A structural investigation into the molecular basis of the hypoxic response in Mycobacterium tuberculosis

Abstract

Central to the pathogenic success of the tuberculosis-causing bacterium, Mycobacterium tuberculosis^ is its ability to persist within its host in a semi-dormant, asymptomatic or latent state for up to 40 years. It is thought that the low oxygen environment found inside the granulomas that form within the lungs of people infected with tuberculosis may stimulate the bacterium to convert into a non-replicating, persistent state. Several microarray analyses have identified M. tuberculosis genes of unknown function with expression significantly altered by hypoxia or exposure to nitric oxide. This project aimed to address the functions of some of these genes through stmctuml analysis. Twelve conserved hypothetical genes of unknown function from the hypoxically upregulated set were cloned and the proteins they encode were overexpressed in E. coli. Six of these proteins were soluble and amenable to further analysis. The crystal structure of one of these proteins, Rv2626c, truncated by 16 amino acids from the C-terminus, was determined by X-ray crystallography. A preliminary onedimensional NMR spectrum was collected from a second protein (Rvl738), and both this and a third protein (Rv0569) are potential targets for future structure solution by NMR. The truncated Rv2626c monomer structure consists of two Cystathionine-([beta]-synthase (CBS) domains and contains an unexpected intramolecular disulfide bond. The full length protein, but not the truncated version, dimerises in an SDS-resistant manner. Site-directed mutagenesis suggests that this is facilitated by the formation of an intermolecular disulfide bond by a Cys residue among the C-terminal residues removed in the truncated protein. This bond is predicted to be protected from the solvent by a-helices formed by the C-terminal tail. CBS domain pairs in human multi-domain proteins have been found to bind adenosine-containing ligands. Various in silico investigations in this project have revealed a potential binding pocket on the truncated Rv2626c structure, although in vitro fluorescent binding assays were unable to detect Rv2626c binding to AMP. It is possible that Rv2626c has a role in stabilising cell components in adverse conditions, since it is expressed at very high levels in response to hypoxic conditions, and is co-expressed with a chaperone. However, no evidence was found for chaperone-like activity using a thermal aggregation assay.