J. Org. Chem. 1989,54, 1543-1548 Investieacidn Cientifica v Tbcnica) for financial suDDort. -_ C.S. t h i n k s CONICET ;or a fellowship.
Note added in proof After the submission of this manuscript for publication, a similar approach to cornpound 6 appeared in the literature (Li, T.-T.; Wu, y.-L. Tetrahedron Lett. 1988, 29, 4039). Registry No. (f)-l,112420-42-5; (f)-6,118798-10-0; (&)-7a, 118798-07-5;(*)-7b,118798-08-6; (f)-7c,118798-09-7; (*)-Sa,
1543
118798-06-4;(*)-8b,60078-94-6;(&)-8c, 60078-92-4; 9,72324-39-1; 10, 118798-11-1;11, 118798-12-2;(&)-12, 118798-13-3;(f)-l9, 118798-14-4;(&)-14, 118798-15-5;(f)-15, 118798-16-6;(f)-16, 118798-17-7; @)-17a, 118798-18-8; (*)-17b,118916-42-0; (f)-17c, 114375-37-0; (*)-I76 118798-19-9; (*)-IS,118798-21-3;(&)-19a, 118798-22-4; (*)-19b,118798-23-5; (&)-19C,118798-20-2; (&)-acyclocitral, 59462-59-8;a,2,6,6-tetramethyl-2-cyclohexene-lmethanol, 118798-24-6;(~)-l-(2,6,6-trimethyl-2-cyclohexen-lyl)ethanone, 72717-26-1;1-(1,3,3-trimethyl-7-oxabicyclo[4.1.0]hept-2-yl)ethanone, 118798-25-7.
Purification and Inhibition of Spinach a,@-DihydroxyacidDehydrataset Michael C. Pirrung,* Hyun-Joon Ha, and Christopher P. Holmes Department of Chemistry, Stanford University, Stanford, California 94305 Received J u l y 15, 1988
The a,@-dihydroxyaciddehydratase (E.C. 4.2.1.9)responsible for the production of a-oxoisovaleric acid in the valine biosynthetic pathway has been purified from spinach leaves. Its properties are similar to those given in a previous report using a less pure preparation. Its monomer mass is estimated to be 55 kDa. Evidence for an enol intermediate in the reaction mechanism has been obtained by a deuterium labeling study. Several inhibitors have been screened against the enzyme. Four of particular effectiveness are 4-fluoro-2,3-dihydroxyisovaleric acid, 1-hydroxy-1-isobutanesulfonic acid, N,N-dimethylglycine N-oxide, and 2-fluoro-3,3-dimethylacrylic acid. As an enol analogue, the latter compound gives further evidence for an enol intermediate. The biosynthetic pathway for the branched-chain amino acids valine, leucine, and isoleucine in higher plants has recently been identified a~ a site of herbicide action. Three classes of compounds, the s~lfonylureas,'-~imidazolinone^,^,^ and triazolopyrimidines,' have been show to inhibit the first and rate-limiting enzyme in the pathway, E.C. 4.1.3.18, acetolactate synthase (ALS) or acetohydroxyacid synthase (AHAS). These compounds have found commercial success as soybean and small grain herbicides, but still lack selectivity for grasses, and some resist soil metabolism. Consequently, two other enzymes in this pathway draw attention as potential targets for developing new meristematic inhibitors. The acetohydroxyacid reductioisomerase6 has thus far not been subjected to serious scrutiny. This work has focused on the subsequent enzyme in the pathway, a,@-dihydroxyacid dehydratase (DHAD). This enzyme catalyzes the transformation of 2,3-dihydroxyisovaleric acid (1) into 2-oxoisovaleric acid (3) with loss of water (eq 1).
1
2
0
3
Much of the detailed information about the valine biosynthetic pathway has come from studies on bacterial enzymes. DHAD has been partially purified from Escherichia Neurospora crassa,l0 and Salmonella typhimurium.l' The stereochemical course of the E. coli12 and Salmonella13 enzymes has been well-studied. I t has been shown that a 2R configuration is uniformly required t Supported by the National Science Foundation (CHE 84-51324) and American Cyanamid Co.
for both the natural substrates and analogues. Evidence from tritium labeling studies implicates an enol intermediate in the reaction catalyzed by the Salmonella deh y d r a t a ~ e . ' ~DHAD has also been identified as the site of hyperbaric oxygen poisoning in E. coli.l5J6 I t has been postulated that this is due to excessive superoxide levels, and superoxide generated from Paraquat has been shown to decrease DHAD activity in vivo." For the purpose of herbicide design, the plant enzyme is required. DHAD activity has been studied in 29 plant species16 and has been found to strongly correlate with seedling growth. A previous publication reported the purification (120-fold, 1% activity yield) of the spinach enzyme.6 I t was shown to require Mg2+ for activity, as further evidenced by inhibitors such as fluoride and (1) Ray, T. B. Plant Physiol. 1984, 75, 827-831. (2) Chaleff, R. S.; Mauvais, C. J. Science 1984,224, 1443-1445. (3) Schloss, J. V.; Van Dyk, D. E.; Vasta, J. F.; Kutny, R. M. Biochemistry 1985, 24, 4952-4959. (4) LaRossa, R. A.; Schloss, J. V. J.Biol. Chem. 1984,259,8753-8757. (5) Shaner, D. L.; Anderson, P. C.; Stidham, M. A. Plant Physiol. 1984, 76, 545-546. (6) Muhitich, M. J.; Shaner, D. L.; Stidham, M. A. Plant Physiol. 1987, 83, 451-456. (7) Pearson, N. R.; Kleschick, W. A. US.Pat. 4605433, 1986. (8) Kanamori, M.; Wixom, R. L. J. Biol. Chem. 1963,238, 998-1005. (9) Meyers, J. W. J. Biol. Chem. 1961,236, 1414-1418. (10) Kiritani, K.; Wagner, R. P. Methods Enzymol. 1970,17,755-764. (11)Arfin, S. M.; Umbarger, H. E. J.Biol.Chem. 1969,244,1118-1127. (12) Hill, R. K.; Yan,S.; Arfin, S. M. J. Am. Chem. SOC. 1973, 95, 7857-7859. (13) Armstrong, F. B.; Lipscomb, E. L.; Crout, D. H. G.; Morgan, P. J. J. Chem. SOC., Perkin Trans. I 1985, 691-696. Armstrong, F. B.; Muller, U. S.; Reary, J. B.; Whitehouse, D.; Crout, D. H. G. Biochem. Biophys. Acta 1977, 498, 282-293. (14) Arfin, S. M. J. Biol. Chem. 1969, 244, 225&2251. (15) Brown, 0. R.; Yein, F. Biochem. Biophys. Res. Commun. 1978, 85, 1219-1224. (16) Brown, 0. R.; Seither, R. L. Fundam. Appl. Toxicol. 1983, 3, 209-214. (17) Kuo, C. F.; Mashino, T.; Fridovich, I. J. Biol. Chem. 1987, 262, 4724-4727. (18) Wixom, R. L.; Hudson, R. J. Plant Physiol. 1961, 36, 598-604.
0022-3263/89/1954-1543$01.50/00 1989 American Chemical Society
1544
J. Org. Chem., Vol. 54,No. 7, 1989
fraction
Pirrung e t al.
Table I. Purification of a$-Dihydroxyacid Dehydratase from Spinach protein, mg mg/mL activity specific activity
vol, mL
19.5 12.5
Sephacryl S-200
AH-agarose
20.2 5.88
1.04 0.47
1693 965
83.8 164.3
purification
yield
119.7 234.7
42 24
EDTA. p-Chloromercuribenzoate was also reported as a strong (0.5 mM) inhibitor. A K , was measured at 6 mM for dihydroxyisovalerate (valine precursor) and 2 mM for dihydroxy p-methylvalerate (isoleucine precursor). The invariance of these values at various purification stages and pH's suggested that a single enzyme was present. This report also showed that the enzyme has n o aconitase or fumarase activity, nor activity on a,p-dihydroxy acids bearing &hexyl or -phenyl groups. During the course of the work described herein, an independent report appeared concerning the spinach e n ~ y m e . ' ~ While important in terms of extending the accepted bacterial biosynthetic pathway to plants, this previous work did not meet our needs for substantial quantities of enzyme for inhibitor screening. Progress toward this goal as well &s some success in DHAD-targeted herbicide design is reported herein. E x p e r i m e n t a l Section General. Poly(ethy1enimine) (Polymin P) was obtained from Sigma as a 50% solution, which was diluted according to Jendrieak and Burgessm DEAEcellulose (DE52) was obtained from Sigma, Sephacryl S-200 from Pharmacia, and aminohexylagarose from Sigma. Xanthine, xanthine oxidase, and superoxide dismutase were from Sigma. 2-Ketc-3-methylbutyric acid, (R)-pantolactone, acid were N,N-dimethylglycine,and dl-2-hydroxy-3-methylhutyric acid was obtained from Aldrich. 2,3-Dihydroxy-3-methylbutyric by the method of d'Angelo.2' 3-Hydroxy-3-methylhutyricacid was prepared by the method of Krapcho and JahngenJ2 N,NDimethyloxamide was obtained by the method of Verm~elen.2~ 2-Fluoro-3-methyl-2-butenoic acid was obtained by the procedure ~ of Gillet.% 1-Hydroxy-2-methylpropanesulfonicacid w a obtained by the procedure of Green and Hi11e.2~ 2,3-Epoxy-3-methylbutanoic acid was obtained by the method of Speziale and Frazier.P6 1-Hydroxy-2-methylpropanephosphonicacid2' was obtained from the diethyl esterz8by cleavage with TMSBr. N,NDimethylglycine N-oxidem was obtained from ethyl N,N-dimethylglycinate by oxidation with hydrogen peroxide and saponification. Tetrahydrofuran (THF) was distilled from sodium/henzophenone under nitrogen immediately prior to use. 'H NMR spectra were recorded on a Nimlet NMC-300 spectrometer. UV spectra were obtained on a Hewlett-Packard 8450A spectrophotometer. High-pressure liquid chromatography was conducted on a Hewlett-Packard 1090 liquid chromatograph using a 4 mm X 15 cm oligonucleotide column operated a t 40 "C and eluted with 15% methanol in 60 mM pH 3 phosphate buffer. Protein concentration was determined by the Bio-Rad modification of the Bradfordm dye-binding assay?' BSA was the (19) Flint. D. H.: Emp-,
M H J . B i d Chcm. I988,263,355%35M. (20) Jendrissk. J. J.; Burgess. R. R. Bioehemisrr) 1975.14.4fi?9-4645. 1211 d'Anzolo. J.: Paves. 0 : Maddaluno.. J.:. Dumas.. F.:. Revisl. C .
TetrahedronLek. i983: 24,5869-5872. (22) Krapcho, A. P.; Jshngen, E. G. E. J. Org. Chem. 1974, 39, 1322-1324. (23) Vermeulen, N. M. J.; Lourens, G. J.; Potgeiter, D. J. J. S. Afr. J. Sei. 1981, 77, 566-569. (24) Gillet, J. P.; Sauvetre,R.; Normant, J. F. Synthesis 1986,35&360. (25) Green, L. R.; Hine, J. J. 0%. Chem. 1974,39,3896-3901. (26) Speziale, A. J.; Frazier, H.W. J . Org. Chem. 1961,26,31763183. (27) Hoffman", M. Pol. J. Chem. 1982,56, 1191-1194. (28) Agawa, T.; Kubo. T.; Ohshiro, Y. Synthesis 1971, 27-28. (29) Ikutani, Y. Bull. Chem. Soe. Jpn. 1968.41, 167+1681. (30) Bradford, M. M. Anal. Biochem. 1976, 72,248-254.
Figure I. DE52 chromatography of DHAD Also,dashed lines; solid lines, gradient and enzyme activity. Aliquots of 100 r L were assayed as described in the Experimental Section.
e
- --
Figure 2. Polyacrylamide gel electrophoresis of DHAD. Molecular weight markers (lanes l and 7 left to right) are a t 97.4, 66.2,42.7,31.0,21.5, and 14.4 kDa. Lane 3 is crude, lane 4 is after DE52, lane 5 is after gel filtration, and lane 6 is after hydrophobic chromatography. standard. Buffer T consists of 50 mM TribHCI, pH 8.00 (25 "C), 2 mM EDTA, 5 mM mercaptoethanol, and 5 mM MgCI,. Purification of Spinach ol$-Dihydroxyacid Dehydratase (DHAD). AU steps were carried out at 4 OC. Fresh spinach leaves (1.2 kg) obtained from a local market were chopped into small pieces and homogenized in a 1 gallon Waring blender with 2 L of buffer T. Filtration through cheesecloth and centrifugation at l2000g for 30 min gave a supernatant to which NaCl was added to a concentration of 200 mM. Polymin P" (0.005 volumes, 15.55 mL of a 12.5% solution adjusted to pH 8.0 with HCI) was added slowly with stirring. After being stirred for 30 min, the solution was centrifuged at lZooOg for 15 min. The supernatant was brought to 30% saturation with ammonium sulfate, stirred for 30 min, and centrifuged and the pellet was discarded. The remaining solution was brought to 55% saturation, stirred for 30 min, and centrifuged, and the Supernatant was discarded. The pellet was suspended in 80 mL of buffer T and dialyzed (Spectra/Por 4) exhaustively against the -e buffer. Insoluble material was removed by centrifugation. The clear yellow dialysate (110.5 mL) was applied to a 225" DEAE-cellulose column equilibrated with buffer T. The column was eluted with a linear 2-L gradient of NaCl (0.0 0.5 M)in
-
(31)Speetor, T. Anal. Biaehem. 1978,86, 142.
J . Org. Chem., Vol. 54, No. 7, 1989 1545
Dihydroxyacid Dehydratase buffer T. Fractions of 7.9 mL were collected. The elution profile is shown in Figure 1. Fractions showing DHAD activity were combined and concentrated to 4-5 mL by ultrafiltration using a PM-30 membrane. This concentrate was applied to a 1.5 X 105 cm column of Sephacryl S-200 equilibrated with buffer T. The column was eluted at a flow rate of 30 mL/h, and fractions of 4 mL were collected. Fractions containing DHAD activity were combined and used without concentration. The enzyme solution after gel filtration was loaded onto a 12.5-mL column of aminohexylagarose equilibrated with buffer T. It was eluted with a 200-mL linear gradient (0.0 0.5 M) of NaCl in buffer T. Four fractions of 3 mL each were pooled, brought to 10% glycerol, and quick-frozen. A representative purification is summarized-in Table I. Gel filtration on a Sepharose 6B column standardized for molecular weight determination [blue dextran (2000 kDa), Pamylase (200 kDa), yeast alcohol dehydrogenase (150 m a ) , bovine serium albumin (66 kDa), carbonic anhydrase (29 kDa), and cytochrome c (12.4 kDa)] gave the M,of DHAD as 55 500. Denaturing polyacrylamide gel electrophoresis of DHAD (specific activity 513 U/mg) against standards indicated a M,of 62000 (Figure 2). Enzyme Assay. A modification of the literature method was used.1° A standard OD curve for a-keto acid (dinitropheny1)hydrazone was obtained by substituting a-ketoisovaleric acid for enzyme and substrate in the procedure below. Solutions of substrate and inhibitors were prepared in buffer T and adjusted to pH 8.00 prior to use. Enzyme solution was added to a 13 X 1cm test tube and the volume was brought to 300 pL with buffer T. A solution of a,P-dihydroxyisovaleric acid (100 pL, 80 mM) at pH 7.5 was added and the mixture was incubated for 30 min at 35 "C. Blanks were prepared by omitting the substrate or by boiling the enzyme solution. The reaction was quenched by the addition of 100 pL of 10% trichloroacetic acid. A saturated solution of 2,4-dintrophenylhydrazine(200 pL) in 2 N HC1 was added and the mixture was allowed to stand at room temperature for 20 min. NaOH solution (2.5 N, 800 pL) WBS added, the mixture was again allowed to stand (30 min), and the coagulated protein was removed by filtration through glass wool. The Am measured against the reference curve gave the amount of a-ketoisovaleric acid produced. Time course studies established that the rate of product formation was linear over the incubation time. K , was determined to be 8 mM. V,, was measured at 4.56 nmol min-' (mg protein)-' X lo3. Inhibition studies (on sodium salts) were conducted as above with substrate at K , (Dixon analysis) or at various substrate and inhibitor concentrations (Hanes-Woolf analysis and least squares by KINFIT, a program kindly provided by Prof. V. Anderson). For those compounds that are chiral, all values refer to the racemate. For HPLC analysis, the reaction mixture consisted of 120 pL of buffer T, 30 pL of 80 mM dihydroxy acid, and 450 pL of enzyme solution incubated at 35 "C for 2 h. Aliquots of 100 pL were quenched into 20 pL of cold methanol and injections of 100 pL were made. Substrate loss under these conditions was 0.438 nmol min-', while product formation was 0.492 nmol m i d . Labeling Experiments. [3-2H]-a-Ketoisovaleric acid was prepared by treating the acid in D20 with catalytic sodium carbonate for 48 h. The reaction mixture was neutralized and concentrated, and the residue was triturated with ethanol. After concentration of the ethanol, the a-keto acid obtained showed >99% deuterium incorporation by 'H NMR. [2-2H]-a,P-Dihydroxyisovalerate was prepared by adding aoxo-P-hydroxyisovaleric acid (7.03 mmol) to sodium borodeuteride (4.6 mmol) in 6 mL of HzO. After acidification and extractive isolation, the acid was esterified with diazomethane and chromatographed on silica gel (1:l EtOAc/hexanes). Saponification gave the sodium salt. This compound showed >90% D incorporation by integration of the residual methine proton signal versus the methyls. Incubation of a 21.6 mM (final concentration) solution of deuterated substrate with 27 pg DHAD (specific activity 3.5 X lo3 nmol (mg protein mi$') was conducted at 35 "C for 2 h in 800 pL of buffer T. After being quenched with 100 pL of 3 M HCl, the reaction mixture was extracted with ether. After drying (Na2S04),the ether extract was treated with excess diazomethane,
-
concentrated to 100 pL, and analyzed by GC/MS (Hewlett Packard 5890, 25 m X 0.2 mm cross-linked Me-Silicone; 50 OC, 2 min; 10°/min to 200 "C; t~ = 5.8 min). Analysis of the ions at m/e 70, 71, and 72 (isobutyroyl ion fragment) permits the determination of the percent deuterium. When deuterated substrate was used, percent deuterium was
