ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS

Vol. 245, No. 1, February 15, pp. 114-124,1986

,&Sulfur Substituted a-Ketoglutarates as Inhibitors and Alternate Substrates for lsocitrate Dehydrogenases and Certain Other Enzymes’

GERHARD W. E. PLAUT,2 TADASHI AOGAICHI, AND JEROME L. GABRIEL

Department of Biochemistry, Tempk University School of Medicine, Philadelphia, Pennsylvania 19140

Received July 29,1985, and in revised form October 17,1985

The &S-isomers of P-mercapto-a-ketoglutarate, fi-methylmercapto-cY-ketoglutarate and /3-methylmercapto-a-hydroxyglutarate have been synthesized. P-Mercapto-a-ke- toglutarate was a potent inhibitor, competitive with isocitrate and noncompetitive with NADP+, of the mitochondrial NADP-specific isozyme from pig heart (Ki = 5 nM; &(DL-

isocitrate)/&(RS-P-mercapto-cY-ketoglutarate) = 650) and pig liver, the cytosolic isozyme from pig liver (I0.5 = 23 nM), and the NADP-linked enzymes from yeast (Ki = 58 nM)

and Escherichia coli (Ki = 58 nM) at pH 7.4 and with Mp as activator. @-Mercapto-a- ketoglutarate was also an effective inhibitor of NADP-isocitrate-dehydrogenase activity in intact liver mitochondria. P-Mercapto-a-ketoglutarate was a much less potent inhibitor for heart NAD-isocitrate dehydrogenase (Ki = 520 nM) than for the NADP-specific enzyme. P-Methylmercapto-a-ketoglutarate (I0.5 = 10 pM) was a much less effective inhibitor than the /3-mercapto derivative for heart NADP-isocitrate dehydrogenase. The P-sulfur substituted a-ketoglutarates were substrates for the oxidation of NADPH by heart NADP-isocitrate dehydrogenase without requiring COz. @Methylmercapto-a- hydroxyglutarate, the expected product of reduction of P-methylmercapto-a-ketoglu- tarate, did not cause reduction of NADP+ but it was an inhibitor competitive with

. isocitrate for NADP-isocitrate dehydrogenase. The P-sulfur substituted cY-ketoglutarate derivatives were alternate substrates for a-ketoglutarate dehydrogenase and the cy- tosolic and mitochondrial isozymes of heart aspartate aminotransferase but had no effect on glutamate dehydrogenase or alanine aminotransferase. o 1986 Academic PF~SS, I~C.

Analogs of isocitrate have been used to study the structural requirements for sub- strates of the NADP-specific and the NAD- specific isocitrate dehydrogenase from heart (1). D-three-a-Methylisocitrate was a potent inhibitor of NADP-isocitrate de- hydrogenase, but not of the NAD-linked enzyme (2,3), and the isocitrate analog l- hydroxy-2-nitropropane-1,3-dicarboxylic acid has been reported to inhibit the NADP-specific enzyme (4). The dead end inhibitors a-methylisocitrate and the IX- ketoglutarate analog oxalyglycine have

1 This work was supported in part by Grant AM 15404 from the National Institute of Arthritis, Dia- betes, Digestive and Kidney Diseases.

‘To whom correspondence should be addressed.

been useful in defining the kinetic mecha- nism of action of NADP-isocitrate dehy- drogenase (5, 6). cY-Methylisocitrate has been used to establish that isocitrate oxi- dation in intact isolated rat liver mito- chondria occurs mainly by way of the NAD-specific dehydrogenase (7) and that the NADP-isocitrate dehydrogenase reac- tion was the principal source of the reduc- ing equivalents required for formation of urea from ammonia in isolated hepatocytes (8). However, studies of the activity of NADP-isocitrate dehydrogenase within intact cell compartments with cy-methyli- socitrate may be limited by the deficiency in membrane transport systems for tri- carboxylates in certain tissues (9). Since the penetration of a-ketoglutarate into

0003-9861/86 $3.00 Copyright 0 1986 by Academic Press. Inc. All rights of reproduction in any form reserved.

114

&SULFUR SUBSTITUTED a-KETOGLUTARATES 115

cell compartments involves a membrane transporter different from that for tricar- boxylates (9), a study of analogs of a-ke- toglutarates as potential inhibitors of iso- citrate dehydrogenases was undertaken. p- Bromo-a-ketoglutarate has been found to be an affinity-directed inhibitor of NADP- isocitrate dehydrogenase (10) and NAD- isocitrate dehydrogenase (11). It seemed possible that other a-ketoglutarate deriv- atives substituted with an electronegative group at the p position may be inhibitors of these dehydrogenases. This was found to be the case in the present studies with P-sulfur substituted cr-ketoglutarate de- rivatives. In this respect, P-mercapto-a- ketoglutarate may be of particular interest since its inhibition constant for soluble NADP-isocitrate dehydrogenase was about 30 times smaller than that of a-methyliso- citrate and it was a potent inhibitor of this enzyme in intact liver mitochondria. The sulfur compounds were also substrates and/or inhibitors of a number of other en- zymes for which a-ketoglutarate is a reac- tant or product.

EXPERIMENTAL PROCEDURES

Materials

NADP-specific isocitrate dehydrogenase was puri- fied from pig heart by the method of Kelly and Plaut

(12). In most of the studies the enzyme purchased from Sigma was used. The enzymes from pig liver cytosol

and mitochondria were purified as reported by Plaut et al. (13). NAD-specific isocitrate dehydrogenase was

purified from bovine heart by the method of Giorgio et al. (14) and a-ketoglutarate dehydrogenase from pig heart as described by Sanadi (15). Bovine liver

glutamate dehydrogenase, pig heart alanine amino- transferase, cytosolic aspartate aminotransferase

from pig heart, pig heart malate dehydrogenase, and rabbit muscle lactate dehydrogenase were purchased

from Boehringer-Mannheim. Mitochondrial aspartate aminotransferase from pig heart was a gift from Dr.

M. Martinez-Carrion. Other chemicals used were re- agent grade and were obtained commercially.

/3-Bromo-a-ketoglutaric acid was prepared by the

method of Hartman (10). The syrup obtained by this procedure was induced to crystallize by the addition of the authentic material.’ A small amount of a brown

a We thank Dr. F. C. Hartman for a sample of the crystalline compound.

impurity could be removed by treatment of the solid

with boiling chloroform in a soxhlet extractor followed

by recovery of the crystals from the cooled extract.

Barium oxalosuccinate was prepared by the method of Ochoa (16). After conversion to the sodium salt by

treatment with NazSOI, the a-ketoglutarate and ox- alosuecinate content of the solution (about 10 mM)

was estimated as follows: a-ketoglutarate was deter-

mined spectrophotometrically by the coupled aspar- tate aminotransferase/malate dehydrogenase assay

(17) before and after incubation of 1 vol of the sample in 3 vol of 0.05 M ethylenediamine-0.05 M succinic acid

buffer (pH 7.0). Ethylenediamine catalyzes the de- carboxylation of oxalosuccinate to a-ketoglutarate and

the difference in a-ketoglutarate concentration de- termined before and after incubation with the

ethylenediamine containing buffer is due to oxalo-

succinate. Thin-layer chromatography. The solvent systems

used for development were I, isopropyl ether:88% formic acid:water (64~2); II, liquid phenol:88% formic

acid:water (83:1:17); III, benzene:tetrahydrofuran: acetic acid (45:2’7:3).

Free acids were separated in systems I and II on Eastman Chromagram No. 13255 (cellulose) and de-

tected on the dried support media with glucose/aniline reagent (18) or as the fluorescent quinoxalinol deriv-

atives formed with the a-keto acids after spraying

with o-phenylenediamine reagent (19). The 2,4-dini- trophenylhydrazones of a-keto acids were separated in system III on Eastman Chromagram No. 6060 (sil-

ica) and detected as yellow zones under visible light.

The R, value for each compound is recorded in pa- renthesis after the roman numeral identifying each

system below. fl-Mercapto-a-ketoglutarate: I (0.18), II (0.26), III (0.19); fl-methylmereapto-a-ketoglutarate:

I (0.90), III (0.27); fl-bromo-a-ketoglutarate: I (0.78), II (0.54); a-ketoglutarate: I (0.59), III (0.52); 2,4-dini- trophenylhydrazine: III (0.93).

Assays

All enzyme activity units refer to micromoles per minute per milligram protein and incubations were

done at 25°C except for the a-ketoglutarate dehydro- genase assay which was at 30°C. The initial velocities were determined by following formation or consump-

tion of reduced pyridine nucleotides at 340 nm in a Cary Model 219 speetrophotometer.

NADP-isocitrate dehydrogenase: The standard as- say mixture contained 167 mM Na-Heper? at pH 7.4,

1 mM dithiothreitol, 10 pM NADPH, 2 mg/ml bovine serum albumin, 1.33 mrd MgSO,, 100 PM NADP+, 25 ELM DL-iS0CitId.e and enzyme. Bovine serum albumin

’ Abbreviation used: Hepes, N-(2-hydroxy- ethyl)piperazine-N’-2-ethanesulfonic acid.

and NADPH were present in the incubation media the conditions described previously (7). The reoxida- because they favor formation of the active dimeric tion of NADH was inhibited by antimycin A and the

form of the enzyme (6). Changes in substrate concen- further oxidation of a-ketoglutarate formed from trations when P-substituted cy-ketoglutarate analogs added isocitrate was prevented by the addition of so- were added are indicated in the text, table, and figure dium arsenite. The activity of NAD-isocitrate dehy-

legends. The oxidation of NADPH (reductase activity) drogenase activity was optimized by maintaining a was measured in a medium containing 167 mM Na- high ADP/ATP ratio by the addition of glucose and

Hepes at pH 7.4, 2.2 mM MgSO* 1.67 mg/ml bovine yeast hexokinase to the reaction mixtures.

serum albumin, 1.0 mM dithiothreitol, 0.15 mM

NADPH, variable concentrations of oxalosuccinate or the @sulfur substituted cr-ketoglutarate derivatives

PREPARATION OF /3-SULFUR SUBSTITUTED

and enzyme. A control without substrate accompanied c~-KETOGLUTARATE AND (Y-

all samples and the change in absorbance at 340 nm HYDROXYGLUTARATE DERIVATIVES

was subtracted from the sample containing all com-

ponents. This correction became especially important P-Mercapto-a-ketoglutarate

when the rate of reaction with a particular substrate (Ammonium Salt)

was very slow and larger than the usual amounts of Solution A: A solution of 1.8 g j3-bromo-ol-ketoglu- enzyme had to be used, resulting in significant rates taric acid (8 mmol) in 4 ml of water is neutralized to of oxidation of NADPH in the absence of substrate. about pH 8.3 by the slow addition of concentrated

NAD-isocitrate dehydrogenase: The assay mixtures NHIOH at ice bath temperature. Solution B: Water- contained 16’7 mM Na-Hepes at pH 7.4,1.0 rnrd dithio- washed HzS is passed through 1 N NH,OH in an ice threitol, 0.50 mM NAD+, 0.50 mM ADF- when present, bath until saturated. and enzyme. Variations in concentration of magne- Solution A is added dropwise with stirring to 9.8 sium ion and DL-iS0CitXat.e have been indicated in the ml of Solution B cooled in an ice bath. After standing specific experiments. for 5 h in the cold, the compound is precipitated by

a-Ketoglutarate dehydrogenase: The incubation the dropwise addition of 2 vol of cold ethanol, filtered, solution contained 67 mM sodium phosphate buffer at washed with 5 ml of absolute ethanol, and dried in a pH 7.2, 0.05 mM CoA, 0.33 mrd NAD+, 1.0 mM dithio- vacuum at room temperature. The crystalline com- threitol, and enzyme. The concentrations of cu-keto- pound was recovered in a yield of 1.37 g (69%). The glutarate and/or a-ketoglutarate analogs are de- analytical sample is recrystallized by dissolving 100 scribed in specific experiments. mg in 2.0 ml of water followed by the dropwise ad-

Aspartate aminotransferase: The reaction mixtures dition of 5.9 ml of ethanol at room temperature. The contained 77 mM sodium phosphate at pH 7.6,0.20 mM crystals recovered by filtration in 74% yield from the NADH, 1.0 mM dithiothreitol, 8.3 rig/ml malate de- refrigerated solution are washed with cold ethanol hydrogenase, and enzyme. The concentrations of L- and dried in a high vacuum at room temperature. aspartate, a-ketoglutarate, and a-ketoglutarate an- The migration of the compound as a single zone on alogs are cited under the specific experiments. thin-layer chromatography system I or II was distinct

Alanine aminotransferase: The reaction mixture from that of the starting material fi-bromo-cy-keto- contained 77 mM sodium phosphate at pH 7.4,0.16 mM glutarate. For the diammonium salt of B-mercapto- NADH, 1.0 mM dithiothreitol, 6.7 pg/ml lactate de- a-ketoglutarate dihydrate, anal. calcd for C.&&O$ hydrogenase, and enzyme. The concentrations of L- C 24.19, H 6.49, N 11.28, S 12.91. Found: C 24.13, H

alanine and a-ketoglutarate were 40 and 0.9 m&i, re- 6.55, N 11.06, S 12.98. spectively. The 2,4-dinitrophenylhydrazone prepared from the

Glutamate dehydrogenase: The reaction mixture above compound (mp 182-183°C) migrated as a single contained 46 mM triethanolamine hydrochloride at pH zone (R, 0.19) on thin-layer chromatography system 8.0,0.15 mM NADH, 1 ml dithiothreitol and enzyme. III. For the 2,4-dinitrophenylhydrazone of j3-mer- The concentrations of cu-ketoglutarate and ammonia capto-a-ketoglutarate dihydrate, anal. calcd for were 2.7 and 8.1 mM, respectively. CIIHIINIOI,,S: C 33.51, H 3.58, N 14.21, S 8.13. Found:

Reactions in nonrespiring intact liver mitochondria: C 34.21, H 3.40, N 14.15, S 8.17. Additional data on the The coupled reactions contents of sulfhydryl and keto groups of the com-

isocitrate + NH3 -+ glutamate + COz PI pound are summarized in Table I.

and fl-Methylmercapbwketoglutarate

isocitrate + acetoacetate -+ (Potassium Salt)

cY-ketoglutarate + COa + fi-hydroxybutyrate [2] A solution of 900 mg (4 mmol) of @bromo-cu-keto-

glutaric acid in 80 ml of 0.1 M KHCOa is gassed for 2

were examined in intact rat liver mitochondria under h with water-washed methylmercaptan at 25°C. The

116 PLAUT, AOGAICHI, AND GABRIEL

fl-SULFUR SUBSTITUTED a-KETOGLUTARATES 117

TABLE I

Parameters compared

Nitrogen/sulfur Potassium/sulfur

Sulfhydryl group/weight

Keto group/weight

fi-mercapto-a-

ketoglutarate”

(atom or mol ratio)

Calcd. Found

2.00 1.95

1.00 0.97

1.00 1.03

P-methylmercapto- a-ketoglutarate*

(atom or mol ratio)

Calcd. Found

2.06 2.09

0.00 0.00

1.00 1.00

Method

Elemental analysis Atomic absorption/

elemental analysis Speetrophotometric’

Spectrophotometricd

a Diammonium salt.

* Dipotassium salt. ’ Boyer (20).

d Determined by the method of Olson (21). The samples (about 50 PM) were incubated for 5.5 h at 25°C in a medium containing 5.0 mM semicarbazide. HCl in 0.2 M sodium acetate buffer at pH 4.45. Concentrations of

the compounds were calculated by comparison to A,M 250 nm for o-ketoglutaric acid of 11.3. This compares to

AFMnm of 10 for cu-ketoglutarate found by Olson (21).

solution is concentrated to about 30 ml in a flash evaporator at 15 Torr and a bath temperature of 20°C titrated with 60 m&i silver acetate in the dark, and

silver bromide is removed by filtration.’ The filtrate is concentrated to dryness in a flash evaporator at 15

Torr and a bath temperature of 25°C. The residue is

washed three times with 40 ml portions of ethanol. The residue is dried in a vacuum at 56.5”C and recov-

ered in a yield of 815 mg (‘76%).The compound at this stage may still be contaminated with a small amount

of potassium acetate, even though only a single spot for nonvolatile acid (R,O.90) could be detected on thin-

layer chromatography system I. For the preparation of the analytical sample, 100 mg of the potassium salt

was refluxed for 10 min in 5 ml of ethanol and filtered at room temperature. The filtrate containing potas-

sium acetate was discarded and the insoluble potas- sium P-methylmercapto-a-ketoglutarate was col- lected. For the dipotassium salt of fl-methylmercapto-

a-ketoglutarate, anal. calcd for CrJ&K20SS: C 26.85, H 2.25, K 29.14, S 11.95. Found: C 26.68, H 2.26, K 29.14, S 11.46.

The 2,4-dinitrophenylhydrazone prepared from the above compound (mp, 184184.5”C) migrated as a sin-

gle zone (R, 0.27) on thin-layer chromatography in

system III. Anal. calcd for C1zH1aN108S: C 38.71, H

‘In case an excess of silver acetate is added, the excess silver ion can be removed from the filtrate by

passing HzS through the solution and removing AgzS by filtration.

3.25, N 15.05, S 8.61. Found: C 37.96, H 3.20, N 14.56, S 8.35. Additional analytical data are summarized in

Table I.

RS-@Methylmercapto-a-hydroxyglutaric Luctone

A solution of 1.2 g (4.5 mmol) RS-methylmercapto- a-ketoglutarate (dipotassium salt) in 45 ml of water

was treated with 5.4 mmol of NaBH4 for 2 h at room temperature and was then allowed to stand for 18 h

at 5°C. The solution was passed through a 118 X 20-

cm column of AG 5OW-X12 (H+ form), the acidic ef- fluent was collected and concentrated to dryness in a flash evaporator at 20°C and 15 Torr. The residue was

mixed with 2.5 ml ether, insoluble material (boric acid) was removed by filtration, and the filtrate was con-

centrated to dryness in a flash evaporator. The residue was taken up in 10 ml of boiling chloroform, cooled

to room temperature and a small amount of insoluble material was removed by filtration. Crystals formed

after concentration of the solution to about one-half

volume (251 mg, 32%), mp 103-104°C. Additional amounts of compound could be recovered upon further

volume reduction of the mother liquor. Anal. calcd for CsHs04S: C, 40.92; H, 4.58; S, 18.17. Found: C, 41.05;

H, 4.73; S, 17.92. IR (KBr pellet):1786 cm-’ (y-lactone), 1715 (COOH). Eq wt (cold) calcd: 176.2; found: 181.5. Eq wt (hot) calcd: 88.1; found: 90.9. Thin-layer chro-

matography in system I gave a single spot with R, 0.90. Upon hydrolysis of the la&one in excess alkali under an atmosphere of Nz, a single spot with Rf0.73 was obtained in system I which was identical in mi-

118 PLAUT, AOGAICHI, AND GABRIEL

gration with that obtained with a sample of crystalline free fi-methylmercapto-cY-hydroxyglutaric acid. The

lactone was hydrolyzed to the free acid before use in enzyme experiments.

RESULTS AND DISCUSSION

Properties of P-Substituted a-Ketoglutarate Derivatives

The elemental analyses of the salts of p- mercapto-a-ketoglutarate and of @-meth- ylmercapto-cY-ketoglutarate and of their corresponding 2,4-dinitrophenylhydrazone derivatives are in good agreement with the expected structures. Spectrophotometric determination of the keto group content of the P-sulfur substituted a-ketoglutarates by the semicarbazone method of Olson (21) indicated the presence of one carbonyl group per molecule of compound when the reaction was allowed to proceed to com- pletion and P-mercapto-cu-ketoglutarate was found to contain one thiol group per molecule by the spectrophotometric method of Boyer (20) (Table I).

Eflects of &%lfur Substituted a-Ketoglutarate Derivatives on Certain Enzymatic Reactions

The analogs were tested with a number of enzymes for which a-ketoglutarate is a product or substrate of the reaction.

Isocitrate Dehydrogenases

Inhibition. B-Mercapto-cY-ketoglutarate was a potent inhibitor of pig heart NADP- isocitrate dehydrogenase (Table II). The inhibition by the mercapto analog was competitive with respect to isocitrate (Fig. 1) and noncompetitive with NADP+. Inhi- bition was also found in the direction of reductive carboxylation of a-ketoglutarate. The inhibition constants for all of these reactions were in the nanomolar range (Table II). When NADP-isocitrate dehy- drogenase was preincubated with p-mer- capto-a-ketoglutarate (19.5 PM) in a me- dium without isocitrate, Mgz+ and NADP+ for up to 6 h, the inhibition was found to be fully reversible when aliquots of the preincubation mixture were tested in the complete assay medium containing a sat-

urating concentration of isocitrate. This suggests that the chemical modification of a group on the enzyme which affects activ- ity, observed with p-bromo-a-ketogluta- rate (lo), does not occur with the p-mer- capto derivative. It has been shown by im- munochemical procedures that animal tissues contain mitochondrial and cytosolic isozymes of NADP-isocitrate dehydroge- nase (13). The purified pig heart enzyme used in most of the studies here is the mi- tochondrial isozyme. When the partially purified cytosolic isozyme from pig liver was tested, the value of I,,+ for P-mercapto- a-ketoglutarate was similar to that for the mitochondrial enzyme from pig heart (Ta- ble II) and the inhibition was competitive with respect to isocitrate for either iso- zyme. However, in contrast to the enzyme from pig heart (Fig. 2A), the inhibition of the activity of the enzyme from pig liver cytosol in response to increasing p-mer- capto-cY-ketoglutarate was sigmoidal and, when represented in the form of a Dixon plot, the response to increasing inhibitor concentration was nonlinear (Fig. 2B, curve 1). When these data were fitted to a form of the equation for competitive inhibition which included the inhibitor Hill coefficient q as described previously (22) the Dixon plot became linear when q was 1.55 (Fig. 2B, curve 2). This difference in cooperativ- ity in binding of /%mercapto-a-ketogluta- rate to the mitochondrial and cytosolic en- zymes seems to be a characteristic property since the same inhibition pattern was also found with the corresponding isozymes from liver of the ox and the rat with values of q of 1.0 and 1.55 for the mitochondrial and cytosolic isozymes, respectively (not shown).

When the effect of enzyme concentration on specific activity was examined (23) it was found that at low enzyme protein con- centrations in the preincubation mixture, where the specific activity of either isozyme declined (6, 24), the presence of 8-mer- capto-a-ketoglutarate protected the cyto- solic isozyme, but not the mitochondrial enzyme, against the effects of dilution (not shown). This may suggest that P-mercapto- a-ketoglutarate binds to one substrate binding site on either isozyme which ac-

&SULFUR SUBSTITUTED wKETOGLUTARATES 119

TABLE II

INHIBITION BY &%JLFUR SUBSTITUTED-WKETOGLUTARATES OF ISOCITRATE DEHYDROGENASES a

Kinetic constant”

Substrate

Inhibitor varied

NADP-isocitrate dehydrogenase from pig heart

None IC

None NADP+

None a-KG (+CO,)’ b-SH-a-KG IC

,&SH-a-KG NADP+ P-SH-a-KG a-KG (+COz)”

@-MeS-a-KG IC

NADP-isocitrate dehydrogenase from pig liver

/3-SHW-KG @-SH-a-KG

NADP-isocitrate dehydrogenase from yeast

None Ic

@-SH-a-KG IC

NADP-isocitrate dehydrogenase from E. coli

None IC

B-SH-(Y-KG IC

NAD-isocitrate dehydrogenase from bovine heart

None Mg.Ic /3-SH-(y-KG Mg.Ic /3-MeS-a-KG Mg+Ic

K(w) Mode of

(PM) action

3.42 AZ 0.28 (42) S

2.25 + 0.45 (35) S

68 + 7 (7) S 0.0050 + 0.0003 (42) C

0.0060 +- 0.0018 (35) NC

0.0090 + 0.0016 (14) I

10.0 + 0.9 (14) C

0.0111 + 0.0014 (17)“6 I

0.0232 + 0.0007 (ll)d*f I

9.6 f 1.2 (8) S 0.058 + ND (25) C

2.7 + 0.24 (8) S 0.058 f ND (18) C

120 + 10 (23) S

0.52 + 0.19 (23) C 870 + 400 (14) C

‘The conditions of incubation are described under Experimental Procedures. The abbreviations for the compounds are /3-SH-a-KG, ES-@-mercapto-a-ketoglutarate; P-MeS-a-KG, RS+methylmercapto-cY-ketoglu-

tarate; Ic, DLisocitrate; a-KG, cY-ketoglutarate; OSA, oxalosuccinate. *The kinetic constants are K,(app) for the substrate varied and Kj for the analog in the absence and

presence of inhibitor, respectively. The number of data points are shown in parentheses. The abbreviations

are S, substrate; C, competitive inhibition; NC, classical noncompetitive inhibition; I, inhibition type not

determined. ‘Reaction in the direction of oxidation of NADPH.

d Value of I0.s when the concentration of DL-isocitrate was constant at 25 PM. e Mitochondrial isozyme. f Cytosolic isozyme.

counts for inhibition competitive with re- spect to isocitrate. In the case of the cy- tosolic isozyme, additional binding of the analog may occur to a second site resulting in stabilization of the active enzyme dimer and a Hill inhibitor coefficient (q) above unity.

fi-Mercapto-cY-ketoglutarate was also an effective inhibitor of NADP-isocitrate de- hydrogenase from yeast and Escherichia coli with inhibition constants about 10 times larger than for the enzyme from pig

heart (Table II). This suggests similarities in binding sites for the enzyme from widely different species.‘j This becomes of partic- ular interest if one considers that the ac- tivity of the enzyme from E. coli is regu-

6 In parallel experiments it was found that a-meth- ylisocitrate, a potent selective inhibitor of NADP-iso-

citrate dehydrogenase from animal tissues (2) also inhibited the enzymes from yeast (I,,.5 = 2 PM) and E.

cdi (I+s = 4 FM) (P. Zervos and J. L. Gabriel, unpub- lished observations).

120 PLAUT, AOGAICHI, AND GABRIEL

‘/[M2 &-ISOCITRATE], J.M-’

FIG. 1. Inhibition of pig heart NADP-isocitrate de-

hydrogenase by ,%mercapto-cY-ketoglutarate as a function of magnesium isocitrate concentration. The

concentrations of the inhibitor were fixed at 0.0 nbi

(O), 10 nM (0), 20 nrd (m), 40 nb! (O), and 60 nM (A). Magnesium DL-iSOCitra~ was varied by increasing the

total DL-isocitrate concentration. The calculated lines were determined by fitting velocity versus [magnesium

DL-iSOCitrate] data to the velocity expression for com- petitive inhibition (22).

lated by a phosphorylation-dephosphory- lation mechanism (25), whereas, such regulation of NADP-isocitrate dehydro- genase from animal tissues has not been reported and phosphate could not be de- tected in the purified enzyme from pig heart (26).

P-Mercapto-a-ketoglutarate was an in- hibitor, competitive with respect to mag-

E” 4 A

7 .E E 3 l .

z

‘E

9 2 .

>- c

:: 1

d 0

-3. 0 10 20 30 40

[f3-SH-a-KGI,nM

nesium isocitrate, of NAD-isocitrate de- hydrogenase. However, the inhibition con- stant for the analog was about two orders of magnitude larger for the NAD-specific than for the NADP-linked dehydrogenase (Table II).

fi-Methylmercapto-tu-ketoglutarate was a much less effective inhibitor than p- mercapto-a-ketoglutarate for either the NADP-specific or the NAD-specific enzyme from heart; for example, the constants for competitive inhibition with respect to iso- citrate for these analogs differed by about 2000-fold for the NADP-specific enzyme from pig heart (Table II).

Substrate activity. The sulfur substituted analogs were alternate substrates for the NADP-specific dehydrogenase, but not the NAD-specific enzyme, in the direction of NAD(P)H oxidation. In the assay for re- ductase activity the maximal velocity with P-methylmercapto-a-ketoglutarate was around 60% of that with oxalosuccinate but the apparent Km of the analog was about 14-fold larger than that of oxalosuccinate (Table III). On the other hand, the maximal velocity with /3-mercapto-a-ketoglutarate was only about 0.2% of that with oxalo- succinate but this analog showed extremely tight binding to the enzyme (Table III). The low value of Km for P-mercapto-cu-ketoglu- tarate precluded the determination of the constant by the usual velocity response to

‘e i’ 7 0 100 200 300 400

E I I I

- ‘6 ’ B

30 40 50

[p-SH-W-KG], MA

FIG. 2. (A) Inhibition of pig heart mitochondrial NADP-isocitrate dehydrogenase by increasing concentrations of @-mercapto-a-ketoglutarate (/3-SH-a-KG). Data are presented as a Dixon plot of l/velocity vs w-SH-a-KG]. (B) Inhibition of pig liver cytosolic NADP-isocitrate dehydrogenase by increasing concentrations of B-mercapto-a-ketoglutarate (B-SH-a-KG). Data are presented as a

Dixon plot of l/velocity vs p-SH-a-KG] (curve 1, lower scale) and l/velocity vs [@SH-a-KG]‘& (curve 2, upper scale).

&SULFUR SUBSTITUTED a-KETOGLUTARATES 121

TABLE III

THE OXIDATION OF NAD(P)H BY ~-SULFUR SUBSTITUTED WKETOGLUTARATES AND

OXALOSUCCINATE CATALYZED BY ISOCITRATE DEHY~R~~JwA~ES~

Substrate

v,OSNb JLlos) (ratio)

NADP-isocitrate dehydrogenase from pig heart OSA 80 + 12 1.0 (13) fl-MeS-cY-KG 1140 + 11 1.6 (7) @-SH-(Y-KG 0.064 + 0.009 491 (lS)d

NAD-isocitrate dehydrogenase from bovine heart OSA NE” P-MeS-(Y-KG NE /3-SH+KG NE

“The conditions of incubation are described under Experimental Procedures. The abbreviations are the same as in Table II. The number of data points are shown in parentheses.

* The ratio for V~~E~/V&,~~ was 10. ‘NE, no substrate activity. dThe number of time points used to estimate the

single reaction progress curve (27). See the text.

substrate concentration procedure. The constant for the analog reported in Table III was estimated by the single reaction progress curve procedure of Fukagawa (27), starting with an initial concentration of 1.0 pM P-mercapto-cY-ketoglutarate. For p- mercapto-cY-ketoglutarate the values of Km for reductase activity (Table III) and of Ki for inhibition of isocitrate dehydrogenase activity (Table II) were in the nanomolar range.

The rate of reduction of RS-P-methyl- mercapto-cY-ketoglutarate (0.50 mM) by NADPH (1.0 mM) catalyzed by pig heart NADP-isocitrate dehydrogenase was fol- lowed by monitoring the change in NADPH fluorescence; an assay mixture including enzyme but without the methylmercapto analog was used as the control. The data points for the time course, calculated from the decrease in NADPH concentration, could be fitted to a curve best described by two simultaneously occurring and equally weighted first-order decay curves with ap- proximate rate constants of 0.013 and 0.002 per minute, respectively (data not shown).

When the reaction was allowed to proceed to completion, 0.46 mM NADP+ was found to have accumulated as measured by the glucose-&phosphate dehydrogenase reac- tion, indicating that complete reduction of added RS+methylmercapto-a-ketogluta- rate had occurred. The data suggest that the enzyme catalyzes the complete reduc- tion of the two major forms of the analog, but at different rates.

The rates of oxidation of NADPH by the P-sulfur substituted a-ketoglutarate ana- logs catalyzed by NADP-isocitrate dehy- drogenase were not increased by the ad- dition of bicarbonate-COz. This is in agreement with the observations of Ochoa (28) with oxalosuccinate and of Hartman (10) with P-bromo-a-ketoglutarate that CO2 was not required for the NADPH-de- pendent reduction of these substrates by this enzyme. Inhibition of NADP-isocitrate dehydrogenase competitive with respect to isocitrate occurred with P-bromo-a-keto- glutarate (10) and the P-sulfur substituted a-ketoglutarate derivatives (Table II), and Ochoa and Weisz-Tabori (29) reported that isocitrate competitively inhibited the de- carboxylation of oxalosuccinate. These ob- servations suggest that an electrophilic substituent at the /3 position of cu-ketoglu- tarate may fulfill the requirements for binding at an oxalosuccinate binding site on the enzyme.

When the reaction with NADP-isocitrate dehydrogenase was tested in the reverse direction, reduction of NADP+ was not found with up to 10 mM RS$-methylmer- capto-a-hydroxyglutarate. However, p- methylmercapto-a-hydroxyglutarate was a rather weak competitive inhibitor (Ki = 1.7 m&l) with respect to isocitrate in the di- rection of NADPH formation. The situa- tion appears to be similar to the finding of Illingworth and Tipton (30) that oxaloac- etate was reduced by NADPH to D-malate by this enzyme but that the reversal of this reaction could not be demonstrated.

NAD-isocitrate dehydrogenase did not exhibit reductase activity with the sulfur containing analogs of a-ketoglutarate (Ta- ble III), which may be consistent with the previous observation that oxalosuccinate was not a substrate for the enzyme (31).

122 PLAUT, AOGAICHI. AND GABRIEL

Bednar et al. (11) reported that oxidation of NADH occurred at a rapid rate, albeit at a rather high concentration, with @- bromo-a-ketoglutarate (K, = 5.2 mM) and at an extremely slow velocity with a-ke- toglutarate (K, = 11 mM) in the absence of COz. The experiments with the NAD- specific enzyme from pig heart of Bednar et al. (11) (1982) were done at pH 6.1 and with Mn2+ as the activator, whereas, the present incubations with the bovine heart enzyme were done at pH 7.4 and with M$+ as the activating ion, which may account for the observed differences in substrate activity for the sulfur and bromo analogs. It is noteworthy, however, that the enzyme was inhibited by both the /3-bromo (11) and the P-sulfur analogs (Table II). This may suggest that P-bromo-a-ketoglutarate and the P-sulfur substituted a-ketoglutarate derivatives interact with the NAD-specific enzyme at other than an oxalosuccinate binding site.

Experiments with intact liver mitochon- dria. When examined with varying con- centrations of @-mercapto-a-ketoglutarate the conversion of isocitrate and ammonia to glutamate (Eq. [l]) was inhibited from 45 to 55% at the extrapolated infinite con- centration of ,L?-mercapto-a-ketoglutarate in nonrespiring rat liver mitochondria (apparent Ki between 0.8 and 1.8 PM),

whereas, no significant inhibition of the reduction of acetoacetate to P-hydroxybu- tyrate by isocitrate (Eq. [2]) was observed with up to 10 pM P-mercapto-a-ketogluta- rate (experiments not shown). These re- sults were similar to those obtained pre- viously with a-methylisocitrate a potent selective inhibitor of NADP-isocitrate de- hydrogenase; however, the apparent Ki for ,&mercapto-a-ketoglutarate was about lOOO- to 2000-fold smaller than that esti- mated for ru,-threo-a-methylisocitrate (7). Thus, the lack of effect of either inhibitor on the intramitochondrial production of NADH from added isocitrate and its COU-

pled reoxidation to NAD+ by acetoacetate reflect the relative insensitivity of NAD- isocitrate dehydrogenase to inhibition by P-mercapto-a-ketoglutarate (Table II) and the seemingly small activity of transhy- drogenase which would be required for the

generation of NADH from NADPH formed via the NADP-isocitrate dehydrogenase reaction. In contrast, the reductive ami- nation of a-ketoglutarate to glutamate can utilize either NADH or NADPH, and the inhibitable portion of the coupled reaction (Eq. [l]) is attributable to inhibition of mi- tochondrial NADP-isocitrate dehydroge- nase activity since p-mercapto-cy-ketoglu- tarate is not an inhibitor of glutamate de- hydrogenase (see next section). These results suggest that P-mercapto-a-keto- glutarate may be useful as a selective in- hibitor of NADP-isocitrate dehydrogenase activity in studies of metabolic processes in mitochondria and, perhaps, more com- plex cell preparations.

Efect of Sulfur Analogs on Enzymes for Which a-Ketoglutarate Is a Substrate

The results obtained with a number of enzymes are shown in Table IV.

fi-Mercapto-a-ketoglutarate was a po- tent inhibitor of a-ketoglutarate dehydro- genase, competitive with a-ketoglutarate (Ki = 24 PM). As judged by NADH for- mation, the /3-methylmercapto derivative could replace a-ketoglutarate as a sub- strate for the enzyme with a slower turn- over for the analog than with the natural substrate. This is consistent with inhibition by the sulfur derivatives competitive with respect to a-ketoglutarate.

P-Mercapto-a-ketoglutarate was not an inhibitor for glutamate dehydrogenase or alanine aminotransferase. However, P- mercapto- and P-methylmercapto-cw-keto- glutarate were alternate substrates for the mitochondrial and cytosolic isozymes of aspartate aminotransferase. The expected products of the reaction, the p-sulfur sub- stituted glutamate derivatives, were not identified. However, the formation of ox- aloacetate from aspartate by transami- nation with the P-sulfur c-u-ketoglutarate analogs could be monitored by following the consumption of NADH when coupled to the malate dehydrogenase reaction. The maximal velocities with the sulfur analogs were about 30-40% of that with cy-ketoglu- tarate; however, under comparable condi- tions the K, values for the B-sulfur sub-

123 @WLFUR SUBSTITUTED a-KETOGLUTARATES

TABLE IV

SUBSTRATE ACXIVITIES OF @-SULFUR SUBSTITUTED a-KETOGLUTARATES WITH VARIOUS ENZYMES“

Kinetic constant*

Substrate K&m) V, Enzyme varied MM)

e (ratio)

KGDH a-KG 0.20 f 0.03 (22) fl-MeS-o-KG 2.19 + 0.17 (8) 0.4

GOT(c) ASP 5.8 -+ 0.4 (8) a-KG 0.21 f 0.01 (6) @-SH-a-KG 7.4 + 3.0 (17)b 0.4 P-MeS-a-KG 1.2 f 0.4 (9)b 0.4

GOT(m) Asp 0.58 + 0.06 (6) a-KG 0.24 -t 0.06 (9) P-SH-(Y-KG 5.1 f 1.7 (17)b 0.3 @MeS-a-KG 4.7 + 1.6 (17)* 0.3

o The abbreviations for the enzymes are KGDH, a-ketoglutarate dehydrogenase; GOT(c), cytosolic isozyme of aspartate aminotransferase; GOT(m), mitochondrial isozyme of aspartate aminotransferase. See Table II for other abbreviations. The number of determinations are shown in the parentheses.

*The concentration of ~-Asp was at its K,,,.

stituted derivatives were from 6- to 35fold larger than that of a-ketoglutarate. A number of amino acids have been found to substitute for aspartate as a substrate for aspartate aminotransferase but the en- zyme seems to have a narrower specificity for the a-ketoglutarate/glutamate pair of the transamination reaction [for review see (32)]. The reactivity of the P-sulfur analogs with the enzyme, particularly of ,&meth- ylmercapto-a-ketoglutarate with the cy- tosolic isozyme (Table IV), therefore, may be of interest for studies requiring an al- ternate substrate since its Km was only sixfold larger and its maximal velocity was about 40% of that for cY-ketoglutarate.

CONCLUSIONS

,&Mercapto-cu-ketoglutarate was a very effective inhibitor of NADP-isocitrate de- hydrogenase. The tight binding of the an- alog to the enzyme is illustrated by the KJDL - isocitrate)/Ki(RS- p - mercapto - CY - ketoglutarate) ratio of about 650 for the dehydrogenase from pig heart (Table II). Under the incubation conditions used, the dimeric form of the enzyme is prevalent

and in accord with previous studies the or- der of addition of substrates should be random (12,33). This is in agreement with the present observation that the inhibition by ,L?-mercapto-a-ketoglutarate was com- petitive with respect to isocitrate and non- competitive with NADP+ (Table II). p- Mercapto-a-ketoglutarate was an alternate substrate for oxalosuccinate for the reduc- tase activity of the enzyme (Table III). However, at the extremely low values of Km and V its effect for practical purposes is equivalent to that of a deadend inhibitor which binds to a central complex of the en- zyme-possibly displacing oxalosuccinate from an oxalosuccinate-enzyme complex. Thus, its actions are those attributable to a transition state analog (34).

REFERENCES

1. PLAUT, G. W. E., BEACH, R. L., AND AOGAICHI, T. (1975) Biochemistry 14,2581-2588.

2. PLAUT, G. W. E., BEACH, R. L., AND AOGAICHI, T. (1975) J. Biol Chem 250,6351-6354.

3. BEACH, R. L., AOGAICHI, T., AND PLAIJT, G. W. E. (1977) J. Biol. Chmrh 252,2702-2709.

4. ALSTON, T. A., PORTER, D. J. T., AND BRIGHT, H. J. (1983) Act. Chem Res. 16,418-424.

124 PLAUT, AOGAICHI, AND GABRIEL

5.

6.

7.

8.

9.

10. 11.

12.

13.

14.

15.

16. 17.

18.

NORTHROP, D. B., AND CLELAND, W. W. (1974) J. Biol Chxem. 249,2928-2931.

KELLY, J. H., AND PLAUT, G. W. E. (1981) J. Biol

Chem. 256,335-342.

19. WIELAND, T., AND FISCHER, E. (1949) N&m&s- sensc~ 36,291.

20. BOYER, P. D. (1954) J. Amer. Chem. SC. 76,4331- 4337.

SMITH, C. M., AND PLAUT, G. W. E. (1979) Eur. J. 21. OLSON, J. A. (1959) Arch. Biochem Biophys. 85,

Biochem 97,283-295. 225-233. PETCU, L. G., AND PLAUT, G. W. E. (1980) Biochem 22. GABRIEL, J. L., AND PLAUT, G. W. E. (1982) J. Biol

J. 190,581-592. Chem. 257,8021-8029. CHAPPELL, J. B., AND ROBINSON, B. H. (1968) in

The Metabolic Roles of Citrate (Goodwin, T. W., ed.) Biochem. Sot. Symp. No. 27, pp. 123- 133, Academic Press, New York.

HARTMAN, F. C. (1981) Biochemistry 20,894-898. BEDNAR, R. A., HARTMAN, F. C., AND C~OLMAN,

R. F. (1982) Biochemistry 21,3681-3689. KELLY, J. H., AND PLANT, G. W. E. (1981) J. Bid

Chen~ 256,330-334.

23. KURGANOV, B. I. (1967) Mel Biol (Mommu) 1,17- 27.

24. CARLIER, M. F., AND PANTALONI, D. (1978) Eur. J.

B&hem 39,571-578.

25. GARNAK, M., AND REEVES, H. C. (1979) J. Bid

Chem, 254,7915-7920. 26. MAS, M. T., AND COLMAN, R. F. (1984) Biochemistry

23.1675-1683.

PLANT, G. W. E., COOK, M., AND AOGAICHI, T. (1983) Biochmx Biophys Actu 760.300-308.

GIORGIO, N. A., JR., YIP, A. T., FLEMING, J., AND PLAUT, G. W. E. (1970) J. Bid Chem 245,5469-

5477.

27. FUKAGAWA, Y. (1980) Biochem J. 185.186-189. 28. OCHOA, S. (1948) J. Bid Chem 174,133-157. 29. OCHOA, S., AND WEISZ-TABORI, E. (1948) J. Bid

Chem 174,123-132.

SANADI, D. R. (1969) in Methods in Enzymology (Lowenstein, J. M., ed.), Vol. 13, pp. 52-55, AC- ademic Press, New York.

OCHOA, S. (1948) J. Bid Chem. 174,115-122. BERGMEYER, H. U., AND BERNT, E. (1963) in Meth-

ods of Enzymatic Analysis (Bergmeyer, H. U., ed.), pp. 837-841, Academic Press, New York.

ZWEIG, G., AND SHERMA, J. (1972) Handbook of Chromatography, Vol. 2, p. 132, Chem. Rubber Co., Cleveland.

30. ILLINGWORTH, J. A., AND TIPTON, K. F. (1970) FEBS Lett. 7, 157-160.

31. PLAUT, G. W. E., AND SUNG, S. C. (1954) J. Biol. Chem. 207,305-314.

32. BRAUNSTEIN, A. E. (1973) in The Enzymes, Vol. 9, pp. 379-381.

33. UHR, M. L., THOMPSON, V. W., AND CLELAND, W. W. (1974) J. Bid Chewy 249.2920-2927.

34. WOLFENDEN, R. (1977) in Methods in Enzymology (Jakoby, W. B., and Wilchek, M.) Vol. 46, pp. 15-28, Academic Press, New York.