Biochemical and functional characterisation of cyclic AMP-binding, universal stress proteins from mycobacteria
The genus Mycobacterium harbours several pathogenic species, including M. tuberculosis, the causative agent of tuberculosis, which alone is responsible for nearly 1.3 million deaths globally every year. Understanding the molecular basis of the pathogenesis is thus of prime importance. Similar to many other pathogenic organisms which successfully exploit cyclic AMP-mediated signalling as an effective virulence strategy by virtue of its ubiquitous presence across all kingdoms of life, several studies have implicated the same in M. tuberculosis, where cAMP is utilised as a potent toxin interfering and modulating host signalling inhibiting bacterial clearance. Mycobacterial genomes, including those from non-pathogenic species, encode a precipitously large number of adenylyl cyclases, enzymes that produce cAMP from ATP, and cAMP effector proteins, in concurrence with high intracellular and extracellular cAMP levels. Barring a few, the cellular functions of most of these proteins remain uncharacterised till date. The present study attempted to systematically measure cAMP levels in both pathogenic and non-pathogenic mycobacteria. Interestingly, it was observed that a significant fraction of intracellular cAMP remains protein bound in mycobacteria. Using a biochemical, as opposed to predictive approaches, abundantly expressed universal stress proteins (USP), Rv1636 and MSMEG_3811 in M. tuberculosis and M. smegmatis, respectively, were identified to bind cAMP, adding to the repertoire of cAMP-binding proteins present in mycobacteria. Intriguingly, Rv1636 and MSMEG_3811 do not harbour either cyclic nucleotide-binding (CNB) or GAF domains, the two domains hitherto known capable of binding cyclic nucleotides, but instead comprises a single USP domain. The biochemical and thermodynamic bases of cAMP interaction were elucidated. Based on the MSMEG_3811 crystal structure bound to cAMP, key residues important for cAMP binding were identified and tested using specific mutants. Further, signature sequence features imparting cAMP specificity to this subset of mycobacterial USPs were delineated. The cellular functions of Rv1636/MSMEG_3811 were next investigated. Besides the absence of any other associated domain(s), the high abundance of Rv1636/MSMEG_3811 and their high affinity for cAMP suggested a unique role for these USPs to act as a cellular ‘sink’ for cAMP. This hypothesis was tested by overexpressing Rv1636 as well by generating a deletion mutant of msmeg_3811 in M. smegmatis. In contrast to fast-growing non-pathogenic M. smegmatis, rv1636 was identified to be an essential gene in slow-growing pathogenic M. tuberculosis. Further, the gene essentiality was found to be dependent on the ability of Rv1636 to bind cAMP. The molecular basis for the essentiality of rv1636 in M. tuberculosis and most likely of orthologues in other slow-growers, in contrast to msmeg_3811 in M. smegmatis, was addressed. Rv1636 has been reported in several proteomics-based studies to be a secreted protein. The mechanism of secretion of Rv1636 was deciphered in M. tuberculosis, by a novel assay to monitor protein secretion in mycobacteria. In summary, the present study reports the identification and characterisation of mycobacterial universal stress proteins which bind cAMP with high specificity, describing how they may act as a cellular ‘sink’ for cAMP - a function essential for the survival of slow-growing pathogenic mycobacterial species. The absence of USPs in vertebrates opens up the possibility of targeting Rv1636 to develop effective strategies to inhibit the growth of slow-growing mycobacteria in future.