CHAPTER 26
Herbicides
W . A. Andreae
I. Introduction 243 II. The Action of Herbicides on the Plant 245
A. Auxin Herbicides 245 B. Inhibitors of Growth Regulator Action 249
C. Inhibitors of Cell Divison 250 D. Inhibitors of Photosynthesis 251 E . Inhibitors of Normal Chloroplast Development 253
F. Inhibitors of Metabolic Synthesis 253 G. Herbicides as Inhibitors of More than a Single Process 253
III. Metabolism of Herbicides by Plants 254 A. Conversion of Precursors to Active Herbicides 255
B. Metabolic Inactivation 256
References 258
I. INTRODUCTION
The term herbicide, as used in agriculture, applies to a heterogeneous group of chemicals with the property of eradicating all vegetation or of selectively killing weeds without seriously injuring the cultivated crops.
Since the turn of the century, the number of known herbicides has in- creased steadily, and at present about 80 are produced commercially. A comprehensive list can be found in the report of the Terminology Com- mittee of the Weed Society of America (1). Almost all commercial herbi- cides were first developed empirically, and studies of their modes of action were undertaken only after their economic benefits were recognized. The results of these investigations are of considerable interest to biology and agriculture, since they furnish new insights into the physiological proc- esses of plants and at the same time provide a rational approach to the development of new chemicals to satisfy specific needs.
This chapter deals with the metabolic relations between the applied 243
244 W. A. ANDREAE chemical and the plant and is mainly concerned with those herbicides which specifically affect plant metabolism. A number of herbicides, such as metabolic inhibitors not specific for plants, inorganic salts, and petro- leum oils, will not be discussed here beyond a brief mention. Metabolic poisons, such as sodium cyanide, sodium arsenite, calcium arsenate, or substituted phenols, when applied as herbicides may act through their established role of blocking or uncoupling specific enzyme systems, as discussed elsewhere in this volume, although other, as yet unknown toxic actions may be involved. In general, enzymes of plant or animal origin are sensitive to the same specific enzyme poisons. However, similar physi- ological processes in plants or animals may be mediated by different metabolic pathways, so that inhibitors may differ in their toxicity to plants or animals; for example, only a small part of the respiration of higher plants is cyanide sensitive (#), whereas the respiration of many animals is almost entirely inhibited by cyanide. General poisons, such as chloropicrin, methyl bromide, phenylmercuric acetate, and pentachloro- phenol, are in use as soil sterilants where they are effective against weeds and weed seeds as well as insects. In practice, because of toxicity hazards, efforts are made to replace these general metabolic inhibitors by safer, plant-specific herbicides.
The first observation of selective weed killing was accidental and dates back to 1896, when a French vineyard was sprayed with copper sulfate as a preventive measure against fungal infection of the vines. When the spray drifted into the adjacent grain field, the selective kill of broadleaf weeds within the cereal crop was noted. Ammonium sulfate and iron sulfate are also selective weed killers, but their use as herbicides has been largely superseded by more effective organic compounds. At present, high concentrations of certain fertilizer salts, such as calcium cyanamide and kainite as well as dilute sulfuric acid, are used as selective weed killers in grain crops. Their herbicidal action may be ascribed to tissue corrosion on physical contact, although some as yet unrecognized metabolic action may also be involved. Ammonium sulfamate, borax, potassium cyanate, and sodium chlorate are herbicidal at relatively low concentrations, but in spite of long usage no satisfactory explanation of their action has been advanced.
Petroleum oils came into use as general plant eradicants about 40 years ago. More recently, it was noted that light oils are selective herbicides;
for instance, Stoddard Solvent is relatively harmless to carrots and other umbelliferous plants but toxic to most other vegetation. The proportion of aromatic hydrocarbons in the oils seems to increase their toxicity. Oils per se are probably not metabolic inhibitors but act by physical plugging
26. HERBICIDES 245 of the tissue vessels during transport. There is, however, good evidence that various oxidation products of aromatic hydrocarbons, when present in the atmosphere, are extremely phytotoxic (3).
The first part of this chapter deals with the physiological and bio- chemical effects of plant-specific herbicides on plant processes. The second part deals with the metabolic transformation of herbicides within the plant.
II. THE ACTION OF HERBICIDES O N THE PLANT
A. Auxin Herbicides
At present, three general types of naturally occurring plant growth regulators are recognized: the auxins, the gibberellins, and the kinins.
Each possesses properties distinctly different from the others, and normal growth depends on their interactions. As yet, herbicidal activity has been found only among the synthetic auxin analogues. At low concentrations all auxins stimulate cell enlargement as measured by the pea stem or Avena coleoptile elongation test. A positive response in such isolated tissue segments is the only accepted criterion of auxin activity. With concentrations which are supraoptimal for growth, the stimulatory effect is progressively decreased. This has been referred to as the growth in- hibitory action, although even with relatively high concentrations growth still exceeds that of the control tissues. Auxins also regulate the meriste- matic activity of intact plants, so that the visible symptoms of auxin herbicide application are the result of their combined actions on cell elongation and cell division.
The first auxin was isolated by Kogel et al. (4) in 1934 and was identi- fied as 3-indoleacetic acid (IAA). IAA is still the only naturally occurring auxin whose structure is irrefutably established. Within a short time other indole acids and certain naphthalene acids were found to possess auxin activity, but not until 5 years later was there any hint, unpublished at the time, that growth regulators might possess selective herbicidal activity. In 1939 Slade, Templeman, and Sexton sprayed oat seedlings with naphthaleneacetic acid (NAA) and noted the selective phytotoxic action on broadleaf plants in a stand of oats. In 1942 substituted benzoic and phenoxyacetic acids were introduced as powerful auxins (5). In 1944 the first references to 2,4-D and 2,4,5-trichlorophenoxyacetic acid
(2,4,5-T) as selective herbicides were published by Hamner and Tukey (6). The independent works on 2,4-D and 2-methyl-4-chlorophenoxy-
246 W. A. ANDREAE CI
CH2—COOH
CI
0 - C H2- C O O H
Η
3-Indoleacetic acid 2,4-Dichlorophenoxyacetic acid (2,4-D)
(IAA)
CH2—COOH
1-Naphthaleneacetic acid iNAA)
acetic acid (MCPA) by Slade, Templeman, and Sexton ( 7 ) , by Nutman, Thornton, and Quastel (8), carried out and reported in 1942 and later by Blackman (9) were not published until 1945 because of wartime restric
tions but predated the work of Hamner and Tukey by several years.
Koepfli et al. {10), in 1938, on the basis of compounds then known, described certain requirements for auxin action: a ring system as a nucleus with at least one double bond in the ring; a side chain possessing a carboxyl group or a group easily converted into a carboxyl group; at least one carbon atom between the ring and the carboxyl group in the side chain; a particular spatial relationship between the ring system and the carboxyl group. With the introduction of substituted benzoic and phenoxyacetic acids it also seemed necessary to postulate that at least one ortho position in the ring must be free. However, a low activity of 2,6-dichlorophenoxyacetic acid in the pea test (11) makes the universality of this requirement questionable. Smith et al. (12) recognized that at least one hydrogen atom on the phenoxyacetic acid side chain is essential for activity. Thus, 2-(2,4-dichlorophenoxy)isobutyric acid is inactive, whereas 2-(2,4-dichlorophenoxy)propionic acid is active and is an effec
tive herbicide with modified selectivity. The discovery by van der Kerk et al. (13) that S-(dimethylthiocarbamyl)thioacetic acid shows apprecia
ble activity throws further doubt on many of these generalizations. Fur
ther details on various aspects of the structure-activity relations can be found in a recent review by Wain (14). At present, it would appear that
26. HERBICIDES 247
except for the substituted benzoic acids all active auxins possess an acetic acid moiety.
In recent years considerable attention has been paid to the cell wall as the primary site of auxin action, since auxin application stimulates cell wall synthesis (15) and increases cell wall plasticity (16). From electron microscope observations it appeared that during wall expansion the cellulose microfibrils did not change their orientation (17). Hence, cell wall expansion could not be caused by the stretching of the cellulose matrix. Attention was then focused on the quantitatively much smaller cell wall component, the pectic fraction, as the matrix-cementing sub- stance. Bennet-Clark (18) first discussed the possible role of poly- galacturonic acid chains on cell wall loosening. From physicochemical considerations it appears that the extensibility of the pectic acid com- ponent is controlled by the carboxyl group. As long as the carboxyl groups remain free, electrovalent bonds and to a lesser extent hydrogen bonds will provide considerable tensile strength and rigidity to the wall, but when the pectic acid is methylated to pectin there will be reduced tensile strength and greater plasticity. There are several experimental observa- tions which support the pectin hypothesis of cell wall plasticity change.
Calcium ions are known to inhibit growth and to cause electrovalent binding of adjacent pectic acid molecules (19). Chelators like ethyl- enediaminetetraacetic acid ( E D T A ) not only counteract the effect of added calcium but, for a limited time, resemble auxins as growth regu- lators (20). IAA has been reported to stimulate the incorporation of the methyl groups of applied methionine into the pectin molecule (21) and to inactivate the pectin methylesterase activity by binding this enzyme to a cell wall fraction (22). Thus, auxins would favor both plasticity, by stimulating the pectin synthesis, and decreased rigidity, by inhibiting the enzymic hydrolysis of pectin to pectic acid. Unfortunately there is no agreement on these hypotheses. The growth response to applied E D T A only superficially resembles a true auxin action (23); moreover, there is controversy about the effect of auxin on the pectin methylesterase ac- tivity; it has been reported to stimulate (24) rather than to inhibit this enzyme (22) and to be without effect on the binding of this enzyme to the cell wall (25).
Besides the action on cell wall metabolism, other metabolic effects of auxins have been investigated. The inhibitory effect of auxin on ascorbic
S-(Dimethylthiocarbamyl)thioacetic acid S
248 W. A. ANDREAE acid oxidase activity (26) and the consequence of this inhibition on growth rates (27) have been the subject of over 20 papers by Marre and co-workers in Milan. These papers have been critically reviewed (28).
2,4-D has recently been shown to have a direct effect on protein reten
tion in detached leaves (29). Fully grown leaves, if inserted into water, lose most of their chlorophyll and more than half of their original protein content after a few days. Application of 2, 4-D, kinetin (6-furfurylamino- purine) (80, 81), or benzimidazole (32) preserves the protein content and the green color.
Auxins have been reported to alter the physical state of the cellular proteins in such a manner as to protect the tissues proteins from heat coagulation (33). The physiological significance of this observation must await further experimentation.
For many years it seemed plausible that auxins may form thio esters with coenzyme A (CoA), since nearly all active auxins possess an acetate side chain. Such auxin-CoA thio esters might be expected to affect plant metabolism at sites where acetyl CoA participates as the essential meta
bolite. There is, in fact, indirect evidence that auxins do form such thio esters in plants. The known β-oxidations of indolebutyric acid (84) and 2,4-dichlorophenoxybutyric acid (85) undoubtedly, by analogy with ani
mal reactions, involve the intermediary formation of CoA thio ester, as would the observed formation of indoleacetylaspartic acid (86). Recently, the synthesis of the CoA thio esters of IAA, 2,4-D, and N A A has been accomplished by Zenk (37) using an enzyme preparation from liver mitochondria. While the CoA thio ester of IAA further reacted with glycine, it is of particular interest that the CoA thio ester of 2,4-D did not. There is, however, no evidence that the auxin thio esters participate in growth stimulatory processes, and, indeed, one could imagine that the reverse is the case. In the formation of the auxin thio ester as an inter
mediate in the continuous synthesis of indoleacetylaspartic acid, it is con
ceivable that IAA competes for CoA with synthetic processes of growth, and with high concentrations of IAA, growth would thus be inhibited.
Whether auxins stimulate or inhibit growth depends on their concentra
tion and duration of their application. Bonner and Foster (88) proposed a scheme which assumes a two-point attachment between the auxin mole
cule and a specific receptor site within the tissues, giving a growth stimu
latory auxin-receptor complex analogous to enzyme-substrate complexes.
High concentrations would favor the condition where two auxin molecules become attached to the same receptor site, forming an inactive auxin- receptor-auxin complex. The kinetics of such a system lends itself to thermodynamic treatment, and a theoretical growth curve has been calcu-
26. HERBICIDES 249 lated for the Avena coleoptile-auxin system. The experimental results reported by Bonner and Foster (88) closely fitted the theoretical growth curve, but other workers, using slightly different experimental conditions, were unable to confirm these data (89). Even if one assumes that auxins react with a single, rate-limiting site the growth effect which can be measured experimentally requires so many diverse secondary reactions that the experimental conditions become very critical.
In contrast to the growth effects on isolated tissue sections where the action is immediate, the herbicidal effects in intact plants develop gradu- ally and undoubtedly involve many other factors. While the effect of auxin herbicides can be ascribed, in part, to an intensification or prolonga- tion of growth stimuli, death is probably due to the cumulative effects of a general metabolic derangement. Many differences in the chemical com- position (carbohydrates, proteins, fats, etc.) of treated tissues have been reported, particularly in long-term experiments or with application of supraoptimal concentrations. These observations have been reviewed in detail (40-42). The reported changes are often minor or inconsistent and, taken by themselves, do little to clarify either growth stimulation or herbicidal activity. However, although individually these results are not impressive, combined they indicate a significant difference in the over-all metabolic balance. Since in the case of auxin herbicides, the effects of a single application extend over weeks, relatively small changes may, in time, bring about extreme metabolic derangements leading to the death of the plant. Indeed, a slow, generalized deterioration marks the final phases of 2,4-D treatment, and death cannot be ascribed to the modifica- tion of any one specific metabolic mechanism.
B. Inhibitors of Growth Regulator Action
Maleic hydrazide counteracts growth stimulation caused by applied IAA in the slit pea curvature test (48). This apparent antiauxin effect conceivably could be caused by lowering of the internal auxin level, since in vitro experiments have shown that maleic hydrazide accelerates the enzymic destruction of IAA (44)- However, direct measurements of the internal auxin level in maleic hydrazide-treated tissues show no signifi- cant changes (45). It is more probable that maleic hydrazide, by acceler- ating the IAA loss from bio-assay solutions, curtails the availability of IAA to the tissues. The stimulatory action of maleic hydrazide on IAA oxidase activity is similar to that of certain monohydric phenols (46).
Although the name maleic hydrazide implies a diketone structure (1,2- dihydropyridazine-3,6-dione), maleic hydrazide in solution exists only as
2 5 0 w . a. a n d r e a e the enol structure [6-hydroxy-3(2H)-pyridazinone] (47). The phenolic nature of maleic hydrazide, is further indicated by the fact that it is con
jugated in the plant with sugars (48). Thus, maleic hydrazide resembles a monohydric phenol both in structure and biological activity.
Maleic hydrazide is an interesting growth retardant, since it acts with
out altering the over-all growth pattern. This type of inhibition stands in striking contrast to the action of the gibberellins, which specifically promote the internodal elongation of treated plants. Whereas maleic
OH I HC,
HC. .NH I I II
ο
M a l e i c h y d r a z i d e N-l-Naphthylphthalamic acid (NAP)
[(CH3)3N+— C H2- C H2- C 1 ] CI"
N-Trimethy 1-2 - chloroethylamine chloride
hydrazide overcomes the stimulatory effects of applied gibberellins, gib- berellin does not overcome the inhibitory action of applied maleic hydra
zide, which makes it difficult to accept maleic hydrazide as an antagonist of endogenous gibberellin (49).
Recently a new compound, iV-trimethyl-2-chloroethylamine chloride, has been described as an antigibberellin, since not only does it decrease internodal elongation of intact plants but this inhibition is overcome by applied gibberellin (50).
Antiauxin activity has been ascribed to iV-l-naphthylphthalamic acid ( N A P ) , a herbicide which inhibits the geotropic response of roots, a response which is considered to be under auxin control (51).
C. Inhibitors of Cell Division
Since studies on cell elongation were of surprisingly practical benefit, other cellular functions, such as cell division, were investigated in the hope of developing new herbicides acting through other basic processes. Ethyl phenylcarbamate, a known inhibitor of cell division, was subsequently found to have a very desirable phytotoxic selectivity (52). In contrast
26. HERBICIDES 2 5 1 to the auxin herbicides, grasses were more susceptible than broadleaf plants. Systematic studies of related carbamates led to the introduc
tion of isopropyl iV-phenylcarbamate (IPC) (52) and, later, isopropyl AT-(3-chlorophenyl) carbamate (CIPC), and 4-chloro-2-butynyl AT-(3- chlorophenyl) carbamate (barban) as selective herbicides.
Soon after the introduction of maleic hydrazide, this compound became known as an inhibitor of cell division (53), and some workers ascribed its action, not only on prolonging dormancy but also on growth retardation, to the inhibition of cell division (54, 55). While the phenylcarbamates and maleic hydrazide both interfere with normal mitosis, maleic hydra
zide differs from other chromosome-breaking agents in not providing a sticky chromosome surface (53).
D. Inhibitors of Photosynthesis
Wessel and van der Veen (56), working with isolated chloroplasts, were the first to recognize that herbicidal phenylureas, such as 3-(p-chloro- phenyl) -1,1-dimethylurea (monuron) and 3 - (3,4-dichlorophenyl) -1,1- dimethylurea (diuron), are extremely powerful inhibitors of the Hill reaction and provided direct evidence that the phenylureas inhibit the mechanism of photosynthesis. The probable site of action of these herbi
cides on the Hill reaction has since been identified by two independent approaches. Bishop (57) showed that photoreduction by adapted algae, in contrast to photosynthesis, was not inhibited by diuron. Photoreduction differs from photosynthesis in that hydrogen replaces water as the ulti
mate electron donor; thus, photoreduction utilizes only part of the enzyme systems involved in photosynthesis, since the enzymes for splitting water are not required. Jagendorf (58), working with preparations of chloro
plasts from higher plants, showed that the formation of A T P , a by-product of the Hill reaction, is not inhibited by monuron if iV-methylphenazonium ions function as the electron carriers. Later, Krall et al. (59) found that if ΛΓ-methylphenazonium ions serve as the electron carriers, there is no oxygen production during the Hill reaction; apparently, the reduced ΛΓ-methylphenazonium ions are capable of entirely replacing water as the
Isopropyl iV-phenylcarbamate (IPC)
4-Chloro-2-butynyl N-(3-chlorophenyl) carbamate
(barban)
252 W. A. ANDREAE electron donor in the Hill reaction just as hydrogen replaces water in photoreduction of adapted algae. The results of these studies on the Hill reaction, therefore, fully confirm Bishop's original hypothesis that the mechanism of molecular oxygen production (i.e., the oxidation of water) is blocked by these phenylureas.
Phenylcarbamate herbicides, such as IPC and CIPC, also inhibit the Hill reaction but are much less potent than the phenylureas (60), whereas a new class of herbicides consisting of the anilides of aliphatic acids, such as iV-(3,4-dichlorophenyl)-2-methyl pentanamide (karsil) cause inhibi-
a
- J
r\ .
i, - L
f, (
c a'
3 - (/>-Chlorophenyl)-1, 1-dimethylurea
(monuron)
CI
\ , Ο C H3
II I
CI—(( NV- N - C - CH—CH2— C H2— C H3
^=J^ Η
Ν- (3,4- Dichloropheny 1)- 2 - methyl pentanamide
(karsil)
CI
I
I II
C HS- C H2- N H/ C ; :N/ C^ N H - CH2—CHS
2 - C h l o r o - 4 , 6 - b i s ( e t h y l a m i n o ) - s - t r i a z i n e (simazin)
tion of oxygen production at similarly low concentrations as diuron. The phenylureas, phenylcarbamates, and acylanilides are structurally related in that they are all anilides. Simazin [2-chloro-4,6-bis(ethylamino)-s- triazine] (61) and propazin [2-chloro-4,6-bis(isopropylamino)-s-triazine]
also inhibit photosynthesis by inhibiting the oxygen production (61a).
26. HERBICIDES 253 Ε. Inhibitors of Normal Chloroplast Development
Herbicide application frequently destroys the green color of treated plants but with certain herbicides this is the major symptom. Amitrole (3-amino-l,2,4-triazole) at low concentrations induces chlorosis in young leaves, although the same low concentrations do not affect the chlorophyll content of the mature leaves. Apparently, leaves are highly susceptible at the stage of chloroplast development; if the plastids are not already formed at the time of treatment, the leaves remain permanently chlorotic
Η
H N ^ N ° ^ C/ 0- C H - C H3
1 I I I
N ° - N H2
°
Η3"~ίΓ°
C V X ) HNH
3 - A m i n o - l , 2, 3-(a-Iminoethyl)- 4-triazole 5-methyltetronic acid (amitrole)
(62). The simultaneous application of riboflavin or other flavinoids re
verses the amitrole effect, but purines such as adenine or guanine are ineffective (63). Derivatives of tetronic acids, such as 3-(a-iminoethyl)- 5-methyltetronic acid also induce chlorosis in treated plants (64), but whether they act on chlorophyll synthesis, chlorophyll destruction, or plastid development is not known.
F. Inhibitors of Metabolic Synthesis
Recent work has shown 2,2-dichloropropionic acid (dalapon), a widely used herbicide, may act as an inhibitor of pantothenic acid synthesis
(65), since growth inhibition by dalapon in yeast or in higher plants is reversed by ^-alanine or pantothenate application. Pantothenate ana
logues, in which the hydroxyl group is replaced by chlorine, show the same effects as dalapon (65).
G. Herbicides a s Inhibitors of More than a Single Process
It is reasonable to assume that the application of any herbicide would affect more than a single vital process; the most sensitive process, that is, the one which succumbs first or at the lowest inhibitor concentration, is considered the primary site of action. The classification of herbicides as
254 W. A. ANDREAE inhibitors of growth, cell division, or photosynthesis is undoubtedly an oversimplification. Carbamates, for instance, were first classified as in- hibitors of cell division (52), and only recently was it realized that certain carbamates may also inhibit photosynthesis (60). Amitrole, which is con- sidered primarily an inhibitor of chloroplast synthesis (62), is an inhibitor of catalase activity in animal (66) and plant tissues (62) and an inhibitor of phosphorylase activity in plants (67). At present, opinions are still divided as to whether the primary site of action of maleic hydrazide is inhibition of auxin action (48), inhibition of gibberellin action (49), or inhibition of cell division (54); it has also been reported to inhibit dehy- drogenase activity (68) and to interfere with carbohydrate metabolism
(69).
The efficacy of a herbicide depends on more than the in vitro inhibition of a vital process; successful uptake by the roots or leaves, transport to the site of action, and metabolic stability within the tissue are also essen- tial. Consequently, there is not always a good agreement between tests on isolated systems and tests on intact plants.
In determining what constitutes selective toxicity a better understand- ing of the uptake and transport mechanisms may be essential. Selectivity between susceptible and resistant species is usually one of degree only and largely depends on the age of the plant. In some cases morphological factors such as depth of root system or position of leaves may explain selectivity. The difference between grasses and broadleaf plants in their sensitivity to 2,4-D was explained, at one time, as a difference in the physical properties of the leaf surface; the spray solution runs off grasses more rapidly than from broadleaf plants, so that the latter remain wetted and in contact with the spray for a much longer period. However, under similar conditions dalapon is toxic to grasses and relatively inocuous to broadleaf plants, and consequently this interpretation of selectivity has become less tenable.
III. METABOLISM OF HERBICIDES BY PLANTS
Herbicides, besides affecting plant metabolism, are themselves meta- bolized within the plant. Metabolism may lead to the activation of other- wise inactive precursors or to the detoxication of active compounds. Plant species differ in their ability to react with certain herbicides, and in some cases, the observed difference in response of susceptible and resistant plant species coincides with the difference in their metabolic activities.
The biochemical mechanisms involved are often quite similar to those
26. HERBICIDES 255 already established in animal tissues: the hydrolysis of esters, amides, or nitriles to their corresponding acids; stepwise shortening of fatty acid side chains by ^-oxidation; glycoside formation; and conjugation of organic acids with endogenous amino acids.
A. Conversion of Precursors to Active Herbicides
As long ago as 1935 naphthaleneacetonitrile ( N A N ) was recognized as possessing growth activity resembling that of NAA, and the activity of the nitrile was attributed to its conversion within the tissues to the active free acid (70). In recent years nitrile hydrolysis has been extensively studied. It was found that if oat tissues are incubated in indoleacetonitrile
(IAN) solutions, IAA accumulates in amounts which can be detected chemically (71), whereas with pea tissues this is not possible (72). Since the growth of oat and wheat tissues is stimulated by IAN, while the growth of pea tissues is not (78), these results strongly suggest that I A N is not an auxin in its own right but must be converted to IAA for activa- tion. Hydrolysis of I A N may not involve the intermediate formation of the amide, since peas respond to indoleacetamide while wheat does not show as good a growth response to the applied amide as to the nitrile (72).
An enzyme, indoleacetonitrilase, has been found in the leaves and stem of barley plants (74). Unlike I A N the 2,4-dichlorophenoxyacetonitrile is hydrolyzed equally well by all plant species (75); the nitriles of 2,4-D and of 2,6-dichlorobenzoic acid (76) are active herbicides. Besides the hydrolyzing mechanism, plants contain a metabolic pathway by which I A N is oxidized to indolecarboxylic acid with the loss of one carbon (75).
The ester herbicides of the 2,4-D type are readily hydrolyzed within the plant to the free acids (77). In practice the esters are widely used, since the free acids form insoluble calcium salts which tend to precipitate from the spray solution. Moreover, the reduced polarity of the esters also increases their lipophilic properties and thus facilitates entry into the tissues. Not all esters are hydrolyzed by higher plants. The esters of chlorophenoxy alcohols, such as sesone (sodium 2,4-dichlorophenoxyethyl sulfate), are not active when applied directly (78). However, in the soil sesone is hydrolyzed by soil organisms to the corresponding alcohol, which is subsequently oxidized to 2,4-D and taken up by the plants through their roots (79).
In an attempt to impart greater selectivity to existing herbicides a num- ber of amino acid derivatives were prepared (80, 81). In general, however, the L - or natural isomers of derivatives such as 2,4-dichlorophenoxy- acetylaspartic acid have activities similar to those of the parent com- pounds, whereas the D-isomers are not active. An interesting exception is
256 W. A. ANDREAE the aspartic acid derivative of IAA, which is not active on pea tissues A successful contrivance by which greater specificity is imparted to auxin herbicides involves ^-oxidation of their higher homologues. If a homologous series of dichlorophenoxy acids is examined for auxin or herbicidal activity, it becomes apparent that as the side chain is lengthened the activity alternates with each consecutive member (83).
Compounds with an even number of carbons in the side chain possess an activity similar to that of 2,4-D, whereas compounds with an odd number of carbons have the same negligible activity as the parent dichlorophenol.
While many plants are susceptible to all even-numbered homologues, there are exceptions, such as clover and celery, which are killed only by 2,4-D. These observations led to the development of 2,4-dichlorophenoxy- butyric acid [4-(2,4-DB)] and other phenoxybutyric acids as new herbi- cides with greater selectivity (84).
Some dipyridyl quaternary salts such as diquat (l,r-ethylene-2,2'- dipyridylium dibromide) have herbicidal properties. All members in this series show a correlation between the redox potential and phytoxicity;
apparently they are activated by a reduction process (85) in the green tissues. The simultaneous application of monuron, a specific inhibitor of reductions by chloroplasts, delays the onset of diquat symptoms in the light. It has been observed that dipyridyl compounds are reduced by illuminated chloroplasts (86). It seems probable, therefore, that photo- chemical processes are involved in the activation of diquat by reduction, but since diquat is also slightly active in the dark, metabolic reduction must play some part (87).
B. Metabolic Inactivation
The fact that the naturally occurring auxin IAA is not a herbicide, whereas the synthetic auxin analogue 2,4-D is a very effective one, may be due to their different metabolic fates in the plant. Applied IAA dis- (82).
2 B r —
l , l ' - E t h y l e n e - 2 , dipyridylium dibromide
(diquat)
26. HERBICIDES 257 appears very rapidly from the tissues, whereas 2,4-D persists long after its initial application. Two main processes have been found to be involved in IAA metabolism by tissues, namely, oxidative degradation and con- jugation with aspartic acid or ammonia (88) ,1
Tissues extracts have long been known to oxidize IAA, and the system involved has been identified as a peroxidase-flavoprotein complex (89).
Decarboxylation of the side chain or ring-opening have both been impli- cated as the initial step (90). In spite of many efforts, the primary products of IAA breakdown are still in doubt. Intact tissues incubated in an IAA solution degrade IAA after a lag period, which is inversely pro- portional to the concentration applied (91, 92). However, it is not clear as yet how much of this degradation can be attributed to oxidation within the tissues. Because of an inverse relationship between growth rate and IAA oxidase activity, some workers have considered that IAA oxidase plays an important part in regulating the level of endogenous IAA (92), while others consider that IAA oxidase plays little or no part in regulating the level of internal IAA (91, 93).
2,4-D is not attacked by IAA oxidase and is only slowly decarboxylated by tissue segments (94). It has been shown, however, that certain resistant plant tissues, such as the leaves of red currants, decarboxylate 2,4-D more rapidly than leaves of the susceptible black currant species (95), and resistance in the red currants has been ascribed to the ability of the plant to destroy 2,4-D. However, most 2,4-D-resistant plant species have a slow rate of 2,4-D decarboxylation, and another mechanism must account for the 2,4-D resistance of such plants.
Applied IAA accumulates in the tissues mainly as indoleacetylaspartic acid or indoleacetamide,1 the former primarily in legumes, the latter in cereals (36, 88). It has been reported that IAA is biologically bound to tissue proteins and that it forms a stable complex (96), but this could not be confirmed by the author (91). IAA may accumulate in pea stem tissues if the acid is applied continuously at concentrations high enough to inhibit growth (82); such supraoptimal concentration for growth satu- rate the mechanism of indoleacetylaspartic acid formation. Removal of such IAA-treated tissues results in the almost quantitative conversion of the accumulated IAA to indoleacetylaspartic acid (82). Since applied indoleacetylaspartic acid does not stimulate or inhibit growth of pea epicotyls, conjugation has been regarded as a detoxication process (82).
2,4-D, unlike IAA, undergoes little conjugation with aspartic acid (34) and, in short term experiments, persists almost entirely as unaltered 2,4-D. In long term experiments, however, 2,4-D does give rise to certain unidentified substances (97-99), one of which could possibly be 2,4-
258 W. A. ANDREAE dichlorophenoxyacetylaspartic acid. Since, however, applied 2,4-dichloro- phenoxyacetylaspartic acid is as effective physiologically as 2,4-D (80), this substance could not be regarded as detoxication product.
Conjugation has not yet been demonstrated in plant extracts. However, Zenk (87) has synthesized indoleacetylglycine in enzyme preparations from liver mitochondria in the presence of the CoA thio ester of IAA and glycine; under the same conditions the thio ester of 2,4-D does not undergo conjugation.
Conjugation has been noted with a variety of substances, although the physiological significance is not clear. Amitrole is conjugated with alanine in dwarf bean plants (100) to a substance which still possesses physiologi
cal activity (101). Tryptophan application to various plant species results in its conjugation with endogenous malonic acid to malonyltryptophan
(102). Glycosides of amitrole (108), of maleic hydrazide (48), and of cinnamic acid (104) have been found, but their physiological activity is unknown.
Other metabolic inactivations of herbicides have been observed, but the mechanism is as yet obscure. Corn, for instance, is remarkably resistant to simazin. This compound, although taken up by the plant, cannot be detected in the tissues. It is suggested that the resistance of corn is due to a detoxication mechanism (42); and indeed, extracts of corn decompose simazin, whereas extracts of wheat are ineffective. Heating of corn ex
tracts destroys their ability to inactivate simazin (105), indicating that an enzymic process is involved.
At present, the main efforts to achieve better plant control are directed toward synthesis and large-scale screening tests of untried chemicals.
Sometimes, unexpected compounds may possess surprisingly effective phy- totoxic properties. For instance, Zytron [0-2,4-dichlorophenyl O-methyl isopropylphosphoramidothioate] was originally developed as an insecti
cide with anticholinesterase activity; as such it was of no practical value.
However, for the selective control of crabgrass in lawns Zytron was found to be a valuable herbicide. While at present many effective compounds can still be discovered empirically, progress in the future will depend more and more on a better understanding of the biochemical functions of al
ready known herbicides in plants.
1 Note added in proof: The indoleacetamide reported to be found in in- doleacetic acid-treated plants (88) has since been shown by Μ. H. Zenk, Nature 191, 493 (1961), to be an artifact formed from indoleacetyl-jS-D-glucose.
REFERENCES
1. Terminology Committee Weed Society of America, Weeds 8, 487 (1960).
2. P . B . Marsh and D . R . Goddard, Am. J. Botany 26, 724 (1939).
26. HERBICIDES 259 3. A. J. Haagen-Smit, Ind. Eng. Chem. 44, 1342 (1952).
4. F. Kogl, A. J. Haagen-Smit, and H. Erxleben, Z. physiol. Chem. Hoppe- Seyler's 22$, 90 (1934).
5. P. W. Zimmerman and A. E. Hitchcock, Contribs. Boyce Thompson Inst.
12,321 (1942).
6. C. L. Hamner and Η. B. Tukey, Science 100, 154 (1944).
7. R. E . Slade, W. G. Templeman, and W. A. Sexton, Nature 155, 497 (1945).
8. P. S. Nutman, H. G. Thornton, and J. H. Quastel, Nature 155, 498 (1945).
9. G. E . Blackman, Nature 155, 500 (1945).
10. J. B. Koepfli, Κ. V. Thimann, and F . W. Went, J. Biol. Chem. 122, 763 (1938).
11. Κ. V. Thimann, Plant Physiol. 27, 392 (1952).
12. M. S. Smith, R. L. Wain, and F. Wightman, Ann. Appl. Biol. 39, 295 (1952).
13. G. J. M. van der Kerk, Μ. H. van Raalte, A. Kaars Sijpesteijn, and R. van der Veen, Nature 176, 308 (1955).
14. R. L. Wain, Advances in Pest Control Research 2, 263 (1958).
15. J. Bonner, Proc. Natl. Acad. Sci. U.S. 19, 717 (1933).
16. A. N . J. Heyn, Rec. trav. botan. neerl. 28, 113 (1931).
17. A. Frey-Wyssling. Protoplasma 25, 261 (1936).
18. T. A. Bennet-Clark, in "The Chemistry and Mode of Action of Plant Growth Substances." (R. L. Wain and F. Wightman, eds.), p. 284.
Academic Press, N e w York, 1956.
19. B. J. Cooil and J. Bonner, Planta 48, 696 (1957).
20. Ο. V. S. Heath and J. E. Clark, Nature 177, 1118 ( 1 9 5 6 ) ; 178, 600 ( 1 9 5 6 ) ; J. Exptl. Botany 11, 167 (1960).
21. L. Ordin, R. Cleland, and J. Bonner, Plant Physiol. 32, 216 (1957).
22. Κ. T. Glasziou and S. D. Inglis, Australian J. Biol. Sci. 11,127 (1958).
23. C. H. Fawcett, R. L. Wain, and F. Wightman, Nature 178, 972 (1956).
24. W. H. Bryan and Ε. H. Newcomb, Physiol. Plantarum 7, 290 (1954).
25. E . F. Jansen, R. Jang, and J. Bonner, Plant Physiol. 35, 567 (1960).
26. E. Marre and O. Arrigoni, Atti accad. nazl. Lincei, Rend., Classe sci. fis.
mat. e nat. 18, 539 (1955).
27. S. Tonzig and E . Marre, Rend. ist. lombardo sci. Pt. I 89, 243 (1955).
28. A. W. Galston and W. K. Purvis, Ann. Rev. Plant Physiol. 11, 239 (1960).
29. D. J. Osborne, Nature 188, 240 (1960).
30. A. E. Richmond and A. Lang, Science 125, 650 (1957).
31. K. Mothes, Naturwissenschaften 47, 337 (1960).
32. C. Person, D. J. Samborski, and F. R. Forsyth, Nature 180, 1294 (1957).
33. A. W. Galston and R. Kaur, Proc. Natl. Acad. Sci. U.S. 45, 1587 (1959).
34. W. A. Andreae and Ν . E . Good, Plant Physiol. 32, 566 (1957).
35. R. L. Wain and F. Wightman, Proc. Roy. Soc. B142, 525 (1955).
36. W. A. Andreae and Ν. E. Good, Plant Physiol. 30, 380 (1955).
37. Μ. H. Zenk, Z. Naturforsch. 15b, 436 (1960).
38. J. Bonner and R. F. Foster, in "The Chemistry and Mode of Action of Plant Growth Substances" (R. L. Wain and F. Wightman, eds.), p.
295. Academic Press, N e w York, 1956.
39. T. A. Bennet-Clark and N. P. Kefford, J. Exptl. Botany 5, 293 (1954).
260 W. A. ANDREAE 40. J. Bonner and R. S. Bandurski, Ann. Rev. Plant Physiol 3, 59 (1952).
41. D. J. Wort, Weeds 3,131 (1954).
42. Ε . K. Woodford, K. Holly, and C. C. McCready, Ann. Rev. Plant Physiol 9, 311 (1958).
43. A. C. Leopold and W. H. Klein, Physiol Plantarum 5, 91 (1952).
44. W. A. Andreae and S. A. Andreae, Can. J. Botany 31, 426 (1953).
45. L. J. Audus and R. Thresh, Ann. Bontany (London) 20, 439 (1956).
46. P. L. Goldacre, A. W. Galston, and R. L. Weintraub, Arch. Biochem.
Biophys. 43, 358 (1953).
47. D. M. Miller and R. W. White, Can. J. Chem. 34, 1510 (1956).
48. G. Η. N. Towers, A. Hutchinson, and W. A. Andreae, Nature 181, 1535 (1958).
49. P. W. Brian and H. G. Hemming, Ann. Appl. Biol. 45, 489 (1957).
50. Ν. E. Tolbert, J. Biol Chem. 235, 475 ( I 9 6 0 ) ; Plant Physiol. 35, 380 (1960).
51. R. L. Jones, T. P. Metcalfe, and W. A. Sexton, J. Sci. Food Agr. 5, 32 (1954).
52. W. G. Templeman and W. A. Sexton, Nature 156, 630 (1945).
53. C. D. Darlington and J. McLeish, Nature 167, 407 (1951).
54. V. A. Greulach and J. G. Haesloop, Am. J. Botany 41, 44 (1954).
55. A. H. Haber and J. D. White, Plant Physiol. 35, 495 ( I 9 6 0 ) .
56. J. S. C. Wessels and R. van der Veen, Biochim. et Biophys. Acta 19, 548 (1956).
57. Ν . I. Bishop, Biochim. et Biophys. Acta 27, 205 (1958).
58. A. T. Jagendorf, Brookhaven Symposia in Biol. 11, 236 (1958).
59. A. R. Krall, Ν. E. Good, and B. C. Mayne, Plant Physiol. 36, 44 (1961).
60. D. E. Moreland and K. L. Hill, J. Agr. Food Chem. 7, 832 (1959).
61. D. E . Moreland, W. A. Gentner, J. L. Hilton, and K. L. Hill, Plant Physiol. 34, 432 (1959).
61a. Ν. E. Good, Plant Physiol. 36, 788 (1961).
62. Η. T. Pyfrom, D. Appleman, and W. G. Heim, Plant Physiol 32, 674 (1957).
63. K. A. Sund, E. C. Putala, and Η. N. Little, J. Agr. Food Chem. 8, 210 (1960).
64. J. Alamercery, C. L. Hamner, and M. Latus, Nature 168, 85 (1951).
65. J. L. Hilton, L. L. Jansen, and A. W. Gentner, Plant Physiol 33, 43 ( 1 9 5 8 ) ; Weeds 7, 381 (1959).
66. W. G. Heim, D. Appleman, and Η. T. Pyfrom, Am. J. Physiol 186, 19 (1956).
67. A. G. Gentile and J. E. Frederick, Physiol. Plantarum 12, 862 (1959).
68. F. M. R. Isenberg, C. O. Jensen, and M. L. Odland, Science 120, 464 (1954).
69. A. W. Naylor, Arch. Biochem. Biophys. 33, 340 (1951).
70. P. W. Zimmerman and F. Wilcoxon, Contribs. Boyce Thompson Inst. 7, 209 (1935).
71. Β. B. Stowe and Κ. V. Thimann, Arch. Biochem. Biophys. 51, 499 (1954).
72. C. H. Fawcett, R. L. Wain, and F. Wightman, Proc. Roy. Soc. B152, 231 (1960).
26. HERBICIDES 261 73. Ε. R. Η. Jones, Η. Β. Henbest, G. F. Smith, and J. A. Bentley, Nature
169, 485 (1952).
74. Κ. V. Thimann and S. Mahadevan, Nature 181, 1466 (1958).
75. C. H. Fawcett, R. C. Seeley, H. F. Taylor, R. L. Wain, and F. Wightman, Nature 167, 1026 (1955).
76. H. Koopman and J. Daams, Nature 186, 89 (1960).
77. A. S. Crafts, Weeds 8, 19 (1960).
78. A. J. Vlitos, Contribs. Boyce Thompson Inst. 17, 127 (1953).
79. L. J. Audus, Nature 171, 523 (1953).
80. J. W. Wood and T. D. Fontaine, J. Org. Chem. 17, 891 (1952).
81. C. F. Krewson, E . J. Saggese, and T. F. Drake. Weeds 8, 107 ( 1 9 6 0 ) ; J. Agr. Food Chem. 8,104 (1960).
82. W. A. Andreae and M. W. H. van Ysselstein, Plant Physiol. 31, 235 (1956).
83. C. H. Fawcett, J. M. A. Ingram, and R. L. Wain, Proc. Roy Soc. B142, 60 (1954).
84. R. L. Wain, Ann. Appl. Biol. 42, 151 (1955).
85. R. F. Homer, G. C. Mees, and Τ. E . Tomlinson, J. Sci. Food Agr. 11, 309 (1960).
86. Ν. E. Good and R. Hill, Arch Biochem. Biophys. 57, 355 (1955).
87. G. C. Mees, Ann. Appl. Biol. 48, 601 (1960).
88. Ν . E. Good, W. A. Andreae, and M. W. H. van Ysselstein, Plant Physiol.
31, 231 (1956).
89. A. W. Galston, J. Bonner, and R. S. Baker, Arch. Biochem. Biophys. 42, 456 (1953).
90. P. M. Ray, Ann. Rev. Plant Physiol. 9, 81 (1958).
91. W. A. Andreae and M. W. H. van Ysselstein, Plant Physiol. 35, 225 (1960).
92. W. A. Galston and L. Y. Dalberg, Am. J. Botany 41, 373 (1954).
93. W. R. Briggs, T. A. Steeves, I. M. Sussex, and R. H. Wetmore, Plant Physiol. 30, 148 (1955).
94. R. L. Weintraub, J. N. Yeatman, J. A. Lockhard, J. H. Reinhart, and M.
Fields, Arch. Biochem. Biophys. 40, 277 (1952).
95. L. C. Luckwill and C. P. Lloyd-Jones, Ann. Appl. Biol. 48, 613 (1960).
96. S. M. Siegel and A. W. Galston, Proc. Natl. Acad. Sci. U.S. 39, 1111 (1953).
97. R. W. Holley, Arch. Biochem. 35, 171 (1952).
98. E . G. Jaworski and J. J. Butts, Arch. Biochem. 38, 207 (1952).
99. S. C. F a n g and J. S. Butts, Plant Physiol. 29, 56 (1954).
100. P. Massini, Biochim. et Biophys. Acta 36, 548 (1959).
101. D. Racusen, Arch. Biochem. Biophys. 74, 106 (1958).
102. Ν . E . Good and W. A. Andreae, Plant Physiol. 32, 561 (1957).
103. A. C. Gentile and J. E . Frederick, Physiol. Plantarum 12, 862 (1959).
104. P. N. Avadhani and G. Η. N. Towers, Can. J. Biochem. Physiol. 39, 1605 (1961).
105. W. Roth. Compt. rend. acad. sci. 245, 942 (1957).
