Synthetic analogues of marine bisindole alkaloids as potent selective inhibitors of MRSA pyruvate kinase

Globally, methicillin resistant Staphylococcus aureus (MRSA) has become increasingly difficult to manage in the clinic and new antibiotics are required. The structure activity relationship (SAR) study presented in this thesis forms part of an international collaborative effort to identify potent and selective inhibitors of an MRSA pyruvate kinase (PK) enzyme target. In earlier work the known marine natural product bromodeoxytopsentin (1.6), isolated from a South African marine sponge Topsentia pachastrelloides, exhibited selective and significant inhibition of MRSA PK (IC₅₀ 60 nM). Accordingly bromodeoxytopsentin provided the initial chemical scaffold around which our SAR study was developed. Following a comprehensive introduction, providing the necessary background to the research described in subsequent Chapters, this thesis has been divided into three major parts. Part one (Chapter 2) documents the synthesis of two natural imidazole containing topsentin analogues 1.40, 1.46, five new synthetic analogues 1.58—1.61, 2.104. In the process we developed a new method for the synthesis of topsentin derivatives via selenium dioxide mediated oxidation of N-Boc protected 3-acetylindoles to yield glyoxal intermediates which were subsequently cyclized and deprotected to yield the desired products. Interestingly we were able to demonstrate a delicate relationship between the relative equivalents of selenium dioxide and water used during the oxidation step, careful manipulation of which was required to prevent the uncontrolled formation of side products. Synthetic compounds 1.40, 1.46, 1.58—1.61 were found to be potent inhibitors of MRSA PK (IC₅₀ 238, 2.1, 23, 1.4, 6.3 and 3.2 nM respectively) with 1000-10000 fold selectivity for MRSA PK over four human orthologs. In the second part of this thesis (Chapter 3) we report the successful synthesis of a cohort of previously unknown thiazole containing bisindole topsentin analogues 1.62—1.68 via a Hantzsch thiazole synthesis. Bioassay results revealed that these compounds were only moderate inhibitors of MRSA PK (IC₅₀ 5.1—20 μM) which suggested that inhibitory activity was significantly reduced upon substitution of the central imidazole ring of topsentin type analogues with a thiazole type ring. In addition in Chapter 3 we describe unsuccessful attempts to regiospecifically synthesize oxazole and imidazole topsentin analogues through a similar Hantzsch method. As a consequence of our efforts in this regard we investigated three key reactions in depth, namely the synthesis of 2.2, 3.38, 3.40, 3.41 via α-bromination of 3-acetylindole and the synthesis of indolyl-3-carbonylnitriles 2.13, 3.45—3.47 and α-oxo-1H-indole-3-thioacetamides 3.48—3.51. The investigation of the latter led to the isolation and elucidation of two anomalous N,N-dimethyl-1H-indole-3-carboxamides 3.52 and 3.53. Finally the third part of this thesis (Chapter 4) deals with in silico assessment of the binding of both the imidazole and thiazole containing bisindole alkaloids to the MRSA PK protein which initially guided our SAR studies. In this chapter we reveal that there appears to be no correlation between in silico binding predictions and in vitro MRSA PK inhibitory bioassay data. Superficially it seems that binding energy as determined by the docking program used for these studies correlated with the size of the indole substituents and did not reflect IC₅₀ MRSA PK inhibitory data. Although this led us to computationally explore possible alternative binding sites no clear alternative has been identified