No abstract

Biosynthesis of the isoprenoid precursor, isopentenyl diphosphate, is a critical function in all independently living organisms. There are two major pathways for this synthesis, the non-mevalonate pathway found in most eubacteria and the mevalonate pathway found in animal cells and a number of pathogenic bacteria. An early step in this pathway is the condensation of acetyl-CoA and acetoacetyl-CoA into HMG-CoA, catalyzed by the enzyme HMG-CoA synthase. To explore the possibility of a small molecule inhibitor of the enzyme functioning as a non-cell wall antibiotic, the structure of HMG-CoA synthase from Enterococcus faecalis (MVAS) was determined by selenomethionine MAD phasing to 2.4 A and the enzyme complexed with its second substrate, acetoacetyl-CoA, to 1.9 A. These structures show that HMG-CoA synthase from Enterococcus is a member of the family of thiolase fold enzymes and, while similar to the recently published HMG-CoA synthase structures from Staphylococcus aureus, exhibit significant differences in the structure of the C-terminal domain. The acetoacetyl-CoA binary structure demonstrates reduced coenzyme A and acetoacetate covalently bound to the active site cysteine through a thioester bond. This is consistent with the kinetics of the reaction that have shown acetoacetyl-CoA to be a potent inhibitor of the overall reaction, and provides a starting point in the search for a small molecule inhibitor.

Recent structural studies of the HMG-CoA synthase members of the thiolase superfamily have shown that the catalytic loop containing the nucleophilic cysteine follows the phi and psi angle pattern of a II' beta turn. However, the i + 1 residue is conserved as an alanine, which is quite unusual in this position as it must adopt a strained positive phi angle to accommodate the geometry of the turn. To assess the effect of the conserved strain in the catalytic loop, alanine 110 of Enterococcus faecalis 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) synthase was mutated to a glycine. Subsequent enzymatic studies showed that the overall reaction rate of the enzyme was increased 140-fold. An X-ray crystallographic study of the Ala110Gly mutant enzyme demonstrated unanticipated adjustments in the active site that resulted in additional stabilization of all three steps of the reaction pathway. The rates of acetylation and hydrolysis of the mutant enzyme increased because the amide nitrogen of Ser308 shifts 0.4 A toward the catalytic cysteine residue. This motion positions the nitrogen to better stabilize the intermediate negative charge that develops on the carbonyl oxygen of the acetyl group during both the formation of the acyl-enzyme intermediate and its hydrolysis. In addition, the hydroxyl of Ser308 rotates 120 degrees to a position where it is able to stabilize the carbanion intermediate formed by the methyl group of the acetyl-S-enzyme during its condensation with acetoacetyl-CoA.