolysis, yield glycerol, fatty
acids and a carbohydrate. The carbohydrates have been
identified as mannose, glucose and arabinose. The cyclic
hexahydric alcohol, inositol, has also been found among
the hydrolysis products of this fraction of the acid-fast
bacilli. A comparison of the composition of the phospha-
tides of some acid-fast bacilli is given in Table 26 : —

?>80

raoteriological chemistry

Table 20
( After U. J . Anderson)

Bacillus

Timothy

Human

Avian

Bovine

Grass

Leprosy

Melting point - . -

210« C.

210" C.

208» C.

1900 c.

23PC.

Per

Per

Per

Per

Per

cent.

cent.

cent.

cent.

cent.

Phosphorus - - - -

2-30

2-18

1-87

2-80

1-75

Nitrogen ....

0-36

0-48

1-00

0-22

trace

Total ether-soluble fraction -

66 to 67

55 to 56

57 to 58

600

62-2

Palmitic acid

30-5

18-4

27-0

200

18-6

Oleic acid . . . .

12-8

18-4

7-0

5-6

13-8

Liquid saturated fattv acids

20-9

141

16-0

18-0

13-5

(tuberculostearic and

phthioic)

Total fatty acids recovered

64-2

53-7

500

43-6

45-9

Water-soluble constituents -

33 to 34

46 to 47

43 to 44

40-0

38-0

Mannose . . . .

9-2

13-3

6-7

9-5

5-2

Inositol ....

8-9

30

3-5

2-2

0-6

Other sugars

12-3

— .

—

—

20-6

Glycerophosphoric acid

5-4

60

9-9

100

~

The distribution of the various lipoid fractions of the
acid-fast bacteria is summarised in Table 27.

Table 27
{After R.J. Anderson)

Bacillus

Timothy

Human

Bovine

Avian

Grass

Per

Per

Per

Per

cent.

cent.

cent.

cent.

Phosphatide ....

6-54

1-53

2-26

0-59

Acetone -soluble fat

6-20

3-34

2-19

2-75

Chloroform-soluble Avax

1103

8-52

10-79

4-98

Total lipoids ....

23-78

13-40

15-26

8-37

Polysaccharides ....

0-87

1-09

1-02

3-90

Bacterial residue

75-01

85-50

83-71

87-70

THE LIPOIDS OF MICRO-ORGANISMS 381

The phosphatides of the diphtheria bacilhis, having the
iiitro gen-phosphorus ratio of a monoamino-monophospha-
tide, on hydrolysis yield aldohexoses, fatty acids, a
compound with a high molecular weight and traces of a
base. The solid saturated fatty acid was exclusively
palmitic acid, and the liquid acids contained a substance,
corynin, C50H100O4, m.p. 70° to 71°, containing one
carboxyl group and two hydroxyl groups.

Phytomonas tiunefaciens contains about 2 per cent,
of total lipoids when grown on a sjoithetic, glycerol-
containing medium and about 6 per cent, when the
glycerol is replaced by sucrose. About 70 per cent, of
the acetone soluble fat consists of free fatty acids, mainly
unsaturated. The phosphatides are lecithin and cephalin
in approximately equal quantities. The fatty acids
comprise normal saturated and unsaturated Cis and
C18 acids as well as liquid saturated branched chain acids
of high molecular weight, similar to those in the tubercle
bacilli. Among the liquid, saturated fatty acids is crystal-
line ph}i:omonic acid, C20H40O2, m.p. 24° C. It is optic-
ally inactive and is probably a homologue of tuberculo-
stearic acid. It constitutes about 14 per cent, of the
total fatty acids. No chloroform soluble wax and no
sterols could be isolated.

The phosphatide of LactohaciUus acidophilus yields
glycerophosphoric acid, choline, palmitic, stearic and
unsaturated fatty acids, together with a crystalline non-
reducing polysaccharide which gives glucose, galactose
and fructose on further hydrolysis.

The yeasts also contain a high proportion of phospha-
tides, but their constitution has not been worked out.

Fat Synthesis. — The mechanism of fat synthesis by
micro-organisms has been most extensively studied in the
case of the yeast, Endomyces vernalis, which was used as
a source of fat in Germany during the 1914-1918 war.

Haehn and Kinntof , following a suggestion by Magnus
Levy that acetaldehyde condensed to give aldol as an

382 BACTERIOLOGICAL CHEMISTRY

intermediate, proposed the following scheme to account
for the production of fats from sugar. Glucose breaks
down, probably by the same mechanism as in alcoholic
fermentation, to give methylglyoxal hydrate, which gives
pyruvic acid and hydrogen : —

/OH

CeHiaOe > 2CH3CO.C^OH > 2CH3CO.COOH + 4H.

\h

In view of more recent knowledge of alcoholic fermentation
methylglyoxal hydrate probably should no longer be
considered as an intermediate, the pyruvic acid being
formed in accordance with the Meyerhof-Embden scheme
of alcoholic fermentation. The pyruvic acid is decar-
boxylated to give acetaldehyde and carbon dioxide.
Two molecules of acetaldehyde condense to give aldol : —

O H\

ch3c— h + h— ^c.cho > ch3choh.ch2.cho.

h/

The latter loses water with formation of the unsaturated
aldehyde, crotonaldehyde : —

CH3CHOH.CH2.CHO ^ CH3CH = CH.CH0 + H2O.

The crotonaldehyde is reduced by hydrogen formed during
the production of pyruvic acid to give butyraldehyde,
CH3CH2.CH2.CHO. The butyraldehyde condenses with
another molecule of acetaldehyde to produce a homologue
of aldol, j3-hydroxycaproic aldehyde : —

CH3CH2.CH2.CHO + CH3.CHO — > CH3CH2.CH2.CHOH.CH2.CHO.

Loss of water gives rise to «j8-hexylene aldehyde,
CHsCHa.CHs.CH^CH.CHO, which is reduced to caproic
aldehyde, CH3(CH2)4CHO. This process of condensation
with acetaldehyde, dehydration and reduction continues
until a chain of carbon atoms corresponding to oleic

THE LIPOIDS OF MICRO-ORGANISMS 383

or stearic acids is hiiilf up, Avlien oxidation of the alde-
hyde group to a carboxyl group gives stearic acid,
CH3(CH2)i6COOH, for instance. The fatty acids so
produced esterify with glycerol, which is formed in the
early stages of the process, as in alcoholic fermentation
(see p. 277).

Very similar is the suggestion of Smedley and
Lubrzynska that acetaldehyde and pynivic acid undergo
aldol condensation to give a-keto-y-hydroxy- valeric
acid : —

CH3CHO + CH3CO.COOH — > CH3.CHOH.CH2.CO.COOH

On decarboxylation, followed by internal oxidation-
reduction, butyric acid is formed : —

CH3.CHOH.CK2-CO.COOH — > CH3.CH2.CH2.COOH + CO2

a-Keto-y-hydroxyvaleric acid could also lose water to
give the unsaturated acid, which on decarboxylation
would give rise to crotonaldehyde : —

CH3.CHOH.CB2.CO.COOH > CH3.CH=CH.C0.C00H

CH3.CH=CH.C0.C00H > CH3.CH=CH.CH0 + COj

Crotonaldehyde then condenses with another molecule of
pyruvic acid to give an aldol acid which in turn would
give rise to caproic acid and also to the homologue of
crotonaldehyde containing two extra carbon atoms : —

CH,.CH=CH.CHO + CH3.CO.COOH — ^

CH3.CH = CH.CH.0H.CH2.C0.C00H

CHo.CH=CH.CH0H.CH2.C0.C00H — ^

CH3.CH=CH.CH2.CH2.COOH + COj

CH3.CH=CH.CH2.CH2.COOH + 2H >

CH3.CH2.CH2.CH2.CH2.COOH

CH3.CH = CH.CH=CH.C0.C00H >

CHg.CH^CH.CH^CH.CHO + CO,

Continuation of such steps would give rise to the fatty
acids starting with butyric acid and increasing in chain
length two carbon atoms at a time. a-Keto-y-hydroxy-
valeric acid and a-keto-y-hydroxy-valeraldehyde have

384 BACTERIOLOGICAL CHEMISTRY

been found among the products of the action of enzymes
from Staph, alhus on glucose. That one of these views

of the production of fats is probably correct is supported
by the fact that fixation of acetaldehyde with sulphite
or dimedon causes a lowering of the fat yield, and also
that the fatty acids found in bacteria and yeasts all
contain chains with an even number of carbon atoms
(corresponding to the building up of the chain by addition
of the two carbon atoms of acetaldehyde at a time) ;
moreover, nearly all the shorter fatty acids from butyric
up to arachidic acid are known to occur in micro-
organisms. It is interesting, and perhaps significant,
to note that this synthesis passes through the j8-hydroxy-
aldehydes, whilst the degradation of fats in the animal
body proceeds through the jS-hydroxy-acids, carbon
atoms being split off two at a time.

The formation of fats from such substrates as alcohol
probably also proceeds through acetaldehyde as inter-
mediate, the aldehyde being produced by oxidation, or
via a reserve carbohydrate.

For further reading : —

R. J. Anderson :

(a) " The Separation of Lipoid Fractions from Tubercle Bacilli.'*

J. Biol. Chem., 74 (1927), 525.
(6) " The Phosphatide Fraction of Tubercle Bacilli." J. Biol Chem.,
74 (1927), 537

(c) " The Chemistry of the Lipoids of the Tubercle Bacilli." Physiol.

Reviews, 12 (1932), 166.

[d) " Structural Peculiarities of Acid Fast Bacterial Lipides." Chem-

Rev., 29 (1941), 225.

R. E. Buchanan and E. I. Fulmer, " The Physiology and Biochemistry of
Bacteria," Vol. I, Chapter III. Bailliere, Tindall & Cox. London,
1928.

CHAPTER XXI
THE PIGMENTS OF MICRO-ORGANISMS

OUR present knowledge of the pigments of micro-
organisms is in an unsatisfactory state. The
constitution of comparatively few of them is known,
and their classification is an arbitrary one depending on
solubility relationships. However, there are three main
chemical t\pes into which they fall, namely : —

(a) Carotenoid Pigments. — These are red, orange or
yellow pigments soluble in the fat solvents, ether, alcohol
and chloroform. They are named after the type pigment
carotene, an unsaturated hydrocarbon, C56H40, present
as the red colouring matter of carrots. Hydroxyl deriva-
tives of carotene, the xanthophylls, also belong to this
group. Usually they occur together as more or less
complex mixtures, which until recently were almost
impossible to separate. Nowadays they are separated by
means of chromatographic analysis, which depends on
differential adsorption of the pigments on an appropriate
adsorbent, such as calcium phosphate, alumina, kaolin and
others, a principle originally developed by Tswett. A
solution of the pigment in a suitable solvent is poured
through a column of the adsorbant and the chromatogram
" developed " by washing with the same or different
solvents. The pigments separate into coloured bands at
various depths in the column, which is then sliced between
the bands and the separated pigments eluted. The caro-
tenoids are usually characterised by the bands in their
light absorption spectra. Many carotenoid pigments are
readily bleached on exposure to atmospheric oxidation.

385

380 BACTERIOLOGICAL CHEMISTRY

Most of Miem gi\^e the lipocyan reac^tion, an intense ))lue
colour with concentrated sulphuric acid.

{h) Quinone Pigments. — Substituted toluquinones,

O 0

II II

CH /\ ■ "/X/N

, naphthoquinones and anthra-

II
0 0

0

il

quinones are found quite frequently as

0
pigments in bacteria and the lower fungi.

(c) Melanins. — The melanins are black or brown pig-
ments which are very insoluble in nearly all solvents,
even hot concentrated hydrochloric acid. They are
soluble in warm concentrated sulphuric acid and are
reprecipitated on dilution of the solution with water.
They are formed as a result of the decomposition of
proteins, either by boiling concentrated acids or by the
action of the enzyme tyrosinase on the amino -acid
tyrosine. Tryptophane and hydroxy-phenylethylamine
are also sources of melanin pigments. They occur in the
black Torula yeasts, in the De7natiacece, fungi with dark
brown or black hyphse, and in the black varieties of
certain bacilli such as B. mesentericus 7iiger.

The following is a classification of the pigments based
on solubility characteristics ; —

THE PIGMENTS OF MICRO-ORGANISMS 387

A. Cellular Pigments not colouring the medium.
I. Soluble in chloroform.

(a) Soluble in alcohol —

(1) Carotenoid —

(a) Red, e.g. x^^g^^^^^^^^ ^^

BJwdococcus, sulphur

bacteria, Actinomyces,

Torula.
(jS) Orange, e.g. pigments of

Staph, aureus, Sarcina

aurantlaca.
(y) Yellow, e.g. pigments of

Staph, aureus, Staph.

citreus, Sar. lutea.

(2) Xo 13 -carotenoid —

(a) Colour change with acids
and alkalies, e.g. pro-
digiosin, red in acid,
yellow in alkali ; bac-
terio-chlorin, the green
pigment of sulphur
bacteria ; the red and
yellow pigments of
Fusarium, Aspergillus,
Penicillium.
(fi) No colour change with
acids or alkalies —

(i) Fluorescent, e.g.
pigments of
some species of
Aspergillus.
(ii) Non - fhiorescent,
e.g. Flavobacte-
rium brunneum.

388 BACTERIOLOGICAL CHEMISTRY

II. Insoluble in chloroform —

(a) Soluble in water, e.g. the anthocyanins

of some species of Fusarium.

(b) Insoluble in water —

(1) Soluble in alcohol, e.g. violacein

from Chr. violacemn ; violet
and purple pigments of similar
species .

(2) Insoluble in alcohol —

(a) Soluble in alkali, e.g. the
yellow pigments of some
Micrococci ; aspergillin
from the spores of A.
niger.

{{i) Insoluble in alkali, e.g. the
black pigment of B.
mesentericus niger.

B. Extra-cellular Pigments colouring the medium.
I. Soluble in water —

(a) Soluble in chloroform, e.g. pyocyanin,

the blue pigment of Ps. ceruginosa.
(h) Insoluble in chloroform —
^ (1) Colour change with acid and

alkali, e.g. the green fluores-
cent pigment of Ps. cerugiiiosa
(colourless in acid), red pig-
ment of Sacch. pulcherrimiis
(colourless in alkali) ;
methoxydihydroxytoluquinone
from P. spiniilosum (blue in
alkali, purple when neutral,
yellow in acid).
(2) No colour change with acid or
alkali, e.g. the red pigment of
some species of Actinomyces.

THE PIGMENTS OF MICRO-ORGANISMS 389

II. Insoluble in water —

{a) Soluble in other solvents, e.g. brown
pigment of Flavobacteriiun sicaveolens.

(6) Insoluble in other solvents, e.g. the
black and brown pigments of Actino-
myces, Azotobacter cJiroococcum ; the
melanin from Ps. ceruginosa.

Certain organisms, notably of the genera Pseudomonas
and Acetobacter, are capable of oxidising tyrosine, quinic
acid and similar substances in the medium with formation
of black or brown pigments.

The constitution of a few of these pigments has been
established completely, and fragmentary knowledge is
available about some others. Prodigiosin, 020^^250X3,
the red pigment of Sermtia viarcescens (B. prodigiosus) ,
has been shown to have the structure : —

^^OCHa

Violacein, C42II35O5X5 or C50H42O8N6, the violet pigment of
Chr. violaceum, contains one or more pyrrole nuclei with
hydrocarbon side chains, being similar in constitution to
prodigiosin.

The purple bacteria of the genus RJiodovibrio contain
complex mixtures of carotenoid pigments, including
rhodopin, containing one hydroxyl group and two double
bonds ; rhodovibrin, a polyene alcohol ; rhocloviolascene,
C42H60O2J containing two methoxyl groups and thirteen
double bonds, probably having the structure : —

390 BACTERIOLOGICAL CHEMISTRY

OCH3 CH3
(CH3)2C=CH.C=CH.C =

(CH3)2C = CH.CH = CH.C = CH.CH =
OCH,

CH3
CH.CH = CH.C

CH.C= CH.CH"
CH,

a=CH

I

a=CH

rhodopurpiirin ; flavorhodin, a hydrocarbon ; and j3-
carotene.

Sarcina lutea produces bacterioxanthophyll ; a crys-
talline xanthophyll pigment, sarcina-xanthine, m.p.
149° C, with absorption maxima at 480, 451 and 423 m/x
in chloroform, has been isolated from 8. lutea ; 8.
aurantiaca gives j8-carotene and zeaxanthin ; 8taph.
aureus gives zeaxanthin as the only pigment ; M. phlei
gives lutein (a xanthophyll ester), kryptoxanthin and a-,
j8- and y-carotenes ; M. leprce gives a pigment, leprotin,
which is very like j8-carotene ; some strains of Myco-
bacterium have been shown to produce four carotenoid
pigments when grown on media containing mineral oil
as the sole carbon source. Two of the pigments had
vitamin A activity and one was shown to be astacin ;
8pirillum ruhrum gives the purple pigment, spirillo-
xanthin, C48H66O3, containing one hydroxyl group and
fifteen double bonds, and also other carotenoid pigments.
Bacterium cocovenenans gives a yellow pigment, toxo-
flavin, C6H6O4N2, which is isomeric with methylxanthine.
Anaerobic bacteria apparently do not produce carotenoid
pigments.

The purple sulphur bacteria give bacteriopurpurin,
which is a mixture of the red pigment, bacterioerythrin,
and the green pigment bacteriochlorin or bacterio-
chlorophyll, C55H7206N4Mg.H20, which is very similar to
plant chlorophyll ; on removal of the magnesium it yields
bacteriophseophytin. Bacteriochlorophyll is like chloro-
phyll-a but contains two more hydrogen atoms and has

THE PIGMENTS OF MICRO-ORGANISMS 391

an acetyl group instead of a vinyl group on one of the
carbon atoms. Bacteriochlorophyll is probably carried
on different proteins in the Thiorhodacece, Athiorhodacece
and the green sulphur bacteria, since they give different
absorption spectra. C. diphtherice produces porphyrins,
possibly derived from cytochrome.

Azotobacter chroococcum and CI. ivelchii produce
black melanin pigments from t^TOsine, whilst the latter
also forms a pigment of the thio -amino type.

Actinomyces tvaksmanii is said to give an anthocyanin
pigment, but the chemical properties of the pigment
are not altogether those of an anthocyanin. A. coelicolor
and A. viola ceus-ruher give pigments which are blue in
alkaline solution and red in acid solution, closely resem-
bling azolitmin.

Several phenazine pigments are known. Pj^ocyanin,
the blue, chloroform soluble pigment of Pseudomonas
ceruginosa (B. pyocyaneus) has the constitution
N 0-

1 yi y[ ^' , or possibly a dimeric form of it. A

N +

N OH

CH3
second, yellow pigment, a-hydroxj^phenazine, I I ll I

N

occurs in older culture of Ps. ceruginosa. Pyocyanin
can act as a hydrogen carrier in the reduction of cyto-
chrome, and can act as a hydrogen acceptor in the
formation of phosphoglyceric acid from glucose by the
action of the apozymase-cozjTnase system (see p. 202).
Pyocyanin is bactericidal (see p. 177). Pseudomonas
chloromphis yields a green, crystalline pigment, chloro-
raphin, which, on exposure to air, changes to the crystal-
line, yellow pigment, oxychlororaphin, m.p. 241° C. The

392 BACTERIOLOGICAL CHEMISTRY

latter is the amide of phenazine-1-carboxylic acid,
N CONH2

. On reduction with zinc dust in water

N
it gives the orange -yellow, crystalline dihydrophenazine-
1-carboxylamide, m.p. 192-4° C. Chlororaphin is com-
posed of one molecule of oxychlororaphin and one molecule
of dihydrophenazine-1-carboxylamide and can be pro-
duced synthetically by combining the components in
acetic acid solution. Chromobacterium iodinum gives the
pigment, iodinin, which is the N,N' dioxide of a dihy-
droxyphenazine

O

II

N OH

/ 8

%.

OH

2 I

3

4

N

The positions of the two hydro xyl groups is not known
but they are probably not at positions 2:3 or 2 : 5.
Iodinin, like pyocyanin and chlororaphin, is inhibitory to
bacteria, 2/xg/ml. being sufficient to inhibit the gro\vth of
Streptococcus pyogenes. The effect can be reversed by the
action of hydroxy-anthraquinones or by 2 -methyl- 1 : 4-
naphthoquinone. It is possible that iodinin and the
other phenazine derivatives interfere with the mechanism
of hydrogen transfer which involves quinones, by reacting
at the same enzyme centres, in virtue of their similarity
in structure.

The human tubercle bacillus produces a yellow
crystalline pigment, phthiocol, shown to have the
constitution, 3-hydroxy-2-methyl-l:4-naphthoquinone,

THE PIGMENTS OF MICRO-ORGANISMS 393

0

CH3

. Ill alkaline solution it can undergo a tAvo-
OH

O
stage oxidation-reduction reaction and may be concerned
in the metabolism of the organism. It has some vitamin
K (antihsemorrhagic) activity and raises the prothrombin
content of the blood, on injection.

The red yeasts, such as Torula rubra , produce caro-
tenoid pigments, including jS-carotene, tonilene, and a
polycarboxylic acid pigment. The blue fluorescent
pigment thiochrome, C12H14ON4S,
N=C— N==C— S

I I I ^C.CHa.CHjOH

CH3.C C— CH2— N— C

II II I
N— CH CH3

is derived from aneurin, or vitamin-Bi, which occurs in

considerable amounts in yeasts.

The pigment aspergillin, from the black spores of

A . niger, is a melanin type of pigment soluble in alkali .

Fumigatin, 3-hydroxy-4-methoxy-2 : 5-toluquinone,

0

CH3/ ]jOH

, maroon coloured crystals, m. p. 116° C, from
" 'ioCH '

o
A.fumigatus and spinulosin, 3 : 6-dih3Tlroxy-4-methoxy-
0

' CHg

II 11

2 : 5-toluquinone, jiq\\ IIqch ? purple plates, m.p

II !l II li

394 BACTERIOLOGICAL CHEMISTRY

201° C, from P. sjnnulosum and A. fmnigatus have
already been described as antibiotics (see pp. 163, 179).
Phoenicin, 2 : 2'-dihydroxy-4 : 4'-di-methyldiqiiinone,

yellow brown crystals, m.p.

230° C, produced by P. phceniceiim and P. ruhrum is a
condensed toluquinone pigment. Under appropriate
conditions of growth, fumigatin and phoenicin occur in
the colourless reduced quinol form, in the culture media.
As suggested on page 393 the quinones and quinols form
an oxidation-reduction system and may serve as a
hydrogen transfer mechanism in respiration. Flavo-
glaucin, C19H28O3, lemon yellow needles, m.p. 105° C. and
auroglaucin, C19H22O3, orange crystals, m.p. 152° C,
which occur in the mycelium of A. glaucus furnish
examples of quinol pigments. Flavoglaucin has one of
the structures : —

"CHg OH OH

\C.CH2.CH0. |-^^, CO(CH2)eCH3, CH2=CH.,/\ CO.(CH2)eCH3

CH2

II (CH3)2CH.I 1

OH OH __

or

OH

(CH3)2CH.
CH2 = CH.

^\ CO(CH2)eCH3 ^ . • • ^

, and auroglaucni is the

OH

corresponding unsaturated analogue in ^vhich the side
chain — 00.(CHo)6()H3 is replaced l)y — (J0(CH=CH)3
CH3.

THE riOMENTS OF MICRO-ORGANISMS 305

Pigments wliirji are derivatives oi 2-nicthyl anthra-
()

quinone

/>.

CH,

31
4 ,^

, arc responsi])lc for

the colours of many of the lower fungi, notably in the
genus Helminifwsporium and certain Aspergillus and
Penicillium species. Some of them are listed in Table 28.

Table 28

PIGMENT

COLOUR

Structure

PRODUCED By

Carviolacin

Light brown

Tr ihydroxy-methoxy-2 -methyl-
authraquinone

P. carmino-violaceum

Carviolin

Chrome

Trihydroxy-methoxy-2-methyl-

P. carmino-violaceum

yellow

anthraquinone

Catenarin

Red -

1:5: 8-Trihydroxy-2-hydroxy-

E. catenarium.

methyl-anthraquinone

H. gramineum,
H. tritici-vulgaris,
H. velutinum

Cynodontin

Bronze

1:4:5: 8-Tetrahydroxy-2-

H. avencs, H. cynodontis

metbyl-anthraquinone

H. euchlcence

Emodic acid -

Orange

4:5: 7-TrihydroxT-anthraqui-

P. citreo-roseum,

none-2-carboxylic acid

P. cyclopium

Erythroglaucin

Dark red -

Trihydroxy - methoxy - methyl -

A. glaiicus and related

anthraquinone

strains

Funiculosin -

Deep red -

Trihydroxy - methyl - anthraqui -

P. funiculosum

Helminthosporiii

Maroon

none
4:5: 8-Trihydroxy-2-methyl-

H. catenarium,

anthraquinone

H. cynodmtis,
H. gramineum,
U. tritici vidgariH

(o-Hydroxy-emodin

Dull orange

4:5: 7-Trihydroxy-2-hydroxy-

P. citreo-roseum

methyl-anthraquinone

P. cyclopium

Physcion, parietin,

Keddish

4 : 5-Dihydroxy-7-methoxy-2-

A. glaucus and related

or emodin mono-

orange

methyl-anthraquinone

strains and from

methyl ether

lichens

Tritisporiu

Red-brown -

1:3:5: 8-Tetrahydroxy-6 (or
7)-hydroxymethyl-anthraqui-

H. tritici-vulgaris

none

It is possible that the anthraquinones, like the
toluquinones, may play the part of hydrogen carriers in
mould respiration. Some species related to A. glciiicus

396 BACTERIOLOGICAL CHEMISTRY

were shown to contain the reduction products. 4 ; 5-
dihydroxy-7-methoxy-2-methyl-9-anthranol and the
corresponding 10-anthranol in addition to the oxidised
form, physcion. Instead of the usual polyhydroxy-
anthraquinones, H. leersii gave two reduced compounds,
luteoleersin, C26H38O7, yellow rods, and colourless, albo-
leersin, C26H40O7, which are regarded as the semi-quinone
and quinol respectively and which can easily be converted
into one another by oxidation-reduction processes.

Fairly closely related to the anthraquinone pigments
are the yelloAv pigment ravenelin, 1:4: 8-trihydroxy-3-

HO CO OH

met hylxant hone, > which is

\ /\ /\ /^^'
\/ \/ \/
O OH

produced by H. rave^ielii and H. turcicum, and rubrofus-
arin, C15H12O5, red crystals, which is a dihydroxy-
methoxy-methylxanthone produced by Fusarium
culmorum.

The yellow crystalline pigment of the mycelium of
Oospora sulphur ea-ochracea, sulochrin,

COOCH3 0 OH

^\ II /\

II c 1 I

HO I jloCHa HO I 'cHa

is a benzophenone which is related to ravenelin.

A number of other mould pigments are known but
their constitution is unknown for the most part. The
structure of citromycetin, the yellow pigment from P.
citromyces-glabriim, is partially known : —

THE PIGMENTS OF MICRO-ORGANISMS 397

COUH

I
C CO

^\ /\

HO.C C Cv

I II II \

I II II }c,-a,o

II 11/

HO.C C C/

CH O

whilst citrinin, a yellow pigment produced by P. citrinum
and A. terreus, is : —

CgHg

C
0 /%

% / %

C C.OH

CHo.C C

% /%

C C.COOH

CH,.CH 0

For further reading : —

R. E. Buchanan and E. I. Fulmer, " The Physiology and Biochemistry of
Bacteria," Vol. I, Chapter III. Bailliere, Tindall & Cox. London,
1928.

H. Raistrick and collaborators, Papers in Biochemical Journah 1931 onwards

I

CHAPTER XXII

ANTIGENS, HAPTENS, ANTIBODIES
AND COMPLEMENT

N this chapter it is proposed to describe some of the
substances which enter into the reactions involved
in immunological phenomena.

Antigens

An antigen is a substance which, when introduced
parenterally into the animal body, calls forth the pro-
duction of another substance, known as an antibody,
capable of reacting specifically with the antigen. Anti-
gens always react with their corresponding antibodies,
but everything which reacts with an antibody is not
necessarily an antigen. In order to be antigenic it appears
that a substance must be (a) foreign to the animal into
which it is injected, (b) colloidal and (c) introduced beyond
the epithelial tissues of the animal. For our present
purpose we can divide antigens into those which occur
naturally and those which do not ; the latter we may
term artificial or synthetic antigens.

Natural Antigens. — The natural antigens fall into
three classes, proteins, polysaccharides and lipoids.

Proteins. — The vast majority of antigens are proteins
or contain a protein component. Almost all known
proteins are antigenic provided that they are soluble.
The notable exception is gelatin. It should be pointed
out that gelatin, strictly speaking, is not a naturally
occurring protein since it is produced by the hydrolysis

398

ANTIGENS, HAPTENS, ANTIBODIES, ETC. 399

of collagen ; it may thei'efore be degraded jjelow the
limits of colloidal dimensions necessary for antigenic
power. It differs from the majority of proteins in not
containing tyrosine or tryptophane among the amino -
acids of which it is built up and in being devoid of carbo-
hydrate, and it has been suggested that its lack of anti-
genic properties may be due to this deficiency. Insulin
which is also non-antigenic lacks carbohydrate, but is
rich in tyrosine.

If proteins are rendered insoluble, by heat denaturation
or by treatment with alcohol, for instance, they are no
longer antigenic. If the denaturation has not been carried
too far and is reversible the regenerated undenatured
protein regains its antigenic properties. Such proteins as
casein which are not rendered insoluble by heating do
not lose their antigenicity on such treatment.

The breakdown of a protein with loss of its colloidal
properties is accompanied by a loss of antigenic properties .
Thus a mixture of protein constituents obtained by
hydrolysis is not antigenic. If, however, the fragments
are re-united by enzyme action to form the colloidal
plasteins, these are antigenic although they may have a
specificity different from that of the original protein ;
the plasteins obtained by recombination of the amino -
acids of different proteins usually give cross reactions,
that is, they have a certain degree of common specificity.

Proteins from different sources differ from one another
in the proportions and internal arrangement of their
constituent amino -acids. Even such closely related pro-
teins as the albumins of hens' and ducks' eggs can be
distinguished by using anaphylactic shock in a sensitised
animal as an indicator, although precipitin reactions are
not sufficiently sensitive. It has been shown that these
two albumins possess different amino -acids in the terminal
positions of their molecules although their gross structure
is the same. The fibrinogens and haemoglobins of different
species can be similarly distinguished. As a rule there is

400 BACTERIOLOGICAL CHEMISTRY

a certa-iii amount of cross -react ion l»etween such hetero-
logous antigens and antisera (that is, between the
haemoglobin, say, of one animal and an antiseiTim prepared
against the haemoglobin of another animal), especially if
the antiserum is employed in high concentration and if
the two species are closely related. However, homologous
pairs of antigen and antiserum always react to a con-
siderably higher titre than do heterologous pairs.

The non-structural proteins like globulin or albumin
which circulate in the body are usually highly species
specific. Structural proteins or depot proteins such as
keratin, eye-lens protein, casein and the proteins of
seeds are less specific and give wider cross reactions.
The highly specialised proteins, for example insulin,
which are common to many species are not only not
specific in their immunological reactions but are not
even antigenic.

Proteins of different chemical types, even if they are
obtained from the same species, give distinct and specific
reactions and show no cross -reactions. Thus the serum
proteins, globulin and albumin, of the rabbit differ
from one another immunologically as well as chemically,
and they also differ from haemoglobin and the protein of
the lens of the eye and other proteins.

In view of the fact that most of the chemical methods,
such as halogenation, nitration or the introduction of
azo -compounds, by which proteins can be altered to give
different immunological reactions, involve changes in the
amino -acids, tyrosine, tryptophane or phenylalanine,
which contain a benzene ring, it has been considered that
these constituents of the protein play a particularly
important role in determining antigenic properties. The
fact that the non-antigenic protein, gelatin, contains none
of these amino -acids lends a certain amount of support to
the view. These amino -acids alone, however, cannot
account for all the activity, since the amino, hydroxyl
and carboxyl groups of the aliphatic amino -acids can also

ANTIGENS, HAPTENS, ANTIBODIES, ETC. 401

J>e altered in various ways with corresponding changes in
specificit}'. Aoetylation, which involves the hydroxyl
groups of tyrosine as well as amino groups, has a greater
determining influence on specificity than does treatment
with formaldehyde which acts primarily on amino groups.
Acetylation usually eliminates species specificity but the
action of formaldehyde does not. It may be mentioned
here that the chemical alteration of a protein from a
given animal may change it sufficiently to cause it to
react as a protein foreign to that animal ; thus rabbit
serum treated with formaldehyde is sufficiently different
from the original serum to elicit the production of anti-
bodies when it is injected into the rabbit which supplied it.

Polysaccharides. — The majority of polysaccharides,
bacterial and otherwise, which have been examined are
not antigenic although they are haptens, that is, they
can react with antisera prepared against a complete
antigen of w^hich they formed a part. It was shown by
Zozaya, that starch, dextran, glycogen and the poly-
saccharides of several bacteria, including B. anthracis,
the dysentery bacilli, streptococcus and pneumococcus,
became antigenic if they were adsorbed on to collodion
or aluminium hydroxide as a colloidal carrier. Since
that time several polysaccharides have been suspected of
being in themselves antigenic, without requiring any
colloidal carrier, in spite of being free from proteins. The
first of these was the acetyl polysaccharide isolated from
the Type I pneumococcus by Goebel. This substance
differs from the originally isolated soluble specific sub-
stance in the possession of one acetyl group, which is
apparently sufficient to convert the hajiten into a com-
plete antigen. Some doubt has been cast on this finding
by the work of Felton who could find no correlation
between the acetyl content of various samples and their
antigenicity.

The next supposedly antigenic polysaccharide to be

402 BACTERIOLOGICAL CHEMISTRY

isolated was the "complete antigen" which Boivin and
his co-workers isolated from smooth strains of Salmonella
typhirnurium by extraction with dilute trichloracetic
acid. It was found to be toxic, capable of provoking
antibody production and of immunising mice against
subsequent injection of many times the fatal dose of the
toxin or the living organisms. By the action of hot,
dilute acetic acid the antigenic character and toxicity
were rapidly lost and a polysaccharide, the " residual
antigen," was obtained which was a hapten only, the
hydrolysis having split off fatty acids and a phosphatide.
It has since been shown that a protein fraction is also
removed by hydrolysis and that the antigenicity of the
" complete antigen " depends on its presence (see p 361).
Rough strains of Sal. typJiimurium contain the " residual
antigen " only ; they possess an enzyme which hydro lyses
the " complete antigen." The same antigen was dis-
covered almost simultaneously by Raistrick and Topley,
who isolated it by treating the bacteria with trypsin in
order to digest the bacterial protein and then precipitating
the polysaccharide from solution with alcohol.

More recently, in a similar way, Raistrick and Topley
have isolated a toxic, completely antigenic polysaccharide
from the typhoid bacillus . It confers protective immunity
on mice when injected in extremely small doses and elicits
antibody formation in rabbits. On hydrolysis by very
weak acid this too loses acid groups and yields a neutral
polysaccharide which is non-toxic and non-antigenic (see
p. 363).

It is, therefore, becoming more certain that pure
polysaccharides are not antigenic, but that when com-
bined with a protein they act as the determinant groups
of an antigenic complex.

Lipoids. — There is still considerable controversy as to
the antigenic character of lipoids, although there is no
doubt of the very important part which they play in

ANTIGENS, HAPTENS, ANTIBODIES, ETC. 403

immunological reactions. Although they are not capable
of such great variation in composition as the proteins
and polysaccharides there is still a fairly wide range of
possible lecithins and kephalins (containing different
saturated and uixsaturated acids) and sterols. It seems
almost certainly established that they cannot act as
antigens alone but that they may do so when mixed
with serum ; that is, like most polysaccharides, they are
haptens. Synthetic distearyl lecithin was shown to give
complement fixation with an antisenim prepared against
the compound mixed with pig serum ; cross -reactions
with commercial lecithin preparations were also obtained.
S}Tithetic and purified lecithins were found to be weaker
antigens (when mixed with serum) than crude lecithins.

The Wassermann and flocculation reactions used in
the diagnosis of syphilis employ lipoid " antigens "
obtained by alcoholic extraction of heart muscle tissue.
Their specific activity with s}q^)hilitic sera appears to
depend on a substance, cardiolipin, isolated from the
phosphatide fraction. It contains 4-11 per cent, of
phosphorus but no nitrogen ; and is isolated as a sodium
salt. On saponification it gives 62 per cent, of fatty acids,
a non-reducing carbohydrate and phosphoric acid.
Glycerol is absent, and the substance is analogous to
the carbohydrate containing lipoids of the tubercle
bacillus and Lactobacillus acidophilus (see p. 371).
Lecithin and cholesterol are also necessary in the " anti-
gen " for complement fixation to occur. Other lipoid
fractions, fats and fatty acids, also afford reactions
which, however, are not specific, occurring with normal
as well as syphilitic sera. Bacterial lipoids, like those of
plant and animal origin, behave in a similar w^ay with
sera, giving the same types of flocculation reaction.

The sterols, when mixed with pig serum, also seem to
be antigenic and give antisera which show complement
tixation but not flocculation reactions. Cholest-erol,

404 BACTERIOLOGICAL CHEMISTRY

CH3 I /^^3

CH2 ! CH— CH(CH2)3CH

/\ ^^ \CH

CHg C CH2

^H3 I I I

CH2 I CH CH CHj

CHg C CH

HO.CH C CHj

\ /% /
\/ %/
CHo CH

dihydrocholesterol, oxycholesterol (in which the double
bond is saturated by hydrogen and oxygen respectively)
and ergosterol,

CH3 CH3 .,TT

CH3 1 1 /-^^3

CH2 CH— CH— CH-CH— CH— CH

CH2 C CH2

CH3 1 1

1

CH2 CH CH CHa

CH2 c c

1 1 II

i.CH C CH

CH2 CH

give cross -reactions, the reactions with homologous sera,
however, being stronger than with heterologous sera.
Cholestan (in which the hydroxyl group of cholesterol is
replaced by a hydrogen atom), dibromocholesterol (in
which the double bond is saturated by bromine) and
cholesterol esters showed differences from one another,
due to differences at the doul)le l)ond or the hydroxyl
group, but also gave cross-reactions due to the rest of
the molecule.

ANTIGENS, HAPTENS, ANTIBODIES, ETC. 405

Artificial Antigens. — Our knowledge of artificial anti-
gens has developed largely as a result of Landsteiner's
investigations into the chemical basis of immunological
specificity. He showed that if an antiserum is prepared
by the injection of, say, horse serum into a rabbit that
antiserum will react with horse serum but not with
chicken serum or egg albumin. Similarly an antiserum to
chicken serum will react only with the homologous serum
and not with horse serum or egg albumin, which also
gives a specific antiserum. Landsteiner made artificial anti-
gens by introducing various known chemical groups into
the proteins. For example, he cliazotised the compound

atoxyl, p-aminophenyl-arsinic acid, NH2;(^ \AeO3Hj,

and coupled it with proteins, presumably through the
tyrosine, histidine or tryptophane groups : —

CHo.CH.KH,.COOH

+ 2R.N=N.C1. —
+ 2NaOH

OH

(Tyrosine)

2NaCl.

R.N=Nl 'N=N.R 1- ^xxjv

OH

(3 : 5 di-azo-derivative)

He used the resulting atoxyl-azo -proteins as antigens
and found that any of the antigens reacted with any of
the antisera irrespective of the protein (horse or chicken
serum or egg albumin) which was present in the atoxyl
derivative. In other words, the atoxyl group had
aboHshed (or masked) the original specificity and con-
ferred a new one determined by itself. If tlie proteins

406 BACTERIOLOGICAL CHEMISTRY

were coupled with siilphanilic acid, NH2<^ \sO3H,

instead of with atoxyl, again the original protein specificity
disappeared, nor would sulphanilic acid-azo -proteins
react with antisera to atoxyl-azo -proteins and vice versa.

This specificity can be demonstrated in another way.
The formation of the precipitate by the reaction of the
antiserum with the atoxyl-azo -protein antigen, for
instance, can be prevented by the previous addition of
simple atoxyl derivatives to the antiserum. Atoxyl
diazotised and coupled with tyrosine, or even atoxyl
itself, can act in this way. These simple derivatives are
not themselves antigenic nor do they give any visible
reaction with the antiserum. They are named simple
haptens. The inhibition of precipitin reactions by haptens
is also specific ; atoxyl haptens inhibit the reactions of
atoxyl antisera but not those of sulphanilic acid antisera,
whilst the sulphanilic acid haptens inhibit only reactions
between sulphanilic acid-azo -proteins and the corres-
ponding antisera and not those between any other antigens
and antibodies.

The groups, like atoxyl or sulphanilic acid, which
modify the specificity of antigens in these ways are called
determinant groups. Landsteiner studied a large number
of aromatic amino -acids from this point of view and found
that the specificity which they introduced depended
partly on the substituent and partly on its position in
the ring. Thus antisera to o-amino -benzene -sulphonic
acid-azo -proteins gave reactions with both the 0- and
m-derivatives but not with the ^-derivative, showing the
effect of position. The same antisera also reacted with

o-amino -benzoic acid, \ / ^ , as hapten, or, when

COOH

coupled with protein, as antigen, ])ut they showed no
reactions with the m- and ^-amino -benzoic acids. This
illustrates the com])ined effect of the substituent and its

ANTIGENS, HAPTENS, ANTIBODIES, ETC.

407

position. The electric fields clue to the polarity of the
groupings (see p. 33) are sufficiently alike in the case of
o-amino -benzoic acicl and o-amino -benzene -sulphonic acid
to allow of either of them reacting with the antiserum
prepared against the other ; but altering the position of
the substituent to the 7/i-position is sufficient to change
the electric field so much that cross -reactions can no
longer occur. The difference is, of course, even more
enhanced when the p- and o-derivatives are compared.
The addition of a second substituent is also sufficient
to alter the charge distribution to a great enough extent
to abolish the reactions ; thus antisera to o-amino -
benzene -sulphonic acid-azo -proteins will give no reaction
with chloro -amino -benzene -sulphonic acid-azo -protein (1),

>N=N— Protein

SO,H

, nor with the corresponding methyl

derivative (2),

CH,

)X=X — Protein

SO,H

, and vice versa.

The fields of (1) and (2) are nearly enough alike, however,
for antisera to (1) to react with (2), and vice versa.

Landsteiner also made some interesting experiments
by coupling peptides to proteins. He diazotised 2^-amino-
benzoyl-glycyl -leucine,

CO.

NHa

giycyi [G]

.XH.CH.COOH

I
CH,

I "
CH

Clfg CM3

leucine [L]

coupled it to proteins and made antisera in the usual
way. He also prepared similar compounds with different
arrangements of the amino -acids in the peptide, namely,
LG, LL and GG, where L and G represent leucine and
glycine respectively. The various combinations of

27

408

BACTERIOLOGICAL CHEMISTRY

antigen and antiserum were then tested ; typical results
are summarised in Table 29.

Table 29

Antigen

Antiserum

Hapten

Reaction

GLP

GLS

+ +

LLP

^^

+

GGP

—

—

LGP

—

—

GLP

GL

—

,,

LL

-

,,

GG

+ +

,,

LG

+ +

It was found that GL antiserum reacted strongly
with GL antigen, less strongly with LL antigen and not
at all with GG and LG antigen. The reaction between
GL antiserum and GL antigen was inhibited by GL and
LL as haptens but not by GG or LG haptens. Similar
results were also found for the other antisera. It is seen
that both the amino -acids in the peptide influence the
specificity, but that the terminal amino-acid is the most
powerful factor ; when the terminal amino-acid in the
test antigen is different from that used in preparing the
antiserum no reaction occurs, whilst a different amino-acid
in the intermediate position merely weakens the reactions.

Landsteiner showed that the peptides having molecular
Aveights between 600 and 1000, consisting of 8 to 12
amino-acid residues, obtained by hydrolysis of silk
fibroin, were capable of inhibiting the reaction between
silk fibroin and the antibody to it ; the inference is that
the determinant groups in silk fibroin are not larger than
these peptides. It is probable that the determinant
groups of other proteins are of similar dimensions.

In cases where the determinant group has no very
marked polarity the specificity is less sharply defined, and
the actual nature of the particular groups involved has little

ANTIGENS, HAPTENS, ANTIBODIES, ETC. 409

or no effect . Thus it has been showTi that the determinant

groups NH^/ }-^-\ /' ™2\ /~™~\ /

andNHa/ ^— CHg— / y , linked in the usual way to

proteins, are immunologically equivalent, that is, the
antiserum to one will react equally well \sdth any of them
as antigen or hapten, the slight differences of field due
to the comparatively inert groups — 0 — , — NH — and
— CHg — not being sufficient to influence the reactions.

— C—
If, however, the strongly polar group, 1| , replaces the

0
inei-t group, the equivalence is no longer apparent ;
antisera to the above antigens will not react with anti-

gens or haptens containing ^ ^x

as the determinant group.

As would be expected from what was said about
surface charges in Chapter III, spatial configuration plays
an important part in the specificity induced by deter-
minant groups. Landsteiner, for instance, coupled the
amino -tartranilic acids, which exist in the optically active
d- and Z-forms and the optically inactive me5o-form, with
proteins and prepared the corresponding antisera. The
antisera to the dextro -compound reacted strongly with
the cf-antigen,

OH H

/ \ I I

Protein— X =N< >NH.CO.C C"— COOH

^ ^ I I

H OH

Ijut only very slightly with the /-antigen,

H OH

Pi-otoin— N -n/ ^NH.CO.C C— COOII

^ ^ I I

OH H

410

BACTERIOLOGICAL CHEMISTRY

whilst it gave weak reactions with the meso-antigen,

Protein — N=N<

)NH.CO.C C-

I I
OH OH

COOH

Similarly, the /-antiserum reacted strongly with the
/-antigen, hardly at all with the (/-antigen and weakly
with the me^o- antigen. The me^o-antiserum reacted with
the homologous antigen strongly and gave weak cross -
reactions with the d- and /-antigens. The amino-
tartranilic acids used as haptens gave the corresponding
specific inhibition reactions. If, however, d- and /-malic
acids,

H

OH

I

C, '

and COOH.CH2.C.COOH,

I
OH

having only one optically active carbon atom instead of
the two of the tartranilic acids, were used as haptens the
inhibition reactions were much weaker, but took place
with the antiserum to the corresponding isomer of tartra-
nilic acid. Succinic acid, COOH.CH2.CH2.COOH, having
no optically active carbon atom had no effect as a hapten.
The effects of spatial distribution are also well illus-
trated in the case of the synthetic carbohydrate antigens
studied by Avery and Goebel. They synthesised
23 -amino -phenol- j8-glucoside.

CH2OH

ANTIGENS, HAPTENS, ANTIBODIES. ETC.

and 7:>-amino -phenol- ^-galactoside,

-til

CH„OH

HO

H OH

XH,,

diazotised them and coupled them with albumm and
globulin in the usual way to make antigens. It will be
seen that the only difference between the glucose and
galactose derivatives is in the arrangement of the hydrogen
atom and hydroxyl group on carbon atom 4 of the
sugar molecule. Yet this small difference is sufficient
to make the new antigens specific and to obliterate the
specificity of the proteins to which they are coupled ; thus
the glucoside -globulin antigen gives no reaction with the
galactosicle -globulin antiserum but does react with the
glucoside-albumin antiserum. The inhibition reactions
by the homologous haptens were equally specific, the
glucose-derivative hapten having no effect on the reactions
between galactose-derivative antigens and antisera, and
vice versa. When ;;-amino-phenol-« -glucoside,

HO

412

BACTERIOLOGICAL CHEMISTRY

was diazotisecl and coupled with proteiiLS, it aLsu produced
specific antigens in which the glucose molecule acted as
the determinant group, but marked cross -reactions took
place l)etween the a- and ^-giucoside antigen-antiserum
pairs. It thus appears again that the terminal group,
— CHOH, at position 4 in these cases, has a greater effect
than groups situated within the molecule, such as the
a- and ^-glucoside links in these examples.

Artificial antigens containing glucuronic or galac-
turonic acids as determinants react with various anti-
pneumococcal horse sera, but their injection into animals
does not provoke antibodies which protect the animal
against iniection with pneumococci. If, however, an
antigen made by coupling synthetic, diazotised 2?-amino-
benzyl cellobiuronide (6- jS-glucuronosidoglucose),

H

OH

HO

H

H

OH H
H

COOH

-CHj

I

Os

H A

l/H

|\ OH H
0 OH

-CH2.C6H4.NH2

H

OH

with horse serum globulin is injected, antibodies are
formed which give precipitin reactions with the Type III
polysaccharide, agglutinate Type III pneumococci and
confer on mice passive immunity against virulent Types
II, III and VIII pneumococci. The corresponding
antigen containing gentiobiuronic acid (4- ^-glucuronosido-
glucose).

H

OH

HO

CH2OH

J 0

/ \ H H /.

OH H \| l/H

H / n l\^^ "

0 CHo.CeH^.NH.

COOH H OH

gives rise to antisera which have no protective effect in

ANTIGENS, HAPTENS, ANTIBODIES, ETC. 413

mice against Types III and VIII pneumococci, but which
confer immunity against Type II organisms. Antigens
containing cellobiose or gentiobiose in place of the
aldobionic acids give antibodies which are devoid of pro-
tective effect against infection with pneumococci. Anti-
gens made in the same w^ay using the specific polysac-
charides give antibodies with sharp type specificity
whilst those containing aldobionic acids give antibodies
with a wider specificity covering all types of pneumococci
(for example Types II, III and VIII) which contain the
same aldobionic acid in their specific polysaccharides
(see p. 355).

Artificial antigens containing strychnine or various
sulphonamides, have been prepared by coupling the
diazotised substance with a protein. The resulting
antibodies react specifically with the antigens. The anti-
sera prepared against the strychnine antigen were too
weak to protect mice against the toxic effects of strychnine.
The specificity of the sulphonamide antigens is determined
by the nature of the sulphonamide ; thus sulphanilic
acid, sulphanilamide and sulphacet amide azo -proteins
give cross reactions with the antisera, whilst sulpha-
thiazole and sulphapyridine are much more sharply
specific, as would be expected from the difference in
the structure of the substituents carried by the sul-
phonamide group.

All the artificial antigens which we have considered
so far have been produced by coupling proteins with
known simple chemical compounds by means of the diazo
reaction. This coupling almost certainly occurs w^ith
those amino -acids in the protein, such as tyrosine, histi-
dine and tryptophane, which contain aromatic groups.
The aromatic nuclei can also be modified in other ways ;
for instance, by the introduction of halogens or nitro
groups, wdth similar abolition of the original protein
specificity and the formation of a new specificity depending

414 BACTERIOLOGICAL CHEMISTRY

on the introduced determinant group. Thus iodo- and
bromo -proteins were found to have lost their protein
specificity and reacted only with antisera prepared against
themselves, although there was strong cross -reaction
between the iodo- and bromo -derivatives. The reactions
between both iodo- and bromo -antigens and antisera

I

were inhibited by di-iodotyrosine, H0<^ ^CH^.CH.COOH,

which is apparently the corresponding hapten. Neither
tyrosine, H0<^ ^CHa-CH.COOH itself, nor di-iodophenol,

NH2
I

HO^ y, nor potassium iodide acted as haptens.

I
Dibromotyrosine behaved as a hapten, but less strongly
than the iodo -derivative.

Harrington has made use of tyrosine for coupling
various determinants to proteins. He prepared the
0-^-glucosidyl-tyrosyl-derivatives of gelatin and insulin
by condensing glucosidyl-tyrosine,

with the free amino groups of lysine in the proteins.
These normally non -antigenic proteins were thus con-
verted into antigens which provoked specific antibodies
when injected into rabbits, although rather poorly.
This affords further evidence that gelatin and insulin
are non-antigenic due to lack of tyrosine and carbo-
hydrate and of carbohydrate respectively. Glucosidyl-
tyrosine coupled to ordinarily antigenic proteins, such as
globulin or albumin, gave very good antigens of a sharp
specificity determined by the introduced groups.
Harrington also made antigens containing thyroxine by

ANTIGENS, HAPTENS, ANTIBODIES, ETC. 415

coupling the azide of N-carbobenzoxy-3 : 5-di-iodothyro-

nine,

I

H0<^ \-0-( \cH2.CH.COX3

^ NH.OC.OC.H;

with proteins in alkaline solution and iodinating the
product to convert the di-iodothyronine residues to thy-
roxine and to convert the tyrosine residues initially in
the protein to di-iodotyrosine. Antisera prepared hy the
injection of these antigens were highly specific in their
reactions and were able to prevent the metabolic activity
of thyroglobulin or of thyroxine when these were injected
into animals. Similar results were obtained by coupling
aspirin to proteins ; passive immunisation of animals
with antisera prepared against aspirin antigen prevented
the ordinary pharmacological action of aspirin.

A method of affecting the specificity of proteins which
does not necessarily affect the benzene nuclei is to substi-
tute the hydroxyl and amino groups of the amino -acids
by acyl groups"! A certain amount of cross -reaction
between different acyl proteins occurs ; for instance,
acetyl-proteins react to some extent with antisera prepared
against propionyl-proteins but not with those prepared
against proteins containing longer chain substituents like
butyryl, CH3(CH2)oCO— , or valeryl, CH3(CH2)3CO— .
Proteins containing aromatic substituents like the anisoyl

CHgO/^ ^CO— , or cinnamyl, <^ \cH=CH.CO— groups

do not give cross -reactions with antisera to the aliphatic
derivatives. Methyl or similar alkyl groups can be intro-
duced into the hydroxyl or amino groups of the amino -
acids with similar but weaker effects, a result which is to be
expected in view of the less actively polar character of
such groups. Methylated proteins react with the antisera
to ethyl-proteins but not with those to untreated proteins

416 BACTERIOLOGICAL CHEMISTRY

or to acyl- or nitro -proteins (which have strong pokir
groups).

The amino groups of proteins may be altered in other
ways. Hopkins and Wormall treated proteins with phenyl

isocyanate, <^ \N=C = 0, , which reacts with amino

groups to give phenylureides or substituted ureas : —
R.CH2.NH2 R.CH2.NH

+ > ^CO

C6H5N = C = 0 CeHgNH

The protein-phenylureides were found to have lost the
specificity of the original proteins and to have acquired
a new one due to the phenylureide group. It was found
that the precipitin reactions were inhibited by lysine
phenylureide as hapten, indicating that the amino-acid
lysine, NH2.CH2.(CH2)3CH.COOH, is involved in the

I

NH2
formation of the protein phenylureide derivatives.

Isocyanate derivatives have been used for the intro-
duction of other determinants into proteins. For instance
the isocyanate derivative of histamine can be made by
treating histamine with carbonyl chloride, and then
coupled with proteins to give antigens : — ■

CH==C.CH,.CH.,.NH2 CH==C.CH,.CH,.N = C = 0

i I " " > I I " "

N NH + COCI2 N NH protein

CH CH

(Histamine) ( j3-5-imidazolylethyl isocyanate )

CH==C.CHo.CH,.N =CO.NH

I I " " I

N NH Protein

CH
The coupling probably occurs through the free terminal

ANTIGENS, HAPTENS, ANTIBODIES, ETC. 417

amino groups of lysine. The antisenini prepared agaiiLst
this antigen is specific for histamine and gives cross
reactions witli hist amine -azoproteins. Guinea-pigs im-
munised by injection of the histamine antigen were
protected against the physiological effects of histamine.

In contrast to the effect of isocyanates Landsteiner
found that when the amino groups of proteins are con-
densed with formaldehyde with introduction of a
methylene group,

R— NHa + HCHO > R— N = CH2 + HgO,

there is no apparent change in the specificity of the
proteins, which behave immunologically exactly as the
untreated proteins . This fact is made use of in preparing
anti-toxins ; the toxin is converted to the toxoid by the
action of formaldehyde. The toxoid, although non-toxic,
is still capable of provoking the formation of antibodies
which react specifically with the original toxin.

It was mentioned on p. 362 that the somatic antigens
of Shigella dysenterice and Eherthella typhosa could be
broken down by the action of 90 per cent, phenol into a
polysaccharide hapten and an antigenic protein, and that
the two components could be recombined by solution in
formamide and precipitation of the complex by alcohol.
The protein is also capable of being coupled with other
polysaccharides in the same way to give antigenic com-
plexes. Thus with agar, gum acacia, cherry gum, and
the blood group A specific polysaccharide (isolated from
commercial pepsin, peptone or gastric mucin) antigens
are formed in which the specificity is determined by the
polysaccharide moiety. Gum acacia gives no precipitin
reaction with cherry gum antiserum, although cherry
gum gives a weak reaction with gum acacia antiserum.
Other polysaccharides such as kanten (a degradation
product of agar), gum tragacanth, hyaluronic acid, the
specific polysaccharides of Types I and II pneumococci

418 BACTERIOLOGICAL CHEMISTRY

aud the blood group A polysaccharide give no reaction
with the two gum antisera.

Antibodies
Our knowledge of the chemistry of antibodies is at
present rather vague. For the most part we only know
antibodies by the reactions which they give, and we have
but little insight into their chemical differentiation. As
we have already stated, they are produced as a result of
the injection of foreign colloidal substances into the
animal system. It has been conjectured that they are not
even new substances but merely an altered physical state
of the normal serum proteins . Practically all the evidence,
however, points to the fact that they are entities and
capable of a separate existence ; they can be removed,
either specifically by the corresponding antigen or by such
non-specific agents as kaolin or alumina, from the anti-
serum and then recovered from the complex by suitable
means .

Composition of Antibodies. — That antibodies are pro-
tein in nature is shown by the fact that the specific
precipitates contain more protein than can be accounted
for by the antigen. This is particularly striking when
the antigen is a soluble specific polysaccharide, such as
that of Type I pneumococci, which itself contains no
protein. That the protein in these precipitates is not due
to non-specific adsorption is shown by an experiment in
which an antiserum was coupled with diazotised benzi-
dine-R-salt to give a bright -red dye ; the dyed antiserum
was used in specific and non-specific precipitin reactions,
the red colour appearing only in those precipitates in
which the antiserum was one of a homologous pair.
Again, bacteria or red blood corpuscles sensitised with an
antibody acquire a new isoelectric point and move in an
electric field as though they were coated with globulin.
Such systems, as well as toxin-antitoxin mixtures, con-
taining, say, an antibody prepared in a horse, if injected

ANTIGENS, HAPTENS, ANTIBODIES, ETC. 419

into a rabbit will produce antibodies corresponding to the
protein (horse globulin) of the first antibody. Antibodies
appear always to be associated with the globulin fraction
of the serum proteins, but w^e do not know whether they
are normal globulins modified in some way by the
presence of the antigen or whether they are entirely new
globulins .

The antibodies, generally speaking, are so closely
related to the globulins chemically that it is almost im-
possible to distinguish between them except by serological
reactions. They have the same distribution of amino -
acids, the same nitrogen content and the same isoelectric
point ; it has been shown that the carbohydrate fractions
of normal seiiim globulin, diphtheria antitoxin and the
toxin-antitoxin floccules are identical, but different from
that in albumin. Differences between the precipit ability
of ferric hydroxide sols by normal sera and antisera have
been noted, but the results are not at all constant. The
antibody globulin forms but a small portion of the total
serum globulin, as the figures quoted on p. 427 show, so
that it is not surprising that the antibodies cannot be
distinguished chemically.

During immunisation the serum globulin content in-
creases by about 5 per cent, and the albumin content falls
slightly. The increase of globulin, however, bears no
constant relationship to the antibody titre, and in any
case only a very small proportion of it can be due to
antibody globulin since precipitation results show, for
example, that the unit of diphtheria antitoxin is associated
with only 0-01 mg. of globulin.

The antibody globulins must, obviously, be different
in* some way from normal globulins, and it is generally
thought that they carry active prosthetic groups which
differ in stability from normal globulin. The group may
survive treatment, considerably altering the globulin,
such as the addition of iodo groups or acetylation, or it

420 BACTERIOLOGICAL CHEMISTRY

may be eliminated by treatment not substantially affecting
the globulin as a whole, depending on the point of attack.

Production of Antibodies. — We still do not Imow how
antibodies are produced in response to the injection of
antigens. It has been suggested that antibodies differ
from normal serum globulins in that they incorporate
the antigen in their structure and owe their specificity to
that fact. This theory seems hardly tenable for several
reasons ; an extremely small quantity of an antigen can
induce the formation of almost limitless amounts of
antibody. Animals repeatedly bled after injection of a
single small dose of antigen go on manufacturing antibody
in quantity which would exhaust the supply of antigen
many times over, however little of it were incorporated.
More positive evidence is afforded by experiments made
with artificial antigens containing an easily detectable
hapten group such as the atoyxl group or certain dye
groups ; in no case has a trace of that group been detected
in the antibodies resulting from the injection of such
antigens. A suggestion, made by Breinl and Haurowitz
and by Mudd, which seems more possible is that the anti-
bodies are synthesised from amino-acids and peptides at
the surface of the antigen which, due to its stereochemical
spacing or distribution of surface charges, acts as a sort
of template and impresses its specificity on the newly
formed protein in that way. The antibody is supposed
to differ from normal globulin in the order in which the
amino-acids are incorporated into the protein chain, the
new order being imposed by the presence of the antigen.
The complex of antigen and antibody so formed is as-
sumed to dissociate, the antibody being liberated into the
blood-stream and the antigen being left free to influence
the synthesis of further amounts of the antibody.

Work by Cannon and his collaborators has shown that
the formation of antibodies is associated with the syn-
thesis of globulins in the l)ody, and depends on an
adequate supply of amino-acids in the form of protein

ANTIGENS, HAPTENS, ANTIBODIES, ETC. 421

or other dietary nitrogen compounds. It has also been
shown by the use of amino -acids containing isotopic
nitrogen as a " marker " or " label " that the proteins
of the body, including globulin and albumin, are con-
stantly being broken down and resynthesised and that
the amino -acids themselves undergo similar changes.
Pauling has suggested that antibody globulin molecules
do not differ from normal globulin molecules in their
chemical composition, and that they are sjoithesised in
the normal way, but that they differ in the way that the
peptide chains, particularly^ at the ends, are folded into
a stable configuration. He regards the centre part of the
chain of amino -acids to be held at the s;^Tithesising
centres in the cell whilst the ends of the chain are more
or less mobile, as pictured at (i) in Fig. 8. These free

1

'^Iaaaj^'' ^^0>

B

(II) (ill) Ovj

Fig. 8, — •Synthesis of Normal Globulin.

ends fold into their natural stable configuration and
remain so because held by hydrogen bonds and similar
intramolecular weak bonds (at ii). In the course of
time the molecule becomes liberated into the blood-
stream (as at iii) where the whole of the peptide chain
settles down to the stable configuration of normal
globulin, represented at (iv). When globulin is syn-
thesised in the presence of antigen, which is taken up
by the cell at the site of synthesis, the polypeptide
chain is built up as usual (i. Fig. 9), but the free ends
now assume a configuration which is modified by the
presence of the polar determinant groups on the antigen,
Avhich exert an attraction on groups in the polypeptide
chain (see ii), and cause folding of the ends of the chain

422 BACTERIOLOGICAL CHEMISTRY

in a pattern complementary to that of the particular
part of the antigen in contact with them. The newly
synthesised modified globulin chain becomes released
from the cells as illustrated by (iii), and ultimately one
end of the chain becomes dissociated from the antigen as
at (iv), and the central part of the chain can then take
up its normal globulin folding (v). In time the other
end of the molecule is dissociated from the antigen and

B

111

c Antibodu

B,.-— ^ R B \

G"#

(ivj (v) (vi)

Fig. 9 — Synthesis of Antibod}^ Globulin.

the complete antibody globulin molecule is set free (vi)
into the blood-stream, leaving the antigen molecule
available to influence the folding of the ends of the peptide
chains of freshly formed globulin molecules. The rate
of antibody synthesis will depend on the strength of the
attraction between the coiled ends of the globulin chain
and the determinant groups on the antigen ; the weaker
the determinant groups the less time will occur before the
attraction is overcome in the ordinary dynamic inter-
change which is constantly going on between molecules.
Antil^ody molecules Avill continue to be produced until
their concentration becomes so great that the antigen

ANTIGENS, HAPTENS, ANTIBODIES, ETC. 423

molecules are almost constantly associated Avith already
modified ends of antibody globulin chains and therefore
not free to influence the production of further antibody
molecules .

This theory accounts for the fact that more than one
sort of antibody can be produced by a single antigen if
that antigen carries more than one determinant group
which can act as a template for the coiling of ends of the
peptide chains. Thus it is known that the heat stable
antigen and the heat labile antigen of the vaccinia virus
are two parts of a single antigen molecule. That is,
antigens may be monovalent or, more usually, since they
are complex molecules, multivalent, in the sense of being
able to combine with one or more antibodies. Thus if an
antigen had two hapten groups, A and B, it would give
rise to divalent antibodies carrying the specific end
groups A' — A', B' — B' and A' — B' as well as monovalent
antibodies, A' and B'. The fact that antibodies formed
as a result of prolonged immunisation usually have a
broader specificity and give wider equivalence zones of
reaction with antigens (see Chapter XXIII) is also
accounted for on this basis. The theory also accounts
for the cross reactions which occur between the antisera
to different antigens since several antigens may contain
a common determinant group. The apparent paradox
that antigens with powerful polar determinant groups
are frequently poor antigens, giving only low concentra-
tions of antibody although of sharp specificity, whilst
antigens with weak polar groups, that is of low specificity,
usually give high titre antisera is explained by the fact
that the stronger the polar group the more firmly will
it attract and hold the complementary antibody group
and therefore the less frequently is the antigen molecule
available to influence the formation of fresh antibody
molecules .

On the basis of this theory Pauling predicted that if
normal glolnilin Avere j^laced under mild denaturing

28

424 BACTERIOLOGICAL CHEMISTRY

conditions, such as heating at 50° to 60° C, or solution in
urea or alkali, which cause the protein chains to unfold,
and the conditions then restored to normal in the presence
of an antigen, the polypeptide chains would refold in a
manner complementary to the antigen ; in other words
antibodies would be formed. He claims, on reasonable
but not rigid evidence, to have produced in vitro in this
way antibodies to 1 : 3-dihydroxybenzene-2 : 4 : 6-tri-jt)-
azophenyl arsonic acid, methyl blue and the specific
polysaccharide of Type III pneumococcus.

The Separation of Antibodies. — Many attempts have
been made to separate antibodies from the inactive con-
stituents of the antiserum. These depend on (1) non-
specific methods and (2) specific methods : —

1 (a) AUemtio7i of the Salt Content. — If serum is half-
saturated with ammonium sulphate the globulin is
precipitated, leaving albumin in solution ; the albumin
is precipitated on complete saturation of the serum with
ammonium sulphate. Globulin can be further split into
euglobulin which is insoluble in distilled water, and into
pseudoglobulin which is soluble in distilled water. Dialysis
of serum to remove all the salts, therefore causes the,
precipitation of euglobulin. Dilution of serum with 9 or
10 volumes of distilled water and slight acidification, for
example by passing carbon dioxide into the solution,
precipitates most of the euglobulin fraction and a small
proportion of the pseudoglobulin. Euglobulin, which
can also be precipitated by one-third saturation with
ammonium sulphate, comprises chiefly ^- and y-globulins
Avhich are the slower moving components in an electric
field (see p. 425). The pseudo -globulin is composed
mainly of the fast moving « -globulin fraction. It has been
found that the antibodies are almost entirely associated
with the glol3ulin fraction, but there is little agreement as
to the distribution between euglobidin and pseudo-
globulin, even when dealing with the same antibody.

ANTIGENS, HAPTENS, ANTIBODIES, ETC. 425

Depending on the source of the antibody, from horse or
rabbit senim, for example, on the intensity and length
of the course of immunisation, and on the exact method
of fractionation, the antibody may be found to be
associated with one or other of the globulins or divided
between them. This is not surprising when it is remem-
bered that the proportions of albumin, pseudoglobulin
and euglobulin which can be separated even from normal
serum are very variable.

(b) Precipitation ivith Alcohol. — By this method, also,
the antibodies are separated with the globulin fractions.
Denaturation of the proteins by alcohol is avoided by
working at 4° C. or lower, or by bringing the alcohol
concentration rapidly above 90 per cent. By pre-
cipitating the antibody from an anti-pneumococcus
serum in the cold with 10 per cent, alcohol in the
presence of N/200-sodium chloride at pH 6-7, Felton
succeeded in removing about 90 per cent, of the inactive
protein.

(c) Adsorption Methods. — Methods similar to those
introduced by Willstiitter for separating enzymes have
also been found effective in separating antibodies. For
instance the antibody to the typhoid bacillus can be
adsorbed on alumina and then eluted with dilute (N/100)
alkali. Diphtheria antitoxin, the flagellar typhoid
antibody and the 0 -antibody of Salmonella enteritidis are
adsorbed by kaolin from which they can be eluted with a
solution of 2 per cent, glycine in 2 per cent, sodium
chloride. The eluted typhoid antibody solution gave
negative reactions for proteins but contained 0-6 mg.
of nitrogen per ml. ; it was not affected by proteolytic
enzymes. Only 15 to 20 per cent, of the antibody could
be recovered in this way, but that which was obtained
was about six times as concentrated as the original
serum .

((/) Electrophoresis. — Antibodies have an isoelectric
puint at about /)H 5-5 and accordingly move in an

426 BACTERIOLOGICAL CHEMISTRY

electric field with the globulins, which have about the
same isoelectric point.

As a result of analysis of serum proteins by the use
of Tiselius' apparatus for electrophoresis it is known
that albumin shows greatest mobility in an electric field
whilst normal globulin separates into three fractions,
a-, j3- and y-, which move progressively more slowly.
All these fractions contain cholesterol, phospholipoids
and carbohydrates (glucosamine and mannose), the a-
and j8-globulins being richest in all of them. The anti-
bodies formed as a result of the injection of polysaccharide-
containing antigens, such as the soluble specific substances
of the pneumococcus or the somatic 0-antigens, are
usually associated with the y-globulin. Tetanus and
diphtheria antitoxins are associated with a new com-
ponent, T, not present in normal serum. The T com-
ponent has a mobility of 2xl0~^ cm. per second per
volt per cm., which is midway between the mobilities of
j3- and y-globulins. The antitoxins to CI. ivelchii, CI.
sordelli and CI. cedeynations also have a T component. The
antitoxins of hsemolytic streptococci. Staphylococcus,
CI. botulinum, CI. histolyticum and CI. septicum contain
both T and y-globulins in varying proportions, althougl.
the amount of y-globulin is usually greater than in
normal serum. So far all antitoxins which have been
tested contain the T component. Diphtheria antitoxin
shows a progressive increase of j3-globulin as the course
of immunisation proceeds, the y-globulin increasing at
first but soon reaching a steady value. The toxin-
antitoxin floccules from y-globulin antitoxin contain
about twice as much nitrogen as do the floccules from
^-globulin. The complexes can be expressed as having
the composition (TA4)n and (TAg),! respectively. y-Glo-
bulin antitoxin combines with toxin more rapidly than
does )8-globulin antitoxin, but the complex is less stable.

2. Specific Methods. — The specific methods depend on
sepai'ating tlie anti])ody fiom tlio inactive part, of tlie

ANTIGENS, HAPTENS, ANTIBODIES, ETC. 427

antiserum by allowing it to react with the corresj^onding
antigen and dissociating the antigen-antibody complex
by an- appropriate means. H^emolysins have been elnted
from sensitised red blood corpuscles with dilute acid or
with glycine ; the eluates contained about 40 to 80 X 10~^
mg. of solid per hsemolytic unit. Agglutinins have been
recovered from sensitised typhoid bacilli by extraction
with dilute alkali, but with considerable loss ; no protein
could be detected, but the nitrogen content was 0-4 mg.
per 100 ml. Ramon dissociated the diphtheria toxin-
antitoxin complex and obtained solutions containing
0-012 mg. of protein per unit of toxin. Northrop has
obtained crystalline diphtheria antitoxin by digesting
the toxin in the floccules with tiypsin and crystallisation
from ammonium sulphate solution. The purified anti-
toxin has molecular weight about 80,000, has an electro-
phoretic mobility of 4x 10~^ cm. per second per volt per
cm., at pH 7-3. It contains about one million antitoxin
units per gram of protein nitrogen. Pneumococci have
been agglutinated by the corresponding antiserum and
the complex dissociated by extraction wdth dilute alkali ;
the eluates were colloidal, did not give the ordinary
protein reactions, and were not attacked by trypsin ; the
antibody was not precipitated by 30 j)er cent, sodium
chloride nor by dilution as w^ould be globulins, but as the
solutions contained only 0-00015 mg. of nitrogen per unit
this lack of reactions is not surprising.

The most satisfactory results on these lines are those
in ^vhich pneumococcus antisera have been precipitated
by the protein-free soluble specific polysaccharides and
the antibodies then separated from the precipitates.
Felton decomposed the precipitates from Types I and II
antisera with calcium or strontium hydroxide solutions
in which the protein is soluble but which give insoluble
precipitates wdth the polysaccharides. The antibody was
then precipitated from solution by dialysis, behaving like
euglobulin ; the activity was destroyed by proteolytic

428 BACTERIOLOGICAL CHEMISTRY

enzymes aiul ))y denatiiratioii ; IM) to 1)5 per cent, uf the
protein (0-002 to 0-006 mg. of nitrogen per unit) in these
preparations could be precipitated by the soluble specific
polysaccharide. Heidelberger and Kendall dissociated
similar precipitates with 15 per cent, sodium chloride
and recovered the antibody by dialysis. The remaining
undissociated precipitate was decomposed by barium
hydroxide or barium chloride, which precipitate the
Types I and III specific polysaccharides, leaving the
antibody in solution ; they succeeded in removing all
the inactive protein in this way, 95 to 100 per cent, of
the nitrogen in the final solutions being precipitated by
the polysaccharides.

The Properties of Antibodies. — The size of the anti-
body molecule is about the same as that of normal
serum globulins, as is shown by sedimentation rates in
the ultra-centrifuge and by filtration through membranes
of known pore size. Normal human serum globulin has
a molecular weight of about 170,000, that of the rabbit
about the same and that of the horse has two fractions,
one with molecular weight about 170,000 and the other
with molecular weight about 900,000. Antibodies to the
pneumococcus in horse serum are associated with the
heavy globulin fraction and have molecular weight
about 900,000, and a diameter about 44 m/x. The rabbit
antipneumococcus globulins have a diameter about
11 m/x and molecular weight about 170,000. This dif-
ference between the two antibodies is not restricted to
size ; the horse antibody is said to be associated with
the pseudoglobulin and the rabbit antibody with the
euglobulin. A further difference lies in their lipoid
content ; if the antisera are extracted with fat solvents,
about 1-3 per cent, of lipoid is removed and they no
longer give precipitin reactions, although their protective
effect in mice is unaltered ; the precipitin properties
can be restored by the addition of lecithin to horse

ANTIGENS, HAPTENS, ANTIBODIES, ETC. 429

antiserum and of kcphaliii to rabbit antisemm, but not

vice versa.

Globulins are usually considered to have approximately
spherical molecules but as a result of comparison of their
behaviour in the ultracentrifuge, and studies of viscosity
and electrophoresis it is becoming obvious that they
must be elliptical or even rod shaped. Serum globulin in
surface films opens up into fibre like molecules very
readily — a state which is associated with denaturation.
The normal ratio of length to breadth seems to lie
between 7 to 1 and 10 to 1, although for horse anti-
pneumococcus globulin the ratio is reported as 20 to 1.

The molecules have a short axis 37 A long whilst the long

axis varies between 270 and 350 A long.

Antibodies are comj)aratively labile substances ; they
are much weakened by heating at 60° to 70° C, and are
rapidly destroyed on boiling. They are more stable to
dry heat (as are most proteins) and are not affected by
cold. Prolonged heating at 57° C. precipitates antitoxins
along with the globulins. The effect of heat mns parallel
with denaturation and is independent of the particular
serum in which the antibody is present. Antibodies are
most resistant to heat at neutrality, the rate of destruction
being increased by acid or alkali. If denaturation of the
protein is hindered by the addition of sodium chloride
(above 2N), glycerol or sucrose or by dilution of the
serum, the destruction of the antibody is retarded. The
resistance of antibodies to heat ajijparently varies from
antibody to antibody ; thus the flagellar antibody of
the hog cholera bacillus will withstand 90° C. for twenty
minutes, whilst the somatic antibody is destroyed at
75° C, and loses half its activity at 65° C. in the same
time. The differences are due, however, not to differences
in heat stability but to the fact that the apparently more
susceptible antibodies form complexes with non-specific
nitrogen compounds, such as albumin, in the serum

430 BACTERIOLOGICAL CHEMISTRY

more readily than do the others ; these complexes com-
bine with antigens, as shown by complement fixation,
but do not flocculate, and thus antibody appears to have
been destroyed by the heating. Antisera to rod shaped
viruses, for example tobacco -mosaic virus or potato -X
virus, behave like the flagellar, H -antibodies, whilst
antisera to albumin and other more " globular " antigens
behave like those to somatic, 0 -antigens. Some anti-
bodies are destroyed at the earliest stages of denatura-
tion, whilst others may withstand complete denaturation
if coagulation is prevented by addition of urea or by
dilution with physiological saline.

The deleterious effect of alcohol on antibodies also
runs parallel with the denaturation of the proteins. Below
the critical temperature of 4° C. there is no denaturation
and no destruction, nor is there if the alcohol concentration
is rapidly brought above 90 per cent. The addition of
acid or alkali accelerates the effect of alcohol.

The Effect of Chemical Changes on Antibodies. —

Antibody globulins, like normal globulins, are only slowly
attacked by trypsin but are much more readily destroyed
by pepsin. Brief treatment with pepsin at pK 4 and 37° C.
causes the antibody globulin of diphtheria antitoxin, for
example, to break into two parts, one of which does not
carry antitoxic activity and is easily denatured and
coagulated by heating at 60° C. for fifteen minutes,
whilst the active part is much more resistant to heat and
remains in solution. By precipitation of the active
fraction by ammonium sulphate after removal of the
easily coagulated fraction by heating, a considerable
concentration and purification of the antibody can be
achieved. During the digestion by pepsin the T com-
ponent (see p. 426) disappears and is replaced by the
slower moving y-globulin. More prolonged digestion
causes further breakdown into fragments which can no
longer flocculate with antigen but can still react with it

ANTIGENS. HAPTENS, ANTIBODIES. ETC. 431

as shown by the fact that such digested antibody inliibits
normal toxin-antitoxin flocciilation. The euglobulin of
Types I and II antipneumococcus horse sera can be
digested by pepsin to give smaller active molecules having
a sedimentation constant of 5-2x10"^^ cm. per second
per dyne, corresponding to a molecular weight somewhat
less than 100,000, and able to combine with twice as
much specific poh^saccharide per mg. of antibody nitrogen
as the original antibody.

The introduction of azo-compounds or of iodo -groups
or the action of formaldehyde lowers the titre of an
antiserum, but is said to sharpen the specificity. Eagle
and his co-workers have shown that coupling pneumo-
coccus antiserum with diazotised atoxyl has a differential
effect on the reactions of the antiserum ; the power of
agglutinating bacteria and of precipitating the specific
polysaccharide ma}^ be lost, but not the power to protect
mice against infection. The action of the azo -compound
proceeds at different velocities for the different mani-
festations of the antibody, but a sufficiently long contact
with the reagent causes a complete loss of all the anti-
body reactions. With the pneumococcus antiserum the
precipitin reaction is lost before the agglutinating power,
protective properties or ability to fix complement. In
the case of diphtheria antitoxin the flocculating property
is lost before the power to neutralise toxin. These
effects are illustrated in Table 30. The addition of 1
part of formalin to a thousand parts of antiseiiim brings
about similar effects. It is suggested that the addition
of just enough formaldehyde to prevent aggregation
increases the solubility of the antibody so that it no
longer i^recipitates at the normal conditions of /jH of
serum.

In Type I pneumococcus antiserum replacement of a
hydrogen atom in the amino group by acetyl, by the
action of ketene, reduces its activity, whilst the action of
formaldeh^'de causes loss of specificity. Esterification

432

BACTERIOLOGICAL CHEMISTRY

of the oarbuxyi groups destroys the activity of tlic anti-
body very readily, indicating that the carboxyl group plays
an important part in the reaction with the antigen.
Substitution in the hydroxyl and amino groups has less
effect.

Table 30

( After Eagle)

Time for

Antibody

Reaction

Loss

Hours

Syphilis reagin

Wassermann and flocculation

001

Diphtheria antitoxin

Ramon flocculation

002

»> ■ -

Protection

0-25

Horse pneiimocoecus antiserum

Polysaccharide precipitation

0-07 to 0-2

,, jj ,,

Agglutination

0-24 to 1-0

,, ,, ,,

Mouse protection

0-5 to 2-0

Rabbit v. horse antiserum

Precipitation

0-25

Horse typhoid antiserum -

Agglutination

24

The antitoxic power of tetanus antitoxin remains
intact after the acetylation of 30 to 40 per cent, of the
amino groups by ketene, but further acetylation causes
a raj)id loss of activity.

The action of ninhydrin is analogous to that of
formaldehyde in that it reacts with the terminal amino
groups of lysine residues : —

CO

CO.CHO

C6H,<

\

C(OH),

alkali

C^^

CgHi

CO
(ninhydrin)

CO.CHO
/
/
\
\
COOH

+ NH2CH2R CVH4<

\

COOH
CO.CHiN.CHaR

COOH

+ H,0

ANTIGENS, HAPTENS, ANTIBODIES, ETC. 433

It has l)et?n sliuwji tJiat t37)]ioi(l H -aggiutiniiis in lioi'se
antisera can be concentrated by treatment with nin-
hydrin and precipitation by acid or salts. The O-
agglutinins of typhoid and the Type I pneiimococcus
agglutinins in horse antisera are partially destroyed by
ninhydrin. In rabbit antisera the apparent tit re of
typhoid H -agglutinins is increased 2 to fourfold by the
action of ninhydrin, whilst the typhoid 0 -agglutinins
and pneumococcus antibodies are not affected, unless the
treatment is prolonged.

Complement

In the various lytic reactions, bacteriolysis and
haemolysis, for example, it has been established that two
serum factors are involved — (1) the comparatively heat
stable immune body which is increased in amount during
immunisation and (2) the very labile complement, or
alexin, which is present in normal as well as in immune
sera and which is not increased in amount during
immunisation.

In the case of haemolysis, Ehrlich considered that the
immune body, haemolysin or amboceptor, had two
affinities, one for the cell and the other for the haptophore
or combining group of the complement. The complem' :it
then acted on the cell causing lysis in virtue of its active
or ergophore group, as Ehrlich called it. Bordet regarded
the immune body as a sensitiser with which the cell formed
a complex which had an affinity for complement and
adsorbed it ; the immune body itself has no visible effect
on the cells ; it is the complem^ent which brings about
the lysis . It has been considered that in the lysis of cells
the complement acts as a proteoljrtic enzyme, but it
difff^rs in one respect at least ; its effects are governed by
the Law of Mass Action, that is, it appears not to act as a
catalyst. Moreover, the cell walls of red blood corpuscles,
for example, are not destroyed nor is the liberated haemo-
globin attacked, which appears to be in contradiction to

434 BACTERIOLOGICAL CHEMISTRY

a proteolytic effect of complement. It seems much more
probable that the lysis is due to an alteration of the
physical state of the cell walls, making them permeable
to the cell contents. This view is rendered the more likely
since " artificial " complements, either colloidal silicic
acid or one made by mixing sodium oleate, methyl
alcohol and calcium chloride, have the same effect,
which would appear to be due to a lowering of the surface
tension.

Complement appears to be quite non-specific ; the
complement in the serum of any animal is effective in all
reactions, but the activity of complement in the sera of
different animals varies appreciably and also varies in
any one animal depending on the system tested ; the
guinea-pig is a particularly good source of complement
for hsemolytic systems.

Complement is very susceptible to heat, being des-
troyed by heating at 56° C. for fifteen minutes, resembling
in this respect many enzymes. It is also very susceptible
to physical and chemical changes, even mechanical
shaking is sufficient to inactivate it. Complement is
destroyed by acid and alkali and by ultra-violet light.
If the activation has not been carried too far it is partially
reversible and standing restores a part of the activity.
This suggests that the inactivation is due to a change
in the colloidal condition, probably to an increased
aggregation of the particles which may spontaneously
disperse again. Complement is almost certainly protein
in nature since it is attacked by trypsin, which destroys
the end-piece (see below) first. It is readily adsorbed by
non-specific adsorbants, even by filtration through a
Berkefeld candle.

The Structure of Complement. — The complex nature
of complement was first demonstrated in 1907 by Ferrata,
who showed that on removing the salts from serum it lost
its complementary activity ; the euglobulin was pre-
ci Imitated and the albumin remained in solution. Neither

ANTIGENS, HAPTENS, ANTIBODIES, ETC. 435

of these fractions alone was active, but on mixing them
the activity was restored. The two fractions were called
mid-piece and end-piece respectively, because it was
found that the albumin (end-piece) was not taken up by
sensitised red blood corpuscles unless the globulin mid-
piece had first been adsorbed. Although the mid-piece
is adsorbed in the absence of the end-piece, it cannot
induce lysis until the latter is also adsorbed. In Ehrlich's
terminology the mid-piece carries the haptophore group
and the end-piece the ergophore group.

Liefmann split complement into mid-piece and end-
piece by diluting the serum ten times with distilled water
and passing carbon dioxide through the chilled solution,
whereupon the mid-piece (euglobulin, with some pseudo-
globulin) was precipitated. He showed that the mid-piece
was adsorbed in the Wasserman reaction but not the
end-piece, which is left quantitatively in solution, and
confirmed the finding that both mid-piece and end-
piece are necessary for hsemolysis. Both mid-piece
and end-piece are heat labile and can be adsorbed by
kaolin.

In 1907 it had been shown that the complementary
power of a serum was removed by treatment with yeast,
and in 1914 Coca showed that the component adsorbed
by yeast was heat stable and that adsorption by yeast
leaves the mid-piece and end-piece in solution. Thus
three factors are present in complement, two heat labile,
the mid-piece and end-piece, and one, the " third com-
ponent," heat stable. Ritz, in 1912, had also elicited the
presence in complement of a heat stable third component
which was inactivated by cobra venom. Whitehead,
Gordon and Wormall in 1925 took the problem a step
further when they showed that zymin (a preparation of
acetone dried yeast) is more effective than fresh yeast for
adsorbing the third component ; by examining the effect
of zymin on separated end-piece and mid-j^iece they
sliowed tliat almost all tlie third component was associated

436 BACTERIOLOGICAL CHEMISTRY

with the mid-piece, although a Uttle was sometimes carried
along with the end-piece. The active substance in zymin
is an insoluble carbohydrate called zymosan. The same
workers in 1926 showed that the addition of small amounts
of ammonia to serum destroyed the complement action
and established that it did so by acting on a " fourth
component." The fourth component is heat stable, but
is not identical with the third component since it is not
adsorbed by zymin. The majority of the fourth com-
ponent is associated with the end-piece. It is specifically
removed by treatment with ammonia, methylamine,
ethylamine, hydrazine or phenylhydrazine, other alkalies
completely destroying all the factors, as does ammonia
if it is allowed to react for too long. All these compounds
can react with the potential aldehyde group of carbo-
hydrates and it is suggested that the fourth component
may be a carbohydrate carried by the pseudoglobulin
end-piece. The claim of Takano that the loss of activity
of complement on treatment with ether or chloroform or
viper venom was due to the lipoid nature of the fourth
component is inaccurate since treatment of dried com-
plement with lipoid solvents has no effect on the fourth
component, the loss of activity of liquid serum being
due to denaturation of the proteins.

The symbols C^l, C'2, C'3andC^4 have been suggested
for midpice, endpice, third component and fourth com-
ponent, respectively. Since the ' serves no apparently
useful purpose it is proposed to omit it here.

In hsemolytic reactions components Cl, C2 and C4
combine with sensitised sheep red blood corpuscles but
C3 does not. Although Cl combines with red cells in
the absence of C4 it is hsemolytically inert unless C4
combines previous to, or simultaneously with, it. Al-
though C3 is not fixed by antibody-red-cell aggregates
it is essential for hsemolysis and acts on the sensitised
cells after fixation of the otlier components^ It appears
to liave catalytic activity.

ANTIGENS, HAPTENS, ANTIBODIES, ETC. 437

In complement fixation reactions almost all the C2
and C4, 75 per cent, of the CI and 25 per cent, of the C3
are removed. Elimination of C3 or C4 does not influence
the fixation of the other components, but inactivation
of CI and C2 by heat inhibits the fixation of C3 and C4.
The inactivation of complement by fixation is due mainly
to the removal of CI, C2 and C4 from the serum. In the
absence of Cl, components C2 and C4 are not fixed,
but Cl is fixed in the absence of all the other components.
Human complement has been fractionated in the
same way and shown to contain components Cl, C2,
C3 and C4 which afe almost, but not quite, identical
with the corresponding fractions of guinea-pig com-
plement.

The present state of our knowledge of the structure of
complement can be summarised as follows (illustrated
diagrammatically by Fig. 10) : —

End-piece. — Albumin and pseudo globulin ; heat
labile ; carries 70 to 100 per cent, of C4 and 20 to
40 per cent, of C3 ; constitutes 0-2 per cent, of
total serum protein ; contains about 10 per cent,
of carbohydrate ; very little adsorbed in com-
plement fixation reactions, but necessary for lytic
reactions.

Encl-picc(

Mid-piece

Heat la})ile

n

Fourth Comjionent

Third Component

Heat stable

Fig. 10

438 BACTERIOLOGICAL CHEMISTRY

Mid-piece. — Euglobulin ; heat labile ; carries 60 to
80 per cent, of C3 and 0 to 30 per cent, of C4 ;
constitutes 0-6 per cent, of total serum protein ;
contains about 3 per cent, of carbohydrate ;
completely adsorbed in complement fixation and
lytic reactions.

Third Component. — Mainly associated with mid-
piece ; heat stable ; specifically adsorbed by
zymin or zymosan.

Fourth Component. — Associated with end-piece ; not
adsorbed by zymin ; specifically inactivated by
ammonia ; heat stable ; necessary for complement
fixation and lytic reactions but not for opsonin
action ; possibly a carbohydrate.

Complement, however, is probably even more complex
than this summary suggests ; the separation of these
components is not always sharp and others may be present.
For instance, complement can be reversibly inactivated
by oxidation with iodine ; reduction with ascorbic acid
or glutathione restores the activity. The activity of
oxidised complement can also be restored by the addition
of complement from which the third and fourth com-
ponents have been removed ; the component inactivated
by oxidation is therefore different from C3 and C4. The
activity of complement is also lost on dialysis, but is
restored by addition of a small quantity of the dialysate.
Since the activity of complement devoid of C3, C4 and
the oxidisable component is not restored by the dialysate
whilst the activity of dialysed complement can be
restored by the addition of dialysate from complement
previously deprived of these components, the dialysable
component must be different from them.

In view of this complexity it has been suggested that
complement is not a definite substance but is really a
particular state of the serum colloids, a state which can
be very easily upset even by such means as mechanical

ANTIGENS, HAPTENS, ANTIBODIES, ETC. 439

shaking. It has been maintained that complement is
only active when the colloid particles are of a given size
and when the correct concentrations of electrol^i^es are
present ; changes in the degree of dispersion or the
proportion of the electrolytes disturb the balance with
loss of activity. If the changes are small they may be
reversible and activity spontaneously regained ; large
changes are irreversible. It has been pointed out that
ail changes which inactivate complement lower the surface
tension, due to an alteration of the colloidal conditions
which may be associated with the globulin, albumin,
lipoid or electrolyte components of the system. This is
borne out to some extent by the production of the artificial
complement (p. 434) which is inactivated by heat and can
be used in the Wassermann reaction like ordinary
complement.

It has been claimed that complement may be a
complex of ascorbic acid, proteins and lipoids, since
guinea-pigs fed on a diet devoid of ascorbic acid (vitamin-
C) produced no complement ; the addition of ascorbic
acid to the diet caused the almost immediate appearance
of complement in the serum, only to disappear again if
the vitamin were withdrawn.

For further reading : —

W. C. Bojd, " Fundamentals of Immimology." Chapters II, III, IV and

VII. Interscience Publishers, Inc. New York, 1943.
C. H. Browning, '' Immunochemical Studies." Constable & Co. London,

1925.
F. M. Burnet, '" The Production of Antibodies." Monographs from the

Eliza and Walter Hall Institute of Research in Pathology and Medicine.

No. 1. Macmillan & Co. Melbourne, 1942.
K. Landsteiner, " The Specificity of Serological Reactions." Harvard

University Press. Cambridge, Mass., 1945.
J. R. Marrack, " The Chemistry of Antigens and Antibodies." Medical

Research Council Special Report No. 230. H.M. Stationery Office.

London, 1938.

440 BACTERIOLOGICAL CHEMISTRY

1

W. T. J. Morgan, " A Conception of Immunological Specificity." J.
Hygiene, 37 (1937), 372.

T. W. B. Osborn, " Complement or Alexin." Oxford University Press.
London, 1937.

L. Pauling, " Theory of the Structure and Process of Formation of Anti-
bodies." J. Amer. Chem. Soc, 62 (1940), 2643.

H. P. Treffers, " Some Contributions of Immunology to the Study of
Proteins," Advances i:i Protein Chemistry, 1 (1944), 70.

H. G. Wells, " The Chemical Aspects of Immunity." American Chemical
Society Monographs. The Chemical Catalog Company. New York,
1929.

CHAPTER XXIII

THE MECHANISM OF AXTIGEX-
AXTIBODY REACTIONS

WHEN an antigen and its corresponding antibody
are brought into contact they react and the
reaction manifests itself in a manner depending
on the nature of the antigen and the conditions prevailing
at the time of reaction. If the antigen is a soluble sub-
stance like a toxin or a serum protein precipitation may
occur ; if the antigen is carried by a bacterial cell or a
red blood corpuscle agglutination may result, or the cell
m.ay be rendered sensitive to lysis or phagocytosis, which
Vv'ill take place if complement or leucocytes are also
present. From the majority of evidence available it
appears that antigen-antibody reactions occur in two
stages. The first stage consists in the direct combination
of the two reagents, the reaction being specific ; this is
followed by a non-specific stage, in the sense that one of
several phenomena, such as flocculation, agglutination,
complement fixation or lysis, may occur depending on
the physical conditions operating at the time. Thus if
bacteria are treated with the corresponding anti-serum
in water instead of saline as the medium, no agglutination
occurs, in spite of the fact that the two reagents have
com^bined, as can be shown by separating the organisms
by centrifugalisation, leaving a supernatant fluid devoid
of anti])ody. If the sensitised deposit is suspended in
saline agglutination immediiitely takes place.

In the first stage of the reaction between cintigeji and
antibody the antigen can be replaced by the corresponding

441

442 BACTERIOLOGICAL CHEMISTRY

hapten, following which, as a rule, the secondary reactions
do not take place. There are certain haptens, however,
notabl}^ the soluble specific substances and polyvalent
haptens, sometimes called complex haptens, which
lead to precipitin reactions, the products of their reaction
with antibodies being insoluble. That reaction between
hapten and antibody in the case of simple (non-pre-
cipitating) haptens also occurs is shown by the fact that an
antibody so treated will no longer react with the complete
antigen, its affinity for the active groups of the antigen
having been satisfied by those of the hapten. Moreover it
has been show^n by direct means that the hapten combines
with the antibody even when no visible reaction occurs.
For example, phenyl-azo-p-benzene-arsonic acid was
coupled with horse serum and an antiserum prepared
against the complex. The antiserum was allowed to react
with the hapten (phenyl-azo-77-benzene-arsonic acid) in a
dialysis chamber and the concentration of free hapten
which passed through the membrane measured. Most of
the hapten was shown to be bound to the antibody. If an
antiserum to the corresponding sulphonic acid was used
instead of the homologous antiserum no combination
occurred and the hapten passed through the membrane
until equilibrium was established, showing that the
reaction was specific. If the antibody was allowed to
react with the antigen before introducing the hapten,
no combination with the latter occurred.

Toxin-antitoxin Reactions. — Toxin and antitoxin com-
bine in definite proportions to give a product which is
non-toxic. Ehrlich defined the unit of antitoxin as that
amount of it which would completely neutralise 100
minimal lethal doses (MLD) of toxin. The L+ dose of
toxin is that amount of toxin which when mixed with 1
unit of antitoxin will kill a 250 g. guinea-pig within
ninety-six hours. The L^ dose is that amount of toxin
which when mixed with 1 unit of antitoxin will just not

ANTIGEN-ANTIBODY REACTIONS 44?>

produce any reaction in the guinea-jjig under standard
conditions. It would be expected that L+ — L^ = l MLD,
but this is not the case, the difference being of the order
of 20 to 50 MLD. Ehrlich explained this phenomenon as
being due to the presence in toxin (T) of an epitoxoid (E)
which is non-toxic and has a less affinity than, but the
same combining power as toxin for antitoxin. If it is
assumed, for purposes of illustration, that crude toxin
contains equal parts of toxin and epitoxoid there will be
nT-f nE units of the mixture. If nA units of antitoxin
were added to it, the non-toxic mixture nTA-f nE would
result. This corresponds to the L^ dose. The addition
of 2nA units of antitoxin would yield the mixture
nTA+nEA, also non-toxic. If more crude toxin, say
1 unit, is added, then the true toxin, having a greater
affinity for antitoxin than the epitoxoid, v,'ould turn
some of the latter out of combination : —

nTA + nEA + T-rE > (n + l)TA-f-(n— 1)EA + 2E.

The addition of more and more toxin will turn out more
and more epitoxoid and combine with the resulting anti-
toxin until there is no epitoxoid-antitoxin complex left,
the mixture remaining non-toxic ; after that free toxin
can accumulate and the mixture becomes toxic, corre-
sponding to the L^ dose. Thus the L+ dose will be
bigger than the L^ dose according to the proportion of
epitoxoid in the crude toxin preparation.

Another phenomenon first observed in connection
with toxin-antitoxin reactions, and since found to occur
with all antigen-antibody reactions, is the Danysz phen-
omenon. The amount of toxin neutralised by a given
amount of antitoxin depends on the way in which the
reagents are mixed. If equivalent amounts of toxin and
antitoxin are mixed rapidly the product is non-toxic : —

nA + nT > nTA.

444 bacterioloOtICal chemistry

If, however, oiie-tliird of the toxin is added to the anti-
toxin, the mixture allowed to stand for some time, then
another third of the toxin added, and after a further
interval the last third of toxin added, the mixture is not
non-toxic, as would be expected, l)ut quite strongly toxic.
Ehrlich explained this by assuming that crude toxin
contained a non-toxic fraction, epitoxonoid, which com-
bined only slowly with the antitoxin. If toxin and
antitoxin were mixed rapidly all the antitoxin com.bined
with the toxin and none wath the epitoxonoid. When only
part of the toxin was added and sufficient time allowed,
part of the epitoxonoid combined with the excess of
antitoxin until finally, after adding all the toxin, an
excess of toxin remained corresponding to the quantity
of antitoxin which had combined with epitoxonoid.

Other explanations than Ehrlich's of the toxin-anti-
toxin reactions have been proposed. Arrhenius and
Madsen, for instance, claimed that the reactions followed
the Law of Mass Action just as any ordinary chemical
reaction. Assuming that w^hen one molecule of toxin
and one molecule of antitoxin combine two molecules of
the toxin-antitoxin complex are formed,

A+T r— ^ 2at,

they carried out experiments in which the amount of free
toxin was measured after the addition of various amounts
of antitoxin. From the Law of Mass Action equation : —

[T] [A] = k [at]^

they calculated the dissociation constant, k, to be 0-0093,
and used this value to calculate the a,moimts of free
toxin which should be present for different concentrations
of antitoxin. The agreement between the observed and
calculated values was very close, as may be seen from
Table 31 :—

antigen-antibody reactions
Table 31

445

Antitoxin
added

Amount of Free Toxin

Calculated.

Observed.

0-00 equi

0-25

0-50

0-75

1-00

1-25

1-50

i-alent

1000

75-0

50-5

27-0

8-8

31

1-7

1000

75-0

48-0

260

9-6

31

1-6

Arrhenius and Madsen pointed out that there is a
close similarity between this reaction and the purely
chemical reaction between ammonia and boric acid.
Ammonia is hsemolytic, and may be compared with the
toxin ; boric acid is non-hsemolytic and destroys the
lytic action of ammonia by formation of ammonium
borate, that is, it is analogous to antitoxin : —

3NH4OH+H3BO4 ^==^ (N'H4)3B04 + 3HoO.

They carried out a similar experiment with these reagents,
using the degree of haemolysis as an indication of the
amount of free ammonia, and again found a close agree-
ment between the observed and calculated values (Table
32), although here the degree of dissociation is consider-
ably higher than in the case of toxin and antitoxin.

Table 32

Amount of Boric
Acid Added

Degree of Haemolysis.

Calculated

Observed

0-00 equivalent

0-33

0-66

1-00

1-33

1-66

2-00

1000
750
60-3
50-3
43-2
37-6
33-5

1000
750
63-0
47-5
43-7
360
33-5

446 BACTERIOLOGICAL CHEMISTRY

Ehrlicli's phenomenon, that L,,. — L^^nMLD, is
readily explained, according to Arrhenius and Madsen,
as being due to the fact that sufficient toxin must be
added to the toxin-antitoxin mixture to overcome the
dissociation. This mechanism, however, does not afford
a simple explanation of the Danysz phenomenon, although
they quote a chemical reaction somewhat, though not
completely, analogous. If sodium hydroxide is added to
monochloracetic acid the sodium salt is formed : —

4CH2CI.COOH + 4NaOH > 4CH2Cl.COONa + 4H2O.

If half the sodium hydroxide is added at 70° C, however,
a second reaction takes place, the chlorine atom being
split off with formation of glycollic acid : —

700 C.
4CH2CI.COOH +2NaOH > CHaOH.COONa +NaCl + H2O +3CH2CI.COOH.

On adding the remaining half of the sodium hydroxide
at the normal temperature, one molecule of monochlor-
acetic acid will remain unneutralised (corresponding to
an excess of toxin, as in the Danysz phenomenon).

This mechanism does not account, either, for the dilu-
tion effects observed in toxin-antitoxin reactions. If
equivalent quantities of toxin and antitoxin are mixed
in strong solution to give a non-toxic mixture, the solution
remains neutral on dilution. If, however, the reagents
are diluted before mixing the neutralisation is not com-
plete, although if the Law of Mass Action applied, the
result should be the same in each case.

Bordet regarded the toxin-antitoxin reactions as
following the ordinary course of colloidal adsorption
reactions, where the amount of adsorption depends on
the relative concentration of the reagents. Thus ac-
cording to Freundlich's adsorption isotherm, x=maC'^,
where a;=the amount adsorbed, (7= the concentration of
the remaining unadsorbed substance, m=the amount of
adsorbant, and a and n are constants. According to this
view there should not be any free toxin or antitoxin in

ANTIGEN-ANTIBODY REACTIONS 447

the mixture, but a complex containing more or less of the
reagents according to the proportions in which they
were mixed. This mechanism affords no explanation of the
specificity of such reactions, nor does it account for the
fact that on heating the toxin-antitoxin complex soon
after its formation dissociation occurs, but heating an
hour or more after the reaction has occurred no longer
has any effect. It has been shown that the adsorption of
toxin by colloidal ferric hydroxide and by antitoxin both
follow the adsorption isotherm, but that the iron complex
is just as toxic as the original toxin, Avhilst the toxin-
antitoxin mixture is non-toxic ; in other words, adsorp-
tion alone is not adequate to explain the neutralisation.
Bordet claimed that the neutralisation occurred as a
secondary reaction after the adsorption was complete.
The Danysz phenomenon is readily explained on the
adsorption hypothesis as being due to the complete
covering of the antitoxin by the toxin first added, toxin
subsequently added not being completely adsorbed.

Precipitin Reactions. — When a soluble antigen, such
as a protein, is allowed to react with its corresponding
antibody precipitation follows. The precipitate consists
of 70 to 90 or 95 per cent, of antibody globulin, most of
the remainder being the antigen, although there is often
a variable amount of lipoid material also present. The
preponderance of antibody in the precipitate is not due
to non-specific adsorption by the antigen-antibody .com-
plex, since added normal serum is not taken up and does
not influence the proportion of antibody and antigen in
the precipitate. Dean and Webb showed that the ratio
of antigen to antibody which gave the quickest reaction
(that is, showed precipitation first) was a constant for
any given system. They showed that, in general, in such
mixtures neither an excess of antigen nor of antibody
could be detected in the supernatant solution by the
addition of one or other of the reagents. It was also

448 BACTERIOLOGICAL CHEMISTRY

8liown thai the optiininn [)i'()poi*tioii of a-ntigen and
antibody, although giving the most rapid precipitation,
did not, as a rule, give the greatest amount of precipitate,
but that this usually occurred in the region of antigen
excess.

In the presence of a considerable excess of either
antigen or antibody no precipitation occurs, and a pre-
cipitate already formed may dissolve on adding an excess
of either reagent. This is the so-called zone phenomenon ;
the precipitation only occurs over a limited range of
antigen-antibody proportions. The variation of antigen-
antibody proportions may be brought about in one of
two ways : (a) the amount of antibody may be kept
constant and the amount of antigen varied, as is the
usual practice in precipitin reactions ; or (5) the amount
of antigen may be kept constant and the amount of
antibody varied, as is the normal procedure in carrying out
agglutination reactions and the toxin-antitoxin reactions.
It has been found that the optimal proportions as deter-
mined by these two methods are not the same, but vary
to an extent depending on the particular system being
studied. For example, in the case of the Ramon floccula-
tion reactions of diphtheria toxin and antitoxin, the
optimum proportion is 1 : 8 as determined by the constant
antibody method, and 1 : 64 as determined by the con-
stant antigen method.

Since definite, chemically pure antigens have become
available a considerable amount of accurate quantitative
work on the composition of the precipitates has been
possible. Dean and Webb estimated that horse serum
as the antigen formed about 12 per cent, of the precipitate
at the optimum proportion, and that it fell to about
6 per cent, in the region of antibody excess. Heidel-
berger and Kendall found that Type III pneumococcus
polysaccharide (which contains no nitrogen) formed

ANTIGEN-ANTIBOr>Y REACTIONS 449

about 2 per rent, of the jjiecipitate at the optimum
proportion and about 0-6 per cent, at the end of the zone
of precipitation in the antibody excess region. The
pseudoglobuhn of horse serum, as antigen, forms about
20 per cent, of the precipitate at the optimum proportion
ratio. When antigens which can be separately estimated
in the precipitate and in the supernatant sohition are
employed it is possible to investigate the composition of
the precipitate also in the region of antigen excess. Thus
Heidelberger and Kendall, using R-salt-azo-diphenyl-azo-
egg albumin, a red dye, as antigen and estim^ating its
concentration colorimetrically, found that the precipitate
contained 13 per cent, of antigen at the optimum pro-
portion, 6-7 per cent, at the limit of antibody excess
and 33 per cent, at the limit of antigen excess.

It will be seen that the precipitates do not have a
constant composition, the amount of antigen present
increasing with the proportion of antigen to antibody
in the mixture. The range of composition varies with
the particular antigen-antibody pair concerned. Marrack
and Smith showed that, using diazotised atoxyl coupled
to crystalline egg albumin or to horse pseudoglobulin,
or using iodo-egg albumin or horse pseudoglobulin as
antigens, the proportion of antigen to total protein in
the precipitate increased with the amount of antigen
added to a given amount of antibody. At the optimum
proportion, addition of either antigen or antibody to the
supernatant solution caused precipitation, showing the
presence of small amounts of antigen and antibody in
the solution, due to dissociation of the precipitate ; at
other ratios only antigen or antibody was to be found in
solution. They showed that the change in composition
of the precipitate was not due to non-specific adsorption
of the antigen or the antibody since the introduction of
normal serum or another antigen caused no change in
the composition of the precipitate ; for example, no azo-
protein was carried down in a toxin-antitoxin flocculation.

450 BACTERIOLOGICAL CHEMISTRY

nor with a pseiidoglol)iilin-aiit-ipseu*ioglobuIiii precipitate.
These authors suggested that the change in composition
of the precipitate with the change in proportion of antigen
and antibody in the reacting mixture might be due to the
formation of a series of compounds of the general formula
AniGn, due to the presence of several combining groups
in the antigen and antibody.

Heidelberger and Kendall developed this idea further,
and applying the Law of Mass Action were able to give
a series of equations which could be used to predict the
behaviour of antigen-antibody mixtures over a wide
range of proportions. They used the nitrogen-free Type
III pneumococcus specific polysaccharide as hapten and
purified antibody preparations consisting of euglobulin
as the antibody. They allowed varying proportions of
the reagents to react, separated the precipitate by centri-
fugalisation and analysed it for nitrogen to determine the
amount of antibody it contained. They showed that the
addition of increasing amounts of the polysaccharide to a
constant amount of antibody caused progressive removal
of antibody from solution until the optimum proportion
(or, as they name it, the equivalence point) was reached,
when neither antibody (A) nor polysaccharide (S) could
be detected in solution. On further addition of S it was
taken up by the precipitate until the end of the zone of
precipitation was reached. Actually the equivalence point
was a zone due to a certain amount of dissociation of the
precipitate, giving traces of A and S in solution. The
extent of the zone depended on the particular specimen
of antiserum which was used and on such factors as tem-
perature, but it was fairly constant for any one system.
The zone may be approached from either the hapten excess
or the antibody excess side, giving limiting values. For
the Type III polysacchari de-antibody system the values
for the nitrogen : polysaccharide ratio were 13-5 for the
antibody excess side and 8-6 for the liapten excess end of

ANTIGEN-AXTJBODY REACTIONS 451

the zone, vntli a mean value corresponding to an equiva-
lence point of 10- 8. The ratio of nitrogen to polysaccharide
in the precipitates depended on the ratio of hapten and
antibody reacting and not on the absolute concentrations.
Tlie range of antibody to hapten ratios in the precipitates
was 40 : 1 at the antibody excess end and 5 : 1 at the
hapten excess end of the zone. The soluble compound
which was formed in the inhibition zone of hapten excess
was shown to contain one more molecule of polysaccharide
than the imm.ediately preceding precipitate. They sug-
gested that at the equivalence point a compound AS was
formed. In the region of antibody excess compounds
AgS, A3S, A4S to AmS were formed progressively depend-
ing on the relative excess of antibody. In the region of
hapten excess iVSg was formed as the insoluble precipitate,
which yielded AS3 as the soluble compound occurring in
the inhibition zone. It was found, however, that this
formulation would not fit the requirements of the Law of
Mass Action, and later they suggested that the precipita-
tion resulted from a series of bimolecular reactions. The
first stage they considered to be the formation of the
compound AS,

A + S ^=i AS

having a composition corresponding to the equivalence
point ratio of antibody to polysaccharide. Since both
hapten and antibody are multivalent with respect to each
other as a result of possessing a number of reactive
groups, made up of repeated units of aldobionic acid and
amino -acids respectively, the compound AS is capable of
combining with more antibody or polysaccharide, which-
ever is in excess in the solution. In the region of antibody
excess the following reactions can occur as a second
step : —

AS+A ^-=^ AS. A.
AS+AS ^=i AS. AS.

452 BACTERIOLOGICAL CHEMISTRY

Similarly these new compounds can take part in a third
stage : —

AS.A + A ^=^ AS.A.A.
AS.AS+A ^=^ AS. AS. A.
AS.A+AS.A. ?=^ AS.A.AS.A.
AS.A+AS.AS ^=-i AS.A.AS.AS
AS.AS+AS.AS ^=^ AS.AS.AS.AS.
This process is supposed to continue on similar lines
until insoluble aggregates are built up and precipitation
occurs. When antibody and hapten are mixed in equiva-
lent quantities the compound AS is believed to polymerise,
by a similar mechanism, to give (AS)n, which has the
composition of the precipitate at the equivalence point.
In the region of hapten excess analogous compounds are
formed until the inhibition zone is reached. The latter
only occurs when there is a considerable excess of hapten,
when all the specific groups of the antibody tend to react
with S rather than with AS or similar complexes, so that
no aggregation to form an insoluble precipitate takes
place. A similar explanation accounts for the non-
precipitation with simple haptens which contain only
one or two reactive groups, leading to the formation of
soluble compounds of the type AH^, which show no
tendency to aggregation. If, as in complex haptens like
the azo-dyes studied by Landsteiner, several reactive
groups are present, compounds of the type AH .AH ^re
formed, and aggregation followed by precipitation can
occur.

Applying the Law of Mass Action to the above
reactions, Heidelberger and Kendall deduced that the
equation

R2S2

Mg. of antibody nitrogen precipitated = 2RS . (1)

held in the region of antibody excess, where R is the ratio
of antibody to antigen at the equivalence point, S is the
amount of polysaccharide added and A is the amount
of antibody-nitrogen precipitated at the equivalence

ANTIGEN -ANTIBODY REACTIONS

463

puiiit. This equation covers the range of precipitate
given by mixtures with ratios of antibody to antigen
from R to 2R,. Similar but more complex equations
were deduced for ratios up to 4R. For the region of
antigen excess the equation

(Ki)2A2

Mg. of S precipitated = 2RiA — . . . (2)

Total S

results, where Rj is the ratio of antigen to antibody at
the equivalence point. Close agreement between the
experimental values and the values calculated from the
above equations was found, as the following examples
for two antibody preparations show (Table 33) : —

Table 33

( After Heidelberger and Kendall)

Antibody

1

2

R

13-6

12-4

A

4-08

1-86

Mg. of Nitrogen Precipitated

Mg. of S Added

Calculated

Found

Calculated

Found

0-02

..,

0-46

0-50

005

1-25

1-22

103

103

0-075

1-40

1-41

010

2-27

2-24

1-65

1-66

0-20

3-62

3-62

0-25

3-96

3-87

K the values for the ratio of nitrogen to polysaccharide
in the precipitate at any two points in the region of
antibody excess are known (by experiment), then the ratio
(R) at the equivalence point can be calculated, since
dividing equation (1) by S the equation

N

= 2R-

R2S

~a"

(3)

454 BACTERIOLOGICAL CHEMISTRY

results, in which all the values except R are known.
Similarly the behaviour of the reagents and the com-
position of the precipitates in the region of polysaccharide
excess can be calculated and the quantitative behaviour
of the serum over the whole range can be predicted.

Kendall has derived the same equation (3) from a
consideration of the number of combining groups available
for combination and the proportion of them which are
actually in combination for varying concentrations of
antigen and antibody.

Heidelberger and Kendall showed that these equations
hold not only for the Type III pneumococcus system
(which is a hapten-antibody system) but also for R-salt-
azo-diphenyl-azo-egg albumin and its antiserum., for
crystalline egg albumin and its antiserum and for the
Type I pneumococcus polysaccharide system. There was
found to be a difference between horse and rabbit pneumo-
coccus antisera, possibly due to the difference in molecular
weights of the globulins, which are 500,000 for horse
globulin and 150,000 for rabbit globulin. For rabbit
antiserum to Type III pneumococcus the value of 211 is
13-5 and for the horse antiserum 32, giving ratios of 85
and 200, respectively, for antibody-protein to poly-
saccharide at the equivalence point. Using the above
values for the molecular weights of the globulins, the
molecular weight of the polysaccharide is thus 1,800 to
2,500, corresponding to 5 to 8 aldobionic acid units.
In the case of the Type I pneumococcus system the values
of 2R are 5-4 and 14-4 for rabbit and horse antisera
respectively, giving values for the molecular weight of
the Type I polysaccharide of 4,400 to 4,500. It can be
calculated from these values that the composition of the
precipitate in the equivalence zone of the Type III system
with rabbit antisera would be from S3 A 2 to S2A ; at the

ANTIGEN-ANTIBODY REACTIONS 455

beginning of the inhibition zone it corresponds to S4A, and
the soluble complex is S5A. With Type III horse antisera
the compounds are S3 A to SeA for the equivalence zone
and SioA for the soluble complex in the inhibition
zone.

The combination of antigen and antibody takes place
as though the molecules behaved as fairly rigid bodies,
the antigens being roughly spherical in shape and anti-
bodies more or less ellipitical, with a ratio of length to
breadth of about 7 to 10 (see page 429). Antigen molecules
are probably multivalent in that they contain several
reacting sites or determinant groups, even when the
determinant groups are all the same. They will usually
have a higher valency the larger the molecule. If Pauling's
conception of antibody formation is correct antibody
molecules are for the most part to be regarded as divalent,
although some monovalent molecules are also formed
(see page 423). If the antigen and antibody are of
about the same size then at the equivalence point the
ratio of the weight of antibody to the weight of antigen
in the precipitate would be about A^/2 where N repre-
sents the valence of the antigen and the limiting values
would be 1 for excess of antigen and N-l for excess of
antibody. When the antigen is much larger than the
antibody the ratio would be less than N/2 at the equiva-
lence point. The value of N, the effective valence of an
antigen, is determined by the number of antibody
molecules which can be packed round the antigen molecule.
If the two molecules are spheres of equal size, 12 antibody
molecules can be fitted into place round one antigen
molecule. If the antigen is larger than the antibody
more molecules of the latter can come into contact with
the antigen and N may be larger than 12 ; if the antigen
is the smallei' molecule then N is less than 12. These
expectations are borne out by experiment, as the values
of A^ in Table 34 show :—

456

bacteriological chemistry
Table 34

Antigen

Molecular Weight

N

Ovalbumin -
Serum albumin
Thyroglobulin
Busycon hsemocyanin

40,000

67,000

700,000

7,000,000

6

6—8

30—40

74

Pauling has suggested that the forces which hold the
complex of antigen and antibody together are a com-
bination of the weak van der Waals forces, which need
very close juxtaposition of the atoms concerned if they
are to be effective ; the weak forces exerted by the
polarisation of one atom by the dipolar character of
another ; and the stronger electrostatic attraction
between positively charged amino groups and negatively
charged carboxyl groups, for example, and the even more
powerful hydrogen bonds. The energy of electrostatic
attractions may be quite considerable, of the order of
5 Calories per gram molecule, if the appropriately charged
groups can come into close apposition. The hydrogen
bond, which results from the attraction of a hydrogen
nucleus from one electronegative atom by the unshared
electron pair of another electronegative atom, depends
for its strength on the degree of electronegativity of the
two atoms, the most electronegative, oxygen and nitrogen,
giving the strongest hydrogen bonds, with an energy of
about 5 Calories per bond. Specificity is due to the
complementary configuration and arrangement of the
groups of atoms in the antigen and antibody molecules
which could form hydrogen bonds, or give rise to electro-
static attraction, and this in turn depends on the size
and dispositions of the areas on antigen and antibody
molecules which could come into contact. If the arrange-
ments of groups of atoms in tlie molecules are such that
the molecules could only come into close contact at a
few 2)olnts, combination would bo weals, wJiereas if the

ANTIGEN-ANTIBODY REACTIONS 457

contact was close over a considerable area then firm and
specific combination would occur. The forces which
hold the molecules together are not themselves specific ;
the specificity depends entirely on the appropriate
distribution of the atoms involved. The method by which
such complementary structures might be built up in
antibodies has been outlined on page 421.

The evidence points to the fact that the second
stage of antigen-antibody reactions, especially in agglu-
tination and precipitin reactions, is also specific. Thus
agglutination carried out with mixtures of two bacteria
and the corresponding antisera normally gives clumps
which contain one type of organism only, and not the
mixture of organisms which would be expected if the
aggregation were non-specific. Similarly, precipitin
reactions carried out in the presence of a heterologous
antigen do not give precipitates containing the second
antigen. It is reasonable to suppose, therefore, that the
second stage, the actual formation of a precipitate, is a
continuation of the first stage until aggregates of sufficient
size to be insoluble are formed. This suggestion forms
the basis of the " framework " or " lattice " theory put
forward by Marrack and elaborated by Heidelberger and
by Pauling and their collaborators.

The framework hj^jothesis requires that antibodies
be at least divalent, otherwise antibody molecules could
not form links between two or more antigen molecules
to give the framework structure illustrated in Fig. 11
which represents the state at the equivalence point.

re- Anfigen
Antibody

Fig. 11.

458 BACTERIOLOGICAL CHEMISTRY

In the region of antibody excess the structure of the
framework would be represented as in Fig. 12, in which
only one valence of many of the antibody molecules

X / \ / \ / \ / \ /

• • e e «

/ \ / \J\ /^\ / \ / \

Fig. 12.
is involved and the precipitate has a higher proportion of
antibody than it has in the equivalence zone. The
proportion of antibody may be so high that the complex
is soluble and give rise to the so-called " inhibition " or
" pro -zone." The formation of precipitates by the
interaction of multivalent haptens and antibodies is
accounted for on the same basis, whilst monovalent
haptens could not give rise to such a framework. Thus
OH R R R

haptens such as R<^ ^OH or H0<(^ ^ <( ^QH,

OH R R

where R represents diazotised arsanilic acid,

— N=n/ X^sOgHa , give precipitates with anti-
serum prepared by injection into rabbits of diazotised
arsanilic acid coupled to protein. Haptens like H0<^ ^r,
containing only one determinant group, do not give pre-
cipitates although they combine with the antiserum.

Boyd suggests that antigen-antibody precipitates do
not arise as the result of building up aggregates by the
framework process but that they result from the pre-
vention of the lyophilic polar groups of antibodies and
antigens from exerting their normal " solubilising "
function. The comparatively large antibody molecules
are visualised as l)eing lield closely together by the
antigen or liapten molecides in such a way that the
uncom])ined polar groups of the antibody are " occluded "

ANTIGEN-ANTIBODY REACTIONS 459

and prevented from contact with solvent molecules, so
that the complex composed of antigen or hapten and a
few antibody molecules becomes lyophobic and insoluble.
He accounts for the observed failure of some multivalent
liaptens to cause precipitation as being due to the too
close proximity of the determinant polar groups so that
there is not room for two or more antibody molecules to
be brought into such positions that their polar groups
are occluded, the complex, therefore, remaining soluble.
If the hapten molecule is larger, for instance if R of the

above formulae is — N=N<Q ^N=N

the polar groups are further apart, steric hindrance is
less, more antibody molecules may be able to react with
occlusion of a larger number of their polar groups and
consequent precipitation.

The evidence for making a definite decision between
the " framework " and " occlusion " hypotheses of
precipitation is at present not adequate, but the balance
seems in favour of the former.

Pauling and his collaborators have proposed equations
to account quantitatively for the reactions between
multivalent antigen molecules and divalent antibody
molecules, assuming equilibrium between antigen-
antibody soluble complexes of the types AB, A^B and
AB^ and the precipitate ABp, where A represents
antigen and B represents antibody. The resulting
expression has been shown to hold for relatively simple
systems, such as those composed of divalent antigen
and divalent antibody.

Hershey has also proposed a rather elaborate series
of equations which account fairly well for quantitative
findings and also enable deductions to be made as to the
results to be expected from alteratioixs in the systems.

The Reacting Groups in Precipitin Reactions. — At
present comparatively little is loiown of this subject, but
some data are available. Chow and Goebel showed that

460 BACTERIOLOGICAL CHEMISTRY

if the amino groups of a pui'ifiecl aiiti))0(ly to Type 1
pneumococcus were acetylated by the action of ketene,
the antibody lost much of its power of reacting with the
Type I polysaccharide. If the amino groups are treated
with formaldehyde with introduction of methylene
groups (see p. 417) the power of reacting is completely
lost. Reconstitiition of the amino group by treatment of
the methylene derivative with dilute acid at ^^H 4-0 for
several days at 0° C. restores the activity. It is, there-
fore, very probable that the amino groups of this antibody
are involved in its reactions. It is also very probable
that the strongly polar carboxyl group of the uronic
acid or aldobionic acid of the hapten is involved, since
esterification of the Type I polysaccharide with diazo-
methane causes complete loss of activity, although it
must be remembered that one hydroxyl group and the
amino group are also methylated by this procedure.
Alkaline hydrolysis, however, removes the ester methyl
group but not those attached to the hydroxyl and amino
groups, with almost complete restoration of the hapten
activity, wliich suggests that it is the carboxyl group
which is largely responsible for the action, although the
other two groups contribute to some extent.

It has been suggested that the prominent polar
groups in antigens which determine their specificity
actually fit into hollows or sockets in the antibody
molecules which were modelled round the determinant
groups either by the folding mechanism described by
Pauling or by the actual selection of appropriately shaped
amino-acid groups as in the theories put forward by
Breinl and Haurowitz and by Mudd. It is considered
that electronegative groups like the carboxyl ion on an
antigen molecule would be matched by electropositive
groups such as an amino group on the antibody molecule.
This is in keeping with the fact that it is the terminal
parts of determinant groups which have most influence
on specificity, as with the peptide haptens mentioned on

ANTIGEN-ANTIBODY REACTIONS 401

]iat2;e 408 and tlie ('ai])()hy(lrate determinants dcscril)cd
on page 411 .

Heidelberger and Kendall showed that partially
hydrolysed Type III polysaccharide, giving products with
molecular weights between 550 and 1,800, gave reactions
with horse antisera but not with those from the rabbit ;
the aldobionic acid itself gave no reaction with either
antiserum. This suggests that definite groups and not
the molecule as a whole are concerned. Type III poly-
saccharide which has been methylated by dimethyl
sulphate and sodium hydroxide (reagents which do not
esterify the carboxyl group) reacts with the horse anti-
serum, precipitating about two -thirds of the nitrogen of
the antibody. The remaining one-third of the nitrogen
can only be precipitated by unmethylated polysaccharide
in which the hydroxyl groups are free. This confirms that
different groupings of a hapten may act independently
in stimulating the production of antibodies and in reacting
with. them. That is, a single antigen may j^roduce more
than one antibody with specificities corresponding to
different determinant groups in the antigen.

For further reading : —

W. C. Boyd, " Fundamentals of Immnnology." Chapter VI, Interscience
Publishers, Inc. New York, 1943.

K. Landsteiner, " The Specificity of Serological Reactions," Harvard
Univer&ity Press. Cambridge, Mass,, 1945.

J. R, Marrack, " The Chemistry of Antigens and Antibodies." Medical
Research Council Special Report No. 230. H. M. Stationery Office,
London, 1938.

W, T. J. Morgan, " A Conception of Immunological Specificity." J. Hygiene,
37 (1937), 372.

L. Pauling, " Theory of the Structure and Process of Formation of Anti-
bodies." J. Amer. Chem. Soc. 62 (1940), 2643,

L, Pauling, D, H. Campbell and D, Pressman, " The Nature of the Forces
between Antigen and Antibody and of the Precipitation Reaction,"
Physiological Reviews, 23 (1943), 203,

L. Pauling and co-workers, " The Serological Properties of Simple Sub-
stances," J. Amer. Chem. Soc, 64 (1942), 2994, 3003, 3010, 3015;
65 (1943), 728,

H, G. Wells, " The Chemical Aspects of Immunity," American Chemical
Society Monograph Series. The Chemical Catalog Company. New
York/ 1929.

APPENDIX I

THE ISOLATION AND IDENTIFICATION
OF METABOLIC PRODUCTS

IN carrying out investigations into the metabolic
processes of micro-organisms it is obviously advan-
tageous to start with substances of laiown composition
only ; for this reason synthetic media should be employed
whenever possible for the growth of the organisms,
since the use of broth, meat extract and similar materials
introduces mixtures of substances of unknown nature and
amount. For ease in tracing the course of the metabolic
changes a single source of carbon, such as a sugar or
other suitable substance, and of nitrogen, such as an
ammonium salt, should be used.

The fermented solution is usually an aqueous mixture
which will contain the organism, possibly calcium car-
bonate (which may have been added to maintain
neutrality) and insoluble organic calcium salts, volatile
neutral and basic substances, volatile acids, non- volatile
acids and non-volatile neutral or basic products. Gaseous
products are also often produced, the usual gases en-
countered being carbon dioxide, hydrogen and methane ;
these may be collected during the course of fermentation
in the usual way over water after removal of the carbon
dioxide by baryta or soda-lime.

The course of fermentation can be followed by observ-
ing the disappearance of the carbon source ; glucose, for
instance, can be estimated by the optical rotatory power
of the solution or by the reducing power as determined
by the Shaffer-Hartmann method, for example, or by

462

METABOLIC PRODUCTS 463

Willstatter and 8chudel\s alkaline iodine nietliud. If
more than one of these methods is employed any dis-
crepancy between them may afford valuable information
as to the products formed. Thus a low value for glucose
by the optical method as compared with the reduction
method would indicate the formation of a laevo -rotatory
substance among the products, whilst a high value
would suggest the presence of a dextro-rotatory product.

The fermentation mixture will consist of two parts,
insoluble and soluble. The insoluble constituents are first
removed by centrifugalisation or filtration. Difficult
filtration may often be improved by shaking the meta-
bolism solution with kieselguhr and filtering through a
thin layer of kieselguhr on filter paper prepared by
pouring a suspension of kieselguhr on to a wet paper in a
large Buchner funnel (about 1 g. is adequate for a funnel
15 to 20 cm. in diameter). Filtration is usually effective
for fungi, which form a more or less continuous felt of
mycelium, less effective for yeasts and usually ineffective
for bacteria for which centrifugalisation is usually
employed.

The insoluble residue will contain the organism, such
products as may be insoluble in water (usually only
occurring in mould fermentations) and insoluble calcium
salts and excess of calcium carbonate, if the latter has
been employed. The organism is, of course, usually
known but should be examined in order to detect any
possible contamination with unwanted organisms. The
metabolic products of bacteria and yeasts are almost
invariably soluble in water unless the normal course of
fermentation has been interfered with by addition of
calcium carbonate or fixative agents such as sulphite,
dimedon, or /3-naphthylamine. In the latter cases the
residue will naturally be examined for the presence of
the expected products. If the fermentation has been
carried out in the presence of calcium carbonate the
production of calcium oxalate is possible, especially in

464 BACTERIOLOGICAL CHEMISTRY

the (-ase of mould action. It may be detected by .sohition
of the residue in hydrochloric acid, filtration, addition of
ammonia and acidification with acetic acid ; a precipitate
usually indicates the presence of oxalic acid, although
calcium fumarate or succinate are also possible. If the
latter are suspected the solution of the salt in hydrochloric
acid is extracted with ether in which fumaric and succinic
acids are soluble. Precipitation of the calcium salt on
boiling the aqueous solution usually denotes the presence
of citrate.

The solution obtained by filtration or centrifugalisation
is submitted to the following treatment : —

1. Volatile Neutral Products. — The solution is neu-
tralised with sodium carbonate and a portion distilled
off into a receiver cooled in ice. The distillate is tested
for :

(a) Acetaldehyde, by addition of 2 : 4-dinitro -phenyl-
hydrazine hydrochloride ; .the yellow crystalline dinitro-
phenylhydrazone is recrystallised and its melting point,
162° C, determined.

(6) Acetone, by making the 2 : 4-dinitro-phenylhydra-
zone, recrystallising and determining the melting point,
125° C. ; by saturation of the solution with solid
ammonium sulphate, addition of concentrated ammonia
solution and two or three drops of a freshly prepared
solution of sodium nitroprusside, when a purple colour
indicates the presence of acetone ; the iodoform test
may be carried out, using ammonia instead of sodium
hydroxide.

(c) Ethyl alcohol, by the iodoform test ; or by oxida-
tion to acetaldehyde ; 25 ml. of the distillate is treated
with 0-1 g. of potassium dichromate and 0-5 ml. of 20
per cent, sulphuric acid and distilled slowly to collect
about 5 ml. of distillate which is tested for acetaldehyde ;
a green colour of the residue is confirmatory.

METABOLIC PRODUCTS 465

{<!) Butyl and higher alcohols may ho detected )>y
their odour ; the solution is redistilled and the distillate
saturated with sodium chloride when the alcohols separate
as oily drops. If present in sufficient quantity they can
be submitted to fractional distillation and determination
of the boiling points.

(e) Esters may be detected by their odour, and by
fractional distillation and determination of the boiling
points.

2. Volatile Acids. — The residue from the distillation
of the volatile neutral products is made acid with sul-
phuric acid or, preferably, j)hosphoric acid (to prevent
charring) and distillation continued. The distillate is
tested for :

(a) Formic acid, by reduction of ammoniacal silver
nitrate and the usual tests.

(6) Acetic acid, by the odour ; by the production of
the red-brown colour with ferric chloride.

(c) Propionic acid.

(d) Butyric acid. The latter two acids may be separ-
ated and determined by use of Duclaux's " distillation
ratios " (see Beitrand and Thomas' " Manipulations de
Chimie biologique "). Information as to the approximate
composition of the mixture may be obtained by deter-
mining the acid equivalent by titration with sodium
hydroxide using phenolphthalein as indicator, or by
formation of the silver salts.

3. Non-volatile Acids. — If the fermentation has been
carried out in the presence of calcium carbonate the non-
volatile acids may be precipitated from the metabolism
solution as the calcium salts by the addition of four
volumes of alcohol.

(a) Bacterial products. The precipitate will probably
be calcium succinate or lactate. The salt is acidified to
Congo red with sulphuric acid and the solution extracted

466 BACTERIOLOGICAL CHEMISTRY

with ether. Evaporation of the ether sohition to dryness
gives a crystalline residue of succinic acid or a liquid
residue of lactic acid. Succinic acid may be identified by
its melting point (183° C.) and acid equivalent. Lactic
acid may be detected by evaporation to dryness, solution
of the residue in alcohol, addition of concentrated sulphuric
acid and a drop of dilute copper sulphate solution, heating
on a boiling- water bath for a few minutes, cooling and
addition of an alcoholic solution of thiophen ; a cherry-
red colour indicates the presence of lactic acid. Lactic
acid may also be isolated and identified as the crystalline
zinc salt. Lactic acid may also be found among the
volatile acid products since it is somewhat volatile in
steam. It is not completely precipitated as the calcium
salt by 80 per cent, alcohol, the salt being somewhat
soluble.

The test for 'pyruvic acid may also be described here.
With the sodium nitroprusside test, as carried out for
acetone, a vivid blue colour is given by pyruvic acid.
A second test is the addition of a few drops of an alcoholic
solution of guaiacol followed by concentrated sulphuric
acid to give a separate layer ; a carmine -coloured ring
at the junction indicates the presence of pyruvic acid.

(6) Mould products. Oxalic, citric, gluconic, fumaric,
succinic and malic acids are the most common acid
products of fungi. The crystalline calcium salts may be
obtained. Oxalic acid may be tested for as described
above (p. 464). Gluconic acid may be readily character-
ised as its phenylhydrazide (m.p. 200° C). The calcium
salt of citric acid is less soluble in hot than in cold water,
and is precipitated from fairly strong solutions on boiling.
Citric acid may be identified by esterification with methyl-
alcoholic hydrochloric acid solution when the trimethyl
ester, m.p. 78-5° C, is formed, or by conversion into the
amide, m.p. 207° C. Fumaric acid may be extracted with
ether, crystallised and identified by its sublimation at
about 200° C. and acid equivalent. Malic acid is best

METABOLIC PRODUCTS 467

identified by esterification and fractionation of the esters.
Some acids, for example spiculisporic acid (see p. 286)
from P. spiculisporum, are precipitated on acidification
of the metabolism solution with hydrochloric acid.

4. Non-volatile Neutral Products. — Polysaccharides,
if present, will, in general, have been precipitated with the
calcium salts of the non-volatile acids on addition of four
volumes of alcohol to the metabolism solution. They may
be separated from the calcium salts by solution of the
precipitate in water, acidification with hydrochloric acid
and again precipitating with alcohol, the free acids being
soluble in 80 per cent, alcohol. Neutralisation of the
solution, freed from the polysaccharides, with calcium
carbonate will then result in the precipitation of the
calcium salts.

The residue remaining after the distillation of the
volatile acids (see p. 465) is neutralised and the solution
evaporated to dryness, preferably under reduced pressure,
and the residue repeatedly extracted with hot absolute
alcohol. The alcoholic solution is evaporated to dryness
and the residue once more extracted with alcohol.

(a) Mannitol, if present, crystallises out on cooling
as long needles, m.p. 168° C. It may be characterised by
formation of the tribenzylidene derivative, m.p. 224° C.

(6) Addition of four or five volumes of ether to the
alcoholic extract precipitates glycerol, trimethylene glycol
or butylene glycol, which can be separated by distillation
in vacuo. Glycerol and trimethylene glycol can be identi-
fied also as the l^enzoyl or phenylurethane derivatives,
and butylene glycol as the phenylurethane derivative.

Acetoin may be detected in the metabolism solution
by the Voges-Proskauer reaction or by O'Meara's modifica-
tion (addition of creatine).

5. MiscelUuieoas Products. — The addition of fei'ric
chloride to a metabolism solution (particularly useful in

468 BACTERIOLOGICAL CHEMISTRY

the case of mould fermentations) may give rise to a range
of colours : —

(a) Yellow colours (canary to orange yellow) usually
indicate the presence of hydroxy-acids, such as gluconic,
citric, malic or lactic acids.

(b) Blue or violet colours usually indicate phenolic
acids .

(c) An intense blood-red colour is given by kojic acid
(see p. 294).

{d) An intense green or black-green is given by citro-
mycetin.

(e) A rusty brown-coloured precipitate soluble in
excess of ferric chloride to give an iodine brown-coloured
solution is given by citrinin (see p. 159).

The presence of unsaturated products may be detected
by the addition of bromine water, which is decolorised in
their presence.

I

I

APPENDIX II
SYNONYMS OF MICRO-ORGANISMS

T

HE official name according to Bergey " Manual of
Determinative Bacteriology " 5th Edition is given in
ordinary type ; synonyms are given in italics : —

Acetobacter aceti

ascendens
pasteurianum

suboxydans -
xylinum
Achromobacfer radiobader
Actinomj'ces antibioticus
coelicolor

lavendulse -
violaceus
ivaksmanii -
Aerobacter aerogenes

cloacse -
indologenes
levanicum
Alcaligenes fsecalis -

radiobacter

Azotobacter agile

chroococcuiii

Bacillus a - - -

aceti -
acetic us

aceto-cthvliciis -
acidif leans lomjisnim a-'

Bacillus aceti, B. aceticus, Bacterium

aceti, Mycoderma aceti.
Bad. ascendens
Bact. pasteurianum, Mycoderma yastear-

ianum

Bact. xylinum
Alcaligenes radiobacter

Ad. violaceus. Act. waksmanii, Strepto-
thrix coelicolor

Act. coelicolor
Act. coelicolor
B. cerogenes, B. lactis mrogenes, Bact.

cerogenes, Bact. lactis, Bact. lactis

cerogenes
B. cloacce
Aer. cloacse

B. alcaligenes, B. fcecalis alcaligenes,
Bact. alcaligenes

Achromobacter radiobader, B. radio-
bacter, Bact. radiobacter, Rhizobium
radiobacter

Azotobacter vinelandii

Azotobacter beijerinckii, B. azotobacter,
B. chroococcus

Lactobacillus casei

Acetobacter aceti

Acetobacter aceti

IVobably idcnticjil witli li. iii.uci'.vns

Tjactobacilhis delbriickii

469

470

BACTERIOLOGICAL CHEMISTRY

Bacillus acidipropionici
acidophilu s
cerogenes
certrycke
alcaligenes -
aminovorans
amylobacter
anthracis -
azotobacter -
bulgaricus -
butylicus
butyricus -
casei a

casei Y - - -
cellulosce dissolvens
chlororaphis
chroococcus -
cloacce

coli - - - -
delbrilckii -
diphtherice -
dysenterice Flexner
dysenterice Shiga
enteritidis -
fluorescens liquefaciens
friedldnderi
gramdobacter pectino-

voriim
influenzce -
lactis -
lactis acidi -
lactis cerogenes
lactis pituitosi

Propionibacterium pentosaceuin

- Lactobacillus acidophilus

- Aerobacter serogenes

- Salmonella typhimurium

- Alcaligenes fsecalis

- Clostridium butyricum

- Azotobacter chroococcum

- Lactobacillus bulgaricus

- Clostridium butyricum

- Clostridium butyricum

- Lactobacillus casei

- Lactobacillus brevis

- Clostridium dissolvens

- Pseudomonas chlororaphis

- Azotobacter chroococcum

- Aerobacter cloacae

- Escherichia coli

- Lactobacillus delbriickii

- Corynebacterium diphtherise

- Shigella paradysenterise

- Shigella dysenterise

- Salmonella enteritidis
Pseudomonas fluorescens
Klebsiella pneumoniae
Clostridium acetobutylicum

macerans -
mesentericus
viethanicus -
ynorgani
mycoides -
nitrobacter -
oligocarbophilus
pantotrophus
paratyphosus A
paratyphosus B
pneumoniae
polymyxa -
prodigiosus •
prof e as

protens vulgaris
pyoryaneiis -
radicirola. -

• Haemophilus influenzae

- Lactobacillus lactis

- Aerobacter aerogenes

- Bact. pituitosum

- Mycobacterium leprae

Methanomonas methanica
Proteus morgani

Nitrobacter

Carboxydomonas oligocarbophila
Hydrogenomonas pantotropha
Salmonella paratyphi
Salmonella schottmiilleri
Klebsiella pneumoniae

Serratia marcescens
Proteus vulgaris
Proteus \iilgaris
IVudomoiias acriigiuosa
Mliizobiimi Icgiiminosaruiu

SYNONYMS OF MICRO-ORGANISMS

471

Bacillus radiobader -

subtilis

tetani -

thermocellulyticus

tuberculosis

tumefaciens

typhosus

violaceus

vulgaris

welchii

xerosis
Bacterium aceti

acidipropionici a

cerogenes -

certrycke -

ceruginosum

alcaligenes

ascendens

coli

dysenterice Flexner

dysenterice Shiga

enteritidis

enteritidis Breslau

flexneri -

fluorescens

friendldnderi -

gluconicum

influenzce

lactis

lactis

lactis cerogenes

morgani -

nitrobacter

nitrosomonas -

paratypliosum A

2)aratyphosum B

pasteurianum -

pituitosum

jineumonicB

jyrodigiosum

pyocyaneum

radicicola

radiobacter

shigce

tumefaciens

typhosum

violaceum

vulgare

xerosis -

xylinoides

xylinum •

Alcaligenes radiobacter

Clostridium tetani
Clostridium cellulyticum
Mycobacterium tuberculosis hominis
Phytomonas tumefaciens
Eberthella typhosa
Chromobacterium violaceum
Proteus vulgaris
Clostridium perfringens
Corynebacterium xerose
Acetobacter aceti
Propionibacterium freudenreichii
Aerobacter serogenes
Salmonella typhimurium
Pseudomonas aeruginosa
Alcaligenes fsecalis
Acetobacter ascendens
Escherichia coli
Shigella parad3^senteri8e
Shigella dysenteriae
Salmonella enteritidis
Salmonella typhimurium
Shigella paradysenterise
Pseudomonas fluorescens
Klebsiella pneumoniae

Haemophilus influenzae
Aerobacter aeerogenes
Streptococcus lactis
Aerobacter serogenes
Proteus morgani
Nitrobacter
Nitrosomonas
Salmonella parat3'phi
Salmonella schottmiilleri
Acetobacter pasteurianum
B. lactis j^itxitosi
Klebsiella pneumoniae
Serratia marcescens
Pseudomonas aeruginosa
Rhizobium leguminosarum
Alcaligenes radiobacter
Shigella dysenteriae
Phytomonas tumefaciens
Eberthella typhosa
Chromobacterium violaceum
Proteus vulgaris
Corynebacterium xerose

Acetobacter xylinum

31

472

BACTERIOLOGICAL CHEMISTRY

Beggiatoa

Betahacterium breve -
Betacoccus arabinosaceus

bovis
Borrelia recurrentis -

Brucella abortus

Lactobacillus brevis
Leuconostoc mesenteroides
Leuconostoc dextranicum
SpirochcBta recurrentis
Spironema recurrentis

Carboxydomonas oligocarbophila

Cellfalcicula mucosa

Cellulobacillus myxogenes -

Cellulomonas -

Chlamydothrix ferruginea •
ochracea

Chromobacterium iodinum

prodigiosum -'
violaceum

Clostridium acetobutylicum

botulinum
butyricum

cellulosolvens
cellulyticum -
dissolvens
fossicularum -
methanigenes
perfringens -
sporogenes
tetani -
thermocellum
welchii -

Corynebacterium diphtherise
xerose -

Crenothrix polyspora

B. oligocarbophilus
Cellulobacillus mucosus
Pseudomonas myxogenes

Gallionella ferruginea
Leptothrix ochracea

Serratia marcescens
B. violaceus, Bact. violaceum
B. granulobacter pectinovorum,
CI. acetonigemim

B. amylobacter, B. butylicus, B. butyricus,
Granulobacter butyricum, Or. pectino-
vorum, Gr. saccharobutyricum, Plec-
iridium pectinovorum

B. thermocellulyticus
B. cellulosce. dissolvens

B. welchii, CI. welchii

B. tetani

CI. perfringens

B. diphtherice

B. xerosis, Bact. xerosis

Didymohelix ferruginea
Diplococcus gonorrhcece
pneumoniae

Gallionella ferruginea
Neisseria gonorrhoese
B. pneumonice, Bact. pneumonice. Sir.
pneumonice, Pneumococcus

Eberthella typhosa -
Enterococcus fcecalis -
Escherichia coli

B. typhosus, Bact. typihosum
Streptococcus fsecalis
B. coli. Bad. coli

Flavobacterium brunneum
suaveolens
Flexner's bacillus

Shigella paradysenterise

SYNONYMS

Gallioiiella ferrusinea

OF MICRO-ORGANISMS

Chlamydothrix ferruginea

Didymohelix ferruginea

Gdrtiier^s bacillus - - - Salmonella enteritidis

Granulobader hutyricum - - Clostridium butyricum

pedinovorum - Clostridium butyricum

saccharohutyricum Clostridium butyricum

473

Haemophilus canis - - - H. hsemoglobinophilus

ducreyii

haemoglobiuophilus H. canis

influenzae - - B. influenzce, Bact. influenzce

parainfluenzse

Hydrogenonionas pantotropha - B. pantotrophus

Klebsiella pneumoniae

B. friedldnderi, B. pneumoniae, Bact.
friedldnderi, Bact. pneumonice, Klebsi-
ella friedld/ider, Friedldndefs bacillus

Lactobacillus acidophilus
arabinosus
brevis -

bulgaricus

delbriickii -
lactis -
mesenteroides
pentoaceticus
pentosus
plantarum -

Lactococcus dextranicus
Leptospira biflexa -
Leptothrix ochracea
Leuconostoc citrovorum -
dextranicum

mesenteroides

B. acidophilus

Possibly identical with L, plantarum

B. casei y, Betabacterium breve. Possibly

identical with L. pentoaceticus
B. bulgaricus, Thermobacterium bulgari-

cum.
Bacillus a, B. casei a, Streptobacterium

casei
B. acidificans longissimus, B. delbriickii
B. lactis acidi, Thermobacterium lactis

Possibly identical with L brevis
Possibly identical with L. plantarum
Streptobacterium plantarum. Possibly

identical with L. arabinosus and

L. pentosus
Leuconostoc dextranicum
Spirochceta biflexa
Chlamydothrix ochracea
Streptococcus citrovorus
Betacoccus bovis, Lactococcus dextranicus.

Streptococcus paracitrovorus
Betacoccus arabinosaceus

Methanobacterium aliphatica -
aliphatica-
liquefaciens
omelianski -
INIethanonionas methanica
Methanosarcina methanica

B. methaaicus
Sarcina methanica

474

BACTERIOLOGICAL CHEMISTRY

- Neisseria gonorrhoeae

- Nitrosococcus

Micrococcus gonorrhoece

lysodeikticus
nitrosus
urese
Microspira agar liquefaciens
Mycobacterium avium
leprse

paratuberculosis
phlei
tuberculosis

bovis
tuberculosis

hominis
Mycoderma aceti

jpasteurianuni -
Myxococcus ....

Neisseria gonorrhoeae

Nitrobacter ....
Nitrosococcus ... -
Nitrosomonas ... -

Pasteurella . . . -
Phytomonas tumefaciens -

Plectridiimi pedinovorum -
Proactinomyces
Propionibacterium arabinosum -

freuden-

reichii

pentosaceum
Proteus morganii

vulgaris

Pseudomonas aeruginosa -

chlororaphis
fluorescens -

mj^xogenes -
j)yocyanea -
tumefaciens

Khizobium Icguminosarum

radiobader
Rhodubacillus pahistris
Rhodococcus cajisulatus
Rhodorrhagus capsulatus -
Rhodospirillum rubrum -
Rhodovibrio parvus -

Vibrio agar liquefaciens
Avian tuberde bacillus
B. leprce. Leprosy bacillus
Johne's bacillus

Bovine tubercle bacillus

B. tuberculosis. Human tubercle bacillus
Acetobacter aceti
A. pasteurianum

Diplococcus gonorrhoece. Micrococcus

gonorrhoece,
B. nitrobacter. Bad. nitrobacter
Micrococcus nitrosus
Bad. nitrosomonas

B. tumefaciens, Bact. tumefaciens,

Pseudomonas tumefaciens
Clostridium butyricum

Bact. acidi j^ropionici a

B. acidi propionici

B. morgani, Bact. morgani, Morgans
bacillus

B. proteus, B. proteus vulgaris, B
vulgaris. Bad. vulgare

B. pyocyaneus, Bact. ceruginosum, Bact.
pyocyaneum, Ps. pyocyanea

B. chlororaphis

B. fluorescens liquefaciens, Bact. fluores-
cens

Cellulobacillus myxogenes

Ps. aeruginosa

Phytomonas tumefaciens

B. radicicola, Bact. radicicola, Rhizobium

radicicolum
Alcaligenes radiobacter

Rhodorrhagus capsulatus
Rhodococcus capsulatus
Spirillum rubrum

SYNONYMS OF MICRO -ORG ANISMS

475

Sarcina aurantiaoa -
lutea -
methanica -

Salmonella enteiitidis

paratyphi
schottmiilleri -
typhimurium -

Serratia marcescens -

Shiga s bacillus
Shigella dysenteriae -

paradysenterise

Spirillum rubrum
Spironema recurrentis
Spirochceta biflexa

recurrentis
Staphylococcus albus
aureus
citreus

pyogenes albus
Streptobaderium casei

plantar um
Streptococcus acidi lactici
citrovorus -
cremoris
fsecalis
hcemolyticus
hollandicus
lactis -
lactis B
paracitrovorus
pyogenes -
salivarius -
viridans
Strepiothrix ccelicolor
Sulphomonas -

Methanosarcina methanica, Zymosarcina
methanica

B. enteritidis, Bact. enteriiidis, Gartner's
bacillus

B. paratypjhosus A, Bact. paratyphosum A

B. paratyphosas B, Bact. paratyphosum B

B. certrycke, Bact. certrycke, Bact. en-
teritidis Breslau

B. prodigiosus, Bact. prodigiosum,
Chromobacterium prodigiosum

Shigella dysenteriae

B. dysenierice Shiga, Bad. dysenterice

Shiga, Shiga's bacillus

B. dysenterice Flexner, Bad. dysenteriae
Flexner. Flexner s bacillus

Rhodospirillum rubrum

Borrelia recurrentis

Leptospira biflexa

Borrelia recurrentis

Staph, pyogenes albus

Staph, albus

Lactobacillus casei

Lactobacillus plantarum

Str, lactis

Leuconostoc citrovorum

Str. hollandicus, Str. lactis B

Enterococcus fcecalis

Str. pyogenes

Str. cremoris

Bact. lactis, Str. acidi lactici

Str. cremoris

Leuconostoc dextranicum

Str. hcemolyticus

Str. viridans

Str. salivarius

Actinomyces coelicolor

Thiobacillus

Thermobaderium bulgariciun

lactis
Thiobacillus denitrificans -
thio-oxidans -

thioparus
Thiocystis violacea -
Thiothrix nivea

Lactobacillus bulgaricus
Lactobacillus lactis
Sulphomonas denitrificans
Sulph. thio-oxidans, Thio-baderium thio-
oxidans
Sulph. thioparus

470

BACTERTOLOaiCAL CHEMISTRY

Vibrio agar liqucfacicns
aniylocella
cholerce -
comma -

Microspira agar liquefaricns

V. comma
V. cholerce

Zymosarcinn metlianica

Sarcina metlianica

INDEX

AcETALDEHYDE, condensation. 278

— detection, 464

— fixation, 245

— formation by fungi, 302

— hydrogen acceptor, 202, 204

— intermediate in fat synthesis, 382,

et. seq.
fermentation, 245, 309, 316

— reductase, 212

crystalline, 41

Acetarsol, 130
Acetic acid, 245
detection, 465

fermentation, 322 et seq.

— — formation, 247, 254
Aceto-acetic acid, 313, 316
Acetobacter aceti, water content, 58

— ascendens, 322

— ■ 'pasteurianum, 322

— suhoxydans, growth factors for, 101,

118

oxidation by, 241, 242

■ — xylinmn, cellulose in, 61, 345

glyoxalase in, 246

■ oxidation by, 242, 243

vinegar fermentation, 322

Acetoin, detection, 467

— formation, 252

— reduction, 253, 278

— Voges-Proskauer reaction, 252
Acetone, 242

— detection, 464

— by fermentation, 312 ei seq.
Acetonedicarboxylic acid, 284, 299
Acetylmethylcarbinol, see Acetoin
Acid, definition, 4

— production by bacteria, 245 et seq.
by fungi, 282 et seq.

theories of, 297

Acid-fast bacteria, fatty acids in, 370
phosphorus in ash of, 59

— waxes, 372

Aconitic acid, 285
cis-Aconitic acid, 302
Acridine dyes, drug resistance to, 155
Acriflavine, inhibition of hydrogen
transportases, 135

— woimd antiseptic, 131
Actinomyces albus, 157

— antibioticus, 157

— ccelicolor, 391

— lavendulce, 182

— pigments in, 387, 388, 389

— violaceus, 157

— violaceus-riiber , 391

— waksmanii, 390
Actinomycetin, 157
Actinomycin, 157

— A-, 157

— B-, 157, 158
Acyl proteins, 415
Adenine, 333

— as growth factor, 131, 145

— reversal of sulphonamide activity,

145
Adenosine-3-triphosphate, in Tk. tliio-

oxidans, 72

5-triphosphate, 111

in micro-organisms, 72

as phosphate donor, 257,

275, 320, 367
Adenylic acid, 107, 108, 275, 332
Adsorption, isotherm, 446

— polar groups and, 34, 35
Aeration, effect on lipoid content, 63
Aerobacter aerogenes, 174, 245, 252

action of sulphonamides on, 140

adaptive enzymes in, 91

aneurin in, 103

glyoxalase in, 321

growth factors in, 115

— cloaccB, 226

— indologenes, COg fixation, 259

— levanicum, levansucrase in, 350

478

INDEX

Aerobic oxidation, steps in, 2(Mj
Agar as determinant group, 417
Agglutinins, effect of ninhydrin on,
433

— separation, 427

Alanine, degradation, 230, 237

— S3rnthesis, 339

— P-, 101, 107, 224

and drug resistant C. diphtherice,

153, 154
in bios, 100

— d-, as energy source, 218
Alboleersin, 396
Albucid, 133

Albumin, as antigen, 400

— in protein of hen and duck eggs,

399

— in protein of micro-organisms, 60,

330

— precipitation by ammonium sul-

phate, 60, 424
Alcaligenes fcecalis, activation as

hydrogen donator, 195

anaerobic growth, 208

aneurin in, 103

growth factors in, 115

— radiobacter, nitrogen fixation, 221
Alcoholic fermentation, mechanism,

265, 269 et seq.

Meyerhof 's theory of, 274

Neuberg's forms of, 272

products of, 262

Alcohols, oxidation by Metkanohac-

terimn omelianski, 76, 78
Aldehyde mutase, 314, 322
Aldobionic acids as determinant

groups, 413
Aldol, 279, 313, 316, 382
Alexin, see Complement
Alkyl proteins, 415
Allicin, 185

Amides, utilisation of, 216
Amines, utilisation of, 215
Amino-acids, degradation, 227 et scq.
by bacteria, moulds and yeasts,

234

ease of attack, 216, 237

■ energy source of anaerobes, 218

estimation of. 238

Amino-acids in proteins of micro-
organisms, 60, 330

— requirements of heterotrophs, 81

et seq.

— synthesis, 339 et seq.

— toxicity to bacteria, 217

— utilisation of, 215
^-Aminobenzene sulphonamide, 131
Aminobenzene sulphonic acids, 406
Aminobenzoic acids, 406
jp-Aminobenzoic acid, 99, 101

inhibition of sulphonamides,

101, 137

ionisation of, 141, 143

^9-Aminobenzyl cellobiuronide, 412

— gentiobiuronide, 412

fZ- Amino-oxidase, prosthetic group,

41, 42
j9-Aminophenol-P-galactoside, 41 1

— a-glucoside, 411

— P-glucoside, 410

CO -Amino valeric acid, 228, 237
Ammonium borate, 445

— salts, utilisation of, 215

by autotrophic bacteria, 68

by heterotrophic bacteria, 81

et seq.
Amoebff, chemotherapy of, 130
Amygdalin, 37, 51
(?-Amyl alcohol, 280
i5o-Amyl alcohol, 231, 280
Amylase, activation of, 48

— crystalline, 41

— effect of pH on, 48
Anaerobes, absence of catalase from,

199

— growth in oxj^gen, 207
Aneurin, 393

— diphosphate, 43, 150

— growth factor, 100, 102

— in yeast, 339

— pyrophosphate, as co-carboxylase,

103

Anisoyl proteins, 415

Anthocyanin pigments, 391

Anthranilic acid, in tryptophane syn-
thesis, 343

Anthranols, 396

Anthraquinone pigments, 386, 395

Anti-amylase, 50

i

INDEX

470

Antihiosis, (Icfiiiitioii, l")')
Antibiotics. production bv .4._s7>r/7////(/,y,
184

— production by Penicillium. 184

— use in selective media, 18<)
Antibodies, 418 e^ seq.

— antibacterial, 426

— effect of acetylation, 419, 431, 432,
i 460

azo- compounds, 431

esterification, 432

formaldehyde. 431. 460

— — iodo-groups, 419. 431

ninliydrin, 432

pepsin. 430

— H- and 0- types, heat stability.

430

— molecular weight, 428. 454
■ — nature, 418

— production, 420 et seq.
in vitro, 424

— properties, 428

— resistance to heat, 429

— separation, 424 et seq.

— yalency, 423, 455, 457
Anti-enzymes, 50

— immunological, 50
Antigen-antibody reactions, 441 et

seq.

constant antibody method, 448

■ — antigen method, 448

forces involved, 456

inhibition zones, 448. 451, 452,

458

lattice theory of, 457

occlusion theory of, 458

optimal proportions, 447

pro-zones, 448, 451, 452. 458

specificity in, 456, 457

Antigens, molecular weight, 456

— multiple valency, 423, 451, 455

— natural, 398 et seq.

— reactive groups in, 460, 4()1

— synthetic, 405 et seq.
Anti-lecithinase, 50
Anti-pepsin, 50
Anti-ribonuclease, 50
Anti-trypsin, 50

Antisera, difference between horse
and rabbit, 428, 433, 454, 461

Antitoxins. 'I' compcnicnt, 42'»

Auti-urease, 50

Apozymase, 269

Arabinose in phosphatides, 375

— in polysaccharides, 353
Arginine, 111

— sjTithesis, 343

Arsine derivatives from fungi, 305
Arsenoxides, action on SH groups, 152
Arsphenamine, 127
Ascorbic acid, 244, 288, 297

and complement, 439

in fungi, 65

— dehydi'ogenase, 41

Ash, see Mineral constituents
Asparagine, 111

— utilisation, 218

Z-Aspartic acid, degradation, 230

energy source, 219

syTithesis, 340

Aspergillic acid, 158
Aspergillin, 388, 393
Aspergillus candidus, 159
antibiotic from, 183

— clavatus, 160

— elegans, 303

— fischeri, 378

— flavipes, antibiotic from, 184

— flavus. 158, 161

— • — kojic acid from, 295

■ — ■ — penicillin formation bv, 171

— fumigatus, 162, 163, 164,"^ 167, 393,

394

— giganteus, 160, 163

— glaucus, 394, 395

— mannitol in, 62

— melleus, 308

— nididans, 303

— niger, citric acid in, 292, 327
ergosterol, 378

gluconic acid, 292, 304

— — mannitol, 304

mj^cogalactan, 351

oxalic acid, 292

starch, 346

trehalose in, 62

utilisation of carbon source, 191

■ — ■ ochraceus, 308

— oryzce, 294, 302, 378

— parasiticus, 170

480

INDEX

Asjicrgillus sydowi , 305, 37S
~ terreus, 306, 308, 31>U

— vitamin-C in, 05
Aspirin, antisera to, 415

— as determinant grouj), 415
Assimilation, 240

Astacin, 390
Atebrine, 128, 130

— inhibition of Plasmodium respira-

tion, 151
Athiorhodacece, metabolism of, 74, 76

— pigments, 391
Atoxyl, 405

— azo-proteins, 405, 406, 449

— in chemotherapy, 127

— inhibition by ^-aminobenzoic acid,

138

of lipase, 135

Atropine, inhibition of invertase, 135
Auroglaucin, 394
Autocatalysis, 331, 342
Autotrophic bacteria, 67 et seq.

chemosynthetic, 67, 78

classification, 69

photosynthetic, 67, 78

thermodynamic efficiency, 79

Avidin, inhibition of biotin by, 104
Azotobacter, fixation of nitrogen, 214

— chroococcum, fixation of nitrogen,

221 et seq.

gum of, 348

nucleic acids, 336

pigment, 389, 391

sterol in, 377

— utilisation of nitrate and nitrite,

215

Bacillus ac eto- ethyl icus, 245, 246, 312

fermentation of pentosans by,

312

— acidificans longissimus, 319

— aerogenes, see Aerobacter aerogenes

— aminovorans , 328

— anthracis, capsule of, 183, 338

carbohydrates of, 362, 365

cZ-glutamic acid in capsule, 183,

338

growth factors for, 185

inhibition of, 156

nucleic acids of, 336

Bacillus hrcvis, 165

— faecalis alcaligones, see Alcaligenes

f(jEcalis

— lactis, 351
pitiiitosi, 351

— macerans, 312

— mesentericus , aneurin in, 103

antibiotic from, 183

capsule of, 183, 338

cZ-glutamic acid in capsule, 183,

338
levan, 349

— - — niger, pigment, 386, 388

— mycoides, vitamin -B in, 65

— polymyxa, levansucrase in, 350

— ■ prodigiosus, see Serratia marcescens

— pyocyaneus, see Pseudomonas aeru-

ginosa

— subtilis, adenosine triphosphate in,

72

(Z-glutamic acid in capsule, 183,

338

— — levan, 349

proteolytic enzymes, 225

— ■ — subtilin, 185
vitamin-B in, 65

— typhosus, see Eberthella fyphosa

— volutans, 315
Bacitracin, 185
Bacterial gums, 326, 348
Bacterio-chlorin, 73, 387, 390, 391

— erythrin, 73, 390

— phseophytin, 390

— purpurin, 64, 69, 73, 390

— xanthophyll, 390
Bacterium cocovenenans, 390

— gluconicum, oxidation by, 243, 244

— xylinoides, oxidation by, 244

— — kojic acid from, 294
Ba3yer-205, 126

Base, definition, 4

B.C.G., carbohydrates of, 360

— sterols in, 377
Beggiatoa, 69, 72
Benzaldehyde, reduction of, 278
Benzidine-R-salt, 418
Betabacterium vermiforme, dextran

synthesis, 348
Biological reduction, 278

INDEX

481

Bio.s. 1)9

— fompositiuii. I'M I
Biotics, 117

Biotin, 99, 100, 103, 108, 305
Blood Group A ploysaccharide, 357

as determinant group, 417

Botrytis cinerea. antibiotic from, 185

thio-urea from, 305

Bromo-proteins, 414
Brucella abortus, carbon dioxide re-
quirement, 83, 255

growth factors for, 104, 118

Buffer solutions, 15 et seq,

amino-acids as, 18

■ ■ composition, 18

dilution of, 16

effect of acids, 16,17

of alkalies, 16, 17

— — range, 18

relation to ^^K, 17

proteins as, 18

— universal, 18

Butyl alcohol detection, 464
Butvl alcohol-acetone fermentation,
'314 et seq.

mechanism, 316

products of, 316

1 : 3-Butylene-glvcol,' 279

2 : 3-Butylene-glycol, 253, 278

— detection, 467
ButjTic acid, 245, 314, 369

detection, 465

Byssochlamic acid, 291
Byssochlamys fulva, 303

Calomel electrode, 11

Cannizzaro reaction, 248, 274, 303,

304
Carbasone, 130
Carbohydrates, content of bacteria,

fungi and yeasts, 61

— " reserve," 61

— soluble specific substances, 353

— synthesis, 366
Carboligase, 253, 278

Carbon content of bacteria, fungi and

yeasts, 59, 66
Carbon dioxide, fixation, 83

— • autotrophic bacteria, 67, 68,

73, 76, 77, 78, 256

r'arl)on dioxide, fixation, distril»ution

in acid products, 259
— heterotrophic bacteria, 250,

255

phosphorylation, 257

— succinic acid formation, 251

by A. niger and

Bhizopus nigricans, 294
requirement by heterotrophic

bacteria, 87, 255
Carhoxydomonas oligocarbophila, 75
Carboxylase. 43, 52, 212, 248, 272,

277

— absence from propionic acid bac-

teria, 254

— constitution, 43

— in amino-acid degradation, 230,

234
Carboxypeptidase, crystalline, 41
Cardiolipin, 403
Garlic acid, 287
Carlosic acid, 287
Carolic acid, 287
Carolinic acid, 287
Carotene, 385, 390
P-Carotene, 390, 393
Carotenoid pigments, 64, 385
Carviolacm, 395
Carviolin, 395

Casein, antigenicity, 399, 400
Catalase, 212

— crystalline, 41

— effect on notatin activity, 169

— haem in, 41

— in bacterial respiration, 198

• — in Micrococcus lysodeikticus, 93

— inhibition, 200

— prosthetic group, 41
Catalysts, 38

— mode of action, 53
Catenarin, 395
Cellobiase, 326
Cellobiose, 326
Cellobiui-onic acid, 348, 355
Cellulase, 326
CeUulobacillus mucosas, 325, 326

— myxogenes, 325, 326
Cellulomonas, 325

Cellulose, degradation, 323 ei seq.
thermophilic, 324

4:S2

INDEX

Crllulosc, ill Acdobiirlrr .nfliiinni. (il,

— in bacteira, (U

— staining by zinc chlnriodide, ^^i\
Cerebrin, 378

Cerotic acid, 370
Chemoreceptors, 135
Chemosynthetic bacteria, (SI
Chemotherapeutic agents, modes of
action, 134 ct seq,

— index, 124

Chemotherapy, essential metabolite
theory, 137

— of bacterial infections, 131

— of protozoal infections, 126
Cherry gum as determinant group, 417
Chinoform, 130

Chitin, in micro-organisms, 61
Chlorine metabolism of fungi . 306
Chlororaphin, 179, 391
Cholera red reaction, 210
Cholestan, 404
Cholesterol, 372, 377

— as antigen, 404

— growth factor, 121
Choline sulphate, 305
Chromatographic analysis, 385
Chromobacterium iodinum, 167, 392

— violaceum, 185, 388, 389
Chromoparous organisms, 64
Chromophoric organisms, 64
Chymotrypsin, crystalline, 41

— crystallisation of, 46
Chymotrypsinogen, crystallisation of,

45
Cinnamyl proteins, 415
Citric acid, 286, 292, 297, 327

detection, 466, 468

Citrinin, 159, 283, 397

— detection, 468
Citromycetin, 396

— detection, 468

Citrulline in arginine synthesis, 346
Clasterosporum, 303
Clavacin, 160

— inactivation by SH., 161
Clavatin, 160, 161
Claviformin, 160, 161
Clostridium acetobutylicum, butyl-
alcohol-acetone fermentation, 314

( Uhsfridiiitn (irdoJiiili/Jii-iijii, C( )^ lix-
atioii, 2()0

— — growth factors for, 101, 104, 118

— botulinum, antitoxin, 426
growth factors, 85, 114

— butylicum, growth factors, 104, 118

— butyricum, fermentation by, 314

fixation of nitrogen by, 220, 222

glycogen in, 345

granulose, 62

growth factors for, 101, 118

— — inhibition of, 156, 315
— • cellvlolyticum, 324

— ■ dissolvens, 324

— fossicularum, 323

— histolyticum, antitoxin, 426

— methanigenes , 323

— (xdematiens, antitoxin, 426

— septicum, antitoxin, 426

— sordellii, antitoxin, 426

— sporogenes, aerobic growth, 198,

211

— — growth factors, 114, 118

— tetani, growth factors, 101, 110, 118

— thermocellum, 324

— welchii, antitoxin, 426

CO2 fixation, 260

growth factors, 118

— — hyaluronidase in, 358

pigment of, 391

a-toxin, 50

Cocaine, inhibition of invertase, 135
Co-decarboxylase, 111, 150, 25!)
Co-enzyme I, 49, 263, 267

— constitution, 268

— hydrogen carrier, 202, 203, 204,

268, 276, 320

— identity with V-factor, 108

— in acetic acid fermentation, 322

alcoholic fermentation, 263

lactic acid fermentation, 249,

320

propionic acid fermentation.

249

— magnesium in, 267, 269

— nicotinamide in, 107, 267

— occurrence, 268

Co-enzyme II, as growth factor, 109

— as hydrogen carrier, 204

— nicotinamide in, 107

INDEX

483

Co-enzyme R, identity with biotin. 104
Colloids, 26 et seq.

— cadmium sulphide sol, 27

— coagulation, 28 et seq.
zone of, 29

— critical surface potential, 28, 29,

30, 33

— effect of electrolytes, 28 et seq.

— emulsions, 26

— ferric hydroxide sol, 28

— foams, 27

— fogs. 27

— gold sol, 27

— ij-ophilic, 27 et seq.
— • h'ophobic. 27 et seq.

— particle size, 26, 27

— polj'saccharidfs, 27, 30. 33

— proteins, 27, 30, 33

— smokes, 27

— stability, 27, 28

— suspensions, 26
Comenic acid, 294
Comparator, Lovibond, 15
Complement, 433 et seq.

— artificial, 434, 439

■ — inactivation by dialysis, 438
by iodine, 438

— lability, 434

— role of components, 436, 437

— structure of, 434 et seq.
Continuous phase, 26
Copper protein enzymes, 52
Corynebacterium eliphtheriee, antitoxin,

419, 426, 431

— — — crystalline, 427

■ effect of pepsin on, 430

■ carbohydrates, 61

drug resistant strains, 153, 154

fatty acids in, 371

— — growth factors for, 101, 106.

108, 109. 118
inhibition bv pantovltaurine,

149, 150

lijioids of, 62

metabolism, 80, 85, 216

nucleic acids of, 60, 33<]

pho.sphatides of. 381

porphyrins of, 391

— — waxes of, 377

Cor3'nin, 381

Cozymase, see Co-enzyme

Crenothrix, 70

^-Cresol from tyrosine, 233

Cydopium viridicatum, 176

Cvnodontin. 395

Cystein, 207, 211

— de-amination, 237
Cytidilic acid. 333
Cytochrome, 105, 200

— distribution, 201, 202

— inhibition, 201

— in respiration, 201

— oxidase, 212

Cytophaga hitchinsoni , 324, 326
C\i;osine, 333

Danysz phenomenon, 443. 446. 447

Deaminase, 212

Deamination, desaturative, 229

— hydrolytic, 229

— oxidative, 230

— reductive, 228
Decarboxylase, amino-acid. 111
Decarboxylation of amino-acids, 228
Dehydrase, 197
Dehydrogenase, 197

— inhibition, 49

— in respiration, 20l
Desmolasss, 53
Desoxyribonucleic acids. 336

— in Rough-Smooth conversion, 337
Desoxyribose, 333
Desthiobiotin, 104

Determinant groups, 406. 423
Dextran, 346, 348, 401

— constitution, 347

— serological reactions, 346
Diaphorase, 212

— and co-enzyme, 206

— prosthetic group, 41, 42
Diasone, 134

Diastase, 37

Diazotised arsenilic acid hapten, 458

Didymohelix, 70

— ferruginea. 74
Dibromo-cholesterol, 404
Dibromo-tyrosine, 414
Dihydro- cholesterol, 404

484

INDEX

Dihydro-co-enzyme I, 203, 208, 270,

277, 321
Dihydro-erdin, 306
Dihydro-geodin, 306
Dihydro-penicillic acid, 171
Dihydro-penicillin I, 172
Dihydro -phenazine - 1 - carboxylamide,

392
Dihydroxyacetone from glycerol, 243
— phosphate, 274, 276, 320
(3:5- Dihydroxy - 2 - carboxy benzyl)

methyl ketone, 290
(3:5- Dihydroxy - 2 - carboxy benzoyl)

acetyl carbinol, 290
3 : 5-Dihydroxyphtbalic acid, 291
Dihydroxy-stearic acid, 369
Di-iodophenol, 414
Di-iodotyrosine, 414
ay-Diketo-adipic acid, 298
Py-Diketo-adipic acid, 299
Dimedon, 77, 260, 272, 384
4 : 6-Dimethoxytoluquinone as bac-
teriostatic agent, 163
Dimethylpyruvic acid, 285
Dimethyl-selenium, 305
Dimethyl-tellurium, 305
Diphosphoglyceric acid, 276, 320
Diphosphopyxidine nucleotide, 202,

267
Diphtheria bacillus, see Coryne-

bacterium diphtherice
Diphtheric acid, 371
Diplococcus pneumonice, antibodies to,

412, 427, 428, 431
antisera, cross reactions, 348,

412

autolytic enzymes in, 358, 304

carbohydrate F, 354, 357

carbohydrates of, 353, 401

growth factors for, 105

inhibition by pantoyltaurine,

149

Rough-Smooth conversion, 337

sulphonamide resistant strains,

152

synthetic antigens, 412, 413

Dismutation, 248
Disperse phase, 20

charge on particles, 27

size of particles, 26, 27

Dissimilation, 240

Dissociation constant, acetic acid, 10

acids, 5

bases, 6

• phosphoric acid, 10

water, 6

Drug resistance, 151 et seq.
development of, 154

Eberthella typhosa, carbohydrates of,

362, 363, 402, 417

CO -enzyme in, 107

exacting and non-exacting

strains, 83

fermentation products of, 248

inhibition by indole-acrylic acid,

147

variants, 83

vitamins in, 115

Egg albumin-antiserum system, 454
Ehrlich's phenomenon, 443, 446
fZ-2-Eicosanol, 373, 374, 377
Electrode potentials, calomel, 11

-hydrogen, 11, 12

quinhydrone, 13

Electrodes, calomel, 11

— glass, 12, 13

— hydrogen, 11, 12

— quinhydrone, 12
Electrophoresis, 425, 426
Elution, 45

Emetine, 125, 128

Emodic acid, 395

Emodin monomethyl ether, 395

Emulsin, 37

— specificity, 50

— synthesis by, 40
Endo-enzymes, 53

Endomyces vernalis, lipoid content,
63, 371, 381

resistance to pyrithiamino, 151

End-piece, 435, 437

Energy requirements of micro-organ-
isms, 188

Energy liberated in oxidations, 191

Enolase, 212, 276, 320

Entamoeba histolytica, 125, 128

Enterokinase, 49

Enzymes, 36 et seq.

— activators, 48

INDEX

485

Enzvmes activity, measuremeut of,
■^46

— adaptive, 89

speed of production, 9G

— adsorption of, 45

— anti-enzymes, 50

— as catalysts, 38

— chemical nature of, 40 et seq.

— classification, 52

- — colloidal carriers, 41
nature, 39

— constitutive, 89 et seq.

— crystalline, 41

— crystallisation of, 45

— effect of concentration, 4G
of heat, 47

of^jH, 48

of substrate concentration, 47

-on equilibrium of reactions, 39

on velocity of reactions, 39

— endo-cellular,^ 53, 189

— exo-cellular, 53, 188

— inhibition by carbon monoxide, 49
by chloroform, 49

by cyanide, 49

by drugs, 135

by heavy metals, 49, 135

by substrate analogues, 135, 196

by sulphides, 49

by urethane, 49

■ — isolation of, 38

— nomenclature, 52

— physical properties, 43 el seq.

— prosthetic groups of, 41 et seq.

— purification, 45

— relation to cells, 38

— reversible effect of, 40

— separation, 44

— specificity, 50

— stability, 40

— synthesis by, 40

— theories of action, 53

— units, 46

— yariability of content in micro-

organisms, 95

— Warburg's yellow enzyme, 41, 42
Epitoxoid, 443

Epitoxonoid, 444
Equivalence point, 450
Erdin, 306

Ergosterol, 372

— as antigen, 404

— in fungi and yeasts, 63, 377
Ergosteryl palmitate, 63, 378
Erythroglaucin, 395

EschericJiia coli, action of sulphona-
mides on, 139, 140

activation of hydrogen dona-
tors by, 194, 195, 209

adenine triphosphate in, 72

anaerobic growth, 208

aneurin in, 103

carbohydrates of, 362

carbon dioxide fixation, 25!)

— and succinic acid for-
mation, 260

requirement, 82

co-enzyme in, 107

fermentation products of, 248,

250

growth in sjoithetic media, 82

■ inhibition by indole-acrylic acid,

147

— P-naphthjd-acrylic acid, 148

— — ■ — pyrithiamine, 151

sulphanilamide, 149

metabolism, 81

production of indole, 233

sterols in, 377

— • — v-ariants, 89, 96

yitamins in, 65, 115

water content, 58

Eserine, inhibition of esterase, 135

Ethyl acetate, 303

Ethyl alcohol, detection, 464

production by bacteria, 249,

251, 312

■ — by fungi, 302

— by yeasts, 262 et seq.

Ethyleneoxide-a- P'dicarboxylic acid,

284
Euglobulin, 424

— solubility, 60
Exacting strains, 83, 96
Exo-enzymes, 53

Fats, 369

— of yeast, 63, 371

— staining by dimethyl-amido-azo-

benzene. 62

4S0

INDEX

Fats, staining by dimethyl -^> -
phenylene diamine, 02

a-naphthol, 62

osmic acid, 56

Sudan III, 56

— synthesis, 381 et seq.
Fatty acids, 62

of acid-fast bacteria, 371

of C. diphtherice, 371

of micro-organisms, 369

■ of moulds, 371

— • — of yeasts, 371

Fermentation, nature of, 37, 192, 326

Fibrinogen as antigen, 399

Ficin, crystalline, 41

Flavacidin, 161

Flavicin, 161

Flavine adenine dinucleotide, 42, 248

Flavohacterium brunncurn, 387

— suaveolens, 389
Flavoglaucin, 394

Flavoprotein enzymes, 41, 42, 52, 169,
212

and co-enzyme I, 203

Flavorhodin, 390

Fluoride, inhibition of alcoholic fer-
mentation by, 275, 277

Folic acid, 112

Formaldehyde, by reduction of carbon
dioxide, 260

— effect on protein antigens, 401, 417

— fixation by dimedon, 77, 260
Formic acid, 245

breakdown, 203, 247, 248

detection, 465

formation, 247, 260

Formic dehydrogenase, 212

Forsmann antigen, 354, 357

Fourth component of complement,
436, 438

Friedlander's bacillus, polysaccha-
rides, 354, 358

Fructosan, 348, 351

Fructose- 1 : 6-diphosphate, 266, 275

in Th. thio-oxidans, 72

Fulvic acid, 291

Fumarase in Microrocnis lysodeikticvs,
93

Fumaric acid, 202, 229, 258, 284, 293

detection, 464, 466

Fumaryl-(/Z-alanine, 286
Fumigacin, 162
Fumigatin, 163, 179, 393
Funiculosin, 395
Fusarium, 302

— cuhnorum, 396

— javanicum, antibiotic from, 185

— pigments of, 387
Fusel oil, 231, 279 et seq.

Galactan, 351

Galactose, fermentation by yeast, 89,
96

— in polysaccharides, 353
Galacturonic acid as determinant

group, 412
Gallic acid, 289
Gallionella, see Didymohelix
Gas production by bacteria, 248, 251
Gas ratios, 251
Gelatin, 398

— conversion to antigen, 414
Gelatinase, formation by Proteus vul-
garis, 96

Gentisic acid, 289, 307
Gentisyl alcohol, 307
Geodin, 306
Germanin, 126
Gigantic acid, 163
Glass electrode, 12, 13
Glaucic acid, 291
Glauconic acid, I, 291

II, 291

Gliocladiinn fimhrlatum, 164
Gliotoxin, 164
Globulin, a-, 424, 426

— P-, 424, 426

— Y-, 424, 426

— T-, 426

— as antigen. 400

— carbohydrate in, 426
— • in antibodies, 419

— in protein of micro-organisms, 60,

330

— precipitation by alcohol, 425

— — by ammonium sulphate, 60, 424

— shape of molecule, 428
Gluconic acid, 242, 285, 292, 300, 327
detection, 466, 468

Glucose monophosphate, 264

INDEX

487

Glucose oxidase, 212, 293

1 -phosphate, 72

6-phosphate, 72

0-[3-Glucosidyl-tyrosine, as deter-
minant group, 414

0-P-Glucosidyl-tyrosyl-gelatin, 414

insulin, 414

Glucosone, 295

Glucuronic acid, 285, 348

as determinant group, 412

intermediate in citric acid pro-
duction, 300

Glutamic acid. 111

as energy source. 219

rf-Glutamic acid, 183, 338

^-Glutamic acid, degradation, 230

Glutamine as growth factor, 104

Glutathione, 136, 207

— as prosthetic group of glvoxalase,

208

— in bacterial respiration, 207
Giyceraldehyde, 270
3-Glyceraldehj-de phosphate, 202, 270,

274, 276
Glycerol by fermentation, 310

— detection, 467

— formation in alcoholic fermenta-

tion, 269, 273, 277
by fungi, 302

— oxidation of, 243
Gtycerophosphoric acid, 202
Glycine, 218

— degradation, 228, 232, 237

— methylamine from, 228
Glycogen as antigen, 301

— in Aspergillus, 345

— in bacteria, 345

— in yeast, 62, 345

— staining by iodine, 56
Glycollic acid, 284

intermediate in citric acid pro-
duction, 300
(dyoxalase, 208, 212, 246, 321
Gonococcus, carbohydrates of, 365

— metabolism of, 81

— pjTuvic oxidase in, 43
Gramicidin, 165

— action of formaldehyde on, 166
Gramicidin-S, 166

Gram stain, ribonucleic acid and, 337

Granidose, 62

Griseo-fulvin, 307

Growth factors, 82, 84, 98 et seq.

adenine, 101

[3-alanine, 100, 101

_p-aminobenzoic acid, 101

aneurin, 100, 102

bios, 99

biotin, 100

— — Clostridium, 85, 99, 103

H. influenzce, 86, 105, 108

inositol, 100, 105

Z-leucine, 100, 106

nicotinamide, 106

nicotinic acid, 100

pantothenic acid, 107

pyridoxine, 100

requirement by heterotrophs, 82

sporogenes factor, 114

staphylococcus factor, aerobic,

112

anaerobic. 111

uracil. 111

F-factor, 86

X-factor, 86

Guaiacum, 37

Guanidine, effect on sulphonamide

activity, 145
Guanine, 333
Guanylic acid, 332
Z-Guluronic acid, 243
Gum acacia as determinant group, 417

Haematin as growth factor, 105, 109
Hsemocyanin in enzymes, 41
Haemoglobin as antigen, 399
Hsemolysin, separation, 427
Hcemophilus canis, V- and X-factors,
86, 109

— influenzce, growth factors, 86, 105,

107, 108, 109, 119

metabolism, 81, 216

F-factor, 86

— — Z-factor, 86, 105

— parainfliienzce, amino-acid degrada-

tion by, 230

growth factors, 109, 119

symbiosis, 109

— pertussis, growth factors, 106, 119

488

INDEX

Haptens, 406

— combination with antibodies, 442,

452, 458

— complex, 442,- 458

— simple, 442, 452

" Heavy " carbon, as tracer element,

247, 256
Helminthosporin, 64, 395
Helminthosporium avence, 395
antibiotic from, 185

— catenariiim, 395

— cynodontis, 395
• — euchlance, 395

— geniculatum, 303

— gramineum, 64, 395

— leersii, 396

— ravenelli, 395, 396

— tritici-vulgaris, 395

— turcicum, 395, 396
Helvolic acid, 162, 167
Hemipyocyanin, 178, 179
Heparin, 50
Heterobiotin, 104
Heterotrophic bacteria, 80 et seq.
Hexacosanoic acid, 374, 375
Hexadecenoic acid, 371
Hexose diphosphate, 320
Hexose monophosphate, 266
Hexose monophosphorylase, 212
Hexose phosphates, 264
Histamine, antisera to, 417

— as determinant group, 416
Histidine, 216, 405

— deamination, 230, 237
Histone, bacteriostatic action, 185
Holozymase, 269

Hyaluronic acid, 358
Hyaluronidase, 358

— in CI. welchii, 95
Hydrogenase, 212

— in nitrogen fixation, 224
Hydrogen bacteria, 70, 74, 207

— donators, 194, 208

activation of, 194, 208

inhibition of, 196

mechanism, 195

— electrode, 11, 12

— ions, acidity due to, 7

concentration in acetic acid, 7

in hydrochloric acid, 7

Hydrogen ions, concentration in

water, 7
measurement of, colori-

metric, 13 et seq.
electrometric, 11 et

seq.

in water, 6

Hydrogenlyase, 94, 212, 260
Hydrogenomonas , 70

— metabolism of, 74

Hydrogen peroxide in bacterial res-
piration, 198 et seq.

— production from cellulose, 323
from formic acid, 249

— sulphide, activation of papain, 48
oxidation by sulphur bacteria,

72, 73

— transportase, 197
Hydrolases, 52, 188
^-Hydroxybenzoic acid, 232
Hydroxy-emodin, 395
Hydroxylamine in nitrogen metabol-
ism, 223

Hydroxyl ions, alkalinity due to, 7

concentration in water, 7

in water, 6

2 - Hydroxymethyl - 5 - carboxyfurane,

288
a-Hydroxyphenazine, 391
2)-Hydroxyphenyl-lactic acid, 229
/-Hydroxyproline, 218
Hypoxanthine, 219, 336

Iminazole-acrylic acid, 230
Indicators of oxidation-reduction, 23,
24

— of^pH, 13, 14

Congo-red, 14

litmus, 14

methyl orange, 13

phenolphthalein, 14

ranges of, 15

Indole -acrylic acid, antibacterial ac-
tion, 147
Indole in tryptophane synthesis, 343

— from tryptophane, 209, 233
Indole-carboxylic acid, 232
Indole-lactic acid, 229
Indole-propionic acid, 228
Influenza bacillus, see H. influenzie

INDEX

489

Inositol as growth factor, 105, 108

— in bios, 100

— in phosphatides, 375, 379, 380

— relation to streptomj^cin, 182
Insulin, conversion to antigen, 41-1

— non-antigenicity, 399, 400
Invertase, 40

— inhibition, 49

— purification. 45

— specificity, 51
lodinin, 167", 392
lodo-proteins, 414
Ionic product, 6. 7

determination , 7

lonisation, degree of 7

— of acids, 5, 7

— of bases, 6

— of water, 5, 7
Ions, activity, 5
Iron bacteria, 70

metabolism, 74

Isomerase, 212, 276, 320
Isopropanol, fermentation product ,

318

— oxidation to acetone, 74, 242
Itaconic acid, 285

Johne's bacillus, nutrition of, 87
vitamins in, 113

Kala-azar, chemotherapy of, 129
Kanten, 417
Kephalin. 381, 403

— in antisera, 428
Keratm, as antigen, 400
a-Keto-acids, 230

— in protein synthesis, 340
P-Keto-acids, 232
2-Ketogluconic acid, 242, 244
5-Ketoglutaric acid, 242
a-Ketogluconic acid, 230
2-Keto-Z-gulonic acid, 244
y-Ketopentadecoic acid, 285
Klebsiella 'pneiimonke , carbohydrates

of, 354, 358
Kojic acid, 163, 171, 289, 294, 300

detection, 468

Krj^toxanthin, 390

L+dose of toxin, 442

L^ dose of toxin, 442
Lactase, 39
Lactic acid, 245

as hydrogen donator, 194, 196,

208, 209
bacteria, 318

— — b^eakdo^^^l to propionic acid,

254

detection, 466, 468

enzyme, 212, 321

enzymic breakdown, 135

fermentation, 249, 318 ef seq.

— phosphorylation, 249, 320

— — formation, 208, 246

production by fungi, 247, 248,

251, 282, 284, 293
Lactic dehydrogenase, 212
Lactobacillus acidophilus, action of

sulphonamides on, 140

cholesterol in, 377

fatty acids in, 371

phosphatides of, 381

— arabinosus, amino-acid require-

ments, 216
— • — growth factors for, 104, 110,
112

— bulgaricus, 319

— casei, 146, 319

growth factors, for 104, 1 10,

112, 113

— delbmckii, 41, 43, 248, 319
glyoxalase in, 321

growth factors for, 110, 112

— ■ — pjTuvic oxidase in, 43
" yellow enzyme " in, 41

— drug resistant strains, 153

— growth factors for, 105, 110, 116,

119

— use in assay of growth factors, 116

— pentoaceticus, 321

adaptive enzymes of, 93

Lactoflavin-5-phosphoric acid, 42
Laurie acid, 369
Lecithin, 378, 381, 403

— in antisera, 428
Lecithinase, CI. welchii a-toxin, 50
Leishmania, chemotherapy of, 130
Lentinus lepideus, 307
Leprosinic acid, 374, 375, 377
Leprotin, 390

49U

IxMDEX

Leptothrix, 70

— crassa, 74

— ochracea, 79
250-Leucine, degradation, 280

— toxicity to bacteria, 218
^-Leucine, as energy source, 218

— as growth factor, 106

— degradation, 231, 280

— in bios, 100

— toxicity to bacteria, 217
Leucoflavoprotein as hydrogen car-
rier, 205

Leuconostoc citrovorum, 319

— dextranicum, 319, 348

— mesenteroides, adaptive enzymes in,

92

dextran in, 346

drug resistant strains, 153

glj'oxalase in, 321

growth factors for, 110, 112,

119
Levan, 349
Levansucrase, 350
Linoleic acid, 369
Linolenic acid, 369
Lipase, nature of, 40

— specificity, 51

— synthesis by, 40
Lipocyan reaction, 386
Lipoids, 369

— as antigens, 402 et seq.

— content in bacteria, fungi and

yeasts, 62

— " firmly bound," 376

— in complement, 436

— of acid-fast bacteria, 372 d scq.
Lovibond comparator, 15
Luteic acid, 351
Luteoleersin, 396

Lutein, 390
Luteose, 351
Lysine, 111

— phenyhireidc as hapten, 416
Lysozyme, 157, 365

— crystalline, 41

— substrate of, 157

Malic acid, 258, 284, 293, 300, 410

detection, 467, 468

Malonic acid, 284, 299

Maltase, adsorption of, 45

— specificity, 50

— synthesis by, 40
Mannitol, 300, 303 et seq.

— detection, 467

— in micro-organisms, 62
Mannocarolose, 352
Mannonic acid, 285, 297

Mannose in phosphatides, 375, 379,
380

— in polysaccharides, 352, 353
Mapharsen, 127

Marfanil, inhibition by ^j-aminu-

benzoic acid, 138
Melanin pigments, 64, 386
Melezitose, 349, 350
Mellein, 307

Meningococcus carbohydrates, 365
Mepacrine, 128

Mercuric chloride as antiseptic, 136
Metabolic products, separation, 462

et seq.
Metachromatic granules, nucleic acid

of, 60

staining, 56

Methane bacteria, 75

— production, 323 et seq.
Methanohacterium omelianski , oxida-
tion by, 76, 78

Methanol reduction, 76
3Iethanomonas aliphatica, 75

— aliphatica-liquefaciens, 75

— methanica, 75

Methionine, reversal of sulphonamide

activity, 145
Methoxydihydroxytoluquinone, 179,

388
Methyl anisate, 307

— cinnamate, 307

Methylene blue as hydrogen acceptor,
202

technique, 193

]\Iethylglyoxal, 208, 245, 272

— conversion to lactic acid, 246

— dismutation of, 273

— hydrate, 250, 271
Methylheptanone, 278
Methyl mercaptan, 305

— |j-methoxycinnamate, 307
6-Methyl salicylic acid, 289, 308

INDEX

491

Y-Mcthyltetrouic acid, 286, 294
Micro-aerophilic organisms, absence

of catalase, 199
Micrococcus lysodeikticus , carboxylase

in, 259

— — potysaccharide in, 365
production of enzymes by, 93

— urece, metabolism, 190
Mid-piece, 435, 438

^Mineral constituents in bacteria, fungi

and yeasts, 58

phosphorus in, 59

potassium in, 59

Minioluteic acid, 286
Monilia, lactic acid from, 282
Monochloracetic acid, 446
Mono-iodoacetic acid, inhibition of

alcoholic fermentation by, 275,

278
Mucoproteins, 61
Mucor, alcohol production, 302

— glycerol production, 302

— lactic acid production, 321

■ — rammanianus, growth factor for,
102

— rouxii, 282
Miitase, 193, 273

Mycobacterium leprce, pigment of, 390
trehalose in, 359, 371

— 2Mei, extract, 113

nucleic acids of, 336

— • — nutrition of, 86

■ — — ■ pigment of, 390

soluble specific substances, 358

sterols of, 377

waxes of, 373, 377

— tuberculosis, carbohydrates of, 61,

353, 358
■ — • — glycogen in, 345

lipoids of, 62, 373

nucleic acids of, 336

— — nucleoproteins of, 60

nutrition of, 85, 87

phosphatides of, 64, 379

soluble specific substances, 358

— — sterols in, 377

— — waxes in, 373, 374, 377
Mycodextran, 351
Mj^cogalactan, 351
:\[ycolic acid, 374, 375, 377

Mycolic acid, a-, 375

P-, 375

Y-, 377

Mycophenolic acid, 185, 291, 296
Mykol, 372
Myristic acid, 370
Myxococcus, 357

Xaphthoquinones, 386
[3-Naphthyl-acrylic acid, antibacterial

action, 148
Neisseria gonorrhoce, pyruvic oxidase

in, 248
Neurospora crassa, growth factors for,

101, 121

use in assay of amino-acids, 117

of growth factors, 116

— sitophila, growth factors for, 110,

121
use of mutants in amino-acid

synthesis, 343
Niacin, 107

Nicotinamide, 106, 108
Nicotinic acid, 100, 106

in yeast, 339

Nitrate reduction, 209

— utilisation, 70, 82, 215
Nitrifying organisms, 70

metabolism of, 70

Nitrite reduction, 209

— utilisation, 70, 84, 215
Nitrobacter, 69, 70
Nitrobenzene, reduction of, 278
Nitrogen content of bacteria, fungi

and yeasts, 59, 66

— estimation, 59

— fixation of, 214, 220 et seq.
mechanism, 223

— metabolism of heterotrophs, SI et
seq.

— requirements of micro-organisms,

214
Nitrosococcus, 69
Nitrosomonas, 69, 70, 215

— fixation of formaldehyde from, 71
Nomenclature of bacteria, 3

— of enzymes, 52
Non-exacting strains, 83
Norleucine, toxicity to bacteria, 218
Notatin, 168

492

INDEX

Nucleic acids, constitution, 332

Azotobacter chroococcnm , 330

B. anthracis, 336

CI. welcUi, 337

in C diphtherice, 336

Esch. coli, 337

metachromatic granules, 60

~ — M. phlei, 336

tuberculosis, 336

pneumococci, 337

psittacosis virus, 336

Sacch. cerevisice, 60, 332, 337

.Staphylococcus, 337

Streptococcus, 336

vaccinia virus, 336

Nuclein, 59, 332

Nucleoproteins, content of bacteria,
60, 332, 336

— constitution of, 332 et seq.

— hydrolysis of, 60, 332

— in Jf . tuberculosis, 60, 336

— in tobacco mosaic virus, 332

— in yeast, 60, 337

— properties of, 332

— staining by polychrome methylene

blue, 56
Nucleosides, 332
Nucleotides, 333

Ochracin, 308

d-2-Octadecanol, 373, 374, 377
0-heterobiotin, 104
Oleic acid, 369, 378
Oospora lactis, fat in, 371
vitamin-B in, 65

— sulphurea-ochracea, 306, 390
Optoquin, 131

Ornithine, breakdown, 228

— in arginine synthesis, 343
Orotic acid, 112

Oxalacetic acid, 223, 258, 259, 298, 302
Oxalic acid, detection, 464

production by fungi, 284, 292,

298
Oxidases, 193

— activation of, 49

— and cytochrome system, 201

— in bacterial respiration, 200

— inhibition, 49, 197, 200
by cyanide, 135, 200

Oxidation by oxygen, 197

— energy from, 191, 192

— mechanism, 192 et seq.

— of acetaldehyde, 193

— of carbon, 19

— of ethyl alcohol, 19

— of ferrous salts, 20

— of hydroquinone, 19, 192

— of methane, 19

— transfer of electrons, 20

— types of, 197
Oxidation-reduction potentials, 19

et seq.

interpretation, 24

measurement of, colori-

metric, 23 et seq.

electrometric, 20 et seq.

use of, 19

Oxycellulose, 326
Oxychlororaphin, 391
Oxycholesterol, 404

Palitantin, 308

Palmitic acid, 369, 379

Paludrme, 128, 130

Pantothenic acid, 100, 101, 107, 148

antagonism by salicylate, 153

in bios, 100

— — in human blood, 150

in mouse blood, 150

in rat blood, 150

occurrence, 107

properties, 107

Pantoyltauramide, antibacterial ac-
tion, 148

Pantoyltaurme, 125

— antibacterial action, 148
inhibition of, 149 ^

— resistant strains, 153
Papain, activation of, 48

— crystalline, 41

— synthesis by, 342
Parachromophoric organisms, 64
Parasiticin, 170

Parietin, 395

Pasteurella, growth fjxctors for, 106

Patulin, 160, 161, 170

— effect on common cold, 101
Penaldic acid, 173, 174

I

INDEX

493

Penatin, 168, 170
Penicillamine, 173, 174
PenicilHc acid, 163, 170, 288, 296
Penicillin, 171, 283, 305

— -A, 168

— -B, 168, 176

— -F, 172

— -G, 172

— -X, 172

— -K, 172

— I. 172

— II, 172

— Ill, 172

— activity, 175

— drug resistance to, 155, 170

— in chemotherapy, 175

— purification, 172

— structure, 174

— unit, 174
Penicillinase, 174, 176
Penicillium aurantio-hninneum, 378

— aurantio-griseum, 378

— aurantio-virens , 176

— hrevicaule, 305

— hrevi-compactum , 185, 378

— carminoviolaceum, 395

— charlesii, 296, 352

— chrysogenum, 283, 303, 305
penicillin from, 171

— citreoroseum, 395

— citrinum, 159

— citromyces, 327, 397

— citromyces-glabnim, 390

— claviforme, 160

— cydopiuin, 170, 395

— digitatum, 303

— expansuni, 351

— fu n ic ulosum, 395

— glaucum, 346

— griseo-fulvum, 307, 308

— italicum, 378

— javanicum, 378

— johannioli, 176

— luteimi, 293, 327, 351

— mannitol in, 62

— notatum, 168, 171, 283, 305

— palitans, 308

— patulum, 160, 307

— phoeniceum, 394

— puherulum, 176, 296, 378

PeniciUlum resticulosum, antibiotic
from, 184

— rubrum, 394

— spiculisporum, 286

— spinulosum, 179, 388, 394

— stoloniferum, 296

— varians, 352

— vitamin-C in, 65
Penicilloic acid, 173, 174
Penillamine, 173, 174
Penillic acid, 173, 174
iso-Penillic acid, 173, 174
Penillo-aldehydes, 173, 174
Pentacosanoic acid, 375
Pentamidine, 129
Pentosans, fermentation of, 312
Pentoses, in nucleoproteins, 62
Pepsin, 37

— crystalline, 41

— crystallisation, 45

— effect of ^H on, 48

— specificity, 51

— synthesis by, 40
Pepsinogen, crystallisation of, 45
Peptidases, specificity of, 51, 330
Peptides as determinant groups, 407,

440
Peptone, action of papain, 48
of trj-psin, 49

— in micro-organisms, 60
Peroxidase, 212

— crystalline, 41

— in bacterial respiration, 199

— inhibition, 200

— nature of, 40

- — purification, 45

— test for, 200
^H, definition, 8

— effect on bacteria, 4

on enzymes, 4

on pepsin, 4

trj^sin, 4

on Vibrio comma, 4

— of acids, 8, 9, 10

— of alkalies, 8, 9

— of blood, 8

— of water, 8

— determination of, colorimetric, 13

et seq.
electrometric, 11 et 'seq.

494

INDEX

Phenazine pigments, 168, 177, 178,

179, 391

effect on quinones, 168

Phenol from tyrosine, 233
Phenylacetic acid, 231
Phenylalanine, absence from moiild

proteins, 60

— as energy source, 218

— deamination, 229

— in antigens, 400
Phenylarsenoxide, 127

— reaction with SH groups, 136
Phenyl-lactic acid, 229
Phenylureides, 416
Phenylureido-proteins, 416
Phleimycolic acid, 375
Phoenicin, 394
Phosphatase, 265

— production by propionic acid bac-

teria, 96
Phosphates, in fermentation, 249 et
seq., 263

— in metabolism of autotrophic

bacteria, 72, 78
Phosphatides, constitution, 378

— content of bacteria and yeasts, 63,

378, 379

— of acid-fast bacteria, 379, 380

— of C. diphtherice, 381

— of L. acidophilus, 381

— of M. tuhercni^sis , 379
Phosphogluconic acid enzyme, 212
Phosphoglyceraldehyde, 320

— as hydrogen acceptor, 202
Phosphoglyceric acid, 249, 266, 275
breakdown inhibited by

fluoride, 254, 275

in lactic acid fermentation,

321

in propionic acid fermentation,

254

Phosphoglycero-mutase, 212, 276

Phosphopyridine nucleotides, 108

Phosphopyruvic acid, 247, 248, 249,
250, 266, 275, 277, 321

Phosphorylase, potato, 367

Phosphorylation in citric acid fer-
mentation, 298

Photobacterium fischeri, 158

Phthiocerol. 373, 374

Phthiocol, 392

— as growth factor, 114
Phthioic acid, 370, 374, 376, 379
Phycomyces hlakesleeanus , 102

growth factors for, 121

Physcion, 395, 396
Phytochemical reduction, 279
Phytoglycol, 372
Phytomonas tumefaciens , 362, 381

polysaccharide, 348

Phytomonic acid, 381
Pigments, carotenoid, 385, 387

— classification, 387

— melanin, 386

— production by micro-organisms, 64

— quinone, 386

Pilocarpine, inhibition of invertase,

135
Pimelic acid, 104, 109
plL, definition of, 10

— of acetic acid, 10

— of phosphoric acids, 10
Plasmodia, chemotherapy of, 130

— inhibition of respiration, 151
Plasmoquin, 128, 130

— inhibition of Plasmodium respira-

tion, 151

Plasteins, 399

Pneumococcus, see Diplococcus pneu-
monia

Polar groups, 30 et seq.

and adsorption, 34

and solubility, 33, 34

and specificity, 34

behaviour of, 33

— ■ — distribution of, 34

— ■ — • effect in antigen-antibody

reactions, 458

of masking, 34

on activation of hydrogen

donators, 195

on antibody production, 423

— on determinant groups, 407,

409

in polysaccharides, 33, 34

in proteins, 33, 34

strength of, 32

structure of, 31

Pol3^eptides in micro-organisms, 60

— utilisation of, 219

INDEX

495

Polysaccharides, as antigens, 401, 417

— F68 fractions, 361

— isolation, 360 et seq., 467

— of bacteria, fungi and yeasts, 61,

345 et seq.

— synthesis of, 257, 366
Polyuronides, 326
Porphyrin-protein enzymes, 52
Porphyrins, 391

Power alcohol, 311

— gas, 323 et seq.

from cellulose wastes, 325

sewage, 325

mechanism of production, 326

Precipitin reactions, 447 et seq.

composition of precipitate, 448,

449, 454

reacting groups in, 456, 459

theories of, 450 et seq.

Proactinomycin, 176
Prodigiosin," 387, 389
Proflavine, as wound antiseptic, 131

— drug resistance to, 155

Proline, absence from mould proteins,
60

— as energy source for anaerobes, 218

— degradation, 22 S, 237
Promin, 134
Prontosil, 123, 131, 133
Propamidine, 129

— drug resistance to, 155
Propionibacteriiim pentosaceum, 258
fixation of COo, 258

— — formation of succinic acid, 258
Propionic acid, 245, 319

bacteria, 253

— fermentation of glucose, 254

— of glycerol, 253, 259

phosphorylation, 249, 254,

255

detection, 465

Proseptasine, 133

Prosthetic group in enzymes, 41 et

seq.

growth factors and, 115

Protamine, action of trypsin, 49

— bacteriostatic action, 185

— in micro-organisms, 60
Proteinases, 342

proteins, action of trypsin, 49

Proteins, amino-acids from, 329

— as antigens, 398

— conjugated, 60

— constitution, 328

— content of bacteria, fungi and

yeasts, 59

— degradation, 224 et seq.

— hydrolysis, 329

— method of estimation, 59

— molecular weight, 328

— of tobacco mosaic virus, 330

— sparing action of carbohydrates,

226

— synthesis, 338 et seq.

— utilisation, 219

Proteolytic enzymes, 225, 227, 342

sjTithesis by, 342

Proteoses, utilisation of, 219
Proteus morganii, drug resistant

strains, 153
growth factors for, 108, 119

— vulgaris, action of pyridine-3-

sulphonic acid, 146

— — — of sulphonamides, 140

anaerobic growth, 208

aneurin in, 103

carbohydrates of, 365

— — carbon dioxide fixation, 259
^ growth factors for, 106, 119

— — proteolytic enzj^mes of, 225,

227
Protoplasm, 59
Pseudoglobulin, 424

— solubility, 60

Pseudomonas aeruginosa, action of sul-
phonamides on, 140

carbohydrates of, 362

pigment of, 64, 388, 389, 391

pyocyanase formation, 177

— chlororaphis, 179, 391

— fluorescens, 209, 228

adenosine triphosphate in, 72

Pseudopp-idoxine, 110

Ptyalin, activation of, 48

Puberulic acid, 176, 290

Puberulonic acid, 177

Purines, from nucleoproteins, 60, 332

— utilisation of, 219
Putrefaction, nature of, 102
Pyocyanase, 177

400

INDEX

Pyocyanase, Ivtic acition, 177
Pyocyanin, 177, 388, 391

— as hydrogen acceptor, 202, 391
Pyracin, 113
P3rridme-3-nitrile, 106
Pyridine-3-sulphonamide, 140, 149
Pyridine-3-sulphonic acid, 146
Pyridino-protein enzymes, 52
Pyridoxal, 110
Pyxidoxamine, 110
Pyridoxine, 100, 109

— acetyl derivatives, 110
Pyrimidines, as growth factor, 102

— from nucleoproteins, 60, 332

— utilisation of, 219
Pyrithiamine, antibacterial activity,

151
Pyruvic acid, 245, 284

anaerobic dismutation, 248

breakdown, 247

carboxylation, 259

decarboxylation, 241, 271, 272,

273
■ detection, 466

— — fixation of, 246
Pyruvic oxidase, 212, 248

occurrence, 43

prosthetic group, 43

Quinhydrone electrode, 12
Quinic acid, 389
Quinine, 125, 128, 130

— inhibition of lipase, 135

oi Plasmodium respiration, 151

Quinone pigments, 386

in oxidation-reduction systems,

393, 394

Racemiase, 321
Raffinose, 349, 350
Ravenellin, 306, 396
Reductase, 193

— acetaldehyde, crystalline, 41
Residual antigens, 353
Respiration, definition, 187
Rhizohium, fixation of nitrogen, 214,

223

— growth factors for, 104, 120

— leguminosarum, fixation of nitro-

gen, 223

Rhizohium, polysaccharide, 348

cross reaction with pneumo-

coccus antisera, 348
Rhizojpus chinensis, 282

— lactic acid production, 321

— oryzce, lactic acid from, 282, 293
Rhodobacillus i^alustris, 357
Rhodococcus, pigment of, 387
Rhodopin, 389
Rhodopurpurin, 390

Rhodotorula ruber, growth factor for,

102, 120
Rhodovibrin, 389
Rhodovibrio, pigments of, 389
Rhodoviolascene, 389
Riboflavin adenine dinucleotide as

hydrogen acceptor, 204 et seq.

— as growth factor, 99, 111

— in yeast, 339

- — 5-phosphoric acid, in enzymes, 42
Ribonuclease, crystalline, 41
Ribonucleic acid, 334

in Gram staining, 337

Ribose-3-phosphate, 333
Robison's ester, 264
R-salt-azo-diphenyl-azo-egg albumin,

449, 454
Rubrofusarin, 396

Saccharic acid, 285, 298, 299
Saccharomyces cerevisice, fixation of
nitrogen by, 215

Gram staining, 337

synthesis of proteins by, 339

— growth factors for, 101, 104, 105,

110, 121

— pulcherrimus, pigment of, 388
Salicylate, antagonism to panto-
thenic acid, 153

Salmonella paratyphi, amino-acid de-
gradation by, 232
carbohydrates of, 362

— schottmulleri, 232, 362

— typhimurium, 360. 402

— — action of sulphonamides on, 140
Salvarsan, 124, 127

Sarcina aurantiaca, 387, 390

— lutea, 387, 390

— methanica, 76
— ■ xanthine, 390

INDEX

49'

Scatole from tryptophane, 233
Schizophyll u m com m u n e, 3U5
Serine, deamination, 237

— ill tryptophane synthesis, 343

— toxicity to bacteria, 218
Serratia marcescens, anaerobic growth,

208

carbohydrates of, 362

pigments of, 64, 389

proteolytic enzymes of, 225

vitamins in, 115

Shigella dysenterice, carbohj^drates of,

362, 401, 417

glycogen in, 345

growth factors for, 106, 107,

120
variants, 89

— j)aradysenterice, carbohydrates of,

363
Sideromonas, 70
Silk fibroin, constitution, 329

haptens from, 408

Soluble specific substances. 61, 353

F68 fractions, 361

— hydrolysis by enzymes, 357

of ^. dnthracis, 362, 365

of B.C.G., 360

of Brucella, 366

of CI. welchii, 366

of C. dlphtherioe, 366

of Diplococcus pneumonice,

353
Type I, 354, 355, 401,

454, 460

Type II, 354, 355, 412

Type III, 354, 355,

356, 365, 412, 448, 450, 454, 461

Type IV, 354, 357

Type YIII, 354, 357,

412

Type XIV, 354. 357

of ^. typhosa, 362, 402

— ■ of Friedliinder's bacillus, 354

— — — of gonococcus, 365

oi H. infiuenzce, 366

— of H. parapertussis, 366

oi H. pertussis, 366

of Leptospira biflexa, 366

— of meningococcus, 365

of M. phlei, 358

Soluble specific sub.stances of 31.
tuberculosis, 358, 359, 375

■ — of Pasteurella, 366

of Proteus, 362, 365

— of Sal. typhimurium, 360, 402

• of Shigella dysenterice, 362

of Staph, aureus, 363

of Strep, salivarius, 364

of Vibrio comma, 364

Soluseptasine, 133
Sorbitol oxidation of, 243. 244, 304
Sorbose bacillus, 242
Spiculisporic acid, 286, 467
Spinulosin, 179, 393

— dimeth3d ether, 163
Spirilloxanthin, 390
Spirillum rubrum, 390
Spirochaetes, chemotherapy of, 127,

130
SjDores, water content of, 58
Staphylococcus albus, phosphorylation

enzj-mes in, 250

— antitoxin, 426

— aureus, action of pjTidine-S-sul-

phonic acid on, 146

of sulphonamides on, 140

■ adenosine triphosphate in, 72

carbohydrates of, 363

drug resistant strains, 155

■ growth factors for, 120

inhibition by p3^ridine-3-.sul-

phonamide, 149

by pyrithiamine, 151

pigments of, 387, 390

— candidus, COg fixation, 259

— citreus 387,

— growth factors for, 104, 106, 107,

111
Starch as antigen, 401

— in bacteria and fungi, 340

— staining by iodine, b^

— synthesis of, 367
Stearic acid, 369
Sterols, 371, 403

— in bacteria, 63, 377

— in yeast, 63
Stibacetin, 130
Stibenyl, 130

Stickland reaction, 218, 228, 230
Stilbamidine, 129

408

INDEX

Stipitatic acid, 291
iStreptamine, 181
ytreptidine, 180
Streptobiosaminide, 181
Streptococcus, antitoxin, 426

— hovis, dextran synthesis, 348

— cremoris, 319

— growth factors for, 105, 108, 110,

112, 113

— Group N, antibiotic in, 185

— hcemolyticus , growth factors for,

108, 120
pyruvic oxidase in, 43

— lactis, 246, 259, 319, 321

— paracitrovorus, CO2 fixation, 259

— jyyogenes, drug resistant strains,

153

inhibition by iodinin, 392

pantoyltaurme, 148

sulphanilamide, 149

— respiration of, 198, 211

— salivarius, 358

carbohydrates of, 364

dextran synthesis, 348

levan synthesis, 351

Streptomyces grisens, 180
Streptomycin, 180

— inactivation, 180

— purification, 180
Streptothricin, 180, 182
(Strychnine, antisera to, 413

— as determinant group, 413
Substrate, definition of, 47
Subtilin, 185

Succinic acid, 245, 300, 410

detection, 464, 466

formation of, 250, 251, 284,

293
inhibition of hydrogen donalors

by, 196

— dehydrogenase, 212

inhibition, 135

Sulamyd, 133
Sulochrin, 306, 396
Sulphacetamide, 133

— activity, 144

— as determinant group, 413

— ionisation of, 143
Sulphadiazine, 132

— activity, 144

Sulphadiazine, inhibition })y ^j-aniino-
benzoic acid, 139

— ionisation of, 141, 143
Sulphaguanidine, 133

— inhibition by p-aminobenzoio acid,

139
Sulphamerazine, 145
Sulphamethazine, 144
Sulphanilic acid, 406, 413

azo-protems, 406

Sulphanilamide, 131

— activity, 144

— as determinant group, 413

— inhibition of action, 136, 137. 149
by _p-aminobenzoic acid. 139,

140

— ionisation of, 141, 143
Sulphapyridine, 132, 146

— activity, 144

— as determinant group, 413

— inhibition bvj^-aminobenzoic acid,

138, 139

— ionisation of, 141, 143
Sulphasuxidine, 134
Sulphathiazole, 132

— activity, 144

— and decarboxylation, 147, 150

— as determinant group, 413

— inhibition by p-aminobenzoic acid,

138, 139, 140

— ionisation of, 141, 143
Sulphonamides, 132 et seq.

— and respiration, 147

■ — antibacterial index, 139

— antisera to, 413

• — as determinant groups, 413

— inhibition by adenine, 145

by _p-aminobenzoic acid, 138

by methionine, 145

— ionisation of, 141 et seq.
Sulphones, 134

Sulphur bacteria, 69, 70

bacterio-purpurin in, 64, 390

— - — green, 391

metabolism of. 70 et seq.

pigments of, 387, 390, 391

— compounds in fungi, 305

— oxidation by sulphur bacteria, 71

et seq.

— reduction by yeast, 279

INDEX

499

Sulphydryl groups, in eiiz3^mes, 136

reaction with phenylarsenoxide,

136
8vmbiosis, 102, 109
S^Tithalin, 129
SjTithetic media, 81, 462

Tartranilic acids as determinant

groups, 409
Terrein, 308
Terrestric acid, 288
Tetracosanoic acid, 371, 374, 375
Tetranucleotides, 335
Tetrathionates, oxidation by sulphur

bacteria, 71, 72
Tetronic acids, 171, 296
Thiamine, see Aneurin.
Thiazole as growth factor. 102
Thiohacillus, 69

— denitrificans , 69, 72

— thio-oxidans, 69, 71, 115
adenine triphosphate m, 72

— — aneurin in, 103
growth factors in, 115

— thioparus, 69, 70
Thiochrome, 393
Thiocystis, 69, 72
Thiorhodacece, pigment, 391
Thiosulphate, oxidation by sulphur

bacteria, 71, 72

— reduction bv yeast, 279
Thiothrix, 69 72 "

Thio-urea, effect on sulphonamide
activity, 145

— production by fungi, 305

Third component of complement, 435,

438
Threonine, toxicity to bacteria, 217
ThjTuine, 113, 333
Thymonucleic acids, 333
Thyroxine, antisera to, 415

— as determinant group, 415
Toluquinone derivatives as anti-
biotics, 163

— pigments, 386, 393, 394
Turula lipofera, 371

— — lipoid content of, 63, 371

— pigments in, 387

— rosea, vitamin-B in, 65

— rubra, 393

Tortila iitilis. protein synthesis by, 339

Tonilene, 393

Toxin-antitoxin reactions, 442 ct seq.

colloidal character, 446

Toxoflavin, 390
Toxoid, 417
Transaminase, 224

— in amino-acid sjmthesis, 224
Transamination, 341
Trehalose, in Aspergillus niger, 02

— in 31. phlei, 359, 373

tuberculosis, 359, 371, 373

— in yeasts, 62
Tricarboxylic acid cycle, 302
Trichoderma viride, antibiotic from,

185
Tricosanoic acid, 374
Trimethyl arsine, 304
Trimethyleneglycol, 467
Triphosphopyridine nucleotide, 268
Triose phosphate, dismutation, 204
Triose phosphorylase, 212, 320
Tritisporin, 395
Trypanosomes, arsenic resistant

strains, 126, 152

— chemotherapy of 125 et seq., 130

— growth factors for, 121
Trj-pan-red, 126
Trj'parsamide, 127
Trj.-psin, 37

— activation of, 49

— crystalline, 41

— crystallisation of, 46

— effect of pH on, 48

— specificity, 51
Trypsmogen, 49
Tryptophane, 81, 84, 216, 399. 400

— degradation of, 209, 228, 229, 231,

232, 233

— m antigens, 399, 400, 405

— synthesis of, 343
Tryptophol, 231

Tubercle bacillus, see MycvhaderiiDii

tuberculosis
Tuberculostearic acid, 369, 370, 374,

376, 379
Tyrocidine, 182

— action on fiuigi, 183

— hcemolytic effect, 183

— hydrochloride, 165, 182

500

INDEX

Tyrosinase, 52, 38G

— copper in, 41
Tyrosine, 111

— absence from mould proteins, 51

— degradation of, 229, 231, 232, 233,

386 389

— in antigens, 399, 400, 401, 405, 414
Tyrosol, 231

Tyrothricin, 165, 182

Uleron, 134
Uracil, 333

— as growth factor. 111
Urea, as energy source, 189, 211

— effect on sulphonamide activity, 145
Urease, crystalline, 41

— inhibition of, 49

— in 3Iicrococcus lysodeiJdicus, 93
Uridylic acid, 333

Urocanic acid, 230

?i-Valeric acid, from proline, 237

rf- Valine as energy source, 218

Varianose, 352

Verticillium albo-atrum, 305

F-factor, 86

Vibrio agar-liquefaciens, 325

— amylocella, 325

— comma, carbohydrates of, 353, 362,

364

co-enzyme in, 107

vitamins in, 115

Vinegar fermentation, 322

Violacein, 388, 389

• — as antibiotic, 185

Viridin, 185

Virus, metabolism, 81

— of tobacco mosaic, 330 et seq.
Vitamin-A, absence from bacteria and

yeasts, 65
Vitamin -B in bacteria, 65, 100
A'^itamin-C, absence from bacteria and

yeasts, 65

— production by fungi, 65
"\'itamin-D, absence from bacteria and

yeasts, 65

— from ergosterol, 63, 377
Vitamin-H, identity with biotin, 104
Vitamins, bacterial, 98, 245
Vogcs-Proskauer reaction, 252
Volutin, staining of, 56

Warburg's yellow enzyme, crystalline,

41

constitution, 42

riboflavin-5-phosphoric acid

in, 42
Wassermann reaction, 403 *
Water content of bacteria, fungi,

yeasts and spores, 58

— dissociation, 5

— hydrogen ions in, 6

— hydroxyl ions in, 6
Waxes, 371

— of acid-fast bacteria, 373, 374

— of C. diphtheric^, 377

— of micro-organisms, 63

X-factor, 86

Xanthine oxidase as bacteriostatic

agent, 169
Xanthophylls, 385
Xanthopterin, 113
Xylonic acid, 293

Yeasts, amino-acid degradation h\, 234

— carbohydrate content of, 61
— - fatty acids in, 371

— - fermentation of galactose by, 89, 96

— gums, 345

— fixation of nitrogen by, 215

— juice, 38

— — in fermentation, 263
zymase in, 53

— nucleic acids of, 332, 337

— nucleoproteins of, 60

— phosphatide content, 64

— proteins in, 338

— sterol content, 63

— Torula, pigments of, 64

— w^ater content of, 58
Yellow enzyme (Warburg's), 169

as hydrogen carrier, 206

constitution, 42

crj'stalline, 41

Zeaxanthin, 390
Zephiran, lytic action, 177
Zinc chloriodide, 56
Zone phenomenon, 251
Zvmasc, 53, 268
Zymin, 263, 280, 435
Zymohexase, 212, 275, 320
Zymosan, 436

Printed by MXorquodalc & Co. Ltd., Glasgow.