Structural and Inhibitory Studies of LL-Diaminopimelate Aminotransferase and Investigation of Methods for Small Peptide Crystallization

Abstract

The pyridoxal-5'-phosphate (PLP)-dependent enzyme LL-diaminopimelate aminotransferase (LL-DAP-AT) catalyzes a key step in the biosynthesis of L-lysine in plants and Chlamydia. In this thesis, studies of mechanistic and inhibitory aspects of LL-DAP-AT are described. Two LL-DAP-ATs from Arabidopsis thaliana and Chlamydia trachomatis were studied using X-ray crystallography. Furthermore, synthetic analogues of PLP-glutamate adduct 87, PLP-LL-diaminopimelate adduct 89, and pyridoxamine-5′-phosphate-tetrahydrodipicolinate (THDP) adduct 91 were co-crystallized with LL-DAP-AT. It was found that the hydrolysis product of THDP is the true substrate for this enzymatic transamination reaction. Crystallographic studies provide insights regarding the broader substrate specificity of Chlamydia LL-DAP-AT compared to the Arabidopsis enzyme. Since mammals require lysine from their diet, specific inhibitors of LL-DAP-AT could potentially serve as non-toxic antibiotics. A high-throughput screening study has been performed on a 29,201-compound library. N-(3-(Hydrazinecarbonyl)naphthalen-2-yl)benzenesulfonamide (122) was identified as the best hit (IC50 ~ 5 μM). In addition, three potential pharmacophores (derivatives of rhodanine, barbiturate, and thiobarbiturate) were identified. Structure-activity relationship (SAR) studies were conducted based on lead compound 122 and rhodanine derivatives. N-(5-Fluoro-2-(hydrazinecarbonyl)phenyl)benzenesulfonamide (155) was identified as a two-fold better inhibitor compared to the lead compound. It was found that a free hydrazide and a phenylsulfonamide were essential for inhibition. These two moieties needed to be attached to an aryl system, but further substitution of the aryl group was tolerated. These results may provide insight for future inhibitor design. In the second portion of this thesis, efforts toward development of methods for small peptide (< 5 kDa) crystallization are described. A co-crystallization approach was investigated, which involves linking a small peptide to a small molecule inhibitor to a readily crystallizable enzyme. Upon mixing, the small peptide may be co-crystallized with the enzyme. Subtilosin A (SubA, 191), a 35-amino acid circular bacteriocin, was selected for this co-crystallization study. Two bioconjugates, N-acetyl glucosamine (GlcNAc)-SubA and N,N′,N′′-triacetylchitotriose ((GlcNAc)3)-SubA, were prepared toward co-crystallization with lysozyme. A bioconjugate of arylsulfonamide-SubA was also synthesized for co-crystallization with carbonic anhydrase II (CAII). Current attempts towards co-crystallization were unsuccessful, likely due to the poor solubility of these bioconjugates. Efforts are underway towards co-crystallization by potentially improving the solubility of the bioconjugates.