High-throughput assays for biotin protein ligase: a novel antibiotic target.

Abstract

Antibiotics are defined as chemical substances that inhibit or limit the growth of microorganisms. Since the second world war, antibiotics have been widely used to reduce the morbidity and mortality associated with serious bacterial infections caused by organisms such as Staphylococcus aureus. However, it has become increasingly difficult to treat bacterial infections due to the emergence of antibiotic resistant strains. The first clinical case of drug resistant bacteria was observed in S. aureus in 1947, just four years after the mass production of penicillin. Since then, resistance has been reported to every antibiotic ever employed. According to the Centres for Disease Control and Prevention of the United States, more than 70% of hospital-acquired infections show resistance to at least one commonly used antibiotic. Coupled with the paucity of therapeutic agents in the pipeline, there is now an urgent demand for new antibiotics. One of the strategies employed to combat drug resistant bacteria requires new chemical entities that work through novel drug targets for which there is no pre-existing resistance. This thesis focuses on the essential metabolic enzyme biotin protein ligase (BPL) as one such new drug target. BPL is the enzyme responsible for covalently attaching the cofactor biotin prosthetic group onto the biotin-dependent enzymes such as the carboxylases, decarboxylases and transcarboxylases. Enzymatic biotinylation proceeds via a two-step reaction whereby biotinyl-5'-AMP is synthesized from biotin and ATP before the biotin moiety is transferred onto the side chain of one specific lysine present in the active site of the biotin-dependent enzyme. One example of an important biotin-dependent enzyme is acetyl CoA carboxylase (ACC). ACC catalyzes the first committed step in fatty acid biosynthesis. Through genetic studies, it has been demonstrated that BPL activity is essential for bacterial survival. The aim for this project was to develop a convenient, high-throughput assay to measure BPL activity. This assay would permit 1) quantitative kinetic analysis of ligands and inhibitors and 2) screening of compound libraries for new BPL inhibitors. We propose that BPL inhibitors can be developed into new antibiotic agents. The novel BPL assay was developed employing fluorescence polarization (FP). FP is a light based technique which uses plane polarized light for the detection of tumbling motion of fluorescent molecules in solution. As polarization of the emitted light is relative to the apparent molecular mass of the fluorophore, this technique can be use for quantitation of changes in molecular mass of target molecules. This enabled 1) rapid kinetic analysis, 2) a minimal number of handling steps, 3) no washing steps and 4) automation by robotics. A first generation assay was developed for Escherichia coli BPL using peptide 85-11 that has been shown to be a convenient substrate. Following the BPL reaction, biotinylated peptides will form large molecular mass complexes with avidin. The amount of product could then be quantitated using FP. Here, kinetic analysis of MgATP (Km 0.25 ± 0.01 mM) and biotin (Km 1.45 ± 0.15 μM) binding produced results consistent with published data. We validated this assay with inhibition studies with end products of the BPL reaction, AMP and pyrophosphate, and a compound, biotinol-5'-AMP. Statistical analysis, performed upon both intraassay and interassay results (n = 30), showed the coefficient of variance to be <10% across all data sets. Furthermore, the Z' factors between 0.5 and 0.8 demonstrated the utility of this technology in high-throughput applications. However, the use of peptide 85-11, a substrate specific to E. coli BPL, does limit the application of this methodology to E. coli. In the second generation FP assay, I adapted this technology for S. aureus BPL by employing the biotin domain of S. aureus pyruvate carboxylase. Insertion of a fluorescein label was achieved by first engineering a cysteine residue into the domain by site directed mutagenesis then incubation with fluorescein-5'-maleimide. A series of mutants was created to investigate optimal positioning of the label into the substrate. Furthermore, the minimal size of the functional domain was determined. Our data showed that the placement of the fluorescein label is an important aspect of this project. Using this approach, I identified that a 90 amino acid domain with the label at position 1134 was optimal. Kinetic analysis of ligand binding showed SaBPL had a Km for biotin at 3.29 ± 0.37 μM and Km for MgATP at 66 ± 16.08 μM. This was in good agreement with data obtained from our previous assay measuring ³H-biotin incorporation. Inhibitor studies with pyrophosphate and analogues of biotin and biotinyl-5'-AMP further validated the assay. Various studies have shown cross-species biotinylation activities by a diverse range of BPLs. Therefore, using this methodology with a biotin domain as the substrate potentially provides a convenient assay for all BPLs.