**2. Participation of cyclic adenosine 3′,5′-monophosphate (cAMP) in mycobacterial gene regulation**

Cyclic adenosine 3′,5′-monophosphate (cAMP) is one of the most important second messengers used in bacteria and it has been characterized principally in *Escherichia coli*. However its signaling role in *M. tuberculosis* is beginning to emerge. A large number of AC genes are present in *Mycobacterium tuberculosis* in comparison with other microorganims (Cha *et al.*, 2010, Klengel *et al.*, 2005, Shenoy et al., 2004), and also that *Mycobacterium* just have one class III cNMP phosphodiesterase (Rv0805) that has been identified in *M. tuberculosis* (McCue *et al.*, 2000, Shenoy et al., 2007) with a modest ability to efficiently hydrolyse 3′,5′-cAMP and that responds to H2O2 *in vitro* (Bai et al., 2011, Barba et al., 2010). cAMP may serve as both an extra and intracellular signaling molecule in mycobacteria (Agarwal, 2009). cAMP levels are modulated by stress conditions in *M. smegmatis*, where it has been found in concentrations between 10 µM and 1 mM, suggesting that cAMP participates in signalling events within the bacterial cell, mediating its action by downstream effectors (Dass *et al.*, 2008).

It appears that Mycobacteria need to maintain a steady level of cytoplasmatic cAMP in many conditions. For example, it has been shown that cytoplasmic cAMP levels in *E. coli* are reduced three- to fourfold when the carbon source is ~0.2% glucose rather than glycerol (Bai et al., 2011). In contrast, a recent study showed no significant change in the cytoplasmic cAMP levels of *M. bovis* BCG incubated with 0.2% glucose (Bai *et al.*, 2009), or carbonstarved bacteria (Dass et al., 2008). cAMP levels decrease in both fast- and slow-growing mycobacteria in response to very high levels of glucose (2%) (Bai et al., 2011). cAMP levels have been suggested to be high in *Mycobacterium* cells, exceeding up to 100-fold levels found in other bacteria (Nambi et al., 2010, Stapleton *et al.*, 2010, Shenoy & Visweswariah, 2006b, Rickman *et al.*, 2005). However, it is difficult to make a comparison between studies, due to cAMP variations presents in the conditions tested, as well as the different normalization and reporting methods used.

### **2.1 CRP and cNMP binding proteins**

*In silico* studies predict 10 cNMP binding proteins that encompass a wide range of potential effector functions, suggesting a more complex role for cAMP signalling, probably important during host infection. From these 10 cNMP binding proteins, 7 (Rv0073, Rv0104, Rv2434c, Rv2564, Rv2565, Rv3239c and Rv3728) contain an assortment of putative functional domains, including those associated with transport functions and esterase activities (McCue et al., 2000, Shenoy & Visweswariah, 2006a). Only three cNMP binding proteins have been functionally characterized to date (Bai et al., 2011). Two of these proteins, referred to as CRP (*Rv3676*, for Catabolite Represor Protein) and Cmr (*Rv1675c*, for cAMP and macrophage regulator), contain a HTH DNA binding domains, and belong to the CRP-FNR family of transcription factors (McCue et al., 2000). The third protein from this group, is encoded by

Mycobacterium Tuberculosis Signaling via c-AMP 109

the importance of Cmr for the cAMP- dependent regulation of these genes, Cmr has not been shown to directly bind cAMP *in vitro*, and cAMP did not affect Cmr's binding to any of their promoter sequences (Gazdik et al., 2009). The mechanism by which Cmr responds to cAMP levels has yet to be discovered, and it is possible that a second factor plays a

The cAMP binding transcriptional regulator encoded by *Rv3676* in *M. tuberculosis* has been strongly associated with growth during mycobacterial life cycle. Its deletion resulted in impaired growth in macrophage cell lines and in mice (Hunt et al., 2008, Rickman et al., 2005). Using an exponential enrichment (SELEX) approach, Bai et al. defined the CRPMt´s palindromic binding motif (C/TGTGANNNNNNT CACG/A) based on 58 predicted binding sites from the *M. tuberculosis* genome, using a combination of *E. coli* CRP binding sites and *M. tuberculosis* DNA sequences recovered by affinity capture using CRPMt to seed the computational analyses (Bai *et al.*, 2005). Akhter et al. used the positional Shannon relative entropy method to predict 19 new putative binding sites for *M. tuberculosis* (Rv3676) CRP, in addition to the 73 sites previously predicted by Bai et al. (Akhter *et al.*, 2008, Bai et al., 2005). These additional sites resulted from a difference where Akhter et al. used only the information available from the *M. tuberculosis* CRP-regulon instead of adding up the one available from the *E.coli* CRP-regulon. According to Akhter et al., the *M. tuberculosis* CRPregulon comprises genes required for critical functions like: (i) cell-wall biogenesis, (ii) central metabolism pathways, (iii) putative regulatory elements controlling cAMP signaling,

Experimental validation of these predictions has only been completed for Bai et al. model (Bai et al., 2005). Their research showed that mutation of nucleotides G2 or C15 from the palindromic binding motif, abolished CRPMt´s binding. Both positions are conserved in all predicted binding sites. They also evaluated CRPMt and CRPBCG binding to seven putative CRPMt sites, located within intergenic regions, where 6 out of these 7 were found to be functional *in vitro* and *in vivo* (Bai et al., 2007). This increases confidence in the prediction algorithm that was used to identify CRPMt binding sites, although at this point it is possible to suggest that CRPMt regulon requires further refinement. On the other hand, it has been shown that *rpfA* is directly activated by CRPMt (Rickman et al., 2005). The resuscitationpromoting factor (Rpf) is a growth factor that stimulates the growth of aged *M. tuberculosis*  cultures, and members of this family are thought to play a role in reactivation of dormant *M. tuberculosis* (Mukamolova *et al.*, 2002). Regulation of *rpfA* by CRPMt suggests that CRPMt plays a role in persistence and/or reactivation of tuberculosis, but this is only one of many

Expression of *M. tuberculosis whiB1*, a member of the Wbl (WhiB-like) family, is controlled positively and negatively by CRPMt (Stapleton et al., 2010; Agarwal et al., 2006). A CRPMt binding site (CRPMt1) was detected in *whiB1*'s upstream regulatory region (Smith *et al.*, 2010) (Rickman et al., 2005). Reporter assays with native and mutated promoter sequences indicated that transcription from the native, but not the mutant, promoter was affected by cAMP levels via the direct binding with CRPMt (Agarwal *et al.*, 2006). Other evidences indicate that a second CRPMt binding site (CRPMt2) in the *whiB1* promoter altered CRPMt's

and (iv) antibiotic resistance (Akhter et al., 2008) (Table 2).

biological functions that may be regulated by CRPMt in *M. tuberculosis*.

facilitating role.

**2.2 CRPMt regulon** 

*Rv0998* in *M. tuberculosis*, and regulates lysine acetylation in mycobacterial proteins in a cAMP-responsive manner (Nambi et al., 2010). The biological effects of this acetylation are not yet defined.

CRP (cAMP receptor protein) in *Escherichia coli*, is one of the best-studied prokaryotic transcription factor. Currently, a total of 378 target promoters on the *Escherichia coli* genome are proposed to be under the control of cAMP-bound CRP, using a SELEX approach (Shimada *et al.*, 2011). Some of the CRP regulon genes include those encoding the transporters and the catabolic enzymes of glucose (Perrenoud & Sauer, 2005) and nonglucose sugars (Wickstrum *et al.*, 2010), virulence genes (Espert *et al.*, 2011), motility genes (Hollands *et al.*, 2010), GMP synthesis (Husnain *et al.*, 2009), anaerobic growth and nitrate reductases genes (Stewart *et al.*, 2009). To date, in *E. coli*, 2 differents ways in which CRP leads the sigma 70 promoter trancription have been described: in the Class I promoters, CRP binds upstream of the promoter −35 element, at a site centered at position −61.5 or further upstream, and an activating region (AR1) in the downstream subunit of the CRP dimer makes contact with the C-terminal domain of one of the two RNA polymerase α subunits (αCTD). In class II promoters, CRP binds at a target that overlaps the promoter −35 element and is usually centered at position −41.5. AR1 in the upstream subunit of the CRP dimer interacts with αCTD, while a second activating region (AR2) in the downstream subunit interacts with the N-terminal domain of one of the two RNA polymerase α subunits (αNTD)(Hollands et al., 2010).

In a similar manner to the CRP protein from *E. coli*, the *M. tuberculosis* CRP (Rickman *et al.*, 2005) is the best-studied example of a protein implicated in cAMP-mediated signaling. In *M. tuberculosis* CRP is encoded by the gene *Rv3676* and it is homologous to *E. coli* CRP. In *E. coli*, it has been shown that CRP regulates expression of genes required to control metabolism, as well as growth under hypoxic and nutrient-deprived conditions. Like *E. coli* CRP, *M. tuberculosis* Rv3676 possesses an N-terminal cAMP-binding domain and a C-terminal DNAbinding domain. The crystal structures of *M. tuberculosis* CRP at 2.2 and 2.0 A resolution of cAMP-bound (Reddy *et al.*, 2009) and the apo-form (Gallagher *et al.*, 2009), have been reported. Conformational changes required for DNA-binding do not take place in the absence of the second messenger (Reddy et al., 2009). In fact, as opposed to *E. coli* CRP where cAMP binding follows a cooperative mechanism, c-AMP binding sites are independent, and DNA-binding activity is not as enhanced with *M. tuberculosis* Rv3676 (Stapleton *et al.*, 2009). The CRP orthologous in *M. bovis* BCG (CRPBCG) is a fully functional transcription factor, since CRPBCG overcame the virulence deficiency of an *M. tuberculosis crp*  mutant (Hunt *et al.*, 2008). Although both of them have a similar ability to bind cAMP and DNA, CRPBCG's DNA binding affinity is approximately twice that of *Mycobacterium tuberculosis* CRP (CRPMt) (Bai *et al.*, 2007, Hunt et al., 2008). Interestingly, CRPBCG differs from CRPMt in just two amino acid residues (L47P and E178K).

The other predicted cAMP-dependent transcriptional regulator, Cmr, was found to negatively regulate the expression of five proteins (GroEL2, Rv2971, PE\_PGRS6a, Mdh and Rv1265) (Gazdik & McDonough, 2005). The upstream regions of three of these genes (*mdh*, *groEL2* and *Rv1265*) bound specifically to Cmr in electrophoretic mobility shift assays, consistent with direct regulation of these genes by Cmr. Expression of three of these genes was found to be regulated within macrophages, and this regulation was mediated by Cmr in both *M. tuberculosis* and *M. bovis* BCG (Gazdik et al., 2009). Despite the importance of Cmr for the cAMP- dependent regulation of these genes, Cmr has not been shown to directly bind cAMP *in vitro*, and cAMP did not affect Cmr's binding to any of their promoter sequences (Gazdik et al., 2009). The mechanism by which Cmr responds to cAMP levels has yet to be discovered, and it is possible that a second factor plays a facilitating role.

### **2.2 CRPMt regulon**

108 Understanding Tuberculosis – Deciphering the Secret Life of the Bacilli

*Rv0998* in *M. tuberculosis*, and regulates lysine acetylation in mycobacterial proteins in a cAMP-responsive manner (Nambi et al., 2010). The biological effects of this acetylation are

CRP (cAMP receptor protein) in *Escherichia coli*, is one of the best-studied prokaryotic transcription factor. Currently, a total of 378 target promoters on the *Escherichia coli* genome are proposed to be under the control of cAMP-bound CRP, using a SELEX approach (Shimada *et al.*, 2011). Some of the CRP regulon genes include those encoding the transporters and the catabolic enzymes of glucose (Perrenoud & Sauer, 2005) and nonglucose sugars (Wickstrum *et al.*, 2010), virulence genes (Espert *et al.*, 2011), motility genes (Hollands *et al.*, 2010), GMP synthesis (Husnain *et al.*, 2009), anaerobic growth and nitrate reductases genes (Stewart *et al.*, 2009). To date, in *E. coli*, 2 differents ways in which CRP leads the sigma 70 promoter trancription have been described: in the Class I promoters, CRP binds upstream of the promoter −35 element, at a site centered at position −61.5 or further upstream, and an activating region (AR1) in the downstream subunit of the CRP dimer makes contact with the C-terminal domain of one of the two RNA polymerase α subunits (αCTD). In class II promoters, CRP binds at a target that overlaps the promoter −35 element and is usually centered at position −41.5. AR1 in the upstream subunit of the CRP dimer interacts with αCTD, while a second activating region (AR2) in the downstream subunit interacts with the N-terminal domain of one of the two RNA polymerase α subunits

In a similar manner to the CRP protein from *E. coli*, the *M. tuberculosis* CRP (Rickman *et al.*, 2005) is the best-studied example of a protein implicated in cAMP-mediated signaling. In *M. tuberculosis* CRP is encoded by the gene *Rv3676* and it is homologous to *E. coli* CRP. In *E. coli*, it has been shown that CRP regulates expression of genes required to control metabolism, as well as growth under hypoxic and nutrient-deprived conditions. Like *E. coli* CRP, *M. tuberculosis* Rv3676 possesses an N-terminal cAMP-binding domain and a C-terminal DNAbinding domain. The crystal structures of *M. tuberculosis* CRP at 2.2 and 2.0 A resolution of cAMP-bound (Reddy *et al.*, 2009) and the apo-form (Gallagher *et al.*, 2009), have been reported. Conformational changes required for DNA-binding do not take place in the absence of the second messenger (Reddy et al., 2009). In fact, as opposed to *E. coli* CRP where cAMP binding follows a cooperative mechanism, c-AMP binding sites are independent, and DNA-binding activity is not as enhanced with *M. tuberculosis* Rv3676 (Stapleton *et al.*, 2009). The CRP orthologous in *M. bovis* BCG (CRPBCG) is a fully functional transcription factor, since CRPBCG overcame the virulence deficiency of an *M. tuberculosis crp*  mutant (Hunt *et al.*, 2008). Although both of them have a similar ability to bind cAMP and DNA, CRPBCG's DNA binding affinity is approximately twice that of *Mycobacterium tuberculosis* CRP (CRPMt) (Bai *et al.*, 2007, Hunt et al., 2008). Interestingly, CRPBCG differs

The other predicted cAMP-dependent transcriptional regulator, Cmr, was found to negatively regulate the expression of five proteins (GroEL2, Rv2971, PE\_PGRS6a, Mdh and Rv1265) (Gazdik & McDonough, 2005). The upstream regions of three of these genes (*mdh*, *groEL2* and *Rv1265*) bound specifically to Cmr in electrophoretic mobility shift assays, consistent with direct regulation of these genes by Cmr. Expression of three of these genes was found to be regulated within macrophages, and this regulation was mediated by Cmr in both *M. tuberculosis* and *M. bovis* BCG (Gazdik et al., 2009). Despite

from CRPMt in just two amino acid residues (L47P and E178K).

not yet defined.

(αNTD)(Hollands et al., 2010).

The cAMP binding transcriptional regulator encoded by *Rv3676* in *M. tuberculosis* has been strongly associated with growth during mycobacterial life cycle. Its deletion resulted in impaired growth in macrophage cell lines and in mice (Hunt et al., 2008, Rickman et al., 2005). Using an exponential enrichment (SELEX) approach, Bai et al. defined the CRPMt´s palindromic binding motif (C/TGTGANNNNNNT CACG/A) based on 58 predicted binding sites from the *M. tuberculosis* genome, using a combination of *E. coli* CRP binding sites and *M. tuberculosis* DNA sequences recovered by affinity capture using CRPMt to seed the computational analyses (Bai *et al.*, 2005). Akhter et al. used the positional Shannon relative entropy method to predict 19 new putative binding sites for *M. tuberculosis* (Rv3676) CRP, in addition to the 73 sites previously predicted by Bai et al. (Akhter *et al.*, 2008, Bai et al., 2005). These additional sites resulted from a difference where Akhter et al. used only the information available from the *M. tuberculosis* CRP-regulon instead of adding up the one available from the *E.coli* CRP-regulon. According to Akhter et al., the *M. tuberculosis* CRPregulon comprises genes required for critical functions like: (i) cell-wall biogenesis, (ii) central metabolism pathways, (iii) putative regulatory elements controlling cAMP signaling, and (iv) antibiotic resistance (Akhter et al., 2008) (Table 2).

Experimental validation of these predictions has only been completed for Bai et al. model (Bai et al., 2005). Their research showed that mutation of nucleotides G2 or C15 from the palindromic binding motif, abolished CRPMt´s binding. Both positions are conserved in all predicted binding sites. They also evaluated CRPMt and CRPBCG binding to seven putative CRPMt sites, located within intergenic regions, where 6 out of these 7 were found to be functional *in vitro* and *in vivo* (Bai et al., 2007). This increases confidence in the prediction algorithm that was used to identify CRPMt binding sites, although at this point it is possible to suggest that CRPMt regulon requires further refinement. On the other hand, it has been shown that *rpfA* is directly activated by CRPMt (Rickman et al., 2005). The resuscitationpromoting factor (Rpf) is a growth factor that stimulates the growth of aged *M. tuberculosis*  cultures, and members of this family are thought to play a role in reactivation of dormant *M. tuberculosis* (Mukamolova *et al.*, 2002). Regulation of *rpfA* by CRPMt suggests that CRPMt plays a role in persistence and/or reactivation of tuberculosis, but this is only one of many biological functions that may be regulated by CRPMt in *M. tuberculosis*.

Expression of *M. tuberculosis whiB1*, a member of the Wbl (WhiB-like) family, is controlled positively and negatively by CRPMt (Stapleton et al., 2010; Agarwal et al., 2006). A CRPMt binding site (CRPMt1) was detected in *whiB1*'s upstream regulatory region (Smith *et al.*, 2010) (Rickman et al., 2005). Reporter assays with native and mutated promoter sequences indicated that transcription from the native, but not the mutant, promoter was affected by cAMP levels via the direct binding with CRPMt (Agarwal *et al.*, 2006). Other evidences indicate that a second CRPMt binding site (CRPMt2) in the *whiB1* promoter altered CRPMt's

Mycobacterium Tuberculosis Signaling via c-AMP 111

On the other hand, out of 92 genes found and predicted by bioinformatics to be regulated by CRP in *M. tuberculosis*, only 18 [including *echA6*, *ctpE*, *accD3*, *sucC*, *sucD*, *glnD, fadE9*, *fbpC1*  (Ag85C) and *galU*] have conserved orthologous in *M. leprae*, *M. avium* subsp. *paratuberculosis*, and *M. smegmatis* (Akhter et al., 2008). Incorporation of a plasmid harboring and expressing *M. tuberculosis* Rv3676 in an *M. tuberculosis* strain in which *Rv3676* was absent, induced differential expression of 27 genes when compared to the same mutant harboring the BCG gene orthologous to Rv3676 (Hunt et al., 2008). This constitutes evidence that differences in gene regulatory sequences or the regulators among species exist, and it could explain the number of genes that do not respond in a similar manner. Further characterization of the *M. tuberculosis* CRP regulon as well as analysis of what physiological

Little is known about the role of cAMP in mycobacteria, although it is found in both pathogenic and non-pathogenic species. Ingestion of live *M. microti* or *M. bovis* BCG (but not *M. lepraemurium*) increased macrophage intracellular cAMP levels, whereas no change occurred in cells engulfing dead bacilli, latex beads or colloidal gold (Lowrie *et al.*, 1979). The rise in cAMP levels appears directly related to mycobacterial capacity to interfere with phagolysosome formation, evidence suggesting that these microorganisms modify cAMPdependent signalling pathways as a manner to control virulence and infection (Lowrie *et al.*, 1975, Lowrie et al., 1979). Elevated cAMP levels were correlated with reduced phagolysosome fusion during mycobacterial infection of macrophages (Kalamidas *et al.*, 2006). Increased cAMP levels inside phagocytes were shown to negatively modulate actindependent processes, including chemotactic movement and phagocytosis. Macrophage passage was found to have a stimulatory effect on cAMP production by mycobacteria, as cAMP levels within macrophage-passaged mycobacteria were ~50-fold higher than cAMP

levels within bacteria incubated in the tissue culture medium alone (Bai et al., 2009).

Evidences exist that cAMP regulates gene expression in mycobacteria during bacterial growth *in vitro* (Stapleton et al., 2010, Dass et al., 2008), and during macrophage infection (Rickman et al., 2005) where some studies identified Cmr as a transcription factor that regulates cAIGs (cAMP-iduced genes) within macrophages, and suggest that multiple factors affect cAMP-associated gene regulation in tuberculosis-complex mycobacteria (Gazdik et al., 2009). Even during phagocytosis, expression may be down regulated in response to high cAMP or NO levels inside the macrophage environment, providing a mechanism to integrate the transcriptional response to two important signals associated with infection (Smith *et al.*, 2010). Generally, increases in cAMP levels compromise the

It is likely that each cyclase is associated with a distinct signaling pathway. It is expected that specific cyclases are activated to modify cAMP levels in response to different physiological conditions as for example hypoxia, intramacrophage enviroment or pH changes. cAMP receptor protein (CRP) Rv3676 was found to exist as dimmer and exhibited cAMP binding in a concentration-dependent manner and could bind specifically to the putative CRP/FNR nucleotide sequence elements in response to hypoxia (Akhter et al., 2008). The protein itself is composed of three distinct regions of the polypeptide: a large Nterminal domain that binds cAMP, a long -helix (termed the C-helix) that mediates most of

conditions regulate activation/inactivation are required.

bactericidal activity of the host immune system.

**3. Participation of cAMP in mycobacterial pathogenic processes** 


Table 2. *Mycobaterium* genes belonging to CRP regulon and their associated function

regulatory effect inhibiting *whiB1* expression (Stapleton et al., 2010). Interestingly, only CRP1Mt enhances *whiB1* expression, while CRP Mt 2 alone represses *whiB1* expression. Dnase I footprinting assays allowed to determine the presence at −58.5 of an activating CRPMt binding site (CRP Mt 1), which matches in seven of the eight nucleotides from the proposed CRPMt consensus (NGTGNNANNNNNCACA), and also it overlapped with the repressing CRPMt binding site (CRP Mt 2) centred at −37.5. This second site has a poorer match to the consensus (six of the eight defined bases are matched)(Stapleton et al., 2010, Rickman et al., 2005). CRP Mt 1 site was occupied before the CRPMt 2 site by a titration test done with increasing concentrations of CRPMt. It is worth noting that although it was shown that cAMP enhanced binding of recombinant CRPMt to target DNA, this enhancement was not equivalent to that observed for *E. coli* CRP, where DNA-binding affinity is enhanced by several orders of magnitude in the presence of 0.1 mM cAMP, allowing specific DNA binding at nanomolar concentrations. For CRPMt, a much less significant enhancement of DNA binding was observed, and higher concentrations of cAMP compared with *E.coli* CRP were required (Stapleton et al., 2010).

*Rv0993 galU*, uridine diphosphate-glucose pyrophosphorylase *Rv3031* Members of the family of enzymes *Rv3032* transferring activated sugars

*Rv0643c mma3*, methoxy mycolic acid synthase *Rv0904c* AccD3, a putative acetyl CoA carboxylase

*Rv0992c* CHP with putative 5-formyltetrahydrofolate

*Rv0520* Methyl transferase believed to be involved in ubiquinone pathway

*Rv3505* FadE27, protein possibly involved in

*Rv0104* Hypothetical protein probably implied in

*Rv3617* EphA, putative epoxide hydrolase

*Rv3645* Membrane linked adenylyl cyclase

*Rv0907* Hypothetical proteins belonging to the -lactamase family

*Rv0908* CtpE, methyl-accepting chemotaxis protein

*Rv0103c* Probable cation transporter

*Rv0905 echA6*, enoyl-CoA hydratase

biosynthesis.

cyclo-ligase

*Rv2918 glnD*, uridyl transferase

*Rv3114* Nucleoside deaminase

*Rv3113* Phosphatase

Table 2. *Mycobaterium* genes belonging to CRP regulon and their associated function

regulatory effect inhibiting *whiB1* expression (Stapleton et al., 2010). Interestingly, only CRP1Mt enhances *whiB1* expression, while CRP Mt 2 alone represses *whiB1* expression. Dnase I footprinting assays allowed to determine the presence at −58.5 of an activating CRPMt binding site (CRP Mt 1), which matches in seven of the eight nucleotides from the proposed CRPMt consensus (NGTGNNANNNNNCACA), and also it overlapped with the repressing CRPMt binding site (CRP Mt 2) centred at −37.5. This second site has a poorer match to the consensus (six of the eight defined bases are matched)(Stapleton et al., 2010, Rickman et al., 2005). CRP Mt 1 site was occupied before the CRPMt 2 site by a titration test done with increasing concentrations of CRPMt. It is worth noting that although it was shown that cAMP enhanced binding of recombinant CRPMt to target DNA, this enhancement was not equivalent to that observed for *E. coli* CRP, where DNA-binding affinity is enhanced by several orders of magnitude in the presence of 0.1 mM cAMP, allowing specific DNA binding at nanomolar concentrations. For CRPMt, a much less significant enhancement of DNA binding was observed, and higher concentrations of cAMP compared with *E.coli* CRP

carboxyl transferase, which catalyzes the initial steps of fatty acid and mycolic acid

regulating probable acyl-CoA dehydrogenase

cAMP mediated signaling in *M. tuberculosis*

**Associated function Gene Predicted function** 

*Rv0906*

Cell-wall biogenesis

Central metabolism

Putative regulatory elements controlling cAMP signaling

Antibiotic resistance

were required (Stapleton et al., 2010).

pathways

On the other hand, out of 92 genes found and predicted by bioinformatics to be regulated by CRP in *M. tuberculosis*, only 18 [including *echA6*, *ctpE*, *accD3*, *sucC*, *sucD*, *glnD, fadE9*, *fbpC1*  (Ag85C) and *galU*] have conserved orthologous in *M. leprae*, *M. avium* subsp. *paratuberculosis*, and *M. smegmatis* (Akhter et al., 2008). Incorporation of a plasmid harboring and expressing *M. tuberculosis* Rv3676 in an *M. tuberculosis* strain in which *Rv3676* was absent, induced differential expression of 27 genes when compared to the same mutant harboring the BCG gene orthologous to Rv3676 (Hunt et al., 2008). This constitutes evidence that differences in gene regulatory sequences or the regulators among species exist, and it could explain the number of genes that do not respond in a similar manner. Further characterization of the *M. tuberculosis* CRP regulon as well as analysis of what physiological conditions regulate activation/inactivation are required.
