|Home | About | Journals | Submit | Contact Us | Français|
Signaling through the second messenger cyclic di-GMP (c-di-GMP) is central to the life cycle of Vibrio cholerae. However, relatively little is known about the signaling mechanism, including the specific external stimuli that regulate c-di-GMP concentration. Here, we show that the phosphate responsive regulator PhoB regulates an operon, acgAB, which encodes c-di-GMP metabolic enzymes. We show that induction of acgAB by PhoB positively regulates V. cholerae motility in vitro and that PhoB regulates expression of acgAB at late stages during V. cholerae infection in the infant mouse small intestine. These data support a model whereby PhoB becomes activated at a late stage of infection in preparation for dissemination of V. cholerae to the aquatic environment and suggest that the concentration of exogenous phosphate may become limited at late stages of infection.
Vibrio cholerae, the causative agent of the severe diarrheal disease cholera, is a natural inhabitant of temperate aquatic ecosystems throughout the world, including salt, brackish, and some fresh waters (53). The bacterium is known to associate with various aquatic organisms, including cyanobacteria and copepods; additionally, there is evidence that V. cholerae forms biofilms on chitinous surfaces, such as the exoskeletons of zooplankton and phytoplankton (8, 19, 49, 56).
Upon entry into a human host via ingestion of contaminated food or water, the bacteria pass through the gastric acid barrier of the stomach and colonize the small intestine. As the bacterium transitions from its natural environment to that of the small intestine, it undergoes a shift from environmental to virulence gene expression (16, 25, 27, 29, 50). This transition includes induction of virulence factors, including the toxin coregulated pilus (TCP), required for colonization of the small intestine, and cholera toxin (CT), the A-B toxin responsible for the profuse secretory diarrhea that is characteristic of cholera (20, 50). Additionally, the transition from environmental biofilms to the small intestine is accompanied by a decrease in transcription of Vibrio exopolysaccharide synthesis (vps) genes (42).
While the overall mechanism controlling this response to changing environments remains unknown, there is evidence to suggest that the bacterial second messenger bis-(3′,5′)-cyclic di-GMP (c-di-GMP) plays a major role in mediating this adaptive response in V. cholerae (51, 52). The intracellular concentration of c-di-GMP is regulated by the opposing activities of diguanylate cyclase (DGC) and c-di-GMP phosphodiesterase (PDE) enzymes (3, 46). DGCs, containing a GGDEF domain (named for conserved amino acids), catalyze the formation of c-di-GMP from two GTP molecules, while c-di-GMP molecules are hydrolyzed by PDEs, containing an EAL or HD-GYP domain (6, 9, 33, 38, 39, 41, 48). c-di-GMP metabolic enzymes are well conserved throughout the bacterial kingdom. The V. cholerae genome encodes a total of 61 putative c-di-GMP metabolic enzymes: 30 GGDEF, 12 EAL, 9 HD-GYP, and 10 hybrid GGDEF-EAL domain proteins (13, 14).
c-di-GMP was first identified as an allosteric regulator of cellulose synthase in Gluconacetobacter xylinus and has since been recognized as an important bacterial second messenger involved in the regulation of a number of processes (36, 37, 57). Examples include extracellular polysaccharide biosynthesis in Salmonella enterica serovar Typhimurium, Yersinia pestis, Pseudomonas aeruginosa, and V. cholerae (11, 15, 21, 44, 51); motility in Salmonella serovar Typhimurium, P. aeruginosa, Escherichia coli, Caulobacter crescentus, and V. cholerae (1, 4, 11, 18, 22); and virulence in Salmonella serovar Typhimurium, P. aeruginosa, Y. pestis, and V. cholerae (17, 21, 24, 52).
In the classical biotype of V. cholerae, it has been shown that expression of the PDE VieA positively regulates virulence gene expression and negatively regulates vps expression (51, 52). VieA also positively regulates motility and flagellar synthesis genes (4). Additionally, a recent study using the El Tor biotype of V. cholerae showed that ectopic expression of a DGC reduced the induction of virulence genes during infection using the infant mouse model of cholera (47). This information supports the model that c-di-GMP assists in the transition from environment to host via regulation of V. cholerae behavior.
Despite the extensive knowledge of phenotypes and alterations of transcriptional profiles associated with fluctuating concentrations of c-di-GMP, relatively little is known about the mechanism(s) underlying c-di-GMP-mediated regulation, including how the c-di-GMP concentration within the cell is regulated and sensed. Recent studies have identified and confirmed the function of V. cholerae PilZ domain-containing proteins as c-di-GMP binding proteins (34). However, PilZ proteins do not account for most of the observed c-di-GMP-mediated phenotypes in V. cholerae (34).
To gain further insight into the c-di-GMP regulatory circuit, we performed a transposon mutagenesis screen looking for suppressors of the motility defect of a classical biotype V. cholerae vieA mutant having an elevated c-di-GMP level. Here, we report that transposon mutations within the pst operon, encoding the phosphate-specific transport system (Pst), increase the motility of the vieA mutant to near wild-type levels. Further investigation revealed that suppression is due to induction of acgAB, an operon encoding c-di-GMP metabolic proteins, by the response regulator PhoB, which is derepressed in the pst mutant background. Additionally, we show that PhoB regulates expression of acgAB under physiologic conditions during infection of the small intestine. Thus, PhoB regulates the concentration of c-di-GMP during V. cholerae infection.
Bacteria were grown in Luria-Bertani (LB) broth at 37°C with aeration unless otherwise noted. MOPS (morpholinepropanesulfonic acid) minimal medium supplemented with 0.5% glucose and KH2PO4 was prepared as previously described (51). Expression from the Ptac promoter was induced by addition of isopropyl-β-d-thiogalactopyranoside (IPTG) to a final concentration of 1 mM. Antibiotics were added, when appropriate, at the following concentrations: streptomycin (Sm), 100 μg/ml; ampicillin (Ap), 50 μg/ml; kanamycin (Kn), 50 μg/ml; and tetracycline (Tc), 2 μg/ml.
All strains and plasmids used in this study are listed in Table Table1.1. Plasmids with oriR6K were propagated in E. coli DH5αλpir; all other plasmids were propagated in E. coli DH5α. Plasmids for generating in-frame deletions in V. cholerae were constructed in the allelic exchange vector pCVD442 (10). All deletions were constructed by splicing by overlap extension (SOE) PCR using the primers listed in Table Table22 (43). Briefly, DNA fragments of approximately 800 bp upstream and downstream of each deletion were amplified by PCR from V. cholerae O395 genomic DNA, annealed together by complementary sequences in the R1 and F2 primers, and then PCR amplified with the F1 and R2 primers. The final PCR product was blunt ligated into pCVD442. The respective F1/R1 and F2/R2 primer pairs used for generating deletion alleles of phoB, pstCAB-phoU, and acgAB were phoBF1/phoBR1 and phoBF2/phoBR2, pstF1/pstR1 and pstF2/pstR2, and acgABF1/acgABR1 and acgABF2/acgABR2, respectively. Plasmids were conjugated into V. cholerae vieA(E170A) and vieA(E170A) Δpst strains from E. coli SM10λpir as previously described (26). After one passage in LB broth in the absence of antibiotics, sucrose-resistant colonies were selected and were subsequently screened for the desired deletion by PCR.
For transposon mutagenesis, E. coli SM10λpir harboring pAC2889 containing mini-Tn10 (mTn10) was mated with V. cholerae vieA(E170A) on LB plates for 2 h at 37°C. Cells were then replica plated onto selective medium containing Sm and chloramphenicol to select for V. cholerae vieA(E170A) mTn10 transposon insertion mutants.
Chemotaxis plates composed of 0.5% LB broth or MOPS supplemented with 100 μM (high) or 1 μM (low) KH2PO4 and 0.3% agar were used to assess the motility of V. cholerae strains. Strains were streaked for single colonies on LB broth plus antibiotics or MOPS agar supplemented with 100 μM KH2PO4 plus antibiotics and incubated overnight at 30°C. Single colonies were removed with a toothpick, stabbed into the chemotaxis plate, and incubated for 16 h at 30°C or 24 h at 37°C for LB- and MOPS-based plates, respectively.
Alternatively, for the suppressor screen, pools of four mTn10 insertion strains were screened for increased motility. Using a toothpick, the edges of swarms with increased motility were stabbed and streaked out for single colonies. Individual colonies were then tested for motility. The transposon/chromosome junctions were sequenced using the primer catF to identify the insertion site in suppressor mutants.
Strains were grown for 16 h in LB broth or MOPS supplemented with 100 μM KH2PO4 plus antibiotics, diluted 1:1,000 into 0.5 ml LB broth or 1:10 into 0.5 ml MOPS agar supplemented with 100 μM (high) or 1 μM (low) KH2PO4 plus antibiotics in new 13-mm-diameter borosilicate glass tubes, and incubated at room temperature without aeration for 48 h or at 30°C for 24 h, respectively. The medium was aspirated, and adherent biofilms were gently washed three times with LB broth. Biofilms were then stained with 0.5 mg/ml crystal violet for 5 min and then washed extensively with water. Bound crystal violet was solubilized with 1 ml of 100% ethanol and quantified by absorbance at 570 nm. Each experiment was performed in triplicate.
RNA was isolated from 0.5 ml of mid-exponential-phase V. cholerae cultures grown in LB broth at 37°C and purified following resuspension in 1 ml of RNAprotect reagent (Qiagen) by using an RNeasy mini kit (Qiagen). DNA was removed using a DNA-free kit (Ambion). cDNA was synthesized from 1 μg RNA by using an iScript select SYBR green RT-PCR kit (Bio-Rad). Controls lacking reverse transcriptase were included.
qRT-PCR experiments were performed using IQ SYBR green Supermix (Bio-Rad) and an MxP3005P real-time PCR system with MxPro qPCR software (Stratagene). Primers used in these studies are listed in Table Table2.2. For each sample, the mean cycle threshold of the test transcript was normalized to that of rpoB and presented relative to that of the wild type. Values of less than 1 indicate that the steady-state concentration of the transcript is lower than in the wild type. Three independent samples were tested for each condition.
Expression of acgAB in vitro and during infection (in vivo) was monitored using recombination-based in vivo expression technology (RIVET). Reporter strains containing the transcriptional fusion acgAB::tnpR were made by mating E. coli SM10λpir containing pIVET5-VC1593(acgB) with V. cholerae AC61, Δpst, ΔphoB, Δpst ΔphoB, and ΔphoB::phoB rev strains. Each recipient strain contained a res-tet-res resolvable cassette inserted within the lacZ gene.
For in vitro expression analysis, strains were resuspended from plates to an optical density at 600 nm (OD600) of 0.1 in 1 ml LB broth and incubated for 4 h at 37°C with aeration. Serial dilutions were plated onto LB plates supplemented with Sm and Ap and incubated overnight at 37°C. Colonies were counted and then replica plated onto LB plates supplemented with Tc and incubated overnight. Colonies were counted again, and the resolution frequency was determined by the number of Tcs colonies divided by the number of Smr Apr colonies.
For in vivo expression analysis, single-strain infections were performed. Approximately 2 × 105 CFU of each tnpR fusion strain was inoculated intragastrically into eight 5-day-old CD-1 mice. At 21 h postinfection, mice were euthanized, and bacteria were recovered from the small intestines and plated onto LB Sm Ap plates. Resolution frequencies were calculated as described above.
Bacteria were resuspended in 100 ml LB plus antibiotics to an OD600 of 0.05 from an LB plate and incubated at 37°C with aeration until mid-exponential phase. IPTG was added to induce gene expression, and the cultures were incubated for 4 h more at 30°C. Cells were collected, resuspended in PDE reaction buffer (75 mM Tris [pH 8], 25 mM KCl, 10 mM MgCl2) containing 10% glycerol, lysed by French press at 16,000 lb/in2, and fractionated by centrifugation at 10,000 × g for 30 min at 4°C. Aliquots of the soluble fraction were collected and assayed for protein synthesis via Western blot analysis using mouse anti-His5 primary antibodies (Qiagen) and horseradish peroxidase-conjugated sheep anti-mouse secondary antibodies (Amersham). Western blots were developed with ECL detection reagents (Amersham).
For PDE assays, c-di-GMP was synthesized as previously described (48). Ten-microliter reaction mixtures containing 10 μg of lysate and 1 μl of radiolabeled c-di-GMP in PDE reaction buffer were incubated on ice for 5 min, and aliquots were spotted onto cellulose-polyethyleneimine thin-layer chromatography plates. Reaction products were separated in 1.5 M KH2PO4 (pH 3.65) and visualized by phosphorimaging.
DGC reactions were carried out in a similar fashion, except that cyclase reaction buffer was used (75 mM Tris [pH 7.8], 250 mM NaCl, 25 mM KCl, 10 mM MgCl2) and 3 μl of [α-32P]GTP (3,000 Ci/mmol; Perkin-Elmer) was added to the reaction instead of c-di-GMP.
It has previously been shown that a V. cholerae strain harboring a point mutation within VieA [VieA(E170A)], which abrogates PDE activity, results in an increase in the intracellular concentration of c-di-GMP and a concomitant decrease in motility (34, 48, 51). This strain was used as the parental strain for transposon mutagenesis in a screen to identify suppressors of the motility defect.
Approximately 20,000 mTn10 insertion strains were screened for restoration of motility in chemotaxis plates. We observed a total of 27 unique suppressor mutants, of which 8 fell in the pst operon, encoding Pst (Table (Table3).3). The components of the Pst system are usually encoded within a single operon (pstSCAB-phoU) and have two known functions—as a high-affinity inorganic phosphate (Pi) transport system and a regulator of the Pho regulon. The Pho regulon encompasses genes responsible for the bacterium's adaptation and survival in phosphate limiting conditions (<4 μM in E. coli ). Transcription of the Pho regulon is controlled by the two-component system PhoBR. PhoR functions as a histidine kinase responsive to external Pi concentrations, which phosphorylates the response regulator PhoB under low Pi conditions. Phospho-PhoB is then activated and capable of regulating the expression of a number of genes, generally involved in phosphate homeostasis. By some unknown mechanism, the phosphorylation of PhoB is blocked by the Pst system when environmental phosphate is in excess (54). When phosphate is limited, this repression is relieved, thus allowing for activation of the Pho regulon. Null mutations in the Pst system disrupt this regulation of PhoB activation, which leads to constitutively active PhoB and, thus, expression of the Pho regulon, regardless of environmental phosphate availability (35, 54).
To confirm the effect of the pst transposon mutants in the vieA(E170A) strain, we made a complete deletion of the pst-1 operon pstCAB-phoU (Δpst). We have previously shown that this pst mutation leads to activation of PhoB and expression of the Pho regulon in wild-type V. cholerae irrespective of Pi availability, as expected (J. T. Pratt and A. Camilli, submitted for publication). We assayed the motility of the vieA(E170A) Δpst double mutant on chemotaxis plates and saw an increase in motility compared to that of the parental strain, similar to that of the transposon mutants (Fig. 1A and B). Mutation of phoB in the vieA(E170A) Δpst background reduced motility to the level of the vieA(E170A) strain. This result suggests that constitutively active PhoB, as a result of the Δpst mutation, leads to suppression of the motility defect. Consistent with this hypothesis, the vieA(E170A) ΔphoB strain showed no alteration in motility relative to that of the vieA(E170A) strain. The motility phenotype of the vieA(E170A) Δpst ΔphoB strain could be complemented by addition of phoB in trans. These data indicate that PhoB positively regulates motility in V. cholerae.
To study the role of PhoB in V. cholerae motility under a more natural condition of PhoB activation, we measured the motility of Δpst and ΔphoB mutants in the wild-type background under high and low Pi conditions, 100 μM and 1 μM, respectively. Under high Pi conditions, the wild-type and ΔphoB strains showed similar motility levels, while the Δpst mutant showed an increase in motility compared to that of the wild type (Fig. (Fig.1C).1C). This increase was abolished when phoB was mutated in the Δpst background. Under low Pi conditions, the ΔphoB mutant showed a decrease in motility compared to that of the wild type, which could be complemented by addition of phoB in trans. Once again the Δpst mutant showed a PhoB-dependent increase in motility compared that of the wild type (Fig. (Fig.1C).1C). These data suggest that PhoB positively regulates motility under low Pi conditions. However, one caveat to this experiment is the observation that the ΔphoB mutant grows more slowly than the wild type under low Pi conditions (Pratt and Camilli, submitted). The chemotaxis plate assay measures chemotaxis, motility, and growth; therefore, it is a possibility that the observed motility defect of the ΔphoB strain is due in part or in whole to reduced growth rather than specifically reduced motility.
It has recently been shown that PhoB negatively regulates biofilm formation in the soil bacterium Pseudomonas fluorescens (30). To test if this is also true in V. cholerae, we assayed biofilm formation in the pst and phoB mutant strains compared to that in the vieA(E170A) strain, which is a hyperbiofilm former compared to the wild type. We observed that the vieA(E170A) Δpst strain showed decreased biofilm formation compared to the vieA(E170A) strain (Fig. (Fig.2A).2A). This decrease was dependent on PhoB, since a vieA(E170A) Δpst ΔphoB triple mutant formed biofilm at levels similar to the vieA(E170A) strain. Mutation of phoB alone in the vieA(E170A) background had no effect. The hyperbiofilm phenotype of the vieA(E170A) Δpst ΔphoB strain could be complemented by addition of phoB in trans (Fig. (Fig.2A).2A). These results show that PhoB negatively regulates biofilm formation in V. cholerae.
To further confirm the role of PhoB in V. cholerae biofilm formation, we measured biofilm formation of Δpst and ΔphoB mutants in the wild-type background under high and low Pi conditions. Under high Pi conditions, the ΔphoB mutant formed biofilms at levels similar to the wild type, while the Δpst mutant formed less biofilm than did the wild type (Fig. (Fig.2B).2B). Mutation of phoB in the Δpst background led to a restoration of biofilm formation to levels similar to that of the wild type. Under low Pi conditions, the ΔphoB mutant formed more biofilm than did the wild type, and biofilm could be restored to wild-type levels by the addition of phoB in trans (Fig. (Fig.2B).2B). The Δpst mutant formed biofilm at levels similar to the wild type, while mutation of phoB in the Δpst background led to an increase in biofilm formation compared to wild-type levels, similar to that of the single ΔphoB mutant. These data suggest that PhoB functions as a negative regulator of V. cholerae biofilm formation, as induction of phoB via the inactivation of the Pst system leads to a decrease in biofilm under high Pi conditions, while loss of phoB leads to an increase in biofilm formation compared to that of the wild type under low Pi conditions.
Given the known role of c-di-GMP in repressing motility and inducing biofilm formation, we hypothesized that the observed PhoB-dependent increase in motility and decrease of biofilm formation of the vieA(E170A) Δpst strain is due to altered c-di-GMP concentrations. Specifically, the deletion of pst in the vieA(E170A) strain leads to a decrease in the cellular concentration of c-di-GMP, which then leads to restoration of motility and decreased biofilm formation. A similar role for PhoB in the regulation of c-di-GMP was recently observed in P. fluorescens (30).
We further hypothesized that the effect of PhoB on c-di-GMP regulation was mediated through transcriptional control of one or more PDE genes. To test this idea, we designed primers to all 22 V. cholerae genes predicted to encode an EAL domain protein and measured their expression in LB broth in the vieA(E170A) strain versus the vieA(E170A) Δpst strain. We found that only one EAL gene, acgA, was highly induced in the vieA(E170A) Δpst strain compared to its expression level in the vieA(E170A) strain (~100-fold) (Fig. (Fig.33 and data not shown).
Further investigation of the genome sequence revealed that acgA is within a putative operon with a second gene, acgB, encoding a DGC. Because the encoded proteins have opposing activities, we investigated the expression of these genes and their respective roles in the PhoB-mediated effects on motility and biofilm formation. To confirm that these two genes are indeed an operon and that PhoB is responsible for the increased transcription in the vieA(E170A) Δpst strain, we assayed transcription of acgA and acgB in the vieA(E170A), vieA(E170A) Δpst, and vieA(E170A) Δpst ΔphoB strains. We found that both genes are coregulated; each is induced by the pst mutation, and deletion of phoB in the vieA(E170A) Δpst background led to the restoration of parental-strain levels of transcription (Fig. (Fig.3).3). Transcription of acgAB in the vieA(E170A) Δpst ΔphoB strain could be complemented by addition of phoB in trans. The phoB mutation alone in the vieA(E170A) strain had no effect on the transcription of acgAB.
The acgAB operon in V. cholerae has previously been shown to be regulated via read-through transcription from an upstream operon (alsDSO) despite the presence of a predicted transcriptional terminator within the intergenic space upstream of acgA (23). To determine whether PhoB is regulating transcription of acgAB via a potential promoter proximal to the operon or via read-through from an upstream promoter, the expression of alsD was measured and found to be unchanged in the vieA(E170A) strain versus the level in the vieA(E170A) Δpst strain, suggesting that the acgAB operon is regulated independently of alsDSO by PhoB (Fig. (Fig.33).
To determine if induction of acgAB is responsible for the phenotypes observed in the vieA(E170A) Δpst strain, we made individual deletions of acgA and acgB, as well as a deletion of the entire operon (ΔacgAB) in the vieA(E170A) Δpst strain background. We assayed the motility and biofilm phenotypes of these mutants and found that induction of acgA, but not acgB, is responsible for the observed motility phenotype of the vieA(E170A) Δpst strain. The vieA(E170A) Δpst ΔacgA and vieA(E170A) Δpst ΔacgAB strains showed levels of motility similar to that of the vieA(E170A) strain, suggesting an elevated concentration of c-di-GMP (Fig. 4A and B). In contrast, the vieA(E170A) Δpst ΔacgB strain showed an increased motility compared to that of the parental strain, similar to a wild-type level of motility, suggesting a reduced concentration of c-di-GMP. These data are consistent with AcgA functioning as a PDE and AcgB as a DGC.
To further confirm the role of acgAB in the regulation of V. cholerae motility, we measured the motility of the ΔacgA, ΔacgB, and ΔacgAB mutants in the wild-type background under high and low Pi conditions. The motility levels of all strains were similar under high Pi conditions (Fig. (Fig.4C).4C). Under low Pi conditions, the ΔacgA mutant showed a decrease in motility compared to wild-type levels, whereas the ΔacgB mutant showed an increase in motility. The ΔacgAB mutant showed a decrease in motility similar to that observed for the ΔacgA mutant (Fig. (Fig.4B).4B). These results suggest that AcgA and AcgB regulate motility in V. cholerae and are consistent with the model that the acgAB operon is induced by PhoB activation under low Pi limitation.
However, when we examined biofilm formation, we did not detect a role of acgAB in the biofilm phenotype of the vieA(E170A) Δpst strain. Specifically, we saw no change in the biofilms formed by the vieA(E170A) Δpst ΔacgAB triple mutant compared to those of the vieA(E170A) Δpst double mutant under either high or low Pi conditions (data not shown).
Recently, it was shown by using RIVET that acgB is induced by V. cholerae during late stages of infection of the infant mouse small intestine (40). Given that we have identified PhoB as a regulator of acgAB in vitro, we hypothesized that PhoB could represent the in vivo regulator of this operon.
Using the RIVET methodology, we constructed ΔphoB, Δpst, and Δpst ΔphoB mutants in a V. cholerae parental strain background containing a res-tet-res cassette, which is a tetracycline resistance gene flanked on both sides by resolvase recognition sites, and an acgAB::tnpR transcriptional fusion. Therefore, when acgAB is induced, the tnpR gene encoding resolvase will be coexpressed, leading to excision and loss of the tet resistance gene. We can score expression of acgAB in these reporter strains by monitoring tetracycline sensitivity in the bacterial population. To confirm that these new strains are functioning as expected, we monitored resolution in vitro in LB cultures. Consistent with our qRT-PCR results, 100% of Δpst cells expressed acgAB::tnpR and were Tcs, while less than 1% of parental, ΔphoB, and Δpst ΔphoB strain cells resolved (data not shown).
To determine if PhoB is the regulator of acgAB during infection, we inoculated 5-day-old mice intragastrically with the parental strain and ΔphoB acgAB::tnpR fusion strains and determined the resolution frequency for bacteria isolated from the small intestine. Consistent with PhoB functioning as an in vivo regulator of acgAB, we observed that the resolution frequency of the parental strain was 37% while the frequency for ΔphoB was less than 1%, essentially at background levels of resolution. This expression defect could be complemented by chromosomal reversion of the ΔphoB mutation (Fig. (Fig.55).
To determine if PhoB is a direct regulator of acgAB expression, we performed gel shift experiments using purified constitutively active V. cholerae PhoB [PhoBCA; PhoB(D10A/D53E)] (2) and a 32P-labeled DNA fragment containing the acgAB promoter region. However, we did not observe any band shift (data not shown), despite the observation that PhoBCA is functional, suggesting that PhoB acts as an indirect regulator of acgAB expression. This is consistent with the absence of a detectable PhoB binding sequence (Pho box) in the promoter region upstream of acgAB. These results suggest that there is another, unknown regulator of acgAB expression whose expression or activity is controlled by PhoB.
The motility phenotypes observed in the individual acgA and acgB mutants, as well as previous experiments showing motility phenotypes of V. cholerae strains overexpressing AcgA or AcgB (23), suggest that the proteins do possess the predicted enzymatic activities; however, this has not been directly shown. To address this question, we assayed the PDE and DGC activities of AcgA and AcgB, respectively. As is predicted by the data and their protein sequences, both AcgA and AcgB do possess their respective putative enzyme activities. Using E. coli lysates expressing the respective proteins, we show that AcgA has PDE activity and degrades purified 32P-labeled c-di-GMP, while AcgB has DGC activity and catalyzes the formation of c-di-GMP from [α-32P]GTP (Fig. 6A and B, respectively).
The sum of these data suggest that Pi regulates the intracellular concentration of c-di-GMP in V. cholerae through Pst/PhoBR control of acgAB expression. It has been shown that PhoB can be activated by additional stimuli; therefore, it remains possible that stimuli in addition to Pi could also induce expression of acgAB (12, 31, 45, 55). It is curious that the acgAB operon encodes two enzymes with opposing activities on the intracellular concentration of c-di-GMP. There are numerous examples of dual GGDEF-EAL domain proteins; however, in most cases, only one of the domains functions as a c-di-GMP metabolic enzyme (9, 13, 14, 47). Although it is possible that AcgA and AcgB are differentially regulated posttranscriptionally, a second possibility is that, since AcgA and AcgB are predicted cytosolic and transmembrane proteins, respectively, their localization in different compartments of the cell may exert differential effects on local c-di-GMP concentrations.
Here, we show that PhoB regulates motility and biofilm formation in V. cholerae. The observed increased motility in the vieA(E170A) Δpst strain can be attributed to the induction of acgAB encoding functional PDE and DGC enzymes, respectively. This suggests that an alteration in the intracellular c-di-GMP concentration leads to the observed motility increase and that induction of acgA is at least partially complementing the high intracellular c-di-GMP concentration caused by the vieA(E170A) mutation. Additionally, it suggests that Pi is a stimulus that regulates c-di-GMP in V. cholerae. Furthermore, we show that PhoB regulates the expression of acgAB during infection of the infant mouse small intestine. This finding supports previous results that showed that acgB is induced in vivo and also suggests that the Pho regulon is induced in V. cholerae during infection. Both acgB and phoA have previously been identified as transcriptionally induced genes, specifically at late stages of infection (40). This suggests that, as is the case for iron (5, 28), the concentration of exogenous Pi becomes limited at late stages of infection. Whether this reduction in Pi concentration is growth limiting or not is not known.
We have previously shown that PhoB negatively regulates expression of virulence genes in V. cholerae and is required for V. cholerae survival in pond water (Pratt and Camilli, submitted); therefore, this taken together with our current findings suggests a model whereby PhoB becomes activated at late stages of infection and functions to shut down virulence gene expression and induce the expression of a gene set that is required for survival in the aquatic ecosystem. Induction of genes required for survival in the next environment and repression of unnecessary genes (virulence genes) would allow V. cholerae to more efficiently respond to the transition from the host to the aquatic environment and perhaps increase survival.
A.C. is a Howard Hughes Medical Institute investigator. This research was supported by NIH grant AI45746 to A.C.
We thank Stefan Schild for constructing pAC2889.
Published ahead of print on 4 September 2009.