Cyclic AMP In Mycobacteria Adenylyl Cyclases And Cyclic AMP Receptor Proteins
Abstract
The discovery of cyclic AMP (cAMP), nearly 50 years ago by Sutherland radically altered the appreciation of metabolic regulation. Since then the presence of cAMP and its tremendous physiological impact has been demonstrated in many prokaryotic systems. In fact, virulence mechanisms of several pathogens known today exploit cAMP dependent pathways. Interestingly the genome of Mycobacterium tuberculosis H37Rv, the causative agent of tuberculosis, encodes as many as 16 adenylyl cyclases (enzymes that convert ATP to 3’, 5’-cAMP) and 10 cyclic-nucleotide binding proteins. Recent reports show that bacterial-derived cAMP manipulates host signaling for bacterial survival, suggesting an important role for cAMP in the pathogenesis of M. tuberculosis. A large number of non-pathogenic species of mycobacteria also share and conserve several of these cAMP metabolism genes, suggesting that cAMP is not only important for pathogenesis but also may play a critical physiological role in the genus. The work carried out in this thesis aims at a better understanding of the role of cAMP by studying the adenylyl cyclases and cyclic AMP receptor proteins (CRPs) from Mycobacterium smegmatis, a non-pathogenic member of the genus.
Intracellular cAMP levels in a cell are precisely maintained by modulating the activities of the adenylyl cyclases (cAMP synthesising enzymes), the phosphodiesterases (cAMP hydrolysing enzymes) and the secretion machinery, if any. To assess the role of cAMP in mycobacteria, cAMP levels were measured in M. smegmatis during growth and under various stress conditions. The results show that cAMP levels peak at log phase of growth and decline thereafter. Under acidic conditions or on perturbing the cell-wall, cellular cAMP levels are altered, which indicate a possible role of cAMP in stress adaptation.
Earlier work in our laboratory has led to the identification of multiple adenylyl cyclases in the mycobacterial genomes. These cyclases are similar in sequence to the mammalian enzymes and several of them have been shown to be active in vitro displaying a diverse range of biochemical properties. The M. smegmatis genome encodes 10 adenylyl cyclase-like genes. In order to understand the role of cAMP in M. smegmatis, individual cyclases were analysed for their biochemical properties and physiological functions. The work presented in this thesis is concerned with the functional characterization of MSMEG_3578 and MSMEG_3780, two of the several adenylyl cyclases from M. smegmatis.
MSMEG_3578 encodes for a protein that comprises two transmembrane domains, an extracellular receptor-like domain, a membrane anchoring HAMP domain and an intracellular cyclase domain. The cyclase domain is closely related to mammalian cyclases but lacks the canonical residues that are critical for the catalysis of class III cyclases. Interestingly, the stop codon of this gene overlaps with the start codon of the downstream gene, MSMEG_3579 (a putative cyclic nucleotide gated mechanosensitive ion channel), suggesting a functional link between the two genes. The conservation of this gene pair across the mycobacterial genus indicates the importance of this putative receptor-effector pair in the physiology of mycobacteria. Additionally, microarray analysis by various groups have shown that this gene pair in Mycobacterium tuberculosis is differentially regulated in conditions that mimic stress the bacteria may experience during infection. In order to ascertain the physiological role of MSMEG_3578, a knock-out M. smegmatis strain was generated and tested for growth and cAMP defects. The knock-out strain showed growth and cAMP profiles similar to the wild-type strain. Over-expression of MSMEG_3578 in M. smegmatis resulted in a significant rise in cAMP levels. Interestingly, over-expression of the MSMEG_3578 adenylyl cyclase in E. coli did not lead to an elevation in cAMP levels indicating that other mycobacterial proteins may be required for the activity of MSMEG_3578 in vivo. In agreement with this, the purified adenylyl cyclase domain of MSMEG_3578 was found to be biochemically inactive in vitro. Additionally, the over-expressing strain has altered colony morphology as compared to the wild type strain. Perturbation of cAMP levels by over-expression of other cyclases also leads to a similar colony morphology phenotype, indicating the phenotype to be controlled by cAMP in general rather than by a specific cyclase in the cell.
MSMEG_3780 is a highly conserved, biochemically active adenylyl cyclase, speculated to play a house-keeping function in M. smegmatis. Its orthologs from M. tuberculosis (Rv1647) and M. leprae (ML1399) have been biochemically characterized earlier in our laboratory. To unravel the role of this gene in vivo, the ∆MSMEG_3780 strain was tested for growth and cAMP defects under various conditions. The deletion strain did not show any difference in growth rate or morphology when compared to the wild-type strain. However it showed a reduction in intracellular cAMP levels at the log-phase of growth. Reintroduction of the MSMEG_3780 gene in the deletion strain restored cAMP to wild-type levels, thus indicating a crucial role for this adenylyl cyclase in the maintenance of intracellular cAMP levels during logarithmic growth. In order to investigate the regulation of the MSMEG_3780 gene, its promoter activity was tested under various stress-conditions. Acid-stress specifically resulted in the repression of the MSMEG_3780 promoter activity, a condition which also leads to a decrease in cAMP levels in the cells. Further evidence for the susceptibility of the MSMEG_3780 gene to acid-stress was obtained when the ∆MSMEG_3780 strain failed to reduce intracellular cAMP content upon sustained acid-stress conditions. Since Rv1647 shares similar biochemical properties with MSMEG_3780 and can also heterodimerize with the MSMEG_3780 protein in vitro, it was interesting to test whether the M. tuberculosis ortholog could functionally complement MSMEG_3780. To assess this, a complement strain was generated that contained the Rv1647 gene under the control of MSMEG_3780 promoter, integrated into the genome of ∆MSMEG_3780 strain. Rv1647 protein was able to restore the cAMP phenotype seen on acid stress as well as the cAMP levels in the mutant strain at the log phase of growth. This study indicated the role of cAMP and MSMEG_3780 in acid adaptation and also suggested a non-redundancy of adenylyl cyclases in mycobacteria, where different individual cyclases may have unique functions in the cells. Since Rv1647 could complement the cAMP defective phenotype in ∆MSMEG_3780, this suggests functional parallels between the proteins from the two species.
Bacterial adaptation to environmental stress is brought about by a rapid change in its gene expression profile. Cyclic AMP plays an important role by binding to and activating a transcriptional factor, cAMP receptor protein or CRP. We have identified two CRPs from M. smegmatis, viz., MSMEG_0539 and MSMEG_6189 that possess high similarity at the amino acid level (78% overall sequence identity). The CRP ortholog from M. tuberculosis, Rv3676 has been characterized structurally, biochemically and functionally earlier. Western blot and RT-PCR analyses showed that CRPs in M. smegmatis were present during all phases of growth. Both the CRPs were cloned,
expressed and shown to bind cAMP. Since the DNA binding domains of Rv3676 and the two M. smegmatis CRPs are nearly identical, the CRPs from M. smegmatis were predicted to bind similar target sequences. Interestingly, a CRP site was identified in the promoter of the MSMEG_3780 gene, suggesting a possible feed-forward or feed-back loop, where the enzymatic product of the adenylyl cyclase now governs its own gene expression. We performed Electrophoretic Mobility Gel Shift Assays (EMSAs) with M. smegmatis lysates to show that CRP binds to the MSMEG_3780 promoter at the CRP site. Subsequent Chromatin Immunoprecipitation (ChIP) assays confirmed that CRP binding to the MSMEG_3780 promoter occurred in vivo. To investigate the role of CRP in the regulation of the MSMEG_3780 gene, luciferase reporter assays with the wild-type and CRP site mutant promoters were carried out. Results suggest that CRP regulates the MSMEG_3780 gene under osmotic stress. However, CRP did not play any role in basal expression of MSMEG_3780 during growth. To assess which CRP of the two is functionally linked to the MSMEG_3780 promoter, we carried out a footprint assay with purified CRPs. It was intriguing to note that both the CRPs were in fact able to bind the promoter albeit under different conditions. Whereas MSMEG_6189 bound the promoter both in the presence and absence of cAMP, MSMEG_0539 bound the promoter only in the presence of cAMP. MSMEG_6189 thus deviates from the accepted CRP paradigm that seeks an absolute requirement of cAMP for specific DNA binding.
The present study identifies cAMP as an important stress signal in M. smegmatis. Using MSMEG_3780 as a model gene, the role of cAMP in mycobacteria was studied. The two divergent CRPs that were characterized may function and dictate cAMP-mediated and perhaps independent functions in cells. Taken together, our results provide compelling evidence for the important role of cAMP in the general physiology and stress adaptation in M. smegmatis.