Structure and mechanism of a key enzyme in M. tuberculosis cell envelope biogenesis

Mycobacterium tuberculosis (Mtb) is a pathogenic bacterium that causes tuberculosis (TB) and which is currently a major cause of concern for the World Health Organization, due to the huge numbers of humans infected worldwide (about one third of the world's population) and to the proliferation of types (strains) of the bacterium that are widely resistant to existing antibiotics. This is an often inevitable consequence of the overuse of antibiotics, and means that a continual new pipeline of therapeutic drugs must be produced. In the case of Mtb, targeting its complicated fatty acid metabolism pathways and cell wall structure have traditionally been very effective ways of disabling and killing the bacterium. Mtb has a dense outer layer (termed the envelope or wall) which contains a complex cross-linked mixture of carbohydrates and lipids that provide a formidable barrier around the bacterium and that help to protect it from the immune system, and to survive in the human body in the infective state. Understanding how the complicated envelope is constructed could hold the key to combating Mtb, as well as providing interesting new data on the novel biochemistry involved. In this respect, it was shown recently that two different types of drugs (benzothiazinones and dinitrobenzamides) both act on Mtb by inhibiting an enzyme system that is responsible for providing some of the key 'building blocks' for the cell envelope. This system comprises two proteins named DprE1 and DprE2, whose role is to convert a sugar-linked lipid from one conformation to another, to enable it to be used as an 'anchor' by which peculiar long chain fatty acids unique to the mycobacteria (mycolipids) can be attached to the core of the envelope. This becomes a major structural feature of the envelope that is critical for the bacteria to remain viable and to sustain an infective state. However, until recently the DprE1 and DprE2 proteins had proven almost impossible to produce in a soluble form that would be appropriate for studying their 3-dimensional structure and for interrogating their reaction mechanism. In advance of this application, we have overcome these problems by producing soluble DprE1/E2 enzymes cloned from a related bacterium (Mycobacterium smegmatis) and also by producing more stable, soluble forms of Mtb DprE1/E2 by fusing these proteins together at the genetic level. We are now able to make large amounts of the relevant proteins, and in this proposed programme of research we will exploit these breakthroughs to solve the structures of the DprE1/DprE2 proteins (using the technique of X-ray crystallography, where X-ray irradiation of crystals of the target protein produces a specific diffraction pattern that can pinpoint the locations of the component atoms and allow the structure to be built) and will also perform a careful characterization of their mechanisms and their individual roles in the generation of the key 'building blocks' for cell envelope assembly. In this way, we will provide important new information on an enzyme system crucial for the viability of the TB-causing bacterium, enabling further strategies to target this DprE1/E2 system with novel antibiotics.

Arabinogalactan is a critical component of the cell wall of Mycobacterium tuberculosis (Mtb) and other bacteria, where a major role is in the linking of the complex mycolic acids to peptidoglycan to provide a formidable protective barrier around the bacterium. The source of D-arabinose units in the furanose configuration is decaprenylphosphoryl-D-arabinose (DPA), which itself is generated from decaprenylphosphoryl-D-ribose (DPR) by an epimerase system comprising the products of the dprE1 (or Rv3790) and dprE2 (or Rv3791) genes, and which inverts the sugar 2'-hydroxyl group. Due to difficulties in isolating soluble and tractable proteins (particularly DprE1) by expressing the dprE1/2 genes in E. coli, little is known of the structure and function of these proteins, or of how they act synergistically in redox reactions to produce DPA. However, it is known that this is a target system for antibacterial drugs of the benzothiazinone and dinitrobenzamide class, which inactivate the DprE1 enzyme by covalent labelling. This proposal will exploit breakthroughs in expression of soluble forms of DprE1/2 from M. smegmatis, and success in obtaining a soluble Mtb DprE1/DprE2 fusion protein, in order to provide the first detailed account of structure and enzyme mechanism in this epimerase system (including the respective functions of the DprE1/2 components) and will provide much needed data on the mechanisms of action of an enzyme system critical for the viability of the Mtb bacterium. Through an integrated programme of (i) protein isolation, crystallization and structural determination; (ii) spectroscopic, thermodynamic and hydrodynamic analysis; and (iii) steady-state and stopped-flow kinetic experiments with natural substrate and substrate mimics, we will define respective roles of the DprE1/E2 proteins in the process, establish how they achieve these reactions from a structural/mechanistic basis, and thus define clearly how this important epimerization reaction occurs in vivo.

The programme proposed in this application follows from breakthroughs made in expression and purification of soluble forms of the mycobacterial decaprenylphosphoryl-D-ribose (DPR) epimerase system enzymes DprE1 and DprE2, which participate in conversion of DPR to decaprenylphosphoryl-D-arabinose (DPA), a precursor for arabinogalactan, which is required for attachment of mycolic acids to the peptidoglycan layer and is critical for cell envelope integrity. Aside from generating soluble M. smegmatis DprE1/E2 proteins, we have overcome problems in making a soluble form of the M. tuberculosis (Mtb) DprE1 by fusing it to DprE2, enabling solubilisation and stabilization of the pure fusion protein. The research programme proposed is inter- and multidisciplinary, involving structural, mechanistic and microbiological studies that will provide a clearer picture of the functions of these enzymes in Mtb. Beneficiaries include scientists in both industrial/academic sectors working on Mtb from perspectives of both (i) understanding its unusual lipid biochemistry and physiology, and (ii) developing therapeutic strategies. TB is a global problem and huge numbers of scientists worldwide are engaged in fundamental and applied research in the area. This broad research interest extends from Pharma with major programmes on therapeutics, through to Univ. labs exploring e.g. Mtb's biological chemistry, proteomics and genetics. The proposal features a combination of structural biology, enzymology and microbiologically orientated studies (including lipid substrate isolation). These approaches and their outcomes will be of relevance to academics interested in fundamental biochemistry and structure, and to applied researchers in Pharma and anti-infectives sectors where (particularly for Mtb) new strategies and data relevant to envelope biogenesis continue to be of interest due to leading existing therapeutics successfully targeting similar aspects of Mtb biochemistry. Our research will enhance understanding of mechanisms by which benzothiazinone (BTZ) antibiotics are modified to become potent inhibitors, and may also provide novel ideas (from structural and mechanistic studies) as to how they could be modified to increase potency, or for new scaffold development to target DprE1/E2. TB affects increasing numbers of the world population, and is an important disease of cattle, with badgers as a vector of bovine TB being a controversial area. TB research is of major interest worldwide, with TB endemic in third world countries and with several antibiotics redundant due to proliferation of drug-resistant and MDR Mtb strains. Research here addresses a crucial enzyme system for cell wall biogenesis and will be of great relevance to national/international agencies as well as influencing Pharma (who are already studying the BTZ and dinitrobenzamide [DNB] series) by providing key structural/mechanistic data to enable in silico studies of drug binding, and how this could be improved to generate more potent drugs. Timescale for impact are difficult to predict, but we would expect to have the DprE1 structure in 12-18 months, and the DprE2 structure in 24 months, leading on to structures of ligand-bound forms. We expect this work to lead to interest from relevant Pharma in this time window (particularly once publications emerge), and to inspire new approaches to targeting DprE1/E2. For instance, toxicity issues associated with nitro groups on existing drugs is an issue, and DprE1 mutations that remove the target Cys (Cys387) markedly diminish potency of BTZ (and DNB) drugs. Accordingly, we expect our structural work to inspire research to generate inhibitors that bind tightly to DprE1/DprE2 without relying on drug activation and presence of Cys387, which is not required for DprE1 function. We are also confident that the range of approaches (structural, kinetic, chemical, microbiological etc) will enhance the PDRA's skills to advance their career development.