Plant Physiol. (1980) 66, 830-834 0032-0889/80/66/0830/05/$00.50/0

The Site of the Inhibition of the Shikimate Pathway by II. INTERFERENCE OF GLYPHOSATE WITH CHORISMATE FORMATION IN VIVO AND IN VITRO1

Received for publication January 22, 1980 and in revised form April 23, 1980

NIKOLAUS AMRHEIN2, BRIGITTE DEUS, PETER GEHRKE, AND HANS CHRISTIAN STEINRUCKEN Arbeitsgruppe Hormonphysiologie der Pflanzen, Lehrstuhlffur Pflanzenphysiologie, Ruhr- Universitat, D-4630 Bochum, Federal Republic Germany

ABSTRACT incorporation of labeled shikimate into all three aromatic amino acids. Chorismate, but not shikimate, partially reversed the inhi- In the presence of the nonselective herbicide glyphosate (N-Iphospho- bition of anthocyanin production, and labeled shikimate was nomethyliglycine), buckwheat (Fagopyrum esculentum Moench) hypocotyls taken up, but not metabolized, by the tissue in the presence of and cultured cells of Galium moligo L. accumulate an organic acid, which glyphosate. We inferred from these results that glyphosate blocks was identified as shikimate by mass-spectroscopy of its methyl ester. After a step in the biosynthetic sequence from shikimate to chorismate. growth in 0.5 millimolar glyphosate for 10 days, G. mollugo cells contained To substantiate this hypothesis, we decided to investigate whether shikimate in amounts of up to 10% of their dry weight. Synthesis of glyphosate causes the accumulation of a metabolite of the shiki- chorismate-derived in G. molugo was blocked by glyphos- mate pathway in treated plant tissues. Accumulation of phenyl- ate. Chorismate and o-succinylbenzoate (an anthraquinone precursor) alanine in the presence of the PAL inhibitor L-a-aminooxy-,f- alleviated the inhibition. The conclusion drawn from these experiments, phenylpropionic acid (1) had indicated a poor feedback control of that glyphosate inhibits a step in the biosynthetic sequence from shikimate synthesis in buckwheat hypocotyls (14). It seemed to chorismate, was substantiated by the finding that glyphosate is a likely that, in the case of glyphosate, the search for an accumulated powerful inhibitor of the conversion of shikimate to chorismate in cell-free metabolite would meet with success. Next, if our hypothesis is extracts from Aerobacter aerogenes 62-1. correct, glyphosate would be expected to inhibit the formation of substances that originate from chorismate, independent of the branches of the shikimate pathway that lead to the aromatic amino acids. For this purpose, we studied the effect of glyphosate on anthraquinone production in cultured cells of Galium mollugo L. (Rubiaceae). Anthraquinones of the Rubiacaeae-type are bio- synthetically derived from chorismate via o-succinylbenzoate (5, Interference with the biosynthesis of aromatic amino acids has 16, 20) and, under appropriate conditions, are produced in sub- been postulated as the mode of action of the nonselective, broad stantial amounts in cultured cells of Morinda citrifolia (24) as well spectrum herbicide glyphosate3 in higher plants as well as in as of G. mollugo and other members of this family (M. H. Zenk, microorganisms (15). or prephenate dehydra- personal communication). The third, and most direct, approach tase were suggested as the target for the herbicide (15); however, was to measure the effect of glyphosate on the formation of these enzymes in extracts from Escherichia coli were not inhibited chorismate in a cell-free system. We used an extract of the multiply by glyphosate (18), even though phenylalanine and acted blocked auxotroph, Aerobacter aerogenes, strain 62-1, which forms synergistically to reverse the inhibition of the bacterial growth (10, chorismate from shikimate (9). The results presented here leave 18). Only high concentrations of the herbicide were found to no doubt that glyphosate inhibits the formation of chorismate inhibit 3-deoxy-D-arabinoheptulosonate-7-P synthase and dehy- from shikimate both in vivo and in vitro. droquinate synthase, the first two enzymes in the shikimate path- way (18), but the magnitude of the effect was obviously too low to explain growth inhibition. Thus, an enzymic reaction which is MATERIALS AND METHODS subject to inhibition by low concentrations of glyphosate remained General Methods. Buckwheat (Fagopyrum esculentum Moench) to be demonstrated. Failure of glyphosate to lower free phenylal- seedlings were grown in darkness, and excised hypocotyls were anine concentrations in cultured carrot cells was considered by incubated in the light as described previously (13). Haderlie et al. (1 1) to militate against inhibition of aromatic amino The cell culture of B. mollugo L. was placed at our disposal by acid synthesis. On the other hand, we have shown in the preceding Professor M. H. Zenk. The cells were grown in the modified B5 publication (13) that glyphosate blocks the synthesis of phenylal- medium of Gamborg et al. (8, 21) containing 0.2% N-Z-Amine anine-derived phenylpropanoid compounds in buckwheat hypo- Type A (Sheffield Chemical, Norwich, N. Y.) (24). Since gly- cotyls, reduces the level of free phenylalanine, and inhibits the phosate is stable to autoclaving (1 1), all media were autoclaved at 121 C for 20 min. Cells were collected by filtration in a Buchner- 1 This work was supported by grants from the Deutsche Forschungs- funnel, and 2 cm3 packed cell volume (about 30 mg dry weight) gemeinschaft (to N. A.) and by the Bundesministerium fur Forschung und were used to inoculate 25 ml medium in 100-ml Erlenmeyer flasks. Technologie (to Professor M. H. Zenk). Cells were cultured for 10 days under conditions as described in 2To whom reprint requests should be addressed. reference 24. A. aerogenes, strain 62-1 (Klebsiella pneumoniae, 'Abbreviations: glyphosate, N-[phosphonomethyllglycine; PAL, phen- ATCC 25306) was maintained on 2.3% Bacto-nutrient agar and ylalanine ammonia-lyase (EC 4.3.1.5). was grown as in reference 17 for the preparation of the cell-free 830 Plant Physiol. Vol. 66, 1980 SITE OF GLYPHOSATE INHIBITION. II 831 extract. Radioactivity on chromatograms was localized with a production of ['4C]anthranilate from [14CJshikimate was measured Berthold-Frieseke Scanner LB 2760, and radioactivity in liquid by the same procedure. Distribution of radioactivity on the plates samples was determined in a Berthold-Frieseke BF 5000 scintil- after development was measured with a thin-layer scanner, and lation counter. quantitative conversion of shikimate into products was determined Determination of Endogenous Shikimate Content. The extrac- from peak areas. Anthranilate synthase activity was determined tion and purification of shikimate followed basically the procedure as described by Egan and Gibson (6), except that anthranilate was of Yoshida and Hasegawa (23). For routine determinations, 20 measured spectrophotometrically (see above). The incubation buckwheat hypocotyls (1.2 g fresh weight) or 1 g packed cells of mixture for the assa of shikimate 3-kinase contained 1 mm G. mollugo were extracted with three changes of 20 ml methanol shikimate, 0.02 ACi [1 Cjshikimate, 2.5 mM MgCl2, 2 mm ATP, 50 (buckwheat) or 20 ml 80%1o ethanol (Galium) under reflux. mM Tris-HCl buffer (pH 8.2), and 3 Al dialyzed extract in a total Labeled shikimate (71.630 dpm = 38.5 pmol) was added to the volume of 50 ,ul. After 10 min at 37 C, the reaction was stopped by solvent during the first extraction to allow estimation of recovery. the addition of 10 ,ul 1 N HC1. The entire mixture then was spotted After filtration, the solvent was evaporated in an air stream and on Whatman No. 3 paper, which was developed in 1-butanol/ the residue was taken up in 30 ml H20. The extracts were acetic acid/H20 (76:5:19, v/v). The region containing the radio- transferred to glass columns, each containing 3.5 g Amberlite AG active product (RF = 0-0.1) was eluted with five changes of 0.2 ml (type IRA-410, 20 to 50 mesh; Serva, Heidelberg) which had been H20, and radioactivity was determined in an aliquot of the eluate. converted from the Cl- form to the Co32- form with 50 ml 1 N After treatment of the product with alkaline phosphatase (bovine Na2CO3 and then washed with H20 until the effluent was neutral. (calf) intestine, Boehringer, Mannheim) and rechromatography, After washing with 25 ml H20, organic acids were eluted with 50 only labeled shikimate (RF = 0.45) was detectable. ml 1 N (NH4)2C03. Flow rates were not allowed to exceed 1 ml/ Chemicals. Sources of the chemicals are given in the preceding min. The eluates were reduced to dryness at 90 to 95 C and the paper (13). o-Succinylbenzoic acid, synthesized by the procedure residue was taken up in 0.4 ml H20. An aliquot (0.3 ml) was of Dansette (4), was provided by Professor M. H. Zenk. spotted on Whatman No. 3 chromatography paper, which had previously been exhaustively washed with H20 and then dried. RESULTS The chromatograms were developed (ascending) for 6 to 7 h (for about 20 cm) with isobutyl alcohol/ethyl/acetate/acetic acid/ Effect of Glyphosate on Shikimate Content of Buckwheat Hy- formic acid/H20 (35:35:8:2:20, v/v). Radioactive spots were eluted pocotyls. We reported (13) that 1 mm glyphosate inhibits phenyl- with hot H20 (1 ml for each of seven changes) and the eluates propanoid production by 90%o in illuminated buckwheat hypocot- were reduced to dryness at 90 to 95 C. The residues were taken up yls. Furthermore, labeled shikimate was shown to be taken up but in 2.5 ml H20. Radioactivity of an aliquot (0.3 ml) was measured, not metabolized. This prompted us to examine the effect of and the shikimate content in the samples was determined by the glyphosate on the shikimate content of the hypocotyls. Extracts method of Gaitonde and Gordon (7) and corrected for the loss from control and glyphosate-treated tissues were processed for the during the purification. The recovery was 78.7 ± 6%. detection and assay of shikimate as described. Paper chromato- Identification of Shikimate from Buckwheat Hypocotyls. About grams were sprayed with the periodate-benzidine reagent of Ci- 330 hypocotyls (20 g fresh weight) were incubated in 100 ml I mM fonelli and Smith (2) to observe compounds with glycolic hydroxyl glyphosate in 10 mm K-phosphate (pH 5.5) for 24 h in the light groups (white spots on a blue background). Although all extracts and then processed as described above with adjustments for the gave positive reactions in the lower RF regions of chromatograms, larger amount of plant material. The material co-chromatograph- only extracts from glyphosate-treated tissues gave positive reac- ing with radioactive tracer shikimate was subjected to two further tions at RF = 0.50, the position of chromatographed labeled tracer chromatographic purification steps. The solvents were: 1-butanol/ shikimate. After methylation, the material co-chromatographed acetic acid/H20 (100:6:25, v/v) and ethylmethyl ketone/acetone/ with the labeled product (RF = 0.68), and the methylated product formic acid/H20 (80:4:2:12, v/v). Methanol was used to elute the and authentic methyl ester had identical UV spectra radioactive area of the last chromatogram, and the eluate was in absolute ethanol (Amax = 271 nm), as well as identical mass- subjected to methylation with diazomethane (19). After evapora- spectra [mie (relative intensity) 188(M+, 2); 156(15); 137(18); tion of the solvent, the residue was dissolved in H20 and rechro- 129(34); 97(100); 69(44)]. matographed in the latter solvent. The radioactive area was eluted Quinate, which interferes with the assay of shikimate (23), had with absolute ethanol, and UV and mass spectra were recorded in an RF of 0.31 in the solvent used and, thus, was effectively a Beckman 24 spectrophotometer and a Varian MAT 111 mass separated from shikimate. No positive reaction to the periodate- spectrometer at an electron energy of 80 ev. benzidine reagent was found around RF = 0.3, indicating the Determination of Anthraquinone Content. G. mollugo cells were absence of quinate. Once the identity of the organic acid accu- extracted as described above for shikimate. The pooled extracts mulating in the presence of glyphosate had been established, were made up to 25 ml with 80%o ethanol and the A at 434 nm was shikimate concentrations in the eluates from the paper chromat- measured directly or after appropriate dilution. A molar extinction ograms were routinely determined with the assay of Gaitonde and coefficient of e = 5.500 was used to calculate the total anthraqui- Gordon (7). Table I shows that a 24-h incubation of excised none content (24). hypocotyls either in the light or in darkness had little effect on the Conversion of Shikimate to Anthranilate by Extracts from A. content of endogenous shikimate. Glyphosate (1 mM) causes about aerogenes 62-1. Methods for the preparation of cell-free extracts from A. aerogenes 62-1 and conditions for the conversion of Table I. Effect of Glyphosate on Shikimate Content of Excised shikimate to anthranilate were those of Morgan et al. (17). An- Hypocotyls of 6-Day-Old Buckwheat Seedlings thranilate was measured spectrophotometrically after extraction Treatment of Hypocotyls Shikimate Content into ethylacetate (9). When conversion of shikimate into choris- wt mate was to be measured, glutamine was omitted and 0.01 ,uCi [Lmol/lgfresh 0 control 0.036 ["4C]shikimate was included in the assay mixture. Aliquots of the h, mixture were spotted on cellulose thin-layer plates (Merck, Darm- 24 h, darkness 0.031 stadt), which were developed in either 1-butanol/acetic acid/H20 24 h, light 0.036 (76:5:19, v/v) or 2-methyl/2-butanol/formic acid/H20 (30:20:10, 24 h, I mM glyphosate, darkness 0.649 24 1 mm v/v), with authentic chorismate as reference. For control purposes, h, glyphosate, light 1.865 832 AMRHEIN ET AL. Plant Physiol. Vol. 66, 1980 a 20-fold increase in the shikimate concentration in darkness and Table II. Reversal Treatmentfor Inhibition by Glyphosate of a >50-fold increase in the light, which raises the concentration of Anthraquinone Formation in G. mollugo Cells the acid in the tissue to nearly 2 mm. "Reversal agents" were added to the medium at a fimal concentration Effect of Glyphosate on Shikimate and Anthraqulnone Content of 1 mm, and cells were cultured for 10 days before they were extracted. of Cultured G. moUgo Cells. Glyphosate, even at 1 mm, did not Control produced 47 pmol anthraquinones/g dry weight. inhibit the growth of the cells in the modified B5 medium of Addition to Medium Anthraquinone Content Gamborg et al. (8, 21) during a 10-day culture period (Fig. 1), provided that the medium was fortified with 0.2% N-Z-Amine, a % control pancreatic hydrolysate of casein. In the absence of the exogenous 0.5 mM glyphosate 18 amino acids, 0.3 mm glyphosate inhibited the growth of the cells 0.5 + chorismate 53 about 55%. Attempts were not made to reverse this growth inhi- 0.5 + o-succinylbenzoate 80 bition with individual amino acids because similar experiments 0.5 + Phe 19 have been carried out previously with cultured carrot and soybean 0.5 + Tyr 14 cells (10, 11). Concentrations of glyphosate higher than 0.1 mM 0.5 + Phe + Tyr 11 inhibited anthraquinone production (about 80o inhibition at 0.5 mM) asid produced an enormous accumulation of shikimate in the Chorismate 78 cells (Fig. 1). The shikimate content of cells grown in the presence o-Succinylbenzoate 157 of 0.5 to 1 mm glyphosate reached up to 10%o of their total dry Phe 99 weight, representing a concentration increase (assuming that 1 g Tyr 97 packed cells has 1 ml total volume) of shikimate from 14 to 20 Phe + Tyr 73 ,UM in untreated cells to 20 to 23 mm in 1 mm glyphosate-treated cells. Glyphosate inhibition of anthraquinone production was partially alleviated by 1 mm chorismate and 1 mm o-succinylben- zoate, but not by 1 mm phenylalanine and tyrosine, either alone or in combination (Table II). Chorismate alone and the combi- nation of phenylalanine and tyrosine inhibited anthraquinone formation slightly, whereas o-succinyl-benzoate stimulated pig- ment formation. This latter effect has been reported for cultured cells of M. citr!folia (24). The "reversal agents" had no significant effect on cell growth (data not shown). Effect of Glypbosate on the Conversion of Shikimate to Cho- rismate in Cell-free Extracts of A. aerogenes 62-1. Accumulation of shikimate in plant tissues treated with glyphosate and inhibition by glyphosate of the production of anthraquinoid pigments, which derived left no doubt that --I are biosynthetically from chorismate, 4. 6 5 _ 3 .r- U '5 4 3 glyphosate inhibits a step in the biosynthetic pathway from shi- -log Molarity - log Molarity kimate to chorismate. For final proof, it remained to be shown that glyphosate also inhibits the cell-free conversion of shikimate 1 0 20 3A 4a to chorismate. As an experimental system, we chose a cell-free extract of the multiply blocked auxotroph A. aerogenes, strain 62- CO2H C02H 1, which readily converts shikimate to chorismate and, in the CH2 CH2 H C0H2 N N N N 1 2 H CH2 H2C CH2 H CH2 H CH2 I ~~~I 1800 P03H2 H203P P03H2 P03H2 CO2H 1 0o Glyphosate Glyphosine Aminomethane- Iminodiacetate I Phosphonate 0, ._ FIG. 2. Effect of glyphosate and related compounds on the conversion x 8 600-.C V03 ._1 of shikimate to anthranilate in a cell-free extract of A. aerogenes 62-1 (a) 3: and on the formation of anthocyanin in illuminated excised buckwheat 0, E -- 6 >0 hypocotyls (b). For details of the buckwheat experiments see reference 13. n Controls produced in 14 nmol anthranilate/min in a and 0.37 pmol 10in 400 cm .2 0C OE o anthocyanin/24 h in b. 3' 4 0I'U E presence of glutamine, to anthranilate, which can easily be mea- cC sured or spectrophotometrically (6). Preliminary Ea, 200 w fluorimetrically ._ 2 experiments showed that 1 mm glyphosate did not interfere with cr E the conversion of chorismate to anthranilate. 4 75._ When [14CJshikimate was employed as substrate in the absence (A O0 of glyphosate, the entire radioactivity was recovered in anthrani- late when glutamine was present and in chorismate when gluta- Glyphosate (-log Molarity) mine was absent. For routine measurements, production of an- FIG. 1. Effect of glyphosate on dry weight yield (x---x), anthraqui- thranilate from shikimate, therefore, was used to monitor choris- none production (O-O), and shikimate content (@-) of G. mollugo mate formation. Glyphosate is a powerful inhibitor of the forma- cells. Cells were harvested 10 days after inoculation with 2 cm3 packed tion of anthranilate (chorismate) from shikimate; 50% inhibition cells into 25 ml medium containing glyphosate. Anthraquinone content of was achieved with 5 to 7 pm glyphosate (Fig. 2a). The structurally the cells was corrected for the amount present in the inoculum. related compound glyphosine (N,N-bis[phosphonomethylJgly- Plant Physiol. Vol. 66, 1980 SITE OF GLYPHOSATE INHIBITION. II 833 cine), was much less inhibitory than glyphosate, and aminometha- demonstration that glyphosate is a powerful inhibitor of the nephosphonate and iminodiacetate showed no inhibitory activity conversion of shikimate to chorismate (anthranilate) in cell-free (Fig. 2a), which closely parallels the potency of these compounds extracts of A. aerogenes 62-1 (Fig. 2). This is the first time that an to inhibit anthocyanin synthesis in buckwheat hypocotyls (Fig. enzymic reaction in a cell-free extract has been shown to be 2b; ref. 13). as target of glyphosate action was inhibited by glyphosate at sufficiently low concentration to allow excluded since there was no effect of glyphosate on the formation an explanation for the effects of glyphosate on intact cells or of a metabolite by the dialyzed cell-free extract in the presence of organisms. Of the four enzymes involved in the overall conversion ATP alone, which released shikimate upon treatment with alkaline of shikimate into anthranilate, we were able to definitely exclude phosphatase and was tentatively identified as shikimate-3-P. Stud- shikimate kinase and anthranilate synthase as glyphosate targets. ies are in progress to differentiate between the effect of glyphosate Accumulation of shikimate in vivo, rather than shikimate-3-P or on the remaining two enzymes, 5-enolpyruvylshikimate-3-P syn- 5-enolpyruvylshikimate-3-P, may reflect an equilibrium of the thase and chorismate synthase. overall reaction that is unfavorable for the accumulation of the latter intermediates. In the cell-free Aerobacter system, labeled DISCUSSION shikimate in the presence of glyphosate was quantitatively con- verted into metabolite(s) now under investigation (N. Amrhein The uniformity of the response of widely different organisms to and H. C. Steinrucken, unpublished). This will allow us to deter- glyphosate calls for a common determinant in its mode of action mine the precise site of glyphosate action, but we can now state (18). The hypothesis of Jaworski (15) that glyphosate inhibits conclusively that, as a result of glyphosate application to plants, aromatic biosynthesis at the site of chorismate mutase production of chorismate and its metabolites, such as the aromatic or prephenate dehydratase fulfilled this basic requirement, but it amino acids, is inhibited. The fact that is not essential could not be verified (18). From the feeding experiments with for reversal of growth inhibition by glyphosate (10, 11, 15) would labeled shikimate in the preceding paper (13), we concluded that seem to be incompatible with this mode of action. It has been glyphosate inhibits the conversion of shikimate to chorismate. suggested (22) that beans, for example, have an alternate pathway This conclusion was fully substantiated by the results of the for tryptophan biosynthesis, which does not involve shikimate. present investigation. The striking accumulation of shikimate in However, as pointed out (13), failure to observe the incorporation glyphosate-treated buckwheat hypocotyls (Table I) and G. mollugo of labeled shikimate into tryptophan in vivo is likely to be due to cells (Fig. 1) provided the first proof. In addition, this finding is regulatory mechanisms rather than to an alternate pathway. It is of significance for studies on the regulation of the shikimate more probable that, under conditions of restricted chorismate pathway. Apparently, there is no feedback mechanism operative production (as in the presence of glyphosate), the low amount of in higher plants which prevents the overproduction of shikimate available chorismate is preferentially utilized for the production in the presence of glyphosate. We currently investigate the effect of anthranilate (and hence tryptophan) rather than for the pro- of various intermediates and end products of the shikimate path- duction of prephenate (and, hence, tyrosine and phenylalanine). way on the glyphosate-induced accumulation of shikimate to find This assumption seems justified when one compares the Km values evidence for or against an early regulatory step in the pathway. 3- of anthranilate synthase and chorismate mutase for their common Deoxy-D-arabino-heptulosonate-7-P synthase is an important fac- substrate chorismate: anthranilate synthase from A. aerogenes has tor in the control of the pathway in microorganisms, but appar- a Km of 10 ,UM for chorismate (6), whereas chorismate mutase from ently not in higher plants (12). The in vivo studies described here this organism has an apparent Km of 0.56 mm (3). Feedback may help to gain insight into this neglected question. inhibition of chorismate mutase by added phenylalanine and A second point of interest is the effect of light on the shikimate tyrosine (12) further increases the Km of the enzyme for choris- pathway. We have recently obtained evidence that light stimulates mate, and more substrate is, therefore, available for anthranilate the production of phenylalanine in buckwheat hypocotyls inas- synthase. Apart from these considerations, our results stringently much as, in the presence of the specific PAL-inhibitor L-a-ami- indicate that glyphosate inhibits chorismate formation both in vivo nooxy-,8-phenylpropionic acid, more phenylalanine accumulates and in vitro. This herbicide may thus be a very useful agent in the in the light than in darkness (14). The data of Table I indicate elucidation of unresolved questions of the biosynthesis and phys- that light stimulates the production of shikimate, i.e. it effects iology of shikimate- and chorismate-derived metabolites. either an early step in the shikimate pathway or possibly the transfer of shikimate pathway intermediates from one cell com- Acknowledgments-We wish to thank Dr. E. G. Jaworski, Monsanto Co., for the partment to another. samples of glyphosate and glyphosine, and Dr. J. Stockigt for recording the mass Inhibition of anthraquinone production by glyphosate in G. spectra. We are grateful to Professor M. H. Zenk for providing the G. mollugo cell mollugo cells without concomitant inhibition of growth (Fig. 2) is culture, o-succinylbenzoate, and unpublished results. further evidence that glyphosate inhibits chorismate formation. For the first time, it has been shown here that an intermediate LITERATURE CITED metabolite of the shikimate pathway (chorismate), but not phen- 1. AMRHEIN N, KH GODEKE 1977 a-Aminooxy-18-phenylpropionic acid-a potent inhibitor of L-phenylalanine ammonia-lyase in vitro and in vivo. Plant Sci Lett ylalanine and/or tyrosine (Table II), reversed an inhibitory effect 8: 313-317 of glyphosate. The medium was fortified with 0.2% N-Z-Amine 2. CIFONELLI JA, F SMITH 1954 Detection of glycosides and other carbohydrate and, as calculated from the manufacturer's data sheet, contained compounds on paper chromatograms. Anal Chem 26: 1132-1134 phenylalanine and tyrosine at 0.62 and 0.34 mm concentrations, 3. COTTON RGH, F GIBSON 1970 Tyrosine and phenylalanine biosynthesis; the T and P proteins (Aerobacter aerogenes); chorismate mutase (Pisum sativum). respectively. In the absence of N-Z-Amine, glyphosate inhibited Methods Enzymol 17: 564-574 growth. We did not show that phenylalanine and tyrosine (and 4. DANsETTE P 1972 Etude sur la biosynthese des naphthoquinones vegetales et possibly tryptophan) were the active agents in the case in hydrol- bacteriennes. PhD thesis. University de Paris-Sud, Centre d'Orsay, France ysate but, on the basis of the reversal of growth inhibition in 5. DANSETTE P, R AZERAD 1970 A new intermediate in naphthoquinone and experiments by Haderlie et (1 1) and Gresshoff (10), this can menaquinone biosynthesis. Biochem Biophys Res Commun 40: 1090-1095 al. be 6. EGAN AF, F GIBSON 1970 Anthranilate synthase and anthranilate-5'-phospho- assumed. Additional phenylalanine (final concentration; 1.62 mM) ribosyl- I-pyrophosphate phosphoribosyl transferase (PR transferase) aggregate and tyrosine (final concentration, 1.34 mM) did not alleviate from Aerobacter aerogenes. Methods Enzymol 17: 380-386 glyphosate inhibition of anthraquinone formation either (Table 7. GAITONDE MK, MW GORDON 1958 A microchemical method for the detection II). Therefore, the cells not suffer from depletion of these and determination of shikimic acid. J Biol Chem 230: 1043-1050 did two 8. GAMBORG OL, RA MILLER, K OjIMA 1968 Nutrient requirement of suspension amino acids, but rather from depletion of chorismate. cultures of soybean root cells. Exp Cell Res 50: 151-158 More direct proof of our hypothesis was provided by the 9. GIBSON MI, F GIBSON 1964 Preliminary studies on the isolation and metabolism 834 AMRHEIN ET AL. Plant Physiol. Vol. 66, 1980 of an intermediate in aromatic biosynthesis: chorismic acid. Biochem J 90: certain aromatic compounds by cell-free extracts of Aerobacter aerogenes and 248-256 Escherichia coli. Biochem J 89: 229-239 10. GRESSHOFF PM 1979 Growth inhibition by glyphosate and reversal of its action 18. ROISCH V, F LINGENS 1974 Effect of the herbicide N-phosphonomethylglycine by phenylalanine and tyrosine. Aust J Plant Physiol 6: 177-185 on the biosynthesis of aromatic amino acids. Angew Chem 13: 400 1 1. HADERLIE LC, JM WIDHOLM, FW SLIFE 1977 Effect of glyphosate on carrot and 19. SCHARF KH 1970 Darstellung positionsmarkierter Shikimisiuren und deren tobacco cells. Plant Physiol 60: 40-43 Stoffwechsel in Bakterien, Pilzen, und hoheren Pflanzen. PhD thesis, Ruhr- 12. HASLAM E 1974 The Shikimate Pathway. Butterworth, London University, Bochum, Federal Republic of Germany 13. HOLLANDER H, N AMRHEIN 1980 The site of the inhibition of the shikimate 20. UEDA S, F INOUE, T HAYASHI, H INOUYE 1975 Zur Biosynthese des Catalponols pathway by glyphosate. I. Inhibition by glyphosate of phenylpropanoid syn- und artverwandter Stoffe. Tetrahedron Lett 2399-2402 thesis in buckwheat (Fagopyrum esculentum Moench). Plant Physiol 66: 823- 21. ULBRICH B, MH ZENK 1979 Partial purification and properties of hydroxycin- 829 namoyl-CoA: quinate hydroxycinnamoyl transferase from higher plants. Phy- 14. HOLLANDER H, HH KILTZ, N AMRHEIN 1979 Interference of L-a-aminOOxy-/?- tochemistry 18: 929-933 phenylpropionic acid with phenylalanine metabolism in buckwheat. Z Natur- 22. WEINSTEIN LH, CA PORTER, HJ LAURENCOTT 1962 Role of shikimic acid forschung 34C: 1162-1173 pathway in the formation of tryptophan in higher plants: evidence for an 15. JAWORSKI EG 1972 Mode of action of N-phosphonomethylglycine: inhibition of alternative pathway in the bean. Nature 194: 205-206 biosynthesis. J Agric Food Chem 20: 1195-1198 23. YOSHIDA S, M HASEGAWA 1957 A microcolorimetric method for the determina- 16. LEISTNER E 1973 Biosynthesis of morindone and alizarin in intact plants and cell tion of shikimic acid. Arch Biochem Biophys 70: 377-381 suspension cultures of Morinda citrffolia. Phytochemistry 12: 1669-1674 24. ZENK MH, H EL-SHAGI, U SCHULTE 1975 Anthraquinone production by cell 17. MORGAN PN, MI GIBSON, F GIBSON 1963 The conversion of shikimic acid into suspension cultures of Morinda citrifolia. Planta Medica Suppl: 79-101