ANALYSIS OF 2,4-D METABOLITES IN

HIGHER PLANTS BY GAS

CHROMATOGRAPHY

By
NORMAN CLINE GLAZE

A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF

THE UNIVERSITY OF FLORIDA

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE

DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA
April, 1966

G £T 5" 2 a.

AGKI-

CUITURAL

UNARY

UNIVERSITY OF FLORIDA

3 1262 08552 3339

'fr*

ACKNOWLEDGEMENTS

The author wishes to express his sincere appreciation to Dr.
M. Wilcox, Dr. S. H. West, Dr. T. E. Humphreys, and Dr. H. C.
Harris for their advice and guidance during the course of these
studies.

Special acknowledgement is made to Professor Zygmunt Eckstein,
Katedra Technologii Organicznej II Politechniki, Warsaw, for
supplying reprints and translations of necessary passages from his
works, to Dr. G. R. Powell, Unit of Experimental Agronomy, Univer-
sity of Oxford, who supplied samples and physical properties of
4-hydroxy-2,3-dichloro- and 4-hydroxy-2,5-dichlorophenoxyacetic
acids and to Dr. R. H. Biggs for use of his gas chromato graph.

This work was made possible by a grant from the American
Cancer Society.

ii

TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS ii

LIST OF TABLES iv

INTRODUCTION 1

LITERATURE REVIEW 3

MATERIALS AND METHODS 17

EXPERIMENTAL RESULTS 28

DISCUSSION 34

SUMMARY 40

LITERATURE CITED 42

iii

LIST OF TABLES
Table Page

1 Operating Conditions for Gas Chromatography. . . .

2 Retention Times of Standard Solutions in

Methanol 30

3 Relative Peak Heights of Methyl and Propyl Esters

of the Propyl Ethers 32

iv

INTRODUCTION

The use of herbicides is increasing rapidly year by year.
2,4-Dichlorophenoxyacetic acid (2,4-D) is the most widely used of
the herbicides offered on the market. Although 2,4-D was one of the
first selective herbicides discovered and has been investigated by
many workers over the years, there are still many questions con-
cerning its metabolism.

Hydroxylated derivatives have been found in some microorganisms
and plants . This might be expected as hydroxylation is a prime
method of detoxication in biological systems. 2,4-D has been found
to be hydroxylated predominantly in the 5-position. However, 4-
hydroxy-2,3- and 4-hydroxy-2,5-dichlorophenoxyacetic acids have been
identified in some cases. This is interesting as it involves much
more elaborate mechanisms than hydroxylation in a vacant position.
The formation of these hydroxy acids requires a chlorine shift to
one of the meta positions and hydroxylation in the para position.
The object of this work is to determine whether hydroxylated deriv-
atives are formed by excised roots from 2,4-D or related herbicides.

Some workers have also found muconic acids and gamma-lac tones
as succeeding stages in the metabolism of these herbicides. This
would follow from an oxidative degradation of the ring structure.
The lactones found are of interest from the public health stand-
point. The gamma-lactones isolated contain double bonds in the

even numbered positions and compounds of this type have been found
to be carcinogenic. If these compounds appear in any appreciable
amounts and are not very transitory in nature, the use of the herbi-
cides involved might require reevaluation by regulating organiza-
tions.

LITERATURE REVIEW

The literature review will be restricted to the metabolism of
chloro- and methylphenoxyaliphatic and naphthyl- and naphthyloxy-
aliphatic acids by higher plants and microorganisms. Animal metab-
olism is covered by Williams (1950, 1959)* He states that phenoxy-
acetic acid and the ortho- and para-chloro-derivatives are unchanged
in animals. General classes of compounds which are considered to be
metabolites, such as phenols, catechols, etc., are discussed in his
publications.

Newman and Thomas (19^9) discovered that the persistence of
2,4-dichlorophenoxyacetic acid (2,4-D) in soils was decreased by
pretreatment of the soil with 2,4-D and certain other compounds
having similar substituent groups such as 2,4- dichlorophenol. How-
ever, pretreatment with some compounds such as 2- or 3-chlorophenol
or phenoxyacetic acid had no effect on the persistence of 2,4-D in
the soil. Burger et al. (1962) measured the persistence of some
phenoxyaliphatic acids by a bioassay technique using alfalfa seed-
lings. Acetic, alpha-propionic, alpha-butyric and gamma -butyric
acid homologs were toxic when applied to soil planted to alfalfa,
but beta-propionic acid analogs were not toxic to the test species.
Prolonged persistence was exhibited when the soil was treated with
3,4-dichloro- and 2,4,5-trichlorophenoxyaliphatic acids. The dura-
tion of phytotoxicity in soil receiving 4-chloro-, 2,4-dichloro- and

2-methyl-4-chlorophenoxyaliphatic acids was governed by the type and
linkage of the aliphatic side chain. Audus (1952) found that 2,4-D
enriched soil metabolized 2-methyl-4-chlorophenoxyacetic acid (MCPA)
immediately but did not affect the detoxication of 2,4,5-trichloro-
phenoxyacetic acid (2,4,5-T). MCPA enriched soil degraded 2,4-D
and 2,4,5-T.

Audus (1950) isolated Bacterium globiforme which was capable
of detoxifying 2,k- D. A Flavobacterium sp. was found by Burger e_t
al. (1962) which degraded phenoxybutyric acids. Only phenoxybutyric
acids having no meta-chlorine on the aromatic ring were metabolized
by the organism. These results show that the persistence of these
herbicides in soils is governed by the specific structural charac-
teristics of the molecule. A Corynebacterium sp., capable of de-
grading 2,4-D and MCPA, was isolated by Rogoff and Reid (1956). From
all indications, the ring was ruptured and complete destruction of
the molecule followed.

Audus (1952) suggested that the first step in the degradation
of 2,4-D, 2,4,5-T and MCPA might be the removal of the two carbon
side chain leaving a phenol. However, he found no accumulation of
phenols in soil treated with any of these herbicides. If a phenol
is an intermediate, it must be metabolized as rapidly as it is
formed. This is a possibility because it was found that 2,4-
dichlorophenol decomposed rapidly when added to a 2,4-D enriched
soil.

Leafe (1962) suggested two possibilities for the metabolism
of MCPA by Galium aparine. The first postulate listed for the

degradation of MCPA was the removal of the two carbon side chain
either as a two carbon fragment or its loss as carbon dioxide.
Secondly, he suggested that the metabolism or detoxication might
involve conjugation to another cellular component. Leafe stated
that the detoxication of the molecule by loss of both carbon atoms
of the side chain has been established as the reason Galium aparine
is resistant to MCPA.

Measurements were made by Canny and Markus (i960) of the rate
of evolution of labelled carbon dioxide from the shoots and roots
of Vlcia faba which had been treated on one leaflet with 2,4-D
labelled in the carboxyl group. The carbon dioxide evolved from the
roots was more radioactive than that from the shoot. Labelled car-
bon was shown to be incorporated into many compounds in the root.
This is evidence that the degradation may take place mainly in the
roots. It was also concluded that although metabolism of the alkyl
side chain was rapid, this was not the main pathway for inactiva-
tion of the herbicide.

Bach (I96I), working with stems of Phap^olus vulgaris, isolated
one half of the original radioactivity from carboxyl labelled 2,4- D
in an ether extract. The activity was found to be distributed among
ten components. Tests for functional groups showed that these com-
ponents retained the aromatic character of 2,4-D. Neither 2,4-D
nor 6-hydroxy-2,4-dichlorophenoxyacetic acid were isolated. Some
ten amino acids were also labelled in the treated stem tissue.

When red kidney beans were treated with labelled 2,4-D, Holley
(1952) found that in one week two thirds of the radioactivity was

recovered in a water soluble fraction. He stated that the water
soluble material must be a derivative of an unknown acid rather than
of 2,4-D. Holley suggested that the unknown acid might be 3-> 5-
or 6-hydroxy-2,4— dichlorophenoxyacetic acid.

Steenson and Walker (1956) isolated Flavobacterium peregrin urn
and two Achromobacter sp. which were capable of decomposing 2,4-D,
MCPA or 4-chlorophenoxyacetic acid. They found that pretreatment
of the soil with one of the compounds accelerated the rate of de-
composition of the other two. In a liquid medium, F. peregrinum
metabolized 2fk- D and liberated 76 per cent of the chlorine in ionic
form. They suggested that *4— chloro-2-hydroxyphenoxyacetic acid and
4-chlorocatechol might be intermediates in the degradation of h-
chlorophenoxyacetic acid.

A small, Gram-negative, motile organism was isolated from the
soil by Evans and Smith (195*0 which utilized k- chlorophenoxyacetic
acid and 2,k—B and produced 4-chlorophenol and 2,4-dichlorophenol,
respectively. A phenolic acid metabolite was found and it was
suggested to be 6-hydroxy-2, ^-dichlorophenoxyacetic acid. It was
also observed that the chlorine atoms were retained until the aro-
matic character of the molecule was destroyed.

Using a replacement culture technique, Byrde and Woodcock
(1957) showed that phenoxyacetic and beta-phenoxypropionic acids
were hydroxylated mainly in the para position by Aspergillus niger.
The compounds were also hydroxylated in the ortho position but to
a lesser extent. Gamma -phenoxy-n-butyric acid also was metabolized

mainly to 4-hydroxyphenoxyacetic acid by hydroxylation and beta-
oxidation. Delta-phenoxy-n-valeric acid was degraded in a similar
manner to beta-(4-hydroxyphenoxy)propionic acid.

Wilcox et al. (I963) found that barley, oats and corn had the
ability to produce a ring hydroxylated intermediate from a phenoxy-
n-aliphatic acid with an even number of carbons in the side chain
when incubated in aerated water. Odd chain homologs yielded little
of a hydroxylated metabolite. In soybeans, alfalfa and peanuts, no
hydroxylated intermediates were detected. The phenolic product,
produced by the test plants when treated with various phenoxy-n-
aliphatic acids having an even number of carbons in the side chain,
was identical with 4-hydroxyphenoxyacetic acid. These data suggest
that ring hydroxylation by the resistant grass species coincides
with the suggested role of hydroxylation as a detoxication mech-
anism.

Working with Aspergillus niger, Byrde and Woodcock (1957) demon-
strated that phenoxyacetic and beta-phenoxypropionic acids were
hydroxylated in the para position mainly and to a lesser extent in
the ortho position. Faulkner and Woodcock (l96la, 196lb) also
found phenoxyacetic acid to be hydroxylated in the ortho and para
positions. Clifford and Woodcock (1964), however, found 2-hydroxy-
phenoxyacetic acid to be the main metabolite from phenoxyacetic
acid. Bocks e_t al. (1964) also found 2-hydroxyphenoxyacetic acid
to be the main metabolite from phenoxyacetic acid. All other isomers
were formed but in much smaller amounts. Bocks also found phenol to be
a minor product. Faulkner and Woodcock (I96lb) found that 2-chloro-

8

and 4-chlorophenoxyacetic acids were metabolized to 2-chloro-4-
hydroxy- and 4-chloro-2-hydroxyphenoxyacetic acids, respectively.
The original starting materials were monohydroxylated in all vacant
positions but the above were produced in much larger amounts. Also
found was 2-hydroxyphenoxyacetic acid. This was the first example
of fungal replacement of chlorine by a hydroxyl group. Bocks et al.
used methoxybenzene as a substrate and observed that 2-hydroxy-
methoxybenzene was produced. If the incubation was stopped after
three days, phenol was present in almost the same proportions as
2-hydroxymethoxybenzene. If allowed to proceed for a longer period,
the yield of phenol relative to 2-hydroxymethoxybenzene decreased.

Evans and Moss (1957) t working with a Pseudomonas sp., extracted
2-hydroxy-4-chlorophenoxyacetic acid and 4- chlorocatechol from cul-
tures grown on a medium containing 4-chlorophenoxyacetic acid. This
was a strong indication that these were two of the possible inter-
mediates in the degradation of this herbicide.

Bell (i960), studying an Achromobacter sp., found an adaptive
system that is capable of oxidizing 2,4-D. The organism can also
oxidize many 2,4-D analogs. For the oxidation rate to be comparable
to 2,4-D, the organism appeared to require the following: a free
ortho position; a free carboxyl group on the sj.de chain, prefer-
ably beta to the ethereal linkage; a chlorine atom in the para
position; and no more than two chlorine atoms on the ring whether
or not an ortho position is free. He suggested that the probable
degradation route was to 2,4-dichlorophenol but that action by a
decarboxylase to 2,4-dichloromethoxybenzene might be involved. He

also suggested that the metabolites from 2,4-dichlorophenol might
be 4- chlorocatechol or 3j5-dichlorocatechol.

The metabolism of 2,4-D and MCPA by a strain of Flavobacterium
peregrinum and an Achromobacter sp. was studied by Steenson and
Walker (1957) • The evidence presented indicated that adapted
organisms metabolize 2,4-D through 2,4-dichlorophenol and 4-chloro-
catechol and that MCPA is degraded to ^-chloro-2-methylphenol. Bac-
teria, grown on a medium containing 2,4-D, did not oxidize any of
the five possible isomers but would oxidize 2,4- dibromo-, 4-- bromo-
2-chloro-, 4- chloro- and to a small extent, 2-chlorophenoxyacetic
acids .

Thomas et al. (1963? 1964a) studied the hydroxylation of some
phenoxyacetic acids by stem tissue of Avena sativa. They found that
phenoxyacetic acid was converted to 4-hydroxyphenoxyacetic acid.
Phenoxyacetic acids with an unsubstituted para position were hydrox-
ylated in that position. The phenolic acids were accumulated as the
4-0-beta-D-gluco sides. Phenoxyacetic acids with a chlorine atom in
the para position, however, were not hydroxylated to an appreciable
extent. These acids accumulated as neutral products which proved
to be phenoxyacetylglucoses. In the case of 2,4,6-trichlorophenoxy-
acetic acid, however, hydroxylation took place at the meta position
and the glucoside was isolated.

Thomas et al. (1964b) studied the metabolism of 2,4— D in stem
tissue of Phaseolus vulgaris . They found the predominant metabolite
to be 4-hydroxy-2,5-dichlorophenoxyacetic acid. This shows that the
hydroxylation and chlorine shift discovered first in Aspergillus

10

niger also occurs in higher plants . A minor metabolite was shown
to be 4-hydroxy-2,3-dichlorophenoxyacetic acid which also requires
the hydroxylation and chlorine shift. The products accumulated
within the tissue as beta - glue o sides.

Evans and Moss (1957) > using a Gram-negative, motile organism
of the Pseudomonas type, found that when it was incubated with k—
chlorocatechol, beta-chloromuconic acid accumulated. The beta-
chloromuconic acid was further metabolized and the chlorine lib-
erated in ionic form. Later Fernley and Evans (1959) isolated
aloha-chloromuconic acid during the metabolism of 2,^4— D which was
identical with that prepared by the oxidation of 2-chlorophenol.
They observed the shift of the 283 mp-* peak of 2,4-D in the ultra-
violet range to another peak at 272 mo-.

Schemes for the metabolism of chlorophenoxyacetic acids by soil
bacteria were reported by Rogoff (1961), in which 4-chlorophenoxy-
acetic acid was metabolized through 2-hydroxy-4-chlorophenoxyacetic
acid and 4-chlorocatechol to beta-chloromuconic acid and 2,4-D was
degraded through 2,h~ dichlorophenol and 3»5-dichlorocatechol to
albha-chloromuconic acid. Evans and Smith (l95l)5 Evans et al.
(1951), Dagley et al. (I9o0), Ribbons and Evans (1962) and Rogoff
all show schemes for further microbial oxidation of catechol type
derivatives. In all cases, the schemes show the catechol to cis-
cis-muconic acid step and terminate with beta-ketoadipic acid. The
steps between the cis-cis-muconic acid and beta-ketoadipic acid
appear to be somewhat uncertain. The uncertainty seems to be
whether the intermediates are lactonized or not when the double

11

bond migrates, which must occur if this scheme is to be correct.
The beta-ketoadipic acid is presumed to be split into acetic and
succinic acids which would pass into the tricarboxylic acid cycle.

Working with a Gram-negative soil bacterium, Gaunt and Evans
(1961) found that the ultraviolet peaks of MCPA at 228 and 279 mu..
are replaced by a peak at 277 W-- The properties of this compound
were consistent with the lactonic acid, alpha-meth.yl-gamma-carboxy-
methylene-A -butenolide. Peracetic acid oxidation of 4-chloro-2-
methylphenol produced a compound with the same properties. A
phenolic acid was also isolated and found to have properties con-
sistent with a hydroxy-2-methyl-^- chlorophenoxyacetic acid. It was
believed to be the 6-hydroxy-derivative. 4-Chloro-2-methylphenol
was also isolated from the extracts. The following metabolic path-
way for the degradation of MCPA was suggested: MCPA through 6-
hydroxy-4-chloro-2-methylphenoxyacetic acid; 5-chloro-3-methyl-
catechol; alpha-methyl-gamma-chloromuconic acid; alpha-methyl-parama-
carboxymethylene-A -butenolide; alpha -me thy lmaleylacetate and to
smaller fragments prior to entering a terminal respiratory cycle.

The lactones mentioned in the degradation schemes are of con-
siderable interest because butenolides of this general type with
double bonds in even numbered positions have been shown to be car-
cinogenic by Dickens and Jones (I96I, I963).

Byrde et al. (1956), using a replacement culture technique of
Aspergillus niger, found that omega- ( 2-naphth,yloxy ) -n-alipha tic acids
having an even number of carbons in the side chain were beta-oxidized
to (6-hydroxy-2-naphthyloxy)acetic acid. Beta- (2-naphthyloxy) propionic

12

acid was transformed to beta- ( 6-hydroxy-2-naphthyloxy )propionic acid.
The n-b,utyric homolog was hydroxylated in the six position before
beta-oxidation to (6-hydroxy-2-naphthyloxy)acetic acid. Hydroxyla-
tion of the acetic and propionic acids led to a decrease in toxicity
to the fungus. Byrde and Woodcock (1958), working with Sclerotinia
laxa, found that nuclear hydroxylation was virtually absent with this
organism. They did find that gamma- ( 2-naphthyloxy ) -n-butyric and
delta- (2-naphthyloxy)-n- valeric acids underwent beta-oxidation to
the corresponding acetic and propionic acids, respectively. They
also found beta-naphthol was produced from beta- ( 2-naphthyloxy ) -
propionic acid.

Fawcett et al. (195^) found that omega_-phenoxy-n-aliphatic acids
with an odd number of carbons in the side chain yielded a substantial
amount of phenol when applied to flax plants. Compounds with an
even number of carbon atoms in the side chain were beta-oxidized to
the acetic derivative. The decanoic homolog, however, was found to
produce a large amount of phenol. This was best explained from the
fact that omega -oxidation has been shown in animals for some com-
pounds with fatty acids of nine or ten carbon atoms. It was assumed
that this was also occurring in the flax plant.

Omega- (^—chlorophenoxy ) aliphatic and omega-(2,4-dichlorophenoxy)-
aliphatic acids showed the characteristic alternation of activity
in work by Wain and Wightman (195^) » when used in the Went pea test,
the tomato leaf epinasty and the wheat cylinder test. Only the wheat
cylinder test showed activity with even numbered aliphatic acids of
the 2,4,5-trichlorophenoxy series. When solutions of the

13

2,4,5-trichlorophenoxyaliphatic acids with side chains containing an
even number of carbon atoms were pretreated with wheat cylinders,
however, the alternation of activity was shown by the Went pea test
and the tomato leaf epinasty test. This evidence showed that wheat
possessed the ability to beta-oxidize this series but that pea and
tomato did not. When omega-1-naphthylaliphatic acids were tested,
the Went pea test and wheat cylinder test showed the alternation of
activity but only the acetic homolog was active in the tomato leaf
epinasty test. It was found that the substitution of alkyl groups
in the alpha position did not hinder beta-oxidation, but substitu-
tion in the beta position did prevent beta-oxidation.

Webley e_t al. (1955) studied the breakdown of omega-phenyl-
aliphatic acids by Nocardia opaca. They found that acids with an
odd number of carbon atoms in the side chain were converted to ben-
zoic acid and that cinnamic acid was an intermediate. Ortho-
hydroxyphenylacetic acid was found to be a metabolite when acids
with an even number of carbon atoms in the side chain were used in
the growth medium. Benzoic acid was produced essentially quantita-
tively but ortho-hydroxypheny lace tic acid was not, indicating that
it must arise from some side reaction. They stated that the latter
accumulates because it is not metabolized further by this organism.
The evidence suggested that N. opaca was capable of beta-oxidation
of the fatty acid side chains of these phenylaliphatic acids.

Webley et al. (1958) expanded their study to examine the effect
of change of various group;- in the molecule and how beta-oxidation
would be affected. They found the following five factors to affect

14

beta-oxidation : l) an oxygen bridge between the fatty acid and the
ring; 2) ring substitution, particularly in the ortho position; 3)
the nature of the ring substituents ; 4) the type of ring structure
to which the fatty acid is attached; and 5) the position of attach-
ment of the fatty acid. The side chain of phenylbutyric acid was
beta-oxidized more readily than phenoxybutyric acid. The position
of substituted chlorine atoms in the ring affected beta-oxidation
as follows: gamma- (3-chlorophenoxy)butyric acid was degraded most
rapidly; the 4-chloro-homolog was intermediate; and the 2-chloro-
compound was the least reactive. Monomethyl homologs reacted in the
same manner as the corresponding monochloro-compounds. The methyl
homologs, in general, were beta-oxidized more readily than the
corresponding chloro-compounds. The rate of conversion of the beta-
hydroxy- intermediate is reduced with the chlorinated materials.
2-Methyl-4-chloro- and 2,4- dichlorophenoxybutyric acids were tested
to evaluate the effect of disubstitution. It was found that disub-
stitution decreased the rate of beta-oxidation. The oxidation of
the disubstituted compounds proceeded to the beta-hydroxy-derivative
but the acetic homologs were not produced. The caproic homologs
were rapidly converted to the butyrics but again none of the acetic
homologs were obtained. Gamma- ( 2-naphthyloxy )butyr ic and beta-(2-
naphthyloxy)propionic acids were beta-oxidized to the corresponding
acetate and 2-naphthol, respectively. The 1-naphthyloxy-derivatives
remained unchanged throughout the experiment. Gamma- ( 1-naphthyl )-
butyric acid was converted to the acetate while the corresponding
naphthyloxy- compound was unchanged. Beta -oxidation of the fatty

15

acid side chain was most rapid when attached to a phenyl ring and
slowest when attached to a 1-naphthyl group. The indolyl group was
intermediate .

Fawcett et al . (1959), using the wheat cylinder, pea curvature,
pea . cjgment and tomato leaf epinasty tests, assayed the growth-
regulating activities of 18 homologous series of omega-phenoxy-
aliphatic acids with two to seven carbon atoms in the side chain and
with chlorine and methyl groups substituted in various ring posi-
tions. The degradation of some homologs was investigated and
identification of metabolites made by chromogenic methods and bio-
assay. They noted that there were four patterns of biological
activity as follows: 1) where beta-oxidation was operative, and
the acetic acid was active while the propionic acid was not, only
the even numbered homologs showed activity; 2) where beta-oxidation
was hindered at the butyric stage, and the acetic acid was active
but the propionic acid inactive, only the acetic acid showed activ-
ity; 3) where beta-oxidation was operative, and both the acetic
acid and propionic acid were active, all members showed activity;
and 4) where beta-oxidation was hindered at the butyric stage, and
the acetic acid was less active than the propionic acid, the acetic
acid plus all odd numbered homologs showed activity. The results
indicated that beta-oxidation is hindered at the butyric stage in
pea and tomato tissue. The hindrance at the butyric stage was shown
to be associated with an ortho chlorine or methyl group on the ring.
This effect was largely removed by a further chlorine atom in the
para position. Addition of a chlorine atom in tho nota position

16

had no effect. The beta-oxidation sequence was not changed when a
methyl group was substituted for a chlorine atom.

MATERIALS AND METHODS

The excised roots were prepared in the following manner: a
single layer of seed was placed between double thicknesses of
cheesecloth supported on aluminum mesh over a dishpan of aerated
water. These were germinated in a growth chamber in the dark at
71 F for a period of 4 to 7 days depending on the species. At the
end of this period, except in the case of alfalfa, the roots pro-
truding below the aluminum mesh were excised with a razor blade
and used in subsequent incubations. The alfalfa roots did not grow
through the cheesecloth and, therefore, the top layer of cheese-
cloth was removed and the seed cut off with scissors, after which
the roots were used in the incubation studies. The species used
were as follows: Coker 67 field corn (Zea mays L.), Hadden wheat
(Triticum aestivum L.). Moregrain oats (Avena sativa L.), Manchuria
X Rabat barley (Hordeum vulgare L.) and Hairy Peruvian alfalfa
(Medicago sativa L. ) .

Erlenmeyer flasks (300 ml.) were used as incubation vessels.
Tetracycline and streptomycin at concentrations of 2 and 30 ppm,
respectively, were used to prevent buildup of microbial populations.
Sets were run using 0.4 M tris [2-amino-2-(hydroxymethyl)-l,3-

propanediol] buffer (pH 7.4) and 0.4 M phosphate buffer (pH 5.2) at

_2

a final concentration of 2 X 10 M. The treatment flasks were made

1 X 10 "* M in the potassium salt of 2,4- D. All flasks were brought

17

18

to a final volume of 100 ml. Roots (10 g. fresh weight) were then
added to the flasks. The flasks were covered with a tissue and
put on a shaker at approximately 60 cycles per minute. A total of
12 flasks were used for each species. Six blank flasks were used
as well as six containing 2,4-D.

The incubations using tris buffer were incubated for 10 hours.
At the end of this time, the roots were removed from the flasks and
frozen in cold methanol. The roots were then ground in an omni-
mixer, keeping them frozen by immersing the vessel in a bath at
-50 C or colder. After grinding, the solution was centrifuged at
17,500 rpm (37,000 X g) for 30 minutes. The supernatant fraction
was concentrated to approximately 10 ml. by freeze drying. Each
blank and treatment was split in half. To one portion of each, 40
per cent sodium hydroxide was added so that the final solution was
10 per cent in sodium hydroxide. This solution was refluxed for 20
minutes in order to hydrolyze any esters present. The hydrolyzed
sample was brought to pH 3 with sulfuric acid and freeze dried.
Methanol was added to dissolve any hydroxylated derivatives that
might be present. The hydrolyzed and unhydrolyzed samples were then
reduced under a stream of nitrogen to a volume of approximately 1 ml.
and applied to 8 X 8 inch plates spread to a thickness of 1 mm. with
silica gel G (E. Merck Ag., Darmstadt) for thin layer chromatog-
raphy. These plates were run in chambers with an ether :ligro in
(b.p. 66-75 C):formic acid (50:50:2) solvent system. After drying,
the silica gel G between the origin and the solvent front was
scraped from the plate. This carrier was then eluted with acetone.

19

The acetone extracts were reduced under a stream of nitrogen to
approximately 10 ml. and split again. One portion was set aside.
The others were reduced under a stream of nitrogen to approximately
•5 ml. after which the volume was adjusted to 1 ml. with methanol.
These samples were treated with an ethereal solution of diazo-n-
propane. More of the diazo-n-propane solution was added as necessary
to keep the color of the solutions yellow. After 30 minutes, two
drops of a solution of boron trifluoride (0.7 per cent) in methanol
were added to further catalyze the reaction and the solutions were
kept yellow for an additional 30 minutes. At the end of this time,
the samples were again concentrated under nitrogen and adjusted to
a volume of 1 ml. These solutions were used for analysis by gas
chromatography.

The incubation mixtures containing phosphate buffer were treated
similarly except that the extraction of the roots was carried out
by dropping the roots into boiling methanol and boiling an additional
20 minutes. Both halves of the split samples were also used in this
case as two diazo compounds were used. One portion was treated with
diazome thane while the other was treated with diazo-n-propane as in
the previous case. Otherwise, both sets were handled as described
previously.

Solutions of 2,^-, 6-hydroxy-2,4-, 5-hydroxy-2,4-, 4-hydroxy-
2,3- and *J—hydroxy-2,5-dichlorophenoxyacetic acids in methanol were
prepared and treated with diazomethane and diazo-n-propane as de-
scribed previously. The final concentration of these solutions was
2.5 mg./ml. These solutions were used as standards for gas chromatog-
raphy.

20

Mixtures of some synthesized compounds (10 mg.) and lanolin
(k g.) were prepared for tomato epinasty tests. The following com-
pounds were tested: 5-nitro- and 5-amino-4-chloro-2-methylphenoxy-
acetic acids; ethyl 3-amino-2,4-dichlorophenoxyacetate; i|--hydroxy-
2,3-, Jj-hydroxy-2,5-, 6-hydroxy-2,4- and 5-hydroxy-2,4-dichloro-
phenoxyacetic acids. 2,k-D was also included as a standard.
Approximately 300 mg. of the lanolin mixtures of these compounds
were spotted on the petiole of the youngest fully developed leaf
of two month old tomato plants in the greenhouse. Daily observa-
tions were made to observe the effects of the compounds on the
plants.

All melting points were run on a Kofler micro hot stage apparatus
and the temperatures corrected. The compounds were synthesized as
follows :

Diazomethane. - Potassium hydroxide (^0 ml. of 30 per cent
solution) and anhydrous ethyl ether (160 ml.) were placed in a large
test tube and immersed in an insulated container holding an anti-
freeze solution at -10°C or lower. The contents of the tube were
allowed to cool for 20 minutes. N-methyl-N-nitro-N-nitrosoguanidine
(8 g.) was added to the tube. The bath temperature was allowed to
rise only to 0°C. The reaction is complete when the ether assumes
a yellow color and the solid matter disappears and usually is com-
plete in about 30 minutes. The ethereal solution was poured off
into a polyethylene bottle containing potassium hydroxide pellets.
The solution should be stored in a freezer and used within 2k hours
for best results.

21

Diazo-n-propane. - Potassium hydroxide (20 ml. of 60 per cent
solution) and anhydrous ethyl ether (160 ml.) were placed in a large
test tube and cooled as described above. N-propyl-N-nitrosourea
(8 g.), prepared previously by M. Wilcox, was added to the tube.
The reaction is complete when the ether assumes a deep yellow color
and the solid matter suspended in the potassium hydroxide loses its
yellow color, becoming white. The ethereal solution was poured off
as in the case of diazomethane. This solution should be stored in
a freezer and used within 2k hours for best results.

Monoperphthalic acid. - An ethereal solution of monoperphthalic
acid was prepared as described by Payne (1962). The solution was
standardized by adding 30 ml. of 20 per cent potassium iodide solu-
tion to 2 ml. of the ethereal solution and, after 10 minutes, titrating
the liberated iodine with 0.06268 N sodium thiosulfate. This solution
was used in subsequent attempted preparations.

Isopropyl k- chloro-2-methylphenoxyacetate. - To 95 Vev cent
4-chloro-2-methylphenoxyacetic acid (212 g.) was added 2-propanol
(26^ g.). Boron fluoride ethyl ether (178 g.) was added dropwise
and the mixture refluxed for 4 hours. The excess alcohol was dis-
tilled off and the solution neutralized with a 10 per cent solution
of sodium carbonate. The solution was then extracted with three
portions of ethyl ether and the ether removed under vacuum. The
remaining liquid was vacuum distilled (15 mm. Hg) and the portions
distilling over between 170 and 17^ C collected and combined. The
weight of these portions was 193*6 g. (79*8 per cent yield).

22

5-Nitro-4-chloro-2-methylphenoxyacetic acid. - This compound
was prepared as described by Eckstein et al. (196*0 « The compound
melted at ±'j±-152°C. Faulkner and Woodcock (1965) reported 152-
153°C

5-Amino-4-chloro-2-methylphenoxyacetic acid. - This compound
was also prepared as described by Eckstein e_t al. (1964). The com-
pound melted at 194-197°C Eckstein et al. (1964) reported 197-198°C.
The ethyl ester, prepared by the previously described boron tri-
fluoride method, melted at 76-78°C (Found: C, 54.51; H, 5.96; N,
5.95; CI, 14.38. CnH12N0 requires C, 54.22; H, 5*79; N, 5-75;
CI, 14.55 per cent). Faulkner and Woodcock (1965) reported 79-
80°C.

6-Nitro-4-chloro-2-methylphenol. - This compound was prepared
as described by Zincke (1918). He reported the melting point to be
107°C. Our preparation melted at 106°C (Found: C, 44.96; H, 3*^7;
CI, 18.32; N, 7.42. CJ^CINO requires C, 44.82; H, 3.22; CI, 18. 90;
N, 7.47 per cent).

Ethyl 6-nitro-4-chloro-2-methylphenoxyacetate. - 6-Nitro-4-
chloro-2-methylphenol (30 g.) was added to a solution of sodium (3. 7
g.) in 300 nil of absolute ethanol and heated to reflux. Ethyl bromo-
acetate (20 g.) was added dropwise through the condenser and the
mixture refluxed for 3*5 hours. Most of the ethanol was removed
under vacuum. The mixture was then diluted with water and the pre-
cipitate filtered off. The yield was 32 g. (75 per cent yield).
The compound melted at 70 C. Faulkner and Woodcock (1965) also re-
ported a melting point of 70 C.

23

6-Nitro-4-chloro-2-methylphenoxyacetamide. - Ethyl 6-nitro-4-
chloro-2-methylphenoxyacetate (5 g.) was added to 20 ml. of absolute
ethanol and refluxed with 5 ml. of concentrated ammonium hydroxide.
The mixture was allowed to reflux for 8 hours and 5 ml. of concen-
trated ammonium hydroxide was added at 2 hour intervals. The yield
of product was 1.2 g. (27 per cent yield). The compound melted at
153-156°C.

5-Chloro-3-methylcatechol. - Attempts were made to prepare this
compound through two intermediates described by Zincke (1918). The
structures given for the intermediates were dubious. In one trial,
however, a small amount of the desired product was obtained which
agreed with the melting point of 89°C (Found: C, 53.31; H, 4.5^;
CI, 22.05. C7H7C102 squires C, 53.02; H, k.k5\ CI, 22.36 per cent)
reported by Zincke. The last step in the synthesis was very low
yielding, the product unstable and results inconsistent. Since the
preparation of 2-methyl-l,4-muconolactone from this compound failed,
further attempts to prepare it were terminated.

h- Chloro-2-methylmuconic acid. - 4- Chloro-2-methylphenol (4.0
g.) was added to 175 ml. of ether containing monoperphthalic acid
(17 g«) and allowed to stand at room temperature for four days.
None of the desired product was obtained when the preparation was
treated as described by Testa (1952).

2-Methyl-l,4-muconolactone. - 3-Methyl-5-chlorocatechol (1.08
g.) was dissolved in 50 ml. of ether and 88 ml. of ether containing
monoperphthalic acid (4.3^ g.) added. This mixture was allowed to

24

stand at room temperature for five days. None of the desired product
was obtained when the preparation was treated as described by Testa
(1952).

6-Hydroxy-2,4-dichlorophenoxyacetic acid (6-0H-2,4-D). - This
compound was prepared as described by Cavill and Ford (195*0 using
3,5-dichlorocatechol previously prepared in this laboratory. The
compound after two recrystallizations from water melted at 133 C.
Cavill and Ford reported the melting point to be 132 C.

2-Chloro-l,4-muconolactone. - 2-Chlorophenol (6.0 g.) was added
to 164 ml. of ether containing monoperphthalic acid (9*1 g«). On
the following day, 90 ml. of ether containing monoperphthalic acid
(5»0 g.) was added and the solution allowed to stand for four days
at room temperature. This preparation, when worked up, failed to
yield the desired product.

5-Hydroxy-2,4-dichlorophenoxyacetic acid hydrate (5-0H-2,4- D). -
This compound was prepared as described by Moszew and Wojciechowski
(1945). The product after two recrystallizations from water melted
at 166°C (Found: C, 37.62; H, 3. 03; CI, 27.42. CghVCl O'h^O requires
C, 37.67; H, 3*16; CI, 27.80 per cent). Moszew and Wojciechowski
(1945) reported the melting point to be 175 «5 C.

3-Nitro-2,4-dichlorophenol. - This compound was prepared as
described by Groves e_t al. (1929). After two recrystallizations
from 2,2,4-trimethylpentane, the compould melted at 72 C. Groves
et al. found their preparation to melt at 70-72 C.

Ethyl 3-nitro-2,4-dichlorophenoxyacetate. - This compound
was prepared as described by Faulkner and Woodcock (1965). It was

25

found to melt from 80-82°C. Faulkner and Woodcock reported the
melting point to be 85°C.

Ethyl 3-amino-2,4-dichlorophenoxyacetate. - Ethyl 3-nitro-2,4-
dichlorophenoxyacetate (36 g.), 10 per cent palladium on charcoal
(2 g.) and 150 ml. of tetrahydrofuran were added to a bomb with a
total volume of ^30 ml. The bomb was charged with nitrogen and re-
leased several times to remove air. The bomb was then charged with
hydrogen to a maximum pressure of 150 p.s.i. several times and the
pressure drops recorded. The final total pressure drop was ^78
p.s.i. The theoretical pressure drop required was ^73 p.s.i. The
bomb was fastened on a shaker and shaken during the course of the
reaction. When the temperature of the bomb rose noticeably, it was
cooled to room temperature in an ice bath. At the end of the reac-
tion, the material was poured out of the bomb and the charcoal fil-
tered off. The tetrahydrofuran and the water produced were taken
off under vacuum. The residue was heated with 2,2,4-trimethyl-
pentane and the solvent cooled. After several extractions, a yield
of 20.6 g. (6^ per cent yield) was obtained. After recrystalliza-
tion from ligroine (b.p. 66-75°C), the product melted from 56-58°C.
Trials using Faulkner and Woodcock's (1965) procedure for this com-
pound were unsuccessful. They reported that their preparation melted
from 57-58°C

3-Hydroxy-2,4-dichlorophenoxyacetic acid. - Attempts to diazotize
ethyl 3-amino-2,4-dichlorophenoxyacetate and obtain 3-hydroxy-2,A—
dichlorophenoxyacetic acid have been unsuccessful in this laboratory,
although Faulkner and Woodcock (1965) have prepared this compound.

26

2,3-Dichlorohydroquinone. - This compound was prepared using
the procedure described by Conant and Fieser (1923). The product
melted at 1^3°C which agreed with the author's work. Attempts to
prepare this compound by the rearrangement of 2,5-dichlorobenzo-
quinone and subsequent saponification as described by Dimroth et al.
(1926) failed to yield a usable product.

Ethyl ^-hydroxy-2,3-dichlorophenoxyacetate. - 2,3-Dichlorohydro-
quinone (10 g.) was dissolved in 100 ml. of absolute ethanol con-
taining sodium (1.28 g.) and refluxed for 2 hours during the dropwise
addition of ethyl bromoacetate (9. 3 g«)« The excess ethanol was then
removed under vacuum and the solution diluted with water. The prod-
uct, which precipitated out of the solution, was filtered off. The
yield was 12.7 g. (86.5 per cent yield) and after recrystallization
from ligroine (b.p. 60-90°C) melted at 79°C

4-Hydroxy-2,3-dichlorophenoxyacetic acid (4-0H-2,3-D)« - Attempts
to hydrolyze the above ester with acid or base have failed to yield
a significant amount of product with sufficient purity for use. Con-
densation of 2,3-dichlorohydroquinone with bromoacetic acid has also
failed to give the desired product. A small sample, however, was
supplied by Thomas et al. (1963) and melted at l67-l68°C.

Ethyl ^-hydroxy-2,5-dichlorophenoxyacetate. - This compound was
prepared in the same manner as the corresponding 2,3-dichloro- analog
starting with 2,5-dichlorohydroquinone. The product melted at 95 C.

k- Hydroxy-2,5-dichlorophenoxyacetic acid (4- 0H-2,5-D). - Ethyl
4-hydroxy-2,5-dichlorophenoxyacetate (^ g.) was added to 50 ml. of
10 per cent aqueous sodium hydroxide and refluxed for 2 hours. The

27

solution was then acidified and cooled. White crystals precipitated
from the solution. After two recrystallizations from water, 0.6 g.
(16.6 per cent yield) of product melting at I66-I67 C was obtained.
The infrared spectrum of this material was identical with that of a
sample supplied by Thomas et al. (1963). Their preparation melted
at 161-162 C. Attempts to prepare this compound by condensation of
2,5-dichlorohydroquinone with bromoacetic acid failed, as mentioned
previously in the case of the 2,3-dichloro- analog.

EXPERIMENTAL RESULTS

Portions of the treated solutions were injected into the gas
chromatograph as described in table 1. Ten u.1. injections were
used for the root extracts and 1 ul. for the standards. The reten-
tion times of the standards are given in table 2. 2,^-D was de-
tectable when as much as 0.1 u.g. was present in the sample injected
into the gas chromatograph. A sample containing 0.25 l^g* of the
hydroxy-derivatives was necessary for reasonable detection on the
equipment used in this study. Since each alkylated extract repre-
sented 66 mg. of 2,^-D substrate and recovery was estimated at
75 per cent or better, the detection threshold of 0.25 M.g. hydroxy-
dichlorophenoxyacetic acid represented excellent sensitivity. In
the samples treated with diazome thane, a peak corresponding to
2,4-D was present in all of the 2,4-D treated samples. This peak
was masked by the solvent peak due to the higher sensitivity and
large sample injection used when diazo-n-propane was the reactant.
All other peaks in the 2,4-D treated samples did not correspond to
any of the standards and were also present in the control extracts.

To determine the disappearance of 2,4-D from the incubation
media, three 1 ml. samples were taken from one of the flasks at
3 hour intervals during the incubation. These solutions were con-
centrated under nitrogen almost to dryness and adjusted to 1 ml.
with methanol. They were then treated with an ethereal solution of

28

29

TABLE 1
Operating Conditions for Gas Chromatography

Manufacturer

Model

F and M
ij-00

Detector

Column

diameter
length

Carrier
flow

Hydrogen flame ionization

15 per cent SE-30 on Chromosorb W

(60/80 mesh)
3/16 inch
6 feet

Helium

60 ml./min.

Temperatures
flash heater
column
detector

235°C
190°C
240°C

Electrometer settings
knowns

diazomethane treatment
diazo-n-propane treatment

Range
10
10
10

Attenuation

8

1

30

TABLE 2
Retention Times of Standard Solutions in Methanol

Parent
compound

Methyl ester
methyl ether

Methyl ester
propyl ether

Propyl ester
propyl ether

man.

2,4-D

4.0

6-0H-2,4-D

6.7

5-0H-2,4-D

8.1

4-0H-2,3-D

10.2

4-0H-2,5-D

8.0

min.

9.4
10.9
13.5
11.4

min.

6.5
16.4
19.8
16.0
21.5

31

diazo-n-propane, reduced under a stream of nitrogen and readjusted
to 1 ml. with methanol. These solutions were run on the gas chro-
matograph. The size of the peaks representing 2,4— D were not re-
producible between replicates but there was a continuous decrease
in the average peak size as the length of the incubation time in-
creased. There was no peak representing 2,4— D after 12 hours of
incubation. This showed that 2,4-D was being removed from the solu-
tion.

Samples (2.5 mg.) of the synthesized hydroxylated derivatives
were added to a blank extract and carried through the purification
procedure. It is estimated from the relative size of the resulting
peaks on the gas chromatograph that more than 75 per cent of each
known was recovered in the final solutions. It was observed that
the retention times of the knowns added to the blank did not agree
with the standards when treated with diazo-n_-propane. Since the
methanolic extracts had been agitated for several hours, it was
suspected that the acids had been converted to methyl esters. In
order to resolve this discrepancy, methanolic solutions of the
standards in tightly capped tubes were heated in a water bath at
75 C for 2 hours. The object of this procedure was to determine
whether the phenolic acids could be esterified under very mild
conditions. At the end of this time, they were treated with diazo-
n-propane and analyzed on the gas chromatograph. Partial methyl
esterification was indicated by the appearance of a second peak
other than the propyl ether ester from each of the standards, as
shown in table 3« The retention times of the second peaks agreed

32

1

TABLE 3
Relative Peak Heights of Methyl and Propyl Esters
of the Propyl Ethers

Parent compound Methyl ester Propyl ester

6-0H-2,4-D 11.0 1^.5

5-0H-2,4-D 6.6 22.1

M)H-2,3-D 3.3 23*0

4-0H-2,5-D 2.8 19.6

1 Anhydrous methanolic solutions of the samples, in closed tubes,
were heated in a water bath at 75°C for 2 hours and then alkylated
with the appropriate ethereal diazoalkane.

33

with those subjected to the purification procedure and the diazo-
n-propane treatment. The diazomethane treatment gave retention
times which agreed with the standards.

4-Hydroxy-2,3-, ^--hydroxy-2,5-, 6-hydroxy-2,4- and 5-hydroxy-
2,4-dichlorophenoxyacetic acids produced no epinasty on the test
plants. 5-Nitro- and 5-amino-^-chloro-2-methylphenoxyacetates
showed positive epinasty on the test plants on the day following
treatment. Ethyl 6-nitro-4-chloro-2-methylphenoxyacetate produced
no epinasty during twelve days following treatment. Ethyl 3-amino-
2,k- dichlorophenoxyacetic acid showed positive epinasty within 2
hours of treatment, as did 2,4-D.

DISCUSSION

The greatest difficulty in this work was the lack of reference
compounds for use as standards. The next problem was the deter-
mination of an analytical procedure for detection of minute amounts
of these substances from biological materials. Preparations for
some of the compounds involved in this study were published by
others during the course of this work.

Using the procedure of Faulkner and Woodcock (1965), the
yield of 3-amino-2,4-dichlorophenoxyacetic acid was very low. In
an attempt to obtain a better yield, we used hydrogen in a bomb
with palladium on charcoal as a catalyst and obtained a yield of
64 per cent. Faulkner and Woodcock reported a yield of 56 per cent
but in this laboratory this yield was never approached using their
procedure.

Difficulty was also experienced in the procedures of diazotiza-
tion of aromatic amines to phenols reported by the above workers.
In attempts to follow their procedures of diazotizing ethyl 3-
amino-2,4-dichlorophenoxyacetate and ethyl 5-amino-4- chloro-2-
methylphenoxyacetate to the corresponding hydroxy-acids , none of
the desired products were obtained.

6-Hydroxy-4-chloro-2-methylphenoxyacetic acid has never been
synthesized but Gaunt and Evans (I96I), using a soil bacterium, iso-
lated a hydroxy-4-chloro-2-methylphenoxyacetic acid believed to be

34

35

the 6-hydroxy-analog. They also isolated a gamma- lactone which
presumably would be produced via the former compound. For this
reason, an unambiguous synthesis of this compound is of considerable
interest. We prepared 6-nitro-4-chloro-2-methylphenoxyacetamide as
it is possible that amide substitution would prevent the expected
cyclization to a lactam during the reduction of the corresponding
nitro compound to the amino analog. This amide may prove to be a
useful intermediate in the unambiguous synthesis of 6-hydroxy-4-
chloro-2-methylphenoxyacetic acid .

Difficulty was also encountered in the last step of the syn-
thesis of 5-chloro-3-methylcatechol as described by Zincke (1918).
Due to the dubious structures reported for the intermediates and
the low-yielding last step, the preparation of this compound was
abandoned. This compound would probably be an intermediate between
6-hydroxy-4-chloro-2-methylphenoxyacetic acid and the previously
mentioned gamma-lac tone . Because of these difficulties any future
attempts to prepare 5-chloro-3-methyl-catechol should await an un-
ambiguous synthesis of 6-hydroxy-^-- chloro-2-methylphenoxyacetic
acid such as by the above suggested method. Cleavage of 6-hydroxy-
4-chloro-2-methylphenoxyacetic acid prepared in this manner with
hydriodic acid or a similar reagent would be expected to give in
turn 5-chloro-3-methylcatechol unambiguously.

Under the conditions of this work, 2,4-D was not metabolized to
4-hydroxy-2,3-, 4- hydroxy-2,5-, 6-hydroxy-2,4- or 5-hydroxy-2,4-
dichlorophenoxyacetic acid in detectable amounts. Conversion at
any one time of 0.4 per cent of the original 2,4-D in the incubation

36

mixtures to any of the possible hydroxylated intermediates would
have been readily detectable. If any of these derivatives were
formed, they must have been further metabolized as rapidly as they
were produced. It is possible, however, that under some other con-
ditions these derivatives might be detected.

Preparation of more volatile derivatives is necessary for
analysis of phenolic acids by gas chromatography. The use of solu-
tions of diazoalkanes in ether to alkylate the carboxyl and hydroxyl
groups of the possible metabolites proceeded smoothly during this
work. This method has the advantage that one reactant will quickly
convert both carboxylic and phenolic hydroxyls to more volatile
groups at room temperature. After the initial reaction with the
diazoalkane had subsided, a small amount of 'boron trifluoride was
added to further catalyze the reaction and destroy the excess diazo-
alkane as the methyl ether.

By treating methanolic solutions of standards of the above mono-
hydroxydichlorophenoxyacetic acids with diazomethane and diazo-n-
propane, it is possible to prepare derivatives which allow separation
by means of gas chromatography. Use of only one of the reagents will
not produce separable derivatives of all four of the hydroxy-acids
under the conditions of this work. Treatment with diazomethane did
not separate derivatives of 5-hydroxy-2-^— and 4-hydroxy-2,5-di-
chlorophenoxyacetic acids. Diazo-n_-propane treatment did not sepa-
rate derivatives of 6-hydroxy-2,4- and 4-hydroxy-2,3-dichloro-
phenoxyacetic acids .

37

Attempted preparations of gamma-lactones mentioned as metabo-
lites of 2,4-D and MCPA by Fernley and Evans (1959) and Gaunt and
Evans (196l), respectively, were unsuccessful. Synthesis of these
lactones was attempted by oxidation of the corresponding catechol
or phenols with monoperphthalic acid as described by Testa (1952).
The only compound which could be extracted and identified from
these solutions was phthalic acid.

Hydrolysis of ethyl 2,3-dichlorophenoxyacetate to the corre-
sponding acid also failed to yield a usable product. Hydrolysis by
both acid and base was attempted but the desired hydroxy-acid was
not produced. A high melting compound was extracted which may have
been a polymer. Rearrangement of 2,5-dichlorobenzoquinone to the
2,3-dichloro-analog desired as a possible precursor in the synthesis
of the previously mentioned compound by the method of Dimroth et al.
(1926) also proved unsuccessful. We suspect Dimroth et al. sepa-
rated the isomers in a mixture rather than actually causing a re-
arrangement of the molecule.

No detectable amounts of substrate (l X 10"^ M 2,4-D) were
present after 12 hours incubation of excised roots of the plant
species studied. The root extracts, however, did give a peak
representing 2,4-D which showed that the roots had taken up the
herbicide. This indicated that the incubation had not been long
enough for all of the 2,4-D to have been converted to some product
further removed metabolically from the herbicide than those for
which analysis was being made.

38

One method which might make identification easier would be the
use of methanol as the extraction solvent with addition of boron
trifluoride, and subsequent heating to cause esterification of the
carboxylic groups. The resulting esters could be treated with
various diazoalkanes which would etherify the phenolic groups. This
method would enhance dissimilarities in the molecules and might pos-
sibly give better separation in gas chromatography.

Otherwise, it would seem advisable to use some extraction sol-
vent other than methanol which would not react with the acids being
analyzed. This would eliminate some of the identification problems
encountered in this work. Ethyl acetate would be a good solvent to
consider.

Another modification in the procedure which might improve the
technique would be the use of an electron capture detector on the
gas chromatograph. This would probably eliminate most of the peaks
noted except those of the compounds containing chlorine. Elimina-
tion of peaks for most of the natural products would increase the
worker's ability to identify the compounds being analyzed. Another
advantage would be the higher sensitivity to be expected with an
electron capture detector.

In the tomato epinasty tests, it is possible that the 5-nitro-
and 5-amino-4-chloro-2-methylphenoxyacetic acids were contaminated
with MCPA since they were prepared from the parent herbicide. How-
ever, the purification procedures should reduce this possibility
to a minimum. This is particularly true for the amino compound, as
its preparation from MCPA requires two steps with associated

39

purifications. There should have been no 2,4-D contamination in the
ethyl 3-amino-2,4-dichlorophenoxyacetate as it was prepared from
m-nitrophenol and had no common precursor with 2,4-D. Epinasty was
not observed for 6-hydroxy-2,4-, 5-hydroxy-2,4-, 4-hydroxy-2,3- and
4-hydroxy-2,5-dichlorophenoxyacetic acids . 6-Hydroxy-2,iJ— dichloro-
phenoxyacetic acid previously had been shown to be inactive by
Cavill and Ford (1954). This might be expected as hydroxylation is
known to be a prime method of detoxication by biological systems.

SUMMARY

The object of this work -was to determine whether hydroxylated
metabolites or farther derived compounds were formed during the
degradation of 2,4-dichlorophenoxyacetic acid by excised roots of
several species of higher plants. Excised roots were incubated in
a solution of 2,4-D for periods of 10 and 12 hours at two different
hydrogen ion concentrations. The roots were then extracted with
methanol. After prepurification by thin layer chromatography, the
solutions were treated with diazomethane and diazo-n-propane to
alkylate phenolic acids and analyzed by gas chromatography. Syn-
thetic 6-hydroxy-2,4-, 5-hydroxy-2,4-, A-hydroxy-2,3- and 4-hydroxy-
2,5-dichlorophenoxyacetic acids were alkylated similarly and used
as standards in the analysis.

The extracts from the roots contained no detectable amounts of
the possible metabolites for which the analysis was being made. All
of the extracts from 2,4-D treatments did show a peak with a reten-
tion time identical with 2,4-D.

Tomato epinasty tests showed that none of the previously men-
tioned hydroxylated analogs tested were active. Ethyl 6-nitro-4-
chloro-2-methylphenoxyacetate also showed no activity in the test.
Ethyl 3-amino-2,4-dichlorophenoxyacetate and 2,4-D were very active
and began showing epinasty effects within 2 hours of treatment. 5-
Nitro- and 5-&mino-4-chloro-2-methylphenoxyacetic acids also gave

40

41

positive results in the test. The former compound was almost as
active as 2,4-D. The latter was less active and the effects didn't
appear until the second day after treatment.

LITERATURE CITED

Audus, L. J. 1950. Biological detoxication of 2,4-dichlorophenoxy-
acetic acid in soils. Isolation of an effective organism.
Nature 166:356.

Audus, L. J. 1952. The decomposition of 2,4-dichlorophenoxyacetic
acid and 2-methyl-4- chlorophenoxyacetic acid in the soil. J.
Sci. Food Agr. 3:268.

Bach, M. K. 196l. Metabolites of 2,4-dichlorophenoxyacetic acid
from bean stems. Plant Physiol. 36:558.

Bell, G. R. i960. Studies on a soil Achromobacter which degrades
2,4-dichlorophenoxyacetic acid. Can. J. Microbiol. 6:325.

Bocks, S. M., J. R. L. Smith, and R. 0. C. Norman. 1964. Hydroxyla-
tion of phenoxyacetic acid and anisole by Aspergillus niger (van
Tiegh). Nature 201:398.

Burger, K., I. C. MacRae, and M. Alexander. I962. Decomposition of
phenoxyalkylcarboxylic acids. Soil Sci. Soc. Am. Proc. 26:243.

Byrde, R. J. ¥. , J. F. Harris, and D. Woodcock. 1956. Fungal detoxi-
cation. The metabolism of (ju-(2-naphthyloxy)-n-alkylcarboxylic
acids by Aspergillus niger. Biochem. J. 64:154.

Byrde, R. J. W. , and D. Woodcock. 1957* Fungal detoxication. 2.
The metabolism of some phenoxy-n_-alkylcarboxylic acids by
Aspergillus niger. Biochem. J. 65:682.

Byrde, R. J. W. , and D. Woodcock. 1958. Fungal detoxication. J>.
The metabolism of cju-(2-naphthyloxy)-n-alkylcarboxylic acids by
Sclerotinia laxa. Biochem. J. 69:19*

Canny, M. J., and K. Markus . i960. The breakdown of 2,4-dichloro-
phenoxyacetic acid in shoots and roots. Australian J. Biol. Sci.
4:486.

Cavill, G. W. K., and D. L. Ford. 1954. The chemistry of plant-
growth regulators. Part I. 2:4-Dichloro-6-hydroxyphenoxyacetic
acid and related compounds. J. Chem. Soc.:565«

Clifford, D. R., and D. Woodcock. 1964. Metabolism of phenoxyacetic
acid by Aspergillus niger van Tiegh. Nature 203: 763.

42

43

Conant, J. B., and L. F. Fieser. 1923* Reduction potentials of
quinone. I. The effect of the solvent on the potentials of
certain benzoquinones . J. Am. Chem. Soc. 45:2194.

Dagley, S., ¥. C. Evans, and D. W. Ribbons, i960. New pathways in
the oxidative metabolism of aromatic compounds by micro-
organisms. Nature 188:560.

Dickens, F., and H. E. H. Jones. 196l. Carcinogenic activity of a
series of reactive lactones and related substances. Brit. J.
Cancer 15:85.

Dickens, F., and H. E. H. Jones. 19&3- Further studies on the car-
cinogenic and growth- inhibitory activity of lactones and related
substances. Brit. J. Cancer 17:100.

Dimroth, 0., H. Eber, and K. Wehr. 1926. III. Uber das benzochinon-
dichlorid. Ann. 446:132.

Eckstein, Z., E. Dyszer, and T. Niedzwialowska. 1964. 0 nitrowaniu
estrow kwasow 2-metylo- oraz 2-etylo-4-chlorofenoksy-octowego.
I. Poszukiwanie metody syntezy analogow alkilowych 2,4,5-T.
Roczniki Chemii 38:51.

Evans, W. C, and B. S. ¥. Smith. 1951* The oxidation of aromatic
compounds by soil bacteria. Biochem. J. 49 :x.

Evans, ¥. C, B. S. W. Smith, R. P. Linstead, and J. A. Elvidge.

1951* Chemistry of the oxidative metabolism of certain aromatic
compounds by micro-organisms. Nature 168:772.

Evans, ¥. C, and B. S. W. Smith. 1954. The photochemical inactiva-
tion and microbial metabolism of the chlorophenoxyacetic acid
herbicides. Biochem. J. 57 :xxx.

Evans, ¥. C. 1956. Biochemistry of the oxidative metabolism of aro-
matic compounds by micro-organisms. Ann. Rept. Progr. Chem.
(Chem. Soc. London) 53:279.

Evans, ¥. C, and P. Moss. 1957. The metabolism of the herbicide,
p_-chlorophenoxyacetic acid by a soil micro-organism — the forma-
tion of a P-chloromuconic acid on ring fission. Biochem. J.
65:8P.

Faulkner, J. K., and D. ¥oodcock. 196la. Metabolism of chloro-
phenoxyacetic acids by Aspergillus niger. Chem. Ind. (London):
1366.

Faulkner, J. K., and D. ¥oodcock. 19olb. Fungal detoxication. Part
V. Metabolism of 0- and p_-chlorophenoxyacetic acids by Aspergillus
niger. J. Chem. Soc.:5397«

44

Faulkner, J. K., and D. Woodcock. 1964. Metabolism of 2,4-dichloro-
phenoxyacetic acid ("2,4-D") by Aspergillus niger van Tiegh.
Nature 203:865.

Faulkner, J. K., and D. Woodcock. 1965. Fungal detoxication. Part
VII. Metabolism of 2,4-dichlorophenoxyacetic and 4-chloro-2-
methylphenoxyacetic acids by Aspergillus niger. J. Chem. Soc:
1187.

Fawcett, C. H., J. M. A. Ingram, and R. L. Wain. 1954. The B-

oxidation of phenoxyalkylcarboxylic acids in the flax plant in
relation to their growth-regulating activity. Proc. Roy. Soc.
(London), Ser. B 142:60.

Fawcett, C. H., R. M. Pascal, M. B. Pybus, H. F. Taylor, R. L. Wain,
and F. Wightman. 1959. Plant growth-regulating activity in
homologous series of tu-phenoxyalkanecarboxylic acids and the
influence of ring substitution on their breakdown by B-oxidation
within plant tissues. Proc. Roy. Soc. (London), Ser. B 150:95.

Fernley, H. N., and W. C. Evans. 1959. Metabolism of 2,4-dichloro-
phenoxyacetic acid by soil Pseudomonas : Isolation of a-chloro-
muconic acid as an intermediate. Biochem. J. 73:22P.

Gaunt, J. K., and W. C. Evans. 1961. Metabolism of 4-chloro-2-
methylphenoxyacetic acid by a soil micro-organism. Biochem. J.
79:25P.

Groves, L. G., E. E. Turner, and G. I. Sharp. 1929. LXXVT. - The
scission of diaryl ethers and related compounds by means of
piperidine. Part II. The nitration of 2:4:4'-trichlorodiphenyl
ether, and of 2:4-dichlorophenyl p_-toluenesulphonate and benzoate.
J. Chem. Soc.:512.

Holley, R. W. 1952. Studies on the fate of radioactive 2,4-dichloro-
phenoxyacetic acid in bean plants. II. A water soluble trans-
formation product of 2,4-D. Arch. Biochem. Biophys. 35:171.

Leafe, E. L. I962. Metabolism and selectivity of plant-growth regu-
lator herbicides. Nature 193:485.

Moszew, J., and J. Wo jcie chows ki. 1945* Studia nad synteza regula-
torow wzrostu roslin, pochodnych fenoli duruvodorotlenowych.
Roczniki Chemii 28:445.

Newman, A. S., and J. R. Thomas. 1949. Decomposition of 2,4-

dichlorophenoxyacetic acid in soil and liquid media. Soil Sci.
Soc. Am. Proc. 14:160.

Payne, G. B. I962. Monoperphthalic acid. Org. Syn. 42:77.

45

Ribbons, D. ¥. , and W. C. Evans. 19 62. Oxidative metabolism of
protocatechuic acid by certain soil Pseadomonads : A new ring-
fission mechanism. Biochem. J. 83:482.

Rogoff, M. H., and J. J. Reid. 1956. Bacterial decomposition of
2,4-dichlorophenoxyacetic acid. J. Bacteriol. 71:303*

Rogoff, M. H. 1961. Oxidation of aromatic compounds by bacteria.
Advan. Appl. Microbiol. 3:193.

Steenson, T. I., and N. Walker. 1956. Observations on the bac-
terial oxidation of chlorophenoxyacetic acids. Plant Soil 8:17.

Steenson, T. I., and N. Walker. 1957* The pathway of breakdown
of 2,4- dichloro- and 4-chloro-2-methylphenoxyacetic acid by
bacteria. J. Gen. Microbiol. 16:146.

Testa, E. 1952. Oxydationen durch Wasserstoffsuperoxyd und

Persauren die zun spaltung von C-C bindungen fuhren. Juris-
Verlag, Zurich. 49 p.

Thomas, E. W. , 3. C. Loughman, and R. G. Powell. 1963* Hydroxyla-
tion of phenoxyacetic acid by stem tissue of Avena sativa.
Nature 199:73*

Thomas, E. W. , B. C. Loughman, and R. G. Powell. 1964a. Metabolic
fate of some chlorinated phenoxyacetic acids in the stem tissue
of 1 vena sativa. Nature 204:286.

Thomas, E. W., B. C. Loughman, and R. G. Powell. 1964b. Metabolic
fate of 2,4-dichlorophenoxyacetic acid in the stem tissue of
Phassolus vulgaris . Nature 204:884.

Wain, R. L., and F. Wightman. 1954. The growth-regulating activity
of certain uu-substituted alkylcarboxylic acids in relation to
their B-oxidation within the plant. Proc. Roy. Soc. (London),
Ser. B 142:525.

Webley, D. M., R. B. Duff, and V. C. Farmer. 1955* Beta-oxidation
of fatty acids by Nocardia opaca. J. Gen. Microbiol. 13:361.

Webley, D. M., R. B. Duff, and V. C. Farmer. 1958. The influence
of chemical structure on B-oxidation by soil Nocardias. J. Gen.
Microbiol. 18:733.

Wilcox, M., D. E. Moreland, and G. C. Klingman. 1963. Aryl hydroxy-
lation of phenoxyaliphatic acids by excised roots. Physiol.
Plantar um 16:565*

h6

Williams, R. T. 1950* Biological oxidation of aromatic rings.
Biochem. Soc. Symp. (Cambridge, Engl.) Ho. 5« 96 p.

Williams, R. T. 1959* Detoxication mechanisms. The metabolism
and detoxication of drugs, toxic substances and other organic
compounds. Chapman and Hall, London- 795 P«

Zincke, T. 1918. Uber die einwirkung von salpetersaure auf
halogenderivate von o-alkylphenolen. Ann. kl7: 191 •

BIOGRAPHY
Norman Glaze was born in Washington, D. C, on January 16,
193^« He attended public schools in Hyattsville, Maryland, and
was graduated from Northwestern High School in 1952. He obtained
the Bachelor of Science degree from the University of Maryland in
1957* Upon completion, he was inducted into the U. S. Army and
served for two years. After completion, he reentered the University
of Maryland and received the Master of Science degree in 1963* He
then transferred to the University of Florida for work on the Doctor
of Philosophy degree.

47

This dissertation was prepared under the direction of the
chairman of the candidate's supervisory committee and has bet
approved by all members of that committee. It was submitted to
the Daan of the College of Agriculture and to the Graduate
Council, and was approved as partial fulfillment of the require-
ments for the degree of Doctor of Philosophy.

April, 1966

^^/TTean, College of Agriculture

Supervisory Committee :

hair man /

Cha

Dean, Graduate School

3180