1274
Biochemistry 2002, 41, 1274-1284
Alternate Substrates of Human Glutaryl-CoA Dehydrogenase: Structure and Reactivity of Substrates, and Identification of a Novel 2-Enoyl-CoA Product† K. Sudhindra Rao,‡ David Vander Velde,§ Timothy M. Dwyer,| Stephen I. Goodman,‡ and Frank E. Frerman*,‡,⊥ Departments of Pediatrics and Pharmaceutical Sciences, UniVersity of Colorado Health Sciences Center, DenVer, Colorado 80262, Department of Medicinal Chemistry, UniVersity of Kansas, Lawrence, Kansas 66045, and Department of Chemistry, Towson UniVersity, Towson, Maryland 21252 ReceiVed July 30, 2001; ReVised Manuscript ReceiVed NoVember 26, 2001
ABSTRACT: The dehydrogenation reaction catalyzed by human glutaryl-CoA dehydrogenase was investigated using a series of alternate substrates. These substrates have various substituents at the γ position in place of the carboxylate of the physiological substrate, glutaryl-CoA. The steady-state kinetic constants of the six alternate substrates and the extent of flavin reduction in the anaerobic half-reaction were determined. One of these substrates, 4-nitrobutyryl-CoA, was previously thought not to be a substrate of the dehydrogenase; however, the enzyme does oxidize this substrate analogue with a kcat that is less than 2% of that with glutaryl-CoA when ferrocenium hexafluorophosphate (FcPF6) is the electron acceptor. Anaerobic titration of the dehydrogenase with 4-nitrobutyryl-CoA showed no reduction of the flavin; but instead showed an increased absorbance in the 460 nm region suggesting deprotonation of the analogue to form the R-carbanion. Analysis of these data indicated a binding stoichiometry of about 1.0. Under aerobic conditions, a second absorption maximum is observed with λmax ) 366 nm. The generation of the latter chromophore is dependent on an electron acceptor, either O2 or FcPF6, and is greatly facilitated by the catalytic base Glu370. The 466 nm absorbing species remains enzyme-bound while the 366 nm absorbing species is present only in solution. The latter compound was identified as 4-nitronate-but-2enoyl-CoA by mass spectrometry, 1H NMR, and chemical analyses. Ionization of the enzymatic product, 4-nitro-but-2-enoyl-CoA, that yields the nitronate occurs in solution and not on the enzyme. The variation of kcat with the nature of the substituent suggests that the various substituents affect the free energy of activation, ∆Gq, for dehydrogenation. There is a good correlation between log(kcat) and F, the field effect parameter, of the γ-substituent. No correlation was found between any other kinetic or equilibrium constants and the substituent parameters using quantitative structure-activity relationships (QSAR). 4-NitrobutyrylCoA is the extreme example with the strongly electron-withdrawing nitro group in the γ position.
Glutaryl-CoA dehydrogenase (GCD)1 catalyzes the oxidative decarboxylation of glutaryl-CoA to crotonyl-CoA and CO2 (1). Substrate oxidation and decarboxylation are not obligatorily coupled, and some acyl-CoA thioesters, e.g., pentanoyl-, hexanoyl-, and glutaramyl-CoA, that cannot be decarboxylated, are good alternate substrates of the human enzyme (2, 3). Also, Gomes et al. showed that the γ-methyl ester of glutaryl-CoA is a substrate of Pseudomonas fluorescens GCD (4). Alternate substrates and substrate ana† This research was supported by grants from the U.S. Public Health Service, NS 39339 (F.E.F.), and the National Science Foundation, NSF 99-77422, support to the NMR Laboratory at the University of Kansas. * Correspondence should be addressed to this author at the Department of Pediatrics, Box C233, University of Colorado Health Sciences Center, 4200 E. Ninth Ave., Denver, CO 80262. Phone: 303-315-7269, FAX: 303-315-8080, Email:
[email protected]. ‡ Department of Pediatrics, University of Colorado Health Sciences Center. § Department of Medicinal Chemistry, University of Kansas. | Department of Chemistry, Towson University. ⊥ Department of Pharmaceutical Sciences, University of Colorado Health Sciences Center. 1 Abbreviations: GCD, glutaryl-CoA dehydrogenase; FcPF , ferro6 cenium hexafluorophosphate; QSAR, quantitative structure-activity relationships.
logues are often useful reagents for enzyme mechanistic studies. Byron and colleagues recently reported the synthesis of 4-nitrobutyryl-CoA as a potential substrate analogue for GCD (5). Nitro compounds are excellent analogues for the related carboxylic acids. The nitro group is planar like the carboxylate group and has an extremely high dipole moment, essentially identical to those of carboxylic acids and amides, and about 2-fold greater than the corresponding esters. The strong inductive effect of the nitro group decreases the pKa of adjacent methylene protons, and inductive effects could make hydride transfer from the β-carbon of glutaryl-CoA less favorable. These properties suggested that 4-nitrobutyrylCoA could be a potentially useful dead-end inhibitor since it was previously reported that 4-nitrobutyryl-CoA was not a substrate of human GCD (5). Further, some nitro substrate analogues function as active site directed inhibitors that bind covalently and irreversibly at the active sites (6). Nitro analogues, as nitronate anions, may also serve as transitionstate analogues of carbanionic intermediates (7). Medium-chain acyl-CoA dehydrogenase is the archetype of the acyl-CoA dehydrogenases (8). The design of alternate substrates, substrate analogues, and active site directed
10.1021/bi015617n CCC: $22.00 © 2002 American Chemical Society Published on Web 01/03/2002
Substrate Oxidation by Glutaryl-CoA Dehydrogenase inhibitors for this dehydrogenase has focused on the properties of analogues that demonstrate the acidification of R-protons, the redistribution of charge during catalysis, and, more recently, charge properties of the β-carbon that are related to modulation of flavin redox potential and development of the transition state for the dehydrogenase (9-14). In the experiments described in this work, several alternate substrates of glutaryl-CoA dehydrogenase have been examined that focus on substituents at the γ-carbon that are substituted for a carboxylate in the physiological substrate, glutaryl-CoA. The experiments consider the effects of the interaction of the dehydrogenase with the γ-substituent, and field-inductive effects of the substituents. Previous experiments show that electrostatic catalysis is an important element in the catalytic pathway of glutaryl-CoA dehydrogenase and that Arg94 plays a critical role in this regard in the low dielectric constant of the active site (3). Unlike a previous study (5), we show that 4-nitrobutyryl-CoA is a substrate of the dehydrogenase, albeit a very poor one. 4-Nitrobutyryl-CoA can also be oxidized by an oxidase activity of the enzyme. The oxidase/dehydrogenase ratio of human GCD is about 75-fold greater with 4-nitrobutyrylCoA than with glutaryl-CoA as substrate. The product of enzymatic oxidation of 4-nitrobutyryl-CoA is a novel acylCoA that exhibits an additional absorption maximum at 366 nm besides the usual 260 nm peak. This product was identified as 4-nitronate-but-2-enoyl-CoA by chemical and spectral methods. The low activity of GCD with 4-nitrobutyryl-CoA prompted examination of the effect of other substrates with γ-substituents on the kinetic properties of the enzyme. Substituents were chosen to address structure/ activity relationships that could be mediated through the acyl chain and how each substituent affected substrate binding and dehydrogenation. EXPERIMENTAL PROCEDURES Materials. Ferrocenium hexafluorophosphate (FcPF6), monomethyl-glutarate, and 4-nitrobutyric acid methyl ester were obtained from Aldrich. 5-Hexenoic acid was obtained from Lancaster. Deuterium oxide, CoASH, hexanoyl-CoA, glutaryl-CoA, protocatechuic acid, protocatechuate-3,4-dioxygenase, and horseradish peroxidase were purchased from Sigma. Protocatechuate-3,4-dioxygenase from B. cepacia DB01 was a generous gift of Professor David Ballou, University of Michigan, Ann Arbor. Amplex red reagent was obtained from Molecular Probes. Pentanoyl-CoA and glutaramyl-CoA were synthesized as previously described (2, 3). Enzymes. Wild-type GCD (447nm ) 14.5 mM-1 cm-1) and Glu370Asp (446nm ) 14.3 mM-1 cm-1), Glu370Gln (447nm ) 13.2 mM-1 cm-1), and Arg94Gly (447nm ) 14.2 mM-1 cm-1) GCD mutants were expressed in Escherichia coli and purified as previously described (2, 3). Enzyme Assays. Glutaryl-CoA dehydrogenase activity was routinely assayed at 25 °C in 10 mM Tris-HCl, pH 8.0, 30 µM glutaryl-CoA, and 200 µM FcPF6 as the electron acceptor, using 300nm ) 4.3 mM-1 cm-1 (2). In steady-state kinetic experiments, the acyl-CoA was the varied substrate, and the steady-state constants were determined by nonlinear least-squares fit to the Michaelis-Menten equation. Oxidase activity of GCD was measured by monitoring the increase in absorbance at 570 nm (570nm ) 54 mM-1 cm-1)
Biochemistry, Vol. 41, No. 4, 2002 1275 with 22 µM Amplex red reagent and 6 units of horseradish peroxidase and glutaryl-CoA or 4-nitrobutyryl-CoA as substrate (15). This assay requires continuous stirring of the reaction mixture to obtain reproducible results. Amplex red reagent has been used for fluorometric determination of oxidases because of its high sensitivity. However, the absorbance method used here is also extremely sensitive due to the high molar absorptivity of the dye product, resorufin. There is no interference from components such as phenol in other oxidase assays (16, 17). Oxidase activity was also assayed by measuring the increase in absorbance at 366 nm due to formation of the 4-nitronate-but-2-enoyl-CoA from 4-nitrobutyryl-CoA. The molar extinction coefficient, 366nm ) 14.5 mM-1 cm-1, was determined by titrating FcPF6 with known concentrations of 4-nitrobutyryl-CoA in the presence of 100 nM GCD, and monitoring the increase in absorbance at 366 nm. The change in absorbance at 366 nm was a linear function of the concentration of 4-nitrobutyryl-CoA added until the dye was completely reduced. Synthesis of CoA Thioesters. 4-Nitrobutyric acid was prepared by alkaline hydrolysis of the methyl ester in water at 50-60 °C for 15 min with 1 equiv of NaOH. All acylCoA thioesters were synthesized by the mixed anhydride method (18). Acyl-CoA thioesters were purified by anion exchange chromatography on a column (25 × 180 mm) of DEAE-cellulose eluted with a linear gradient (1200 mL) of 0-250 mM LiCl in 1 mM HCl. Alternatively, the compounds were purified by high-performance liquid chromatography on a semipreparative, reversed-phase C-18 column (10 × 250 mm) which was eluted isocratically with 50 mM potassium phosphate buffer, pH 5.3, and methanol (95:5, v/v). The purified acyl-CoA thioesters were desalted on a column (10 × 900 mm) of Sephadex G-10 eluted with water. The CoA esters were stored lyophilized or frozen at -20 °C and were quantitated using 260nm ) 16.4 mM-1 cm-1 (19). The purity of the synthetic compounds was g98% as demonstrated by analytical high-performance liquid chromatography (20). Mass spectrometry of synthetic 4-nitrobutyryl-CoA (acid form, nonisotopic, C25H41N8O19P3S, MW ) 882.2) showed a single compound with an ion at m/z 883.3 ([M+H]+) as expected. One other species ([M+H]+, m/z 840.4) was also observed (
