PhoP contributes to maintenance of mycobacterial cAMP level.

(A) Intra-mycobacterial cAMP levels were determined by a fluorescence-based assay as described in the Methods, and compared for indicated mycobacterial strains, grown under normal or specific stress conditions. For acid stress, mycobacterial strains were initially grown to mid-log phase (OD600 0.4 to 0.6), and then it was transferred to acidic pH (7H9 media, pH 4.5) for further 2 hours of growth at 37°C. For NO stress, cells grown to mid-log phase were exposed to 0.5 mM DataNonoate for 40 minutes. The data represent average values from three biological repeats (*P≤0.05, **P≤0.01). (B) To compare secretion of cAMP by WT and phoPR-KO, cAMP levels were also determined in the corresponding culture filtrates (CF). (C) Immunoblotting analysis of 10 µg of cell lysates (CL) and 20 µg of culture filtrates (CF) of indicated M. tuberculosis strains. α-GroEL2 was used as a control to verify cytolysis of cells, CFP-10 detected as a secreted mycobacterial protein in the culture filtrates, and RpoB used as the loading control.

PhoP regulates expression of phosphodiesterase (PDE) rv0805.

(A) To investigate regulation of cAMP level, mRNA levels of well-characterized adenylate cyclases, and PDEs were compared in indicated mycobacterial strains by RT-qPCR as described in the Methods. The results show average values from biological triplicates, each with two technical repeats. Note that the difference in expression levels of rv0805 between WT and phoPR-KO was significant (p<0.01), whereas the fold difference in mRNA level between WT and the complemented mutant (Compl.) remains nonsignificant (not indicated). (B) To determine the effect of acidic pH conditions of growth, mycobacterial rv0805 expression was compared in WT grown under normal (pH 7.0) and acidic pH (pH 4.5). Average fold difference in mRNA levels from biological duplicates (each with a technical repeat) were measured as described in the Methods (**P≤0.01). As controls, expression of rv1339 and 16S rDNA were also measured. Non-significant difference is not indicated. (C) In vivo PhoP binding to rv0805 promoter (rv0805up) was compared in WT grown under normal and acidic conditions of growth using anti-PhoP antibody followed by ChIP-qPCR. Fold enrichment data represent mean values of two independent experiments with a statistically significant fold difference (**P-value <0.01; *P-value <0.05). The upstream regulatory regions of 16S rRNA (16S rRNAup) and msl3 (msl3up) were used as negative and positive controls, respectively. The assay conditions, sample analyses, and detection are described in the Methods section.

PhoP dependent repression of rv0805 to maintain mycobacterial cAMP level requires the presence of PhoR.

(A-B) To determine the impact of PhoR (the cognate sensor kinase of PhoP), expression of rv0805 was compared in indicated M. tuberculosis H37Rv strains: (A) phoPR-KO::phoP (phoPR mutant complemented with phoP) and (B) phoPR-KO::phoPR (phoPR mutant complemented with phoP-phoR) under normal and acidic conditions of growth. As expected, phoPR-KO::phoPR (Compl.) shows a significant repression of rv0805 (but not rv1339) under acidic pH (***P<0.001). However, rv0805 expression remains comparable in phoPR-KO::phoP under normal as well as acidic conditions of growth. As a control, rv1339 fails to show a variable expression in indicated mycobacterial strains. (C) To determine the effect of ectopic expression of rv0805 on intra-mycobacterial cAMP level, WT and mutant Rv0805 proteins (Rv0805M, defective for phosphodiesterase activity) were expressed in M. tuberculosis H37Rv (to construct WT-Rv0805, and WT-Rv0805M, respectively) as described in the Methods section. Similar to phoPR-KO, WT-Rv0805 (but not WT-Rv0805M) showed a considerably lower level of cAMP relative to WT bacteria. Significance in variation of cAMP levels were determined by paired student’s t-test (**P<0.01). (D-E) Relative expression of phoP and PDE in phoP-KD and rv0805-KD (phoP and rv0805 knock-down constructs, respectively). In keeping with elevated expression of rv0805 in phoPR-KO, phoP-KD shows an elevated expression of rv0805 relative to WT bacilli. In contrast, phoP expression level remains unaffected in rv0805-KD mutant. Panel E measured corresponding intra-bacterial cAMP levels in the respective knock-down mutants, as described in the legend to Fig. 1A.

Regulation of cAMP level and its effect on mycobacterial stress tolerance and survival in macrophages.

(A) To compare susceptibility to low pH conditions, indicated mycobacterial strains were grown at pH 4.5, and similar to phoPR-KO (grey circles), WT-Rv0805 (red circles) shows a significant growth defect relative to WT (empty circles). However, WT-Rv0805M (green circles) grows comparably well as that of the WT (empty circles). In contrast, at pH 7.0 all four mycobacterial strains (WT, empty triangles; phoPR-KO, grey triangles; WT-Rv0805, red triangles; WT-Rv0805M, green triangles) displayed comparable growth. (B) Microplate-based assay using Alamar Blue was utilized to examine mycobacterial sensitivity to increasing concentrations of diamide. In this assay, reduction of Alamar Blue correlates with the change of a non-fluorescent blue to a fluorescent pink appearance, which is directly proportional to bacterial growth. Survival of indicated mycobacterial strains, under normal conditions and in presence of 5 mM diamide, were determined by plotting fluorescence intensity. The data is normalized relative to WT grown in presence of 5 mM diamide. (C) To compare susceptibility to stress conditions, these mycobacterial strains were next grown in the presence of 50 µM Cumene Hydrogen Peroxide (CHP). In presence of CHP, WT-Rv0805 (red column) but not WT-Rv0805M (green column), shows a significant growth defect [relative to WT (empty column)] in striking similarity to phoPR-KO (grey column). Note that similar to phoPR-KO, WT-Rv0805 shows a comparably higher sensitivity to CHP relative to WT bacilli. However, WT-Rv0805M expressing a mutant Rv0805, shows a significantly lower sensitivity to CHP relative to WT-Rv0805, as measured by the corresponding CFU values. The growth experiments were performed in biological duplicates, each with two technical replicates (**P≤0.01; ***P≤0.001). (D) Murine macrophages were infected with indicated M. tuberculosis H37Rv strains. The cellular organelle was made visible by LysoTracker; mycobacterial strains were stained with phenolic auramine solution, and the confocal images display merge of two fluorescence signals (Lyso Tracker: red; H37Rv: green; scale bar: 10 µm). The insets in the merge panels indicate bacteria which either have inhibited or facilitated trafficking into lysosomes. White arrowheads in the merge panels indicate non-colocalization which remains higher in WT-H37Rv and WT-Rv0805M relative to phoPR-KO or WT-Rv0805. (E) Bacterial co-localization of M. tuberculosis H37Rv strains. The percentage of auramine labelled strains co-localized with Lysotracker was determined by counting at least 100 infected cells in 10 different fields. The results show the average values with standard deviation determined from three independent experiments (***P≤ 0.001). (F) Pearson’s correlation coefficient of images of internalized auramine-labelled mycobacteria and Lysotracker red marker in RAW 264.7 macrophages. Data are representative of mean ± S.D., derived from three independent experiments (*P<0.05; ***P<0.001).

Dysregulation of mycobacterial cAMP level impacts mycobacterial survival in vivo.

(A-B) Survival of mycobacterial strains in mice (A) lung, and (B) spleen after animals were given an aerosol infection with ∼100 CFU / lung. Mycobacterial load represents mean CFU values with standard deviations obtained from at least five animals per strains used (**P< 0.01; ***P<0.001; ****P<0.0001). (C) Histopathology of lung sections after 4 weeks of infection with indicated bacterial strains. Sections were stained with hematoxylin and eosin, observed under a light microscope, and images of the pathology sections collected at x40 magnification display granulomas (filled arrows) and alveolar space (empty arrows) (scale bar, 200 µm).

Increased cAMP level and effective stress response versus decreased cAMP level and reduced stress tolerance of mycobacteria.

In this model, upon activation by an appropriate signal via the cognate sensor PhoR, P∼PhoP binds to rv0805 regulatory region and functions as a specific repressor by preventing access for mycobacterial RNA polymerase (RNAP) to bind to the promoter and initiate transcription. In keeping with PhoP-dependent rv0805 repression, our results demonstrate a reproducibly lower level of cAMP in phoPR-KO relative to WT bacilli. Thus, phoPR-KO (or WT-Rv0805) remains ineffective to mount an appropriate stress response most likely due to its inability to coordinate regulated gene expression because of dysregulation of cAMP level, accounts for their reduced stress tolerance. Together, these molecular events suggest that failure to maintain cAMP level accounts for attenuated phenotype of the bacilli.