Avian cytosolic 3-hydroxy-3-methylglutaryl-CoA synthase: evaluation of the role of cysteines in reaction chemistry.

The pH dependence of avian cytosolic HMG-CoA synthase activity is fit by a titration curve with a pK = 8.6. The observation of optimal activity at alkaline pH and the insensitivity of pK to divalent cation concentration suggest that the pK reflects ionization of an amino-acid side chain (e.g., cysteinyl sulfhydryl) rather than substrate enolization. Upon reaction of 3-chloropropionyl-CoA with HMG-CoA synthase C129S, an enzyme variant lacking the sulfhydryl group normally targeted by this mechanism-based inhibitor, stoichiometric modification occurs. Amino-acid analysis indicates that cysteine is the principal target in C129S enzyme, demonstrating the presence of a second reactive cysteine within this enzyme. To test whether another cysteine functions in reaction chemistry, conserved cysteines were identified by sequence homology analysis. Five cysteine residues (C59, C69, C224, C232, C268), invariant in the nine sequences available for various eukaryotic HMG-CoA synthase isozymes, were individually replaced by alanine in a series of mutant enzymes. Kinetic analyses of the isolated mutant HMG-CoA synthases indicate that none of these is crucial to the chemistry that results in production of HMG-CoA. These results further distinguish the HMG-CoA synthase reaction from the related condensation of acyl-CoA substrates catalyzed by beta-ketothiolase.

Mapping of reactive sulfhydryls in avian liver 3-hydroxy-3-methylglutaryl coenzyme A synthase.

Avian liver mitochondrial 3-hydroxy-3-methylglutaryl coenzyme A synthase contains seven sulfhydryls per 53 kDa subunit. Peptides that harbor these sulfhydryls can be mapped by reverse-phase HPLC separation of tryptic digests of denatured 14C-carboxymethylated enzyme. Native enzyme is inactivated by a variety of reagents that target cysteine residues. Of particular interest is the enzyme's sensitivity to reagents (e.g., CdCl2, copper phenanthroline) that target vicinal thiols. The identity of the cysteines which are modified by these reagents can be determined by peptide mapping after denaturation. 14C-carboxymethylation and trypsin digestion of the sample. While the extent of reaction of any particular cysteinyl sulfhydryl depends on the identity of the reagent employed, three of the protein's seven cysteinyl sulfhydryls are frequently modified upon inactivation of the enzyme. The peptides which contain these reactive sulfhydryls have been isolated and their sequences have been determined by Edman degradation techniques. Comparison of these sequences with the deduced primary structure of the rodent cytosolic enzyme (Gil et al. (1986) J. Biol. Chem. 261, 3710) indicates strong homologies. These homologies allow assignment of the reactive residues as Cys-129, Cys-224 and Cys-268. The sensitivity of these residues to reagents that target vicinal thiols, coupled with the fact that cys-129 is the residue involved in formation of the acyl-S-enzyme intermediate (Vollmer et al. (1988) Biochemistry 27, 4288), suggests that these three residues may be closely juxtaposed within the enzyme's catalytic domain.

Characterization of mevalonate kinase V377I, a mutant implicated in defective isoprenoid biosynthesis and HIDS/periodic fever syndrome.

The list of diseases linked to defects in lipid metabolism has recently been augmented by the addition of hyperimmunoglobulinemia D and periodic fever syndrome (HIDS: MIM 260920), which are correlated with depressed levels of mevalonate kinase activity [1,2] and protein [1]. More specifically, a V377I substitution has been proposed to account for this disease. We observed that V377 appears to be far from invariant in eukaryotic mevalonate kinases. Prokaryotic mevalonate kinases are lower in molecular weight and several terminate prior to residue 377 of the eukaryotic proteins. These observations prompted our direct test of the impact of V377 on activity and protein stability by engineering a V377I mutation in a recombinant human mevalonate kinase. The mutant protein has been isolated and kinetically characterized. In comparison with wild-type enzyme, V377I exhibits only modest differences (notably > or = 6-fold inflation of K(m(MVA))) that do not account for the diminished mevalonate kinase activity assayed in HIDS cell extracts. Moreover, thermal inactivation (50 degrees C) of isolated wild-type and V377I enzymes demonstrates little difference in stability between these proteins. We conclude that a single V377I substitution is unlikely to explain the observation of depressed mevalonate kinase stability and catalytic activity in HIDS.

3-Hydroxy-3-methylglutaryldithio-CoA: utility of an alternative substrate in elucidation of a role for HMG-CoA lyase's cation activator.

(S)-(3-Hydroxy-3-methyl-1-thionoglutaryl)-Coenzyme A (HMG[= S]CoA), a dithioester analog of (S)-(3-hydroxy-3-methylglutaryl)-CoA (HMG-CoA), acts as an efficient alternative substrate for avian HMG-CoA lyase. Detection of product formation by HPLC, UV absorbance and coupled enzyme assays indicates that HMG[= S]CoA cleavage yields acetyl[= S]CoA and acetoacetate. HMG[= S]CoA binds to the lyase with a Km of 13 microM and undergoes the cleavage reaction at a maximal rate which is 20% of that observed with HMG-CoA. The enzyme-catalyzed cleavage of both HMG-CoA and HMG[= S]CoA is stimulated by the divalent cations Mg2+ and Mn2+. Mg2+ produces a 2-fold higher stimulation of HMG-CoA cleavage than that observed with Mn2+. In contrast, stimulation of HMG[= S]CoA cleavage is nearly seven times higher with Mn2+ than with Mg2+. Not only is the stimulation of enzymatic activity dependent on the cation, but also the Km values for Mg2+ and Mn2+ are dependent upon the substrate used. In contrast, the Km values for HMG-CoA and HMG[= S]CoA are not markedly dependent on the identity of the divalent cation. These results are compatible with the initial formation of a binary enzyme-substrate complex prior to binding of the divalent cation to produce a catalytically active enzyme-substrate-metal ternary complex.

Phosphoribulokinase: isolation and sequence determination of the cysteine-containing active-site peptide modified by 5'-p-fluorosulfonylbenzoyladenosine.

Phosphoribulokinase (ATP: D-ribulose-5-phosphate 1-phosphotransferase, EC 2.7.1.19) is stoichiometrically inactivated by the ATP analog, 5'-p-fluorosulfonylbenzoyladenosine (FSBA). The inactivation is reversed upon incubation with dithiothreitol suggesting that a cysteine is modified. Since the adduct formed upon enzyme inactivation is unstable to normal procedures for peptide analysis, the site of modification has been identified by converting the labile adduct to the well-characterized carboxymethylcysteine. The approach takes advantage of the sensitivity of the thiolsulfonyl-containing adduct to reducing agents; thus the adduct acts as a temporary protecting group while the previously unmodified cysteines are blocked. The FSBA-modified cysteine is then unmasked with dithiothreitol and radiolabeled. DEAE and reverse-phase high-pressure liquid chromatography of tryptic digests indicate that a single peptide is radiolabeled. Amino-acid analysis and automated Edman degradation techniques have been employed to confirm cysteine as the site of modification. The sequence of the tryptic peptide was determined to be: Ser.Gln-GIn-Gln-Thr-Ile-Val-Ile-Gly-Leu-Ala-Ala.Asp-Ser-Gly-CM-Cys-Gly-Lys. This sequence is identical with the reported N-terminal sequence, thus identifying Cys-16 as the site of modification.

Avian 3-hydroxy-3-methylglutaryl-CoA lyase: sensitivity of enzyme activity to thiol/disulfide exchange and identification of proximal reactive cysteines.

Catalysis by purified avian 3-hydroxy-3-methylglutaryl-CoA lyase is critically dependent on the reduction state of the enzyme, with less than 1% of optimal activity being observed with the air-oxidized enzyme. The enzyme is irreversibly inactivated by sulfhydryl-directed reagents with the rate of this inactivation being highly dependent upon the redox state of a critical cysteine. Methylation of reduced avian lyase with 1 mM 4-methylnitrobenzene sulfonate results in rapid inactivation of the enzyme with a k(inact) of 0.178 min-1. The oxidized enzyme is inactivated at a sixfold slower rate (k(inact) = 0.028 min-1). Inactivation of the enzyme with the reactive substrate analog 2-butynoyl-CoA shows a similar dependence upon the enzyme's redox state, with a sevenfold difference in k(inact) observed with oxidized vs. reduced forms of the enzyme. Chemical cross-linking of the reduced enzyme with stoichiometric amounts of the bifunctional reagents 1,3-dibromo-2-propanone (DBP) or N,N'-ortho-phenylene-dimaleimide (PDM) coincides with rapid inactivation. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of enzyme treated with bifunctional reagent reveals a band of twice the molecular weight of the lyase monomer, indicating that an intersubunit cross-link has been formed. Differential labeling of native and cross-linked protein with [1-14C]iodoacetate has identified as the primary cross-linking target a cysteine within the sequence VSQAACR, which maps at the carboxy-terminus of the cDNA-deduced sequence of the avian enzyme (Mitchell, G.A., et al., 1991, Am. J. Hum. Genet. 49, 101). In contrast, bacterial HMG-CoA lyase, which contains no corresponding cysteine, is not cross-linked by comparable treatment with bifunctional reagent. These results provide evidence for a potential regulatory mechanism for the eukaryotic enzyme via thiol/disulfide exchange and identify a cysteinyl residue with the reactivity and juxtaposition required for participation in disulfide formation.

Crystallization and preliminary X-ray crystallographic analysis of phosphoribulokinase from Rhodobacter sphaeroides.

A recombinant form of Rhodobacter sphaeroides phosphoribulokinase (PRK), expressed in Escherichia coli and isolated by affinity chromatography, was crystallized by the sitting drop vapor diffusion technique using NH4H2PO4 (pH 5.6) as the precipitating agent. PRK crystallizes in the cubic space group P432, with unit cell parameters a = b = c = 129.55 A. Based on the assumption of one 32-kDa monomer per asymmetric unit, the Vm value is 2.83 A3/Da. The octameric molecular symmetry is consistent with two planar tetramers stacked in a nearly eclipsed arrangement. A native data set has been collected to 2.6 A resolution.

Phosphoribulokinase: current perspectives on the structure/function basis for regulation and catalysis.

Phosphoribulokinase (PRK), an enzyme unique to the reductive pentose phosphate pathway of CO2 assimilation, exhibits distinctive contrasting properties when the proteins from eukaryotic and prokaryotic sources are compared. The eukaryotic PRKs are typically dimers of -39 kDa subunits while the prokaryotic PRKs are octamers of -32 kDa subunits. The enzymes from these two classes are regulated by different mechanisms. Thioredoxin of mediated thiol-disulfide exchange interconverts eukaryotic PRKs between reduced (active) and oxidized (inactive) forms. Allosteric effectors, including activator NADH and inhibitors AMP and phosphoenolpyruvate, regulate activity of prokaryotic PRK. The effector binding site has been identified in the high resolution structure recently elucidated for prokaryotic PRK and the7 apparatus for transmission of the allosteric stimulus has been identified. Additional contrasts between PRKs include marked differences in primary structure between eukaryotic and prokaryotic PRKs. Alignment of all available deduced PRK sequences indicates that less than 10% of the amino acid residues are invariant. In contrast to these differences, the mechanism for ribulose 1,5-biphosphate synthesis from ATP and ribulose 5-phosphate (Ru5P) appears to be the same for all PRKs. Consensus sequences associated with M++-ATP binding, identified in all PRK proteins, are closely juxtaposed to the residue proposed to function as general base catalyst. Sequence homology and mutagenesis approaches have suggested several residues that may potentially function in Ru5P binding. Not all of these proposed Ru5P binding residues are closely juxtaposed in the structure of unliganded PRK. Mechanistic approaches have been employed to investigate the amino acids which influence K(m Ru5P) and identify those amino acids most directly involved in Ru5P binding. PRK is one member of a family of phospho or sulfo transferase proteins which exhibit a nucleotide monophosphate kinase fold. Structure/function correlations elucidated for PRK suggest analogous assignments for other members of this family of proteins.

Ribulose-1,5-biphosphate carboxylase. Evidence in support of the existence of distinct CO2 activator and CO2 substrate sites.

Ribulose-1,5-biphosphate carboxylase forms a complex with Mn2+ and CO2 that exhibits a considerably enhanced water proton relaxation rate. This effect is diminished upon interaction of the complex with the substrate, ribulose biphosphate, or with the competitive inhibitor, ribitol 1,5-biphosphate (Ki=0.55 mM). Included among the several mechanisms which can explain these observations is the possibility that a slow exchange of metal ligands occurs. Attempts at testing the feasibility of a slow exchange mechanism led to demonstration of a stable complex of enzyme with CO2 in the presence of metal and carboxyribitol biphosphate (CRBP), an analog of the hypothetical transition state intermediate. The complex formed upon mixing these components is stable to Sephadex G-75 chromatography and contains a nonexchangeable [14C]O2 bound stoichiometrically with respect to enzyme active sites. Mg2+ or Mn2+ can be used to form the E.M.[14C]O2.CRBP complex, which is stable with respect to CO2 exchange until the enzyme is denatured with sodium dodecyl sulfate. If the tight, functionally irreversible binding of the transition state analog is due to its occupancy of ribulose biphosphate and substrate CO2 sites, then simultaneous stoichiometric binding of [14C]O2 to enzyme indicates that 2 CO2 molecules participate in photosynthetic carbon fixation.

3-hydroxy-3-methylglutaryl-coenzyme A synthase reaction intermediates: detection of a covalent tetrahedral adduct by differential isotope shift 13C nuclear magnetic resonance spectroscopy.

Binding of [1,2-(13)C]acetyl-CoA to wild-type 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) synthase is characterized by large upfield shifts for C1 (184 ppm, Deltadelta = 20 ppm) and C2 (26 ppm, Deltadelta = 7 ppm) resonances that are attributable to formation of the covalent [1,2 -(13)C]acetyl-S-enzyme reaction intermediate. NMR spectra of [1, 2-(13)C]acetyl-S-enzyme prepared in H(2)(16)O versus H(2)(18)O indicate a 0.055 ppm upfield shift of the C1 resonance in the presence of the heavier isotope. The magnitude of this (18)O-induced (13)C shift suggests that the 184 ppm resonance is attributable to a reaction intermediate in which C1 exhibits substantial carbonyl character. No significant shift of the C2 resonance occurs. These observations suggest that, in the absence of second substrate (acetoacetyl-CoA), enzymatic addition of H(2)(18)O to the C1 carbonyl of acetyl-S-enzyme occurs to transiently produce a tetrahedral species. This tetrahedral adduct exchanges oxygen upon backward collapse to re-form the sp(2)-hybridized thioester carbonyl. In contrast with HMG-CoA synthase, C378G Zoogloea ramigera beta-ketothiolase, which also forms a (13)C NMR-observable covalent acetyl-enzyme species, exhibits no (18)O-induced shift. Formation of the [(13)C]acetyl-S-enzyme reaction intermediate of HMG-CoA synthase in D(2)O versus H(2)O is characterized by a time-dependent isotope-induced upfield shift of the C1 resonance (maximal shift = 0. 185 ppm) in the presence of the heavier isotope. A more modest upfield shift (0.080 ppm) is observed for C378G Z. ramigera beta-ketothiolase in similar experiments. The slow kinetics for the development of the deuterium-induced (13)C shift in the HMG-CoA synthase experiments suggest a specific interaction (hydrogen bond) with a slowly exchangeable proton (deuteron) of a side chain/backbone of an amino acid residue at the active site.

3-hydroxy-3-methylglutaryl-CoA lyase: catalysis of acetyl coenzyme A enolization.

3-Hydroxy-3-methylglutaryl-CoA lyase, which performs the cleavage of 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA) to acetoacetate and acetyl-CoA by a Claisen-type reaction, also catalyzes enolization of acetyl-CoA. The rate of detritiation of methyl-labeled acetyl-CoA is proportional to enzyme concentration and is diminished by an antiserum that also inhibits the cleavage reaction. The tritium-exchange reaction requires both divalent cation and acetoacetate. An analogue of HMG-CoA, 3-hydroxyglutaryl-CoA, was prepared by reaction of acetonedicarboxylic anhydride with CoASH and reduction of the ketoacyl-CoA product with cyanohydridoborate. While 3-hydroxyglutaryl-CoA does not appear to be a substrate for HMG-CoA lyase, it competitively inhibits both the cleavage reaction (Ki = 50 microM) and the tritium exchange from acetyl-CoA (Ki = 95 microM). Agreement between the Ki values measured for cleavage and for tritium exchange supports the hypothesis that the slow tritium exchange is a lyase-dependent reaction. Initial attempts to demonstrate complete reversibility of the cleavage reaction have not been successful. However, the data suggest that the cleavage of HMG-CoA is at least partially reversible and indicate that enolization of acetyl-CoA may be dependent upon a conformational change of HMG-CoA lyase, induced by binding of acetoacetate, in a manner analogous to the keto acid dependent tritium exchange catalyzed by malate synthase and citrate synthase.

Amino acid sequence of an active site peptide of avian liver mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase.

Hydroxymethylglutaryl-CoA synthase is irreversibly inhibited by the active site-directed inhibitor 3-chloropropionyl-CoA. Enzyme modification has been postulated to involve alkylation of an active site cysteinyl sulfhydryl group. DEAE-Sephadex chromatography of tryptic digests prepared from enzyme inactivated using chloro[14C]propionyl-CoA suggested that bound radioactivity is localized on one peptide. Specificity of the modification was further demonstrated by reverse-phase high pressure liquid chromatography, which was used to isolate the radioactively labeled peptide in a chemically homogeneous form. Automated gas-phase Edman degradation techniques have been employed to confirm the assignment of cysteine as the inhibitor's target residue and to elucidate the sequence of amino acids which flank the 14C-carboxyethylated cysteine: Glu-Ser-Gly-Asn-Thr-Asp-Val-Glu-Gly-Ile-Asp-Thr-(Thr)- Asn-Ala-S-[14C]carboxyethylcysteine-Tyr-Gly-Gln-Thr-(Ala). These data represent the first assignment of active site structure for hydroxymethyl-glutaryl-CoA synthase.

Active-site-directed inhibition of 3-hydroxy-3-methylglutaryl coenzyme A synthase by 3-chloropropionyl coenzyme A.

3-Chloropropionyl coenzyme A (3-chloropropionyl-CoA) irreversibly inhibits avian liver 3-hydroxy-3-methylglutaryl-CoA synthase (HMG-CoA synthase). Enzyme inactivation follows pseudo-first-order kinetics and is retarded in the presence of substrates, suggesting that covalent labeling occurs at the active site. A typical rate saturation effect is observed when inactivation kinetics are measured as a function of 3-chloropropionyl-CoA concentration. These data indicate a Ki = 15 microM for the inhibitor and a limiting kinact = 0.31 min-1. [1-14C]-3-Chloropropionyl-CoA binds covalently to enzyme with a stoichiometry (0.7 per site) similar to that measured for acetylation of enzyme by acetyl-CoA. While the acetylated enzyme formed upon incubation of HMG-CoA synthase with acetyl-CoA is labile to performic acid oxidation, the adduct formed upon 3-chloropropionyl-CoA inactivation is stable to such treatment. Therefore, such an adduct cannot solely involve a thio ester linkage. Exhaustive Pronase digestion of [14C]-3-chloropropionyl-CoA-labeled enzyme produces a radioactive compound which cochromatographs with authentic carboxyethylcysteine using reverse-phase/ion-pairing high-pressure liquid chromatography and both silica and cellulose thin-layer chromatography systems. This suggests that enzyme inactivation is due to alkylation of an active-site cysteine residue.

Electron spin resonance studies of ribulosebisphosphate carboxylase: identification of activator cation ligands.

Ribulosebisphosphate carboxylase (RuBP carboxylase)forms a stable model complex containing stoichiometric amounts of enzyme sites, activator C0(2), divalent activator cation, and the transition-state analogue carboxyarabinitol bisphosphate (CABP). Incorporation of Mn(2+) in the model complex permits investigation of the environment of the activator cation by electron spin resonance (ESR)techniques. Measurements at 9 GHz on the Mn(2+)-containing complex prepared by using dimeric Rhodospirillum rubrum enzyme produce a spectrum which indicates that the cation is bound in an anisotropic environment. Measurements at 9 GHz on the spinach enzyme model complex produce a spectrum in which several of the fine structure transitions are obvious. In contrast, the spectrum produced from Mn(2+) bound to R. rubrum enzyme exhibits an intense powder pattern for the central fine structure transition; the other four fine structure transitions produce powder patterns that are in homogeneously broadened and therefore are not as apparent.Low-temperature measurements at high field (35 GHz) result in substantially simplified spectra. The spectrum of Mn(2+) bound to the R. rubrum enzyme shows less fine structure than the spectrum of Mn(2+) bound in the octameric spinach enzyme complex, where substantial hyperfine splitting is resolved in three of the five fine structure transitions. Measurements at 35 GHz on Mn (2+) bound in the dimeric R. rubrum enzyme complex produce spectra in which only the central fine structure transition produces a prominent signal. However, these samples are characterized by several narrow spectral features which permit investigation of the identity of Mn(2+)ligands by 170 perturbation techniques. Preparation of the R. rubrum RuBP carboxylase model complex in (17)O-enriched water results in a sample which exhibits an obviously broadened 35-GHz Mn(2+) spectrum in comparison to unenriched samples. Removal of H(2)(17)O by gel filtration abolished the spectral broadening, indicating that the Mn(2+)-coordinated water molecules can slowly exchange. No spectral broadening was detectable due to (17)O in the carbamate oxygens derived from activator C(17)O(2). NMR relaxation rate,measurement sat 24.3 MHz demonstrate that stoichiometric amounts of carboxyarabinitol bisphosphate eliminate enhancement of the proton relaxation rate observed in ternary enzyme-C0(2)-Mn(2+) complexes prepared by using dimeric R. rubrum enzyme. This observation, coupled with results of the H(2)(17)O ESR experiments,is compatible with the suggestion that the water molecules which coordinate directly to bound Mn(2+) are nonexchangeable on an NMR time scale but can be displaced by solvent water within 1-2 h. Carboxyarabinitol bisphosphate was selectively enriched with (17)O in the carboxyl group or in the oxygen on C-2. Mn(2+)-containing complexes prepared with either of the (17)O-enriched analogues produced spectra which were broadened in comparison to matched (l6)O controls. Thus,Mn(2+) coordinates directly to CABP, arguing for the participation of cation in the catalytic process.

The stereochemical course of the ribulose-5-phosphate kinase-catalyzed reaction.

Spinach-leaf ribulose-5-phosphate kinase catalyzes the reaction of (Rp)-[beta, gamma-18O, gamma-18O]adenosine 5'-(3-thiotriphosphate) with ribulose 5-phosphate to form ribulose 1-[18O]phosphorothioate 5-phosphate. This product is incubated with CO2, Mg2+, and ribulose-bisphosphate carboxylase to form the [18O]phosphorothioate of D-glycerate. Reduction of this material using phosphoglycerate kinase/ATP, glyceraldehyde-3-phosphate dehydrogenase/NADH, triose-phosphate isomerase, and glycerol-phosphate dehydrogenase/NADH produces glycerol 3-[18O]phosphorothioate, which is subjected to ring closure using diethylphosphorochloridate. This in-line reaction produces a diastereoisomeric mixture of glycerol 2,3-cyclic phosphorothioates. 31P NMR spectroscopy was used to analyze the 18O content of the products. The anti-diastereoisomer, which is the major isomer formed and corresponds to the downfield 31P NMR signal (Pliura, D.H., Schomburg, D., Richard, J.P., Frey, P.A., and Knowles, J.R. (1980) Biochemistry 19, 325-329), retains the 18O label. This observation indicates that the ribulose-5-phosphate kinase reaction proceeds with inversion of configuration at phosphorus. The reaction is, therefore, unlikely to involve the participation of a covalent phosphoryl-enzyme intermediate.

Pseudomonas mevalonii 3-hydroxy-3-methylglutaryl-CoA lyase: characterization of the isolated recombinant protein and investigation of the enzyme's cation requirements.

Pseudomonas mevalonii 3-hydroxy-3-methylglutaryl-CoA lyase has been expressed in an active form in Escherichia coli and purified to homogeneity. Enzyme activity in crude extracts is 30-fold higher than reported for a homologous expression system. After Q-Sepharose fast-flow anion-exchange chromatography, the enzyme, which represents the first homogeneous preparation of a prokaryotic form of the protein, exhibits a specific activity of 70 units/mg. The purified enzyme is stable when stored in 20% glycerol at -80 degrees C. The recombinant bacterial enzyme cross reacts with antiserum produced against avian liver lyase, indicating some sequence homology between the two proteins. The enzyme exhibits a Km = 20 microM for (S)-HMG-CoA. Divalent cations (Mg2+ and Mn2+) markedly stimulate the enzyme activity under assay conditions; activity is only modestly increased by exogenous mercaptans. The activator constant, K(a), for Mg2+ (6.9 mM) is 3 orders of magnitude greater than that for Mn2+ (2.0 microM). While EDTA does not affect activity, o-phenanthroline treatment markedly inhibits the enzyme. In contrast, m-phenanthroline is ineffective, suggesting that the ortho isomer's effect is attributable to chelation of a tightly bound metal ion. Atomic absorption and EPR analyses of isolated enzyme indicate the presence of tightly bound copper. In enzyme expressed using standard LB broth, copper is detected at stoichiometries of only 0.07-0.10. When the growth medium is supplemented with 1 mM CuSO4, stoichiometry of copper binding increases to over 0.7 per enzyme subunit. Copper-enriched lyase displays enhanced thermal stability in comparison with enzyme that is low in metal content.(ABSTRACT TRUNCATED AT 250 WORDS)

3-Hydroxy-3-methylgutaryl-CoA synthase. Participation of acetyl-S-enzyme and enzyme-S-hydroxymethylgutaryl-SCoA intermediates in the reaction.

Acetyl-CoA reacts stoichiometrically with a cysteinyl sufhydryl group of avian liver 3-hydroxy-3-methylglutaryl (HMG)-CoA synthase to yield acetyl-S-enzyme (Miziorko H.M., Clinkenbeard, K.D., Reed, W.D., and Lane, M.D. (1975) J. Biol. Chem. 250, 5768-5773). Evidence that acetyl-S-enzyme condenses with the second substrate, acetoacetyl CoA, to form enzyme-S-HMG-SCoA has been obtained by trapping and characterizing this putative intermediate. [14C]Acetyl-S-enzyme was incubated briefly at -25 degrees with acetoacetyl-CoA, precipitated with trichloroacetic acid, and the labeled acylated enzyme species were isolated. Performic acid oxidation of the precipitated [14C]acyl-S-enzyme intermediates produced volatile [14C]acetic acid from unreacted [14C]acetyl-S-enzyme and nonvolatile [14C]3-hydroxy-3-methyl glutaric acid from enzyme-S-[14C]HMG-SCoA. Condensation of unlabeled acetyl-S-enzyme with [14C]aceto-acetyl-CoA or acetoacetyl-[3H]CoA also produced labeled enzyme-S-HMG-SCoA. Thus, the acetyl moiety from acetyl-CoA and the acetoacetyl and CoA moieties from acetoacetyl-CoA all are incorporated into the HMG-CoA which is covalently-linked to the enzyme. Enzyme-S-[14C]HMG-SCoA was subjected to proteolytic digestion under conditions favorable for intramolecular S to N acyl transfer in the predicted cysteine-S-[14C]HMG-SCoA fragment. Performic acid oxidation of the protease-digested material yields N-[14C]HMG-cysteic acid indicating that HMG-CoA had been covalently bound to the enzyme via the -SH of an active site cysteine. An isotope trapping technique was employed to test the kinetic competence of acetyl-S-enzyme as an intermediate in the HMG-CoA synthase-catalyzed reaction. Evidence is presented which indicates that the rate of condensation of acetoacetyl-CoA with acetyl-S-enzyme to form enzyme-S-HMG-SCoA is more rapid than either the acetylation of the synthase by acetyl-CoA or the overall forward reaction leading to HMG-CoA. These observations, together with indirect evidence that hydrolysis of enzyme-S-HMG-SCoA is extremely rapid, suggest that acetylation of synthase is the rate-limiting step in HMG-CoA synthesis.