Degree Name

Doctor of Philosophy


School of Chemistry


Bacterial resistance to multiple classes of antibiotics, termed multi-drug resistance (MDR) and resulting treatment failures is an increasing global concern problem. This thesis describes the design, synthesis and biological evaluation of new class of antibacterials which aim to address the problem of antibiotic resistance, including infections arising from Methicillin resistant Staphylococcus aureus (MRSA) and Clostridium difficile.

Drug efflux is a major mechanism used by bacteria to counter the effects of antibiotics and plays a key role in clinically relevant multidrug resistance (MDR). One strategy to overcome resistance is pharmacological inhibition of bacterial efflux pumps. Chapter 1 provides a review of small molecule-based approaches to countering efflux-mediated antibiotic resistance. The review summarises the many different classes of efflux pump inhibitors (EPIs) that have been described along with studies that report their use in combination with antibiotics. The review concludes with a summary of new approaches that link the EPI and antibacterial into a single hybrid molecule as a solution to the problem associated with co-administration of the separate agents. A promising approach of this type involves linking of the EPI INF55 (5-nitro-2-phenyl indole) with the antibacterial alkaloid berberine, as described in previous reports from the Bremner/Kelso groups.

One shortcoming of the berberine-INF55 hybrids was that their minimum inhibitory concentrations (MICs) against S. aureus species remained too high to be clinically useful (>1 μg/mL). In Chapter 2 attempts were made to increase the antimicrobial potency of these hybrids through the creation of action prodrugs. The first prodrug target 90 was designed to release INF55 from the prodrug after interaction of the berberine portion with bacterial DNA. However, the final step in the synthesis of 90 did not proceed and a second hybrid prodrug 96 was subsequently designed and successfully synthesised. Unfortunately the antibacterial potency of 96 was not significantly higher than previous (non-cleavable) hybrids from the Bremner/Kelso groups leading us to focus efforts in other directions.

The previous studies from the Bremner/Kelso groups had demonstrated the potential of the berberine-INF55 class of hybrid anti-bacterials but they hadn‘t established whether the compounds function as originally designed; i.e. with the INF55 moiety functioning to block MDR pumps (e.g. NorA) and the berberine moiety providing antibacterial action. In Chapter 3 a study was conducted which attempted to disconnect these activities and establish whether the proposed mechanism was indeed underpinning the activity of the hybrids. To explore the hypothesis, INF55 44 and two of its analogues which show reduced efflux pump inhibitory activity were designed and synthesized and three berberine-INF55 hybrids 88, 110 and 117 which link 44, 100, 101 to berberine 13-position were also synthesized. The aim of this work was to investigate whether the three hybrids show decreasing antibacterial potency that parallels the decreased efflux pump inhibitory activity of the INF55 analogues they incorporate, which would be consistent with the mechanism by which they were designed to act. The three hybrids 88, 110 and 117 were found to show identical antibacterial effects against S. aureus strains varying in levels of expression of the NorA efflux pump and they were also shown to be taken up in similar amounts in these cells. Additionally, the three hybrid show identical curative effects of C. elegans live infection models where the worms are infected with live MRSA or E. faecalis. These findings confirmed that while there are significant differences in the berberine potentiation potencies of the three attached pump inhibitors, these do not produce corresponding differences in antibacterial effects when incorporated into berberine–INF55 hybrids. It was concluded that the antibacterial mechanisms of action of the hybrids must be different from the mechanisms at play when berberine is co-administered with the pump inhibitors.

Clostridium difficile-associated diarrhoea (CDAD), also known as Clostridium difficile infection (CDI), is the leading cause of infectious nosocomial gastrointestinal illness. Section 2 of thesis investigated a new class of agents that are highly selective for Clostridium difficile over other gut commensals and show activity against stationary phase cells. High-throughput screening carried out at Northeastern University initially identified 4 hits (121-124) against C. difficile CD196 were essentially inactive (MIC > 50 μg/mL) across a gut commensal panel. In contrast, metronidazole 118, vancomycin 119 and fidaxomicin 120, the standard antibiotics used for CDAD, all showed high potency against C. difficile accompanied by significant activity against commensals. In Chapter 4 several structural analogs 140-150 were synthesised to answer specific structure-activity questions about the new class of compounds (i.e. diarylacylhydrazones). This SAR work demonstrated that only certain diarylacylhydrazones act as narrow-spectrum antibacterials with selectivity for C. difficile and C. perfringens over gut commensals. Identifying that CCCP 64 shows similar activity to diarylacylhydrazones against the gut panel and against stationary phase C. difficile cells also suggested that a protonophoric mechanism may play a role in the selectivity.

High-throughput screening of the NCI (set V) compound collection by Northeastern University identified NSC 293884 167 as another potent and selective lead compound against C. difficile (MIC 1.56 μg/mL). Two analogs of 167 (i.e. 168 and 169) were synthesized to establish whether the chemistry was amenable to analogue elaboration and to observe whether structure-activity relationships existed for the class, which would support further questions. Excitingly the 4-phenoxy analogue 169 showed very high potency against CD196 (MIC 0.78 μg/mL) and no activity against any other gut panel members (including C. perfringens) making it a truly C. difficile selective compound. Plans are being put in place to extensively explore this compound and the class.



Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.