Allosteric inhibitors of dihydrodipicolinate synthase

Abstract

Dihydrodipicolinate synthase (DHDPS) is an enzyme which catalyzes the first step of the lysine biosynthesis pathway in bacteria and plants. Deletion of the gene encoding DHDPS results in non-viable bacteria, therefore DHDPS is considered a validated drug target. The enzyme is feedback-regulated by lysine, and structural studies have shown that the tetrameric enzyme contains two allosteric sites, each of which bind two lysine molecules. The Palmer laboratory has previously developed a potent inhibitor "bislysine" that mimics the structure of two bound lysine molecules. Previous work showed that S-aminoethylcysteine ("thialysine") was a much poorer inhibitor than lysine, despite the structural similarity of the two compounds. This thesis describes the synthesis of new allosteric inhibitors of DHDPS, with the goal of defining their structural and chemical properties, such as inhibitor side chain length and pKa, that lead to inhibition. Racemic analogs of lysine were generated using the amidomalonic ester synthesis. Analogs of bislysine were generated from dimethyl 2,5-bis([(tertbutoxy)carbonyl]amino) hexanedioate by treatment with lithium diisopropylamide followed by alkylation using various electrophiles. This alkylation step hampers the overall process because it proceeds in low yield (typically near 10%). Studies were undertaken in an attempt to understand the factors influencing this reaction; however, variations in the reaction times, solvent composition, and additives did not improve the yield appreciably. All the inhibitors were tested using the established DHDPS-DHDPR coupled assay to estimate the IC50 values. The lysine analogue (±)-(E)-2,6-diaminohex-4-enoic acid, which has a double bond in the side chain as the only modification, showed weaker inhibition (IC50 = 3.7 mM) compared to racemic lysine (IC50 = 0.2 mM). The altered pKa of the ε-amino group, which makes a hydrogen bond with His59 when bound to the allosteric site, is proposed to account for the loss of activity. Triazolylmethylglycine, which is predicted to have a pKa value closer to lysine, but contains a shorter side chain, was an even weaker inhibitor. Bis-amino acid versions of these compounds were much stronger inhibitors. A bis-analog of para-aminobenzylglycine showed weak inhibition as well, suggesting this bulkier compound, with a much lower side chain pKa, could still bind to the allosteric cavity.