Design and Synthesis of Novel Antibacterial Agents

Abstract

The latest data on antimicrobial resistance (AMR) related mortality is alarming. A conservative estimate places current global death due to AMR infections at 700,000 annually and projections anticipate that this could rise to 10 million by 2050. Over 50 major initiatives have emerged from the European Union (EU), United Kingdom (UK) and the United States of America (US) since 2007 aimed at improving surveillance, stewardship, and refreshing the drug development pipeline. The identification of priority pathogens, such as the ESKAPE (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter species) pathogens has also assisted in directing global attention and efforts towards developing agents against these dangerous microbes. However, despite the FDA approval of 17 new antimicrobial agents since 2013, no novel class of Gram negative active agents has been introduced. Concerning the antibacterials in the drug development pipeline which are currently in phase 3 development or near obtaining approval (fourteen), only one (cefiderocol, a cephalosporin derivative) displays broad activity against priority Gram negative pathogens whilst seven display very specific activity against the same – none of these are new classes. Fragment-based drug discovery (FBDD) is an efficient strategy to utilise in this research. FBDD involves virtual screening of exceptionally small ligands, termed fragments, against the crystal structure of the target protein. Fragments are known to sample chemical space more efficiently than conventional virtual screening ligand libraries. Fragment hits may have low affinity initially but demonstrate a good amount of free energy of binding per heavy atom, referred to as ligand efficiency. It is then necessary to optimise the best scoring fragments by introducing structural modifications to improve affinity and pharmacokinetic properties towards more drug-like leads. Chapter 2 of this thesis details the design of novel Extended Spectrum β-lactamase (ESBL) inhibitors through FBDD. This begins with a discussion regarding the benefits and rationale of virtual screening, followed by details of the software and methods used by the candidate to generate drug-like small molecules against protein crystal structures of the target carbapenemase, New Delhi Metallo-β-lactamase 1 (NDM-1). Careful and rational optimisation of high scoring fragments resulted in a structurally diverse set of lead compounds with anticipated ability to restore carbapenem activity in pathogenic bacterial which express NDM-1. Furthermore, we aimed to synthesise a structural extension of lead compounds to provide a prodrug form which we hypothesised could improve effectiveness of the lead compound in Gram negative bacteria. Previously published bacteria detection and identification research by the Groundwater research group targeted an aminopeptidase widely distributed in Gram negative bacteria for differentiating between Gram negative and Gram positive species from clinical isolates. This enzyme, L-alanyl aminopeptidase, cleaved chromogenic compounds containing a terminal L-alanine amino acid residue, which released the chromophore and produced colouration of the colony. Therefore, these ‘L-alanyl precursors’ allowed for distinction between Gram positive and Gram negative bacteria in clinical isolates. We proposed that L-alanyl derivatives of the lead compounds could produce the same effect, effectively targeting the antibacterial agents towards Gram negative species, which are of greater clinical urgency. The compounds which were successfully synthesised were tested by international collaborators and, separately, by the candidate against two clinically important Gram negative pathogens. Inhibition of the target carbapenemase was determined by observed improvement of carbapenem MIC (in this case meropenem) and, therefore, required testing of the synthesised inhibitors in tandem with the carbapenem, by simultaneous broth microdilution. All synthesised compounds displayed the ability to improve meropenem MIC by at least two-fold (that is, the MIC of meropenem halved in the presence of a test inhibitor) and demonstrated no growth inhibition in the absence of meropenem at the concentrations tested. Particularly noteworthy was the performance of L-proline-based compounds, derived from fragment 1, which displayed impressive activity. The thiol-containing L-proline derivative 8 was able to improve the MIC of meropenem by sixteen-fold (i.e. from 64 µg/mL to 4 µg/mL). The L-alanyl prodrug forms of the lead compounds also, generally, demonstrated greater potency than their respective parent lead compounds. In the case of one pair, the prodrug form was seven-times more effective than the corresponding lead compound. Chapter 3 of this thesis details the use of rational structure modification to generate improved antibacterials based on agents discovered by previous PhD candidates (Dr Liao and Dr Lin) within the research group. The target of this work was Filamenting temperature-sensitive mutant Z (FtsZ), an enzyme ubiquitous in bacteria which is critical for successful cell division and replication. FtsZ inhibitors are a validated antibacterial target, with one such inhibitor, TXA707 and its prodrug TXA709, currently in Phase 1 clinical trials. Research by Dr Liao and Dr Lin identified curcumin-based compounds 88 and 89 respectively which were successful antibacterial agents in Bacillus subtilis, a Gram positive species. The work described in Chapter 3 of this thesis aimed to introduce structural modifications to these compounds to produce less chemically reactive derivatives which retained, or, ideally, improved antibacterial activity. This resulted in the phenolic derivatives which demonstrated comparable activity against B. subtilis. However, these compounds demonstrated no significant activity against a Gram negative species of bacteria, Escherichia coli. Consequently, an L-alanyl prodrug form of the lead phenol 94 was vehemently pursued. In summary, this thesis details the exploration of a diverse range of synthetic reactions which produced structurally diverse potential antibacterial agents. Many of these compounds exhibited encouraging antimicrobial activity, either by inhibiting a resistance-conferring enzyme (NDM-1) or by inhibiting an enzyme critical for bacterial division (FtsZ).