Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2009 Dec 8.
Published in final edited form as: Mol Microbiol. 2004 Aug;53(3):857–869. doi: 10.1111/j.1365-2958.2004.04155.x

Cyclic diguanylate (c-di-GMP) regulates Vibrio cholerae biofilm formation

Anna D Tischler 1, Andrew Camilli 1,*
PMCID: PMC2790424  NIHMSID: NIHMS156657  PMID: 15255898

Summary

While studying virulence gene regulation in Vibrio cholerae during infection of the host small intestine, we identified VieA as a two-component response regulator that contributes to activating expression of cholera toxin. Here we report that VieA represses transcription of Vibrio exopolysaccharide synthesis (vps) genes involved in biofilm formation by a mechanism independent of its phosphorelay and DNA-binding activities. VieA controls the intracellular concentration of the cyclic nucleotide second messenger cyclic diguanylate (c-di-GMP) using an EAL domain that functions as a c-di-GMP phosphodiesterase. Two-dimensional thin layer chromatography of nucleotide extracts confirmed that VieA reduces the concentration of c-di-GMP, opposing the action of c-di-GMP synthetase proteins. Expression of unrelated V. cholerae c-di-GMP synthetase or phosphodiesterae proteins also modulated c-di-GMP concentration and vps gene expression. We propose that c-di-GMP synthetase and phosphodiesterase domain-containing proteins contribute to regulating biofilm formation by controlling c-di-GMP concentration.

Introduction

Vibrio cholerae, a Gram-negative bacterium that causes the acute intestinal infection cholera, is a facultative pathogen that exists in aquatic habitats. Two major virulence factors contribute to disease; cholera toxin (CT) is responsible for the profuse secretory diarrhoea, while the toxin co-regulated pilus (TCP) is required for colonization of the small intestine (Taylor et al., 1987; Kaper et al., 1994). A cascade of transcription factors including AphA/B, TcpP/H, ToxR/S and ToxT regulate expression of CT and TCP both in vitro (for a review see Reidl and Klose, 2002) and during infection (in vivo) (Lee et al., 1999). Recently, additional regulators including the quorum sensing repressor HapR (Zhu et al., 2002) and the VieS/A/B three-component signal transduction system (Tischler et al., 2002) have been shown to modulate virulence factor expression.

Evidence from two different in vivo screens suggests that VieS/A/B contributes to regulating gene expression during infection (Camilli and Mekalanos, 1995; Lee et al., 2001). The vieSAB genes are in an apparent operon on V. cholerae chromosome I, but are differentially expressed from multiple promoters (Lee et al., 1998). The sensor histidine kinase gene, vieS, was identified in a screen for in vivo regulators of CT (Lee et al., 2001); VieS/A/B also contributes to CT production under in vitro inducing conditions (Tischler et al., 2002). The VieB response regulator has a typical N-terminal phosphoreceiver domain but lacks any recognizable DNA binding motif. The vieA gene encodes a second response regulator, which like VieB has a conserved phosphoreceiver motif at its N-terminus. VieA also has a LuxR-family helix–turn–helix (HTH) DNA binding motif at its C-terminus, which is predicted to be involved in transcriptional regulation, and a conserved domain of unknown function that was recently named the EAL domain after its conserved amino acids (Galperin et al., 2001).

Although it has not been directly demonstrated, the EAL domain is predicted to have phosphodiesterase activity against a novel cyclic dinucleotide, cyclic diguanylate (c-di-GMP, bis(3′,5′)-cyclic diguanylic acid), based on its presence in enzymes with this activity (Tal et al., 1998; Chang et al., 2001) and its occurrence in isoenzymes that regulate production of cellulose in the bacterium Glucon-acetobacter xylinus (formerly Acetobacter xylinum) (Tal et al., 1998; Galperin et al., 2001). In G. xylinus it has been demonstrated that c-di-GMP positively regulates cellulose production by allosterically activating the cellulose synthase enzyme (Ross et al., 1986; 1987; 1990; Weinhouse et al., 1997). Mutational analysis identified six genes that control the intracellular concentration of c-di-GMP by influencing synthesis or hydrolysis of the molecule (Tal et al., 1998). Each of these genes encodes both an EAL domain and a second conserved motif, GGDEF, also named for conserved amino acid residues. Again, although it has not been directly demonstrated, diguanylate cyclase activity has been assigned to the GGDEF domain based on sequence alignment with eukaryotic adenylate cyclase (Pei and Grishin, 2001) and the ability of GGDEF domain proteins from other organisms to activate cellulose production in Rhizobium and Agrobacterium (Ausmees et al. 2001). Proteins containing GGDEF and EAL domains are encoded by several gram-positive and virtually all free-living gram-negative bacterial species sequenced to date, and some species harbour large paralogous families of each domain (Galperin et al., 2001).

Of the proteins containing GGDEF and EAL domains that have been partially characterized, most control properties that affect cell–cell interactions. As in G. xylinus, GGDEF and EAL domain proteins have been implicated in the regulation of cellulose synthesis in Salmonella (Romling et al., 2000; Zogaj et al., 2001), R. leguminosarum (Ausmees et al., 1999), and A. tumefaciens (Ausmees et al., 2001). Cellulose production contributes to the multicellular behaviour of these organisms, facilitating symbiotic or pathogenic interactions with a host (Ross et al., 1991). Proteins with GGDEF and EAL domains have also been implicated in biofilm formation. In P. aeruginosa, WspR, which contains a GGDEF domain, regulates expression of a putative fimbrial adhesin required for biofilm formation (D’Argenio et al., 2002). In addition, PvrR, a P. aeruginosa response regulator that contains an EAL domain, was recently demonstrated to regulate biofilm formation and antibiotic resistance (Drenkard and Ausubel, 2002).

Persistence of V. cholerae in aquatic environments between cholera epidemics is likely enhanced by its ability to form biofilms. Biofilms are more resistant to numerous environmental stresses, including chlorine and antibiotics, and may assist in the acquisition of nutrients (Shirtliff et al., 2002). In addition, the biofilm state increases the infectivity of V. cholerae, possibly affording protection from innate immune system functions (Zhu and Mekalanos, 2003). The cell–cell interactions within V. cholerae biofilms are stabilized by production of an exopolysaccharide (EPS) matrix. A cluster of genes required for EPS production (vpsA-Q) has been identified (Yildiz and Schoolnik, 1999; Ali et al., 2000), and their expression is controlled in part by the transcriptional activator VpsR (Yildiz et al., 2001). HapR also regulates biofilm formation by repressing the vpsA-Q genes independently of VpsR (Hammer and Bassler, 2003). In addition, proteins with the conserved GGDEF and EAL domains have been implicated in regulating EPS production in Vibrio species. V. cholerae mbaA mutants overproduce EPS and exhibit enhanced biofilm formation (Bomchil et al., 2003). V. cholerae RocS is required for the switch from smooth to rugose colony morphology, a phenotype associated with overproduction of EPS (Rashid et al., 2003). Finally, V. parahemolyticus ScrC regulates genes required for production of capsular polysaccharide (Boles and McCarter, 2003). These findings suggest that, in addition to its known role as an allosteric activator, c-di-GMP may act as a second messenger within bacteria, coupling environmental signals to gene regulation.

Here we provide evidence connecting c-di-GMP concentration with transcription of the V. cholerae vps genes. After identifying the response regulator VieA as a repressor of the vps cluster in a genetic screen, we sought to characterize the mechanism of regulation. We demonstrate that the VieA EAL domain is responsible for repression of vps transcription and show by two-dimensional thin layer chromatography (2D–TLC) of nucleotide extracts that VieA reduces the intracellular concentration of c-di-GMP. We propose a model in which proteins containing EAL and GGDEF domains transduce multiple signals into the concentration of the c-di-GMP second messenger to affect vps gene expression and biofilm formation.

Results

VieA regulates extracellular polysaccharide synthesis genes

We previously reported that the V. cholerae vieSAB three-component signal transduction system contributes to regulation of ctxAB, the genes encoding CT (Tischler et al., 2002). To identify other genes specifically controlled by the VieA response regulator, a classical biotype V. cholerae strain (AC-V1352) was constructed in which vieA is overexpressed from the arabinose-inducible ParaBAD promoter (Guzman et al., 1995). We chose to work with the classical biotype because VieA is required for regulation of CT production during growth in M9 plus asparagine, arginine, glutamate, and serine (M9 + NRES) (A.D.T. and A.C., unpublished data), a condition that induces the ToxR-regulon in classical but not El Tor biotype V. cholerae (Miller and Mekalanos, 1988). Random lacZ transcriptional fusions were generated in AC-V1352 by transposon mutagenesis with mTn5lacZ. Approximately 6000 mTn5lacZ mutants were replica plated to Luria-Bertani (LB) and LB arabinose media containing the colorimetric β-galactosidase substrate X-gal and screened visually for altered β-galactosidase activity in the presence versus the absence of arabinose.

Of six putative VieA-regulated lacZ fusions for which the genome-mTn5lacZ junction was determined, two proved to be regulated specifically by VieA after phage-mediated generalized transduction into fresh strain backgrounds. A fusion in VCA0707 was repressed 10-fold upon induction of vieA, while overexpression of VieB had no effect on fusion activity (Fig. 1A). The VCA0707 gene product is annotated as a putative UhpC, a protein involved in sensing extracellular glucose-6-phosphate. A fusion in VC0928, which encodes a hypothetical protein, was repressed 100-fold by vieA induction (Fig. 1B). Overexpression of VieB repressed the VC0928 fusion about two-fold, but this appears to be an effect of arabinose because VC0928::lacZ activity was also reduced twofold by arabinose in a strain harbouring pBAD33 (Fig. 1B).

Fig. 1.

Fig. 1

Open in a new tab

VieA represses activity of mTn5lacZ fusions to VCA0707 and VC0928. β-Galactosidase assays were performed after overnight growth in LB (solid bars) or LB + 0.2% arabinose (hatched bars) at 30°C with shaking. Results are the average ± standard deviation for three independent cultures.

A. VCA0707::lacZ activity in V. cholerae harbouring the indicated plasmids.

B. VC0928::lacZ activity in V. cholerae harbouring the indicated plasmids.

C. VC0928::lacZ activity in wild type and ΔvieA V. cholerae harbouring plasmids with the arabinose-inducible protein domains indicated on the x-axis.

D. VC0928::lacZ activity in wild type and ΔvieA V. cholerae harbouring a plasmid encoding arabinose-inducible VCA0956. Assays were performed after overnight growth in LB ± 0.2% arabinose, or M9 + NRES ± 0.2% arabinose.

Because VC0928::lacZ was strongly repressed by overexpression of VieA, the fusion was transduced into the wild type, ΔvieA, and ΔvieB strain backgrounds to test whether deletion of native vieA could relieve repression. In the wild type background, VC0928::lacZ was not active during growth in M9 + NRES (Table 1). While deletion of vieB had no effect on VC0928::lacZ activity, the ΔvieA mutant exhibited approximately 100-fold increased activity of the fusion (Table 1). Taken together, these data indicate that VieA is a negative regulator of VC0928.

Table 1.

Activity of the VC0928::lacZ fusion in various vieA mutant strain backgrounds.

Strain Relevant genotype β-Galactosidase activity
(Miller units)
AC-V1417 wild type 1.4 ± 1.0
AC-V1418 ΔvieA 97 ± 4
AC-V1419 ΔvieB 1.0 ± 0.8
AC-V1729 ΔvieA135–375 (ΔEAL) 110 ± 23
AC-V1598 vieAE170A 100 ± 16
AC-V1614 vieAD286A 86 ± 28
Open in a new tab

Strains were grown overnight in M9 + NRES at 30°C with shaking and assayed for β-galactosidase activity. Results presented are the average ± standard deviation for three independent cultures.

VC0928 encodes a hypothetical protein with no conserved domains nor orthologues in the NCBI non-redundant database, but is located between two operons, vpsA-K and vpsL-Q, that are required for EPS production and biofilm formation (Yildiz and Schoolnik, 1999; Ali et al., 2000). To confirm that VieA regulates VC0928 at the transcriptional level and to test whether VieA regulates the vps operons, ribonuclease protection assays (RPAs) were performed. Riboprobes were generated for VC0928, vpsA and vpsL, and used to detect RNA transcripts of these genes in wild type, ΔvieA and ΔvieB strains, as well as a strain containing an in-frame deletion of the VieA HTH DNA-binding motif. Transcription of VC0928, vpsA and vpsL was greatly derepressed in the ΔvieA mutant, while transcript levels were unchanged for the ΔvieB mutant compared with the wild type (Fig. 2A). Although modest derepression of all three transcripts was observed for the ΔvieA-HTH mutant, each transcript was present at lower levels in this mutant compared with the full deletion of vieA. These results indicate that regulation of the vps gene cluster by VieA requires a functional domain other than the putative DNA-binding HTH motif.

Fig. 2.

Fig. 2

Open in a new tab

Ribonuclease protection assays (RPAs) to analyse regulation of vpsA vpsL, VC0928, and vpsR by VieA. RNA was isolated from wild type or the mutant V. cholerae strains indicated above each lane after growth in M9 + NRES. Transcripts were detected in 1 µg of total RNA by RPA. Band intensities were quantified and normalized to the intensity of the rpoB band. L indicates RNA ladder (sizes in number of bases = 400, 300 and 200).

A. Average intensity relative to the wild type from at least three independent experiments is reported below each lane. U1 indicates undigested rpoB (bottom band), vpsA (top band), and VC0928 (middle band) probes. U2 indicates undigested vpsL probe (lower band) and an unrelated probe.

B and C. The intensity of the full-length (top) vpsR protected band relative to the wild type for at least three independent experiments is reported below each lane. U indicates uncut vpsR and rpoB probes.

VieA regulates vps genes through control of vpsR transcription

The vpsA-Q genes are positively regulated by VpsR, a transcription factor with homology to sigma-54 dependent activators and two-component response regulators (Yildiz et al., 2001). To test whether VieA might indirectly regulate vpsA-Q by affecting expression of vpsR, RPAs were performed to examine vpsR transcript levels. Similar to the results obtained for vpsA, vpsL, and VC0928, vpsR transcripts were present at much higher levels in the ΔvieA mutant than in the wild type (Fig. 2B and C). In contrast, vpsR transcript levels were unchanged for both the ΔvieA-HTH and ΔvieB mutants (Fig. 2B and data not shown). Two distinct bands were consistently observed associated with the vpsR probe; the smaller band may represent a different transcriptional start site, premature transcriptional termination or a post-transcriptional processing site within the vpsR coding sequence. The terminator hypothesis is favoured by a similar fold-regulation of both bands in the ΔvieA mutant. These results suggest that VieA regulates vpsA-Q and VC0928 by controlling expression of the VpsR positive regulator.

To verify that derepression of vps transcription in the ΔvieA mutant is not resulting from a secondary mutation, the mutant was complemented with the arabinose-inducible vieA plasmid. Transcription of vpsR and vpsA were restored to wild type levels upon induction of VieA with arabinose, but not in a pBAD33 vector control (data not shown). These results confirm that deletion of vieA is responsible for the defect in vpsR expression.

VieA is a repressor of biofilm formation

To confirm that VieA-regulation of vps genes is relevant for biofilm formation, both quantitative crystal violet staining and qualitative fluorescence microscopy were performed. A ΔflaA (aflagellate) strain that does not form mature biofilms because it fails to attach to the surface (Watnick and Kolter, 1999) was included in these experiments as a negative control. We observed twofold enhanced biofilm formation by the ΔvieA mutant compared to the wild type when bacteria were grown in LB without aeration (Fig. 3A). In addition, biofilm formation by the wild type strain was reduced threefold by arabinose-induced expression of VieA (Fig. 3B). For confocal fluorescent microscopy, biofilms were grown on glass coverslips and adherent bacteria were visualized by staining with DAPI. In three independent experiments, the ΔvieA mutant formed thicker biofilms than the wild type; the ΔvieA mutant biofilms retained normal architecture, with pillars of bacteria surrounded by fluid-filled channels [compare Fig. 3C (wild type) and 3D (ΔvieA)]. These results are consistent with VieA acting as a negative regulator of the vps genes that are responsible for synthesis of EPS and biofilm formation.

Fig. 3.

Fig. 3

Open in a new tab

Assays for biofilm formation.

A. The wild type, ΔvieA and ΔflaA strains were grown in LB without aeration and adherent bacteria were stained with crystal violet. Staining was quantified by solubilizing crystal violet in 100% ethanol and determining the absorbance at 570 nm.

B. Crystal violet staining and quantitation was performed as in part A for the wild type strain harbouring either pBAD33 or pBAD33::vieA grown in LB ± 0.2% arabinose.

C and D. Confocal fluorescence microscopy images of wild type and ΔvieA biofilms grown on coverslips and stained with DAPI. The centre of each figure is an XY-section through the bio-film (bar = 25 µm). Vertical sections through the biofilms are shown to the right and bottom.

A phosphodiesterase domain of VieA is responsible for vps regulation

The putative DNA-binding HTH motif of VieA was largely dispensable for regulation of vpsR, vpsA-Q and VC0928, suggesting that VieA has an activity other than DNA binding that is responsible for repressing gene expression. One possibility is that phosphorylation of VieA by the VieS sensor kinase is important for repression of the vps genes. VieA contains two putative phosphoreceiver domains; the N-terminal receiver is predicted to be functional, while a second pseudo-receiver immediately adjacent to the HTH motif is predicted to be inactive because it lacks several residues critical for phosphorylation. Point mutations that convert conserved aspartates to non-phosphorylatable residues were constructed to test whether phosphorylation at these sites is required for VieA activity. The D52A and D403N single and double mutants were generated and were tested for both vpsR and vpsA transcript levels by RPA. Each of the point mutant strains exhibited wild type levels of both transcripts (Fig. 2B, and data not shown), indicating that phosphorylation at the D52 and D403 residues is not essential for VieA-mediated repression of vps gene expression.

Because the domains of known function within VieA were not responsible for regulating the vps genes, we sought to define the minimal repressor domain by cloning truncated versions of VieA under the control of the pBAD33 arabinose-inducible promoter. These truncations were tested for the ability to repress activity of VC0928::lacZ in both the wild type and ΔvieA mutant backgrounds during growth in LB. Initially, we observed that the N-terminal 386 amino acids of VieA lacking the HTH motif were sufficient for repression (Fig. 4, row 3). To confirm that phosphorylation of the D52 residue is not important for VieA activity, a similar vector was constructed with the vieAD52A point mutation. Overexpression of the D52A truncation repressed VC0928::lacZ activity as efficiently as the wild type (Fig. 4, row 4). Next, a series of N- and C-terminal truncations of the active 1– 386 amino acid fragment of VieA were generated. A polypeptide corresponding to amino acids 110–386 of VieA was fully proficient for repressing activity of the VC0928::lacZ fusion (Fig. 4, row 8). Further deletions into this region from either the N- or C-terminus abrogated the ability of the protein to repress transcription of the fusion (Fig. 4, rows 5, 9, and 10). All N-terminal truncations of VieA were tagged at the C-terminus with a His6 epitope, allowing detection of the proteins by immunoblotting with monoclonal antibody against polyhistidine. All proteins were undetectable in the absence of arabinose, and were expressed at similar levels upon induction (data not shown).

Fig. 4.

Fig. 4

Open in a new tab

Repression of the VC0928::lacZ fusion by overexpression of VieA truncations. Regions of vieA encoding the indicated amino acids were cloned under the control of the arabinose-inducible ParaBAD promoter in pBAD33 and were introduced into wild type and ΔvieA V. cholerae containing the VC0928::lacZ fusion. Strains were grown overnight in LB (black and white solid bars) or LB + 0.2% arabinose (dark and light hatched bars) at 30°C with aeration and assayed for β-galactosidase activity. Values reported are the average ± standard deviation of at least three independent cultures. Protein domain abbreviations are phosphoreceiver (Rec), pseudo-phosphoreceiver (Rec*), phosphodiesterase (EAL), helix–turn–helix (HTH), and 6-Histidine tag (H6).

The portion of VieA cloned in plasmid pAT1568 encodes an EAL domain, which is predicted to function as a phosphodiesterase of c-di-GMP (Galperin et al., 2001). To confirm that the VieA EAL domain is necessary for regulation of the vps genes, an in-frame deletion of this domain (amino acids 135–375) was constructed on chromosome I. In addition, two highly conserved acidic residues (E170 and D286) that are predicted to be important for coordinating a metal ion involved in catalysis (Galperin et al., 2001) were mutated to alanine. The ΔEAL, E170A, and D286A mutants were analysed for both vpsA (data not shown) and vpsR (Fig. 2C) transcript levels by RPA. In all three mutants, transcription of vpsA and vpsR is increased compared with the wild type, similar to the full in-frame deletion of vieA. In addition, the ΔEAL, E170A and D286A mutations were constructed in the VC0928::lacZ strain background and tested for β-galactosidase activity. Similar to the results of the RPAs, VC0928::lacZ activity was increased in each of the EAL domain mutants (Table 1). Finally, VieA EAL domains containing the E170A and D286A point mutations were cloned into pBAD33 and tested for the ability to repress VC0928::lacZ activity. Neither the E170A nor the D286A point mutant was able to repress expression of the VC0928::lacZ fusion (Fig. 4, rows 11 and 12), even though the proteins were expressed at high levels (data not shown).

Other diguanylate cyclase or phosphodiesterase domains can regulate vps expression

To address whether regulation of the vps genes is specific to the VieA EAL domain or resulting from its activity as a putative c-di-GMP phosphodiesterase, EAL domains from two unrelated proteins were overexpressed from pBAD33 and assessed for the ability to repress VC0928::lacZ. VC0653 (rocS) was recently shown to be important for switching between the smooth and rugose colony morphologies (Rashid et al., 2003), a phenotype associated with EPS production, and is annotated as a putative c-di-GMP phosphodiesterase. VCA0785 has not been implicated in either vps gene regulation or biofilm formation. Both the VCA0785 and RocS proteins also contain a GGDEF domain N-terminal to the EAL domain. While the EAL domain alone of VCA0785 was able to repress VC0928::lacZ activity similar to overexpression of the VieA EAL domain, the EAL domain of RocS caused only a modest repression of VC0928::lacZ (Fig. 1C). However, overexpression of the RocS GGDEF and EAL domains together caused a dramatic decrease in VC0928::lacZ activity (Fig. 1C), suggesting that the RocS EAL domain is not functional when taken out of the context of the intact protein. These data indicate that repression of vps genes is not a function specific to the VieA EAL domain, but that other proteins with similar enzymatic activity can fulfil this role.

Because GGDEF domains are putative diguanylate cyclases that generate c-di-GMP from GTP, we tested whether overexpression of a GGDEF-family protein could activate VC0928::lacZ, similar to deletion of vieA. Expression of full-length VCA0956, a cytoplasmic protein that contains a GGDEF domain as the only conserved motif, resulted in increased activity of VC0928::lacZ in both the wild type and ΔvieA strain backgrounds (Fig. 1D). Notably, the wild type strain expressing VCA0956 exhibited VC0928::lacZ activity similar to activity in the ΔvieA mutant. Induction of VCA0956 caused a severe growth defect in the ΔvieA mutant background, compromising an accurate measurement of VC0928::lacZ activity. The growth defect could be because of high intracellular concentration of c-di-GMP resulting from synthesis in the absence of VieA-mediated hydrolysis. These results suggest that c-di-GMP synthesis induces vps gene expression, similar to deletion of the VieA EAL domain.

VieA phosphodiesterase activity regulates the intracellular concentration of c-di-GMP

Based on mutational analysis of vieA and on overexpression of other EAL and GGDEF domains, VieA appears to control vps gene expression by affecting c-di-GMP concentration. To detect this molecule, 2D-TLC was carried out on nucleotide extracts from cells pulsed with 32P-labelled orthophosphate in a phosphate-limiting minimal medium. VieA represses vps gene expression in this medium, as assessed by measuring VC0928::lacZ activity (data not shown). In extracts from the ΔvieA strain, a spot was reproducibly detected with Rf values (0.16 in NH4HCO3 dimension, 0.32 in KH2PO4 dimension, Fig. 5B and D) similar to published values for c-di-GMP [0.19 and 0.31, respectively (Ross, 1986)]. This spot was below the limit of detection in extracts from the wild type strain (Fig. 5A and C). In addition, a spot corresponding to c-di-GMP was observed when the VCA0956 GGDEF-family protein was expressed in the wild type background. The c-di-GMP spot was only present when protein expression was induced with arabinose and increased in intensity with time of induction and labelling (Fig. 5E – G). c-di-GMP was undetectable after arabinose induction of a strain harbouring the pBAD33 vector, indicating that VCA0956 expression is required for its synthesis (data not shown). The c-di-GMP is intracellular, as it was not detected in culture supernatants from labelled cells (data not shown).

Fig. 5.

Fig. 5

Open in a new tab

Two-dimensional thin layer chromatography to detect c-di-GMP.

A and B. Nucleotides were extracted from the wild type (A) and ΔvieA mutant (B) grown in MOPS + NRES with 32P-orthophosphate, spotted on TLC plates (origin at lower left corner), and developed in 0.2 M NH4HCO3, pH 7.8 in the first dimension (bottom to top) and 1.5 M KH2PO4, pH 3.65 in the second dimension (left to right). The white arrow indicates the spot corresponding to c-di-GMP. Black arrows indicate spots with Rf values similar to published values for GTP, GDP, and GMP.

C and D. Boxed area indicated in A and B, respectively, magnified and overexposed to show the c-di-GMP spot.

E–G. Nucleotides were extracted from the wild type strain harbouring pBAD33::VCA0956 grown in MOPS + NRES with 32P-orthophosphate and separated by 2D-TLC as in A and B. White arrows indicate the c-di-GMP spot.

E. uninduced, 1 h total labelling time.

F. induced 30 min (1 h total labelling).

G. induced 2 h (2.5 h total labelling).

H. 1D–TLC of snake venom phosphodiesterase (SVPD) and alkaline phosphatase (CIP) treated c-di-GMP eluted from 2D–TLC plates. Lane 1, no enzyme; lane 2, SVPD; lane 3, SVPD + CIP; lane 4, CIP.

To confirm that the spot observed by 2D-TLC is c-di-GMP, properties of the molecule were tested by enzyme digestion. Because c-di-GMP is a cyclic dinucleotide, it is resistant to phosphatase but sensitive to phosphodiesterase (Ross et al., 1986). While the spot identified by 2D-TLC was resistant to alkaline phosphatase (CIP), treatment with phophodiesterase (SVPD) resulted in a product that comigrates with GMP and that is sensitive to CIP (Fig. 5H). Taken together, these data indicate that the spot observed by 2D-TLC is c-di-GMP, and that its concentration is increased upon either deletion of vieA or overexpression of the VCA0956 GGDEF-family protein.

Discussion

In this report, we describe a screen for genes controlled by the two-component signal transduction response regulator VieA. We identify two genes, VCA0707 (putative uhpC) and VC0928, that are repressed 10- and 100-fold by VieA, respectively. Regulation of VC0928 was studied in detail because it is strongly repressed by VieA and is linked to genes implicated in EPS production and biofilm formation. Our results show that VieA is a negative regulator of VC0928 and of two large, neighbouring operons, vpsA-K and vpsL-Q, that are required for EPS production. In addition, VieA inhibits transcription of VpsR, a positive regulator of the vps genes. We therefore suggest that VieA controls expression of the vps operons indirectly by affecting transcription of vpsR. VieA does not regulate vpsR transcription by directly binding the vpsR promoter, however, because deletion of the VieA HTH DNA binding motif had little effect on vpsR transcript levels. Rather, based on truncation and deletion analysis, it is the VieA EAL domain, a putative c-di-GMP phosphodiesterase, that is required for transcriptional regulation of the vpsR gene. The modest (less than twofold) increase in vpsR transcription observed for the ΔvieA-HTH mutant may be because of auto-regulation of vieA transcription (Lee et al., 1998). Consistent with this, we observed that vieA transcript levels are reduced fourfold in the ΔvieA-HTH mutant compared to the wild type (data not shown). Presumably, VieA protein, and thus c-di-GMP phosphodiesterase activity, is also marginally reduced in the ΔvieA-HTH mutant, resulting in modest changes in vps transcription.

Several lines of evidence indicate that the enzymatic activity of the VieA EAL domain is required to repress EPS production. First, VC0928::lacZ activity was not repressed by EAL domains with either the E170A or the D286A point mutations, which should abrogate metal ion coordination (Galperin et al., 2001), indicating that the proteins are inactive. Second, overexpression of the EAL domains from either VC0653 (RocS) or VCA0785 repressed activity of VC0928::lacZ. Our experiments highlighted a difference between RocS, VCA0785, and VieA, because RocS required the GGDEF domain in cis for its activity. It is unclear why some EAL domain proteins might need a GGDEF domain for phosphodiesterase activity, but all three of the putative G. xylinus c-di-GMP phosphodiesterases contain a GGDEF domain N-terminal to the EAL domain (Tal et al., 1998). Third, we were able to alleviate repression of VC0928::lacZ activity by overexpression of a putative guanylate cyclase enzyme, VCA0956. Finally, we were able to demonstrate by 2D-TLC of nucleotide extracts that c-di-GMP concentration is increased in both the ΔvieA mutant and a strain overexpressing the VCA0956 GGDEF-family protein.

To confirm that the VieA EAL domain has c-di-GMP phosphodiesterase activity, we have attempted in vitro assays using the His6-tagged VieA EAL domain purified by several different affinity purification methods. This protein is active as a c-di-GMP phosphodiesterase in vivo because it reduces the concentration of c-di-GMP to a level undetectable by 2D-TLC when expressed in the ΔvieA background (data not shown). Thus far, however, we have been unable to recover this activity in vitro. It is possible that the VieA EAL domain requires a cofactor for activity that is not present in our in vitro reaction conditions. Alternatively, VieA may form a complex with another protein(s) in vivo to hydrolyse c-di-GMP.

Taken together, our data support the model depicted in Fig. 6. Increased c-di-GMP concentration, resulting either from reduced activity of the VieA EAL domain phosphodiesterase or increased activity of one or more GGDEF diguanylate cyclases activates expression of the vps genes. Because deletion of the VieA EAL domain derepressed transcription of the VpsR positive regulator, we suggest that the c-di-GMP signal modulates vpsR transcription to affect EPS production. A protein, or alternatively a regulatory sRNA, may sense c-di-GMP to regulate vpsR expression.

Fig. 6.

Fig. 6

Open in a new tab

Model of V. cholerae vps gene regulation by VieA and other GGDEF/EAL domain proteins. GGDEF domains synthesize c-di-GMP from GTP, while EAL domain proteins act as phosphodiesterases of c-di-GMP, degrading it to GMP. VieA is a major contributor to c-di-GMP hydrolysis under the growth conditions we have tested, so deletion of vieA leads to increased intracellular concentration of c-di-GMP. This affects transcription of vpsR, possibly by influencing the activity of a c-di-GMP sensing protein (factor X). Increased vpsR transcription leads to induction of the vpsA-Q gene cluster, production of EPS, and biofilm formation.

Our model predicts that any protein with diguanylate cyclase or c-di-GMP phosphodiesterase activity could be involved in modulating vps gene expression. V. cholerae encodes 41 proteins with a GGDEF domain and 22 proteins with an EAL domain (Galperin et al., 2001), yet VieA appears to be a major regulator of the vps genes under the growth conditions we have tested; other EAL domain proteins must either be poorly expressed or enzymatically inactive under these conditions. Although the experiments reported here were carried out in the V. cholerae classical biotype, data from others suggest that regulation of EPS production by c-di-GMP also occurs in the El Tor biotype (Bomchil et al., 2003; Rashid et al., 2003). VieA does not, however, play a central role in vps gene regulation in the El Tor biotype because deletion of vieA had no effect on VC0928::lacZ activity in several El Tor strain backgrounds during growth in common laboratory media (A.D.T. and A.C., unpublished data). We suggest that functionally redundant EAL domain proteins expressed in El Tor V. cholerae can compensate for loss of VieA.

The existence of multiple GGDEF and EAL domain proteins in V. cholerae suggests the potential for integration of multiple signals to modulate c-di-GMP concentration. Based on our data, amino acids are one such signal that is relayed through VieA. The VieS sensor kinase contains two motifs in its periplasmic domain with homology to amino acid binding proteins, and it is possible that VieS phosphorylates VieA in response to amino acid signals altering the enzymatic activity of the VieA EAL domain. Phosphorylation of the Caulobacter crescentus GGDEF-containing response regulator PleD is required for regulation of stalked cell development in this species, and presumably controls GGDEF domain activity (Aldridge et al., 2003). Because conserved aspartate residues within the phosphoreceiver domains of VieA were not required for vps regulatory activity, we suggest that unphosphorylated VieA is active as a c-di-GMP phosphodiesterase.

The c-di-GMP signal may coordinate regulation of other genes that are important for biofilm formation, such as those involved in metabolism and transport of the EPS sugars. The putative uhpC glucose-6-phosphate sensor, which was identified as VieA-repressed in our screen, may have an indirect role in biofilm formation because the uhpC::lacZ transposon insertion mutant exhibited a modest defect in biofilm formation (data not shown). We are in the process of identifying other genes and proteins that are regulated by c-di-GMP. Our preliminary results indicate that greater than 20 secreted and cell envelope proteins, in addition to those reported here, are regulated by c-di-GMP (A.D.T., A.C. and F.H. Yildiz, unpublished data).

Although our data show that c-di-GMP regulates vpsR transcription, we are lacking information about how this is accomplished. We suggest that the activity of a protein necessary for regulation of vpsR is affected by binding c-di-GMP, similar to the case for cellulose synthase of G. xylinus (Weinhouse et al., 1997). VpsR itself may bind c-di-GMP because it contains a nucleotide binding motif typical of sigma-54 dependent transcriptional activators, but does not require sigma-54 to activate transcription (Yildiz et al., 2001). Perhaps VpsR senses c-di-GMP through this motif, and activates transcription of its own gene and that of the vps genes. Because VpsR has a phosphoreceiver motif, a second possibility is that phosphorylation controls VpsR activity. In this scenario, a sensor kinase may activate VpsR by phosphorylation in the presence of c-di-GMP. Studies are in progress to identify the factor that senses c-di-GMP to regulate EPS production.

Experimental procedures

Growth conditions

Bacteria were grown in LB broth with aeration at 37°C unless otherwise noted. M9 salts plus 0.5% glycerol, trace metals (1 ml l−1 of 5% MgSO4, 0.5% MnCl24H2O, 0.5% FeCl3, 0.4% trinitriloacetic acid) (Callahan et al., 1971) and l-asparagine, l-arginine, l-glutamate, and l-serine each at 25 mM (M9 + NRES), was prepared as previously described (Miller and Mekalanos, 1988). For 32P labelling of nucleotides, bacteria were grown in a similar medium using MOPS salts as the base (Sambrook and Russell, 2001). Arabinose was added at 0.2% final concentration to induce expression from the ParaBAD promoter. Antibiotics were added at the following concentrations unless otherwise noted: streptomycin (Sm) 100 µg ml−1; ampicillin (Ap) 50 µg ml−1; kanamycin (Km) 100 µg ml−1; chloramphenicol (Cm) 10 µg ml−1.

Plasmid and strain construction

All strains and plasmids used in this study are listed in Table 2. Plasmids with oriR6K were propagated in E. coli DH5αλpir; all other plasmids were propagated in E. coli DH5α. Detailed descriptions of strain and plasmid construction are provided in the Supplementary material. In addition, sequences of oligos are provided in Table A1 in the Supplementary material.

Table 2.

Bacterial strains and plasmids used in this study.

Strain or plasmid Relevant genotype and phenotype Source or reference
E. coli strains
    DH5α F-Δ(lacZYA-argF)U169 recA1 endA1 hsdR17 supE44 thi-1 gyrA96 relA1 (Kolter et al., 1978; Hanahan, 1983)
    DH5αλpir F-Δ(lacZYA-argF)U169 recA1 endA1 hsdR17 supE44 thi-1 gyrA96 relA1 λ::pir (Kolter et al., 1978; Hanahan, 1983)
    SM10λpir thi recA thr leu tonA lacY supE RP4-2-Tc::Mu λ::pir Laboratory strain
    AC-E1380 SM10λpir (pAT1379), Apr This work
V. cholerae strains
    O395 classical biotype, Smr (Taylor et al., 1987)
    AC-V61 O395 lacZ res-tet-res, spontaneous partial deletion of lacZ, Smr, Tcr (Camilli and Mekalanos, 1995)
    AC-V1198 AC-V61ΔvieAB, Smr, Tcr This work
    AC-V1200 AC-V61ΔvieB, Smr, Tcr This work
    AC-V1212 AC-V61ΔvieA-HTH, Smr, Tcr This work
    AC-V1386 AC-V61ΔvieA, Smr, Tcr This work
    AC-V1487 AC-V61 vieAD52A, Smr, Tcr This work
    AC-V1543 AC-V61 vieAD403N, Smr, Tcr This work
    AC-V1544 AC-V61 vieAD52A,D403N, Smr, Tcr This work
    AC-V1728 AC-V61ΔvieA135–375 (ΔEAL), Smr, Tcr This work
    AC-V1596 AC-V61 vieAE170A, Smr, Tcr This work
    AC-V1597 AC-V61 vieAD286A, Smr, Tcr This work
    AC-V1539 AC-V61ΔflaA, Smr, Tcr This work
    AC-V1351 AC-V1198 pBAD33, Smr, Tcr, Cmr This work
    AC-V1352 AC-V1198 pAT1337, Smr, Tcr, Cmr This work
    AC-V1353 AC-V1198 pAT1338, Smr, Tcr, Cmr This work
    AC-V1409 AC-V1352 uhpT::mTn5-lacZ, Smr, Tcr, Kmr, Cmr This work
    AC-V1420 AC-V1351 uhpT::mTn5-lacZ, Smr, Tcr, Kmr, Cmr This work
    AC-V1421 AC-V1353 uhpT::mTn5-lacZ, Smr, Tcr, Kmr, Cmr This work
    AC-V1407 AC-V1352 VC0928::mTn5-lacZ, Smr, Tcr, Kmr, Cmr This work
    AC-V1415 AC-V1351 VC0928::mTn5-lacZ, Smr, Tcr, Kmr, Cmr This work
    AC-V1416 AC-V1353 VC0928::mTn5-lacZ, Smr, Tcr, Kmr, Cmr This work
    AC-V1417 AC-V61 VC0928::mTn5-lacZ, Smr, Tcr, Kmr This work
    AC-V1418 AC-V1386 VC0928::mTn5-lacZ, Smr, Tcr, Kmr This work
    AC-V1419 AC-V1200 VC0928::mTn5-lacZ, Smr, Tcr, Kmr This work
    AC-V1729 AC-V1593 VC0928::mTn5-lacZ, Smr, Tcr, Kmr This work
    AC-V1598 AC-V1596 VC0928::mTn5-lacZ, Smr, Tcr, Kmr This work
    AC-V1614 AC-V1597 VC0928::mTn5-lacZ, Smr, Tcr, Kmr This work
    AC-V1460 AC-V61 pBAD33, Smr, Tcr, Cmr This work
    AC-V1461 AC-V61 pAT1337, Smr, Tcr, Cmr This work
    AC-V1463 AC-V1386 pBAD33, Smr, Tcr, Cmr This work
    AC-V1464 AC-V1386 pAT1337, Smr, Tcr, Cmr This work
    AC-V1726 AC-V61 pAT1662, Smr, Tcr, Cmr This work
Plasmids
    pUTmTn5Km2-STM oriR6K mobRP4 mTn5Km2, Kmr Apr (Hensel et al., 1995)
    pAT1379 pUTmTn5Km2::lacZ, Kmr Apr This work
    pCVD442 oriR6K mobRP4 sacB, Apr (Donnenberg and Kaper, 1991)
    pAC274 pCVD442::ΔvieAB, Apr (Lee et al., 1998)
    pSL108 pCVD442:ΔvieB, Apr (Lee et al., 1998)
    pAT1211 pCVD442::ΔvieA-HTH, Apr (Tischler et al., 2002)
    pAT1385 pCVD442::ΔvieA, Apr This work
    pAT1486 pCVD442::vieAD52A, Apr This work
    pAT1542 pCVD442::vieAD403A, Apr This work
    pAT1591 pCVD442::ΔvieA135–375 (ΔEAL), Apr This work
    pAT1594 pCVD442::vieAE170A, Apr This work
    pAT1595 pCVD442::vieAD286A, Apr This work
    pSL140 pCVD442::ΔflaA, Apr (Lee et al., 2001)
    pCR-script f1(+) ori, Apr Stratagene
    pGEM-T f1 ori, Apr Promega
    pAT856 pGEM-T:: ‘rpoB’, Apr (Tischler et al., 2002)
    pAT1454 pGEM-T:: ‘vpsA’, Apr This work
    pAT1455 pGEM-T:: ‘VC0928’, Apr This work
    pAT1456 pGEM-T:: ‘vpsL’, Apr This work
    pAT1509 pGEM-T:: ‘vpsR’, Apr This work
    pAT1537 pGEM-T:: ‘vieA’, Apr This work
    pBAD33 pACYC184 ori araC ParaBAD, Cmr (Guzman et al., 1995)
    pAT1337 pBAD33::vieA, Cmr This work
    pAT1338 pBAD33::vieB, Cmr This work
    pAT1535 pBAD33::vieA-His6, Cmr This work
    pAT1536 pBAD33::NTvieA-His6, Cmr This work
    pAT1561 pBAD33::NTvieAD52A–His6, Cmr This work
    pAT1564 pBAD33::NTR1vieA, Cmr This work
    pAT1567 pBAD33::NTF2vieA-His6, Cmr This work
    pAT1568 pBAD33::NTF3vieA-His6, Cmr This work
    pAT1575 pBAD33::NTF4vieA-His6, Cmr This work
    pAT1581 pBAD33::NTF5vieA-His6, Cmr This work
    pAT1582 pBAD33::NTF1vieA-His6, Cmr This work
    pAT1583 pBAD33::NTF6vieA-His6, Cmr This work
    pAT1615 pBAD33::NTF3vieAE170A–His6, Cmr This work
    pAT1616 pBAD33::NTF3vieAD286A–His6, Cmr This work
    pAT1645 pBAD33::VC0653-GGDEF-His6, Cmr This work
    pAT1646 pBAD33::VC0653-EAL-His6, Cmr This work
    pAT1647 pBAD33::VC0653-GGDEF + EAL-His6, Cmr This work
    pAT1648 pBAD33::VCA0785-EAL-His6, Cmr This work
    pAT1662 pBAD33::VCA0956-His6, Cmr This work
Open in a new tab

Screen for VieA-regulated genes

Strain AC-V1352 [V. cholerae O395 ΔvieAB lacZ::res-tet-res (pBAD33::vieA); Smr, Tcr, Cmr] was grown to mid-exponential phase in LB and mixed 1:1 with exponentially growing E. coli Sm10λpir carrying the mTn5lacZ mini-transposon delivery vector. A total of 100 µl of the mating mix was spread on an LB plate and incubated for 1 h at 37°C. Bacteria were replica plated using Whatman paper to LB containing Sm (150 µg m−1), Km (180 µg ml−1), and Cm and incubated 16 h at 37°C to select V. cholerae exconjugates harbouring mTn5lacZ transpositions. Colonies were replica plated to LB plus Sm, Km, and Cm containing 40 µg ml−1 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside (X-gal) ± 0.2% arabinose and incubated 16 h at 30°C Replica plated colonies were examined for blue/white phenotype. Strains that exhibited colour differences were tested for sensitivity to Ap to confirm that they harboured an mTn5lacZ transposition and not an integration of the donor plasmid. The mTn5lacZ-chromosomal junction was identified in each Aps strain by arbitrary primed PCR and DNA sequencing using primers Arb1/5Lac and Arb2/5Lac2 as described (Merrell et al., 2002). β-galactosidase activity was measured after overnight growth in LB ± 0.2% arabinose at 30°C as described (Miller, 1972).

Ribonuclease protection assay

RPAs were performed essentially as described (Tischler et al., 2002) using total RNA isolated from strains grown in M9 + NRES at 30°C to late exponential phase (OD600 = 1.0). Riboprobes for VC0928, vpsA, vpsL, vpsR, vieA, and hapR were generated as described in supplementary materials and methods. The rpoB riboprobe was included in all hybridizations to serve as a loading control as described (Tischler et al., 2002). Data were collected with a phosphorimager and analysed with ImageQuant (Molecular Dynamics). Band intensities were normalized to the intensity of the rpoB band, and are reported as the average fold difference compared to wild type for at least two independent experiments.

Crystal violet staining of biofilms

Strains were grown 16 h in LB, diluted 1:50 into 0.5 ml LB plus antibiotics in borosilicate glass tubes, and incubated at room temperature without aeration for 24 h. The medium was aspirated and adherent biofiolms were gently washed three times with LB. Biofilms were stained with 0.5 mg ml−1 crystal violet for 5 min, then washed extensively with water. Bound crystal violet was solubilized with 1 ml of 100% ethanol and quantified by absorbance at 570 nm. All experiments were performed in triplicate. Images were acquired with a digital camera.

Confocal microscopy of biofilms

Bacteria were inoculated at OD600 = 0.01 in 7 ml LB plus antibiotics in a 50 ml Falcon tube containing a borosilicate glass coverslip and incubated at room temperature for 24 h without aeration. Biofilms were stained with 1 µg ml−1 4′,6-diamidino-2-phenylindole (DAPI) for 10 min, rinsed with LB, and mounted on hanging drop slides in LB. Biofilms were visualized at 60× magnification using an Odyssey confocal microscope (Noran). Images were captured with a CCD camera (Photometrics) and processed using METAMORPH software (Universal Imaging Corp.). The images presented are representative of three independent experiments.

2D–TLC

Bacteria were grown for 5 h in MOPS + NRES + 0.75 mM KH2PO4 at 30°C with aeration until an OD600 = 0.6 was reached. Cells were pelleted, resuspended in medium plus 100 µCi ml−1 32P orthophosphate (PerkinElmer), and incubated for 1 h to label nucleotides. For strains containing pBAD33 or a derivative, cultures were incubated without induction for 30 min to label nucleotides, after which the culture was split, and protein expression was induced in one half with 0.2% arabinose. Nucleotides were extracted in 1 M formic acid as described (Bochner and Ames, 1982). Polyethyleneimine(PEI)-cellulose TLC plates (Selecto Scientific) were prewashed by soaking for 15 min in dH2O and developing in dH2O. Samples (10 µl) were spotted in 0.5 µl aliquots on 10 cm × 10 cm plates. Plates were developed in 0.2 M NH4HCO3, pH 7.8 in the first dimension, soaked in 100% methanol for 20 min, air dried, and developed in 1.5 M KH2PO4, pH 3.65 in the second dimension. Plates were air dried, exposed to phosphor screens (Kodak), and data were collected with a phosphorimager and analysed with ImageQuant (Molecular Dynamics).

To confirm the identity of the c-di-GMP spot, it was cut from the 2D-TLC plate and eluted in 20 µl of 0.5 M (NH4)2CO3. The eluate was tested for sensitivity to calf intestinal alkaline phosphatase (CIP, New England Biolabs) and snake venom phosphodiesterase I (SVPD, Amersham) in buffer containing 50 mM Tris, pH 8.5 and 10 mM MgCl2. Samples were spotted on PEI-cellulose TLC plates, developed in 1.5 M KH2PO4, pH 3.65, and detected as detailed above.

Acknowledgements

We thank P. Watnick, F. Yildiz, B. Belitsky, D. Merrell, and S. Butler for helpful discussions, K. Kierek for assistance with confocal microscopy, and M. Fisher for sequencing of mTn5lacZ insertions. We thank A. Sonenshein for critical reading of the manuscript. This material is based on work supported under an NSF Graduate Research Fellowship to A.D.T. This research was supported by NIH grant AI45746 to A.C. and the Center for Gastroenterology Research on Absorptive and Secretory Processes, NEMC (P30 DK34928).

Footnotes

Note added in proof

While this manuscript was in press, Paul et al., (2004) demonstrated that purified PleD, a C. crescentus GGDEF-containing response regulator, has guanylate cyclase activity in vitro. Cyclase activity was enhanced by phosphorylation of the receiver domain and required an intact GGDEF output domain. Paul, R., Weiser, S., Amiot, N.C., Chen, C., Schirmer, T., Giese, B., and Jenal, U. (2004) Cell cycle-dependent dynamic localization of a bacterial response regulator with a novel di-guanylate cyclase output domain.

Supplementary material

The following material is available from http://www.blackwellpublishing.com/products/journals/suppmat/mmi/mmi4155/mmi4155sm.htm

Appendix S1. Plasmid and strain construction and His6 Immunoblotting.

Table A1. Sequences of oligonucleotide primers used in this study.

References

  1. Aldridge P, Paul R, Goymer P, Rainey P, Jenal U. Role of the GGDEF regulator PleD in polar development of Caulobacter crescentus. Mol Microbiol. 2003;47:1695–1708. doi: 10.1046/j.1365-2958.2003.03401.x. [DOI] [PubMed] [Google Scholar]
  2. Ali A, Mahmud ZH, Morris JG, Jr, Sozhamannan S, Johnson JA. Sequence analysis of TnphoA insertion sites in Vibrio cholerae mutants defective in rugose polysaccharide production. Infect Immun. 2000;68:6857–6864. doi: 10.1128/iai.68.12.6857-6864.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Ausmees N, Jonsson H, Hoglund S, Ljunggren H, Lindberg M. Structural and putative regulatory genes involved in cellulose synthesis in Rhizobium leguminosarum bv. Trifolii Microbiol. 1999;145:1253–1262. doi: 10.1099/13500872-145-5-1253. [DOI] [PubMed] [Google Scholar]
  4. Ausmees N, Mayer R, Weinhouse H, Volman G, Amikam D, Benziman M, Lindberg M. Genetic data indicate that proteins containing the GGDEF domain possess diguanylate cyclase activity. FEMS Microbiol Lett. 2001;204:163–167. doi: 10.1111/j.1574-6968.2001.tb10880.x. [DOI] [PubMed] [Google Scholar]
  5. Bochner BR, Ames BN. Complete analysis of cellular nucleotides by two-dimensional thin layer chromatography. J Biol Chem. 1982;257:9759–9769. [PubMed] [Google Scholar]
  6. Boles BR, McCarter LL. Vibrio parahaemolyticus scrABC, a novel operon affecting swarming and capsular polysaccharide regulation. J Bacteriol. 2003;184:5946–5954. doi: 10.1128/JB.184.21.5946-5954.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bomchil N, Watnick P, Kolter R. Identification and characterization of a Vibrio cholerae gene, mbaA, involved in maintenance of biofilm architechture. J Bacteriol. 2003;185:1384–1390. doi: 10.1128/JB.185.4.1384-1390.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Callahan LT, Ryder RC, Richardson SH. Biochemistry of Vibrio cholerae virulence. II. Skin permeability factor/cholera enterotoxin production in a chemically defined medium. Infect Immun. 1971;4:611–618. doi: 10.1128/iai.4.5.611-618.1971. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Camilli A, Mekalanos JJ. Use of recombinase gene fusions to identify Vibrio cholerae genes induced during infection. Mol Microbiol. 1995;18:671–683. doi: 10.1111/j.1365-2958.1995.mmi_18040671.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Chang AL, Tuckerman JR, Gonzalez G, Mayer R, Weinhouse H, Volman G, et al. Phophodiesterase A1, a regulator of cellulose synthesis in Acetobacter xylinum, is a heme-based sensor. Biochemistry. 2001;40:3420–3426. doi: 10.1021/bi0100236. [DOI] [PubMed] [Google Scholar]
  11. D’Argenio DA, Calfee MW, Rainey PB, Pesci EC. Autolysis and autoaggregation in Pseudomonas aeruginosa colony morphology mutants. J Bacteriol. 2002;184:6481–6489. doi: 10.1128/JB.184.23.6481-6489.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Donnenberg MS, Kaper JB. Construction of an eae deletion mutant of enteropathogenic Escherichia coli by using a positive-selection suicide vector. Infect Immun. 1991;59:4310–4317. doi: 10.1128/iai.59.12.4310-4317.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Drenkard E, Ausubel FM. Pseudomonas bio-film formation and antibiotic resistance are linked to phenotypic variation. Nature. 2002;416:740–743. doi: 10.1038/416740a. [DOI] [PubMed] [Google Scholar]
  14. Galperin MY, Nikolskaya AN, Koonin EV. Novel domains of the prokaryotic two-component signal transduction systems. FEMS Microbiol Lett. 2001;203:11–21. doi: 10.1111/j.1574-6968.2001.tb10814.x. [DOI] [PubMed] [Google Scholar]
  15. Guzman L, Belin D, Carson MJ, Beckwith J. Tight regulation, modulation, and high-level expression by vectors containing the arabinose pBAD promoter. J Bacteriol. 1995;177:4121–4130. doi: 10.1128/jb.177.14.4121-4130.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hammer BK, Bassler BL. Quorum sensing controls biofilm formation in Vibrio cholerae. Mol Microbiol. 2003;50:101–114. doi: 10.1046/j.1365-2958.2003.03688.x. [DOI] [PubMed] [Google Scholar]
  17. Hanahan D. Studies on transformation of Escherichia coli with plasmids. J Mol Biol. 1983;166:557–580. doi: 10.1016/s0022-2836(83)80284-8. [DOI] [PubMed] [Google Scholar]
  18. Hensel M, Shea JE, Gleeson C, Jones MD, Dalton E, Holden DW. Simultaneous identification of bacterial virulence genes by negative selection. Science. 1995;269:400–403. doi: 10.1126/science.7618105. [DOI] [PubMed] [Google Scholar]
  19. Kaper JB, Rasano A, Trucksis M. Toxins of Vibrio cholerae. In: Wachsmuth KI, Blake PA, Olsvik O, editors. Vibrio Cholerae and Cholera: Molecular to Global Perspectives. Washington, D.C: American Society for Microbiology; 1994. pp. 145–176. [Google Scholar]
  20. Kolter R, Imizuka M, Helinski DR. Trans-complementation-dependent replication of a low molecular weight origin fragment from plasmid R6K. Cell. 1978;15:1199–1208. doi: 10.1016/0092-8674(78)90046-6. [DOI] [PubMed] [Google Scholar]
  21. Lee SH, Angelichio MJ, Mekalanos JJ, Camilli A. Nucleotide sequence and spatiotemporal expression of the Vibrio cholerae vieSAB genes during infection. J Bacteriol. 1998;180:2298–2305. doi: 10.1128/jb.180.9.2298-2305.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lee SH, Hava DL, Waldor MK, Camilli A. Regulation and temporal expression patterns of Vibrio cholerae virulence genes during infection. Cell. 1999;99:625–634. doi: 10.1016/s0092-8674(00)81551-2. [DOI] [PubMed] [Google Scholar]
  23. Lee SH, Butler SM, Camilli A. Selection for in vivo regulators of bacterial virulence. Proc Natl Acad Sci. 2001;98:6889–6894. doi: 10.1073/pnas.111581598. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Merrell DS, Hava DL, Camilli A. Identification of novel factors involved in colonization and acid tolerance of Vibrio cholerae. Mol Microbiol. 2002;43:1471–1491. doi: 10.1046/j.1365-2958.2002.02857.x. [DOI] [PubMed] [Google Scholar]
  25. Miller JH. Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press; 1972. [Google Scholar]
  26. Miller VL, Mekalanos JJ. A novel suicide vector and its use in construction of insertion mutations: osmoregulation of outer membrane proteins and virulence determinants in Vibrio cholerae requires Toxr. J Bacteriol. 1988;170:2575–2583. doi: 10.1128/jb.170.6.2575-2583.1988. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Pei J, Grishin NV. GGDEF domain is homologous to adenylyl cyclase. Proteins. 2001;42:210–216. doi: 10.1002/1097-0134(20010201)42:2<210::aid-prot80>3.0.co;2-8. [DOI] [PubMed] [Google Scholar]
  28. Rashid MH, Rajanna C, Ali A, Karaolis DKR. Identification of genes involved in the switch between the smooth and rugose phenotypes of Vibrio cholerae. FEMS Microbiol Lett. 2003;227:113–119. doi: 10.1016/S0378-1097(03)00657-8. [DOI] [PubMed] [Google Scholar]
  29. Reidl J, Klose KE. Vibrio cholerae and cholera: our of the water and into the host. FEMS Microbiol Rev. 2002;26:125–139. doi: 10.1111/j.1574-6976.2002.tb00605.x. [DOI] [PubMed] [Google Scholar]
  30. Romling U, Rohde M, Olsen A, Normark S, Reinkoster J. AgfD, the checkpoint of multicellular and aggregative behaviour in Salmonella typhimurium regulates at least two independent pathways. Mol Microbiol. 2000;36:10–23. doi: 10.1046/j.1365-2958.2000.01822.x. [DOI] [PubMed] [Google Scholar]
  31. Ross P, Aloni Y, Weinhouse H, Michaeli D, Weinberger-Ohana P, Mayer R, Benziman M. Control of cellulose synthesis in Acetobacter xylinum, a unique guanyl oligonucleotide is the immediate activator of the cellulose synthase. Carbohydr Res. 1986;149:101–117. [Google Scholar]
  32. Ross P, Weinhouse H, Aloni Y, Michaeli D, Weinberger-Ohana P, Mayer R, et al. Regulation of cellulose synthesis in Acetobacter xylinum by cyclic diguanylic acid. Nature. 1987;325:279–281. doi: 10.1038/325279a0. [DOI] [PubMed] [Google Scholar]
  33. Ross P, Mayer R, Weinhouse H, Amikam D, Huggirat Y, Benziman M, et al. The cyclic diguanylic acid regulatory system of cellulose synthesis in Acetobacter xylinum. J Biol Chem. 1990;265:18933–18943. [PubMed] [Google Scholar]
  34. Ross P, Mayer R, Benziman M. Cellulose biosynthesis and function in bacteria. Microbiol Rev. 1991;55:35–58. doi: 10.1128/mr.55.1.35-58.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Sambrook J, Russell DW. Molecular Cloning: a Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press; 2001. [Google Scholar]
  36. Shirtliff ME, Mader JT, Camper AK. Molecular Interactions in biofilms. Chem Biol. 2002;9:859–871. doi: 10.1016/s1074-5521(02)00198-9. [DOI] [PubMed] [Google Scholar]
  37. Tal R, Wong HC, Calhoon R, Gelfand D, Fear AL, Volman G, et al. Three cdg operons control cellular turnover of cyclic di-GMP in Acetobacter xylinum: Genetic organization and occurrence of conserved domains in isoenzymes. J Bacteriol. 1998;180:4416–4425. doi: 10.1128/jb.180.17.4416-4425.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Taylor RK, Miller VL, Furlong DB, Mekalanos JJ. Use of phoA fusions to identify a pilus colonization factor co-ordinately regulated with cholera toxin. Proc Natl Acad Sci. 1987;84:2833–2837. doi: 10.1073/pnas.84.9.2833. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Tischler AD, Lee SH, Camilli A. The Vibrio cholerae vieSAB locus encodes a pathway contributing to cholera toxin production. J Bacteriol. 2002;184:4104–4113. doi: 10.1128/JB.184.15.4104-4113.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Watnick PI, Kolter R. Steps in the development of a Vibrio cholerae El Tor biofilm. Mol Microbiol. 1999;34:586–595. doi: 10.1046/j.1365-2958.1999.01624.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Weinhouse H, Sapir S, Amikam D, Shilo Y, Volman G, Ohana P, Benziman M. c-di-GMP-binding protein, a new factor regulating cellulose synthesis in Acetobacter xylinum. FEBS Lett. 1997;416:207–211. doi: 10.1016/s0014-5793(97)01202-7. [DOI] [PubMed] [Google Scholar]
  42. Yildiz FH, Schoolnik GK. Vibrio cholerae O1 El Tor: Identification of a gene cluster required for the rugose colony type, exopolysaccharide production, chlorine resistance, and biofilm formation. Proc Natl Acad Sci. 1999;96:4028–4033. doi: 10.1073/pnas.96.7.4028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Yildiz FH, Dolganov NA, Schoolnik GK. VpsR, a member of the response regulators of two-component regulatory systems, is required for expression of vps biosynthesis genes and EPSETr-associated phenotypes in Vibrio cholerae O1 El Tor. J Bacteriol. 2001;183:1716–1726. doi: 10.1128/JB.183.5.1716-1726.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Zhu J, Mekalanos JJ. Quorum sensing-dependent biofilms enhance colonization in Vibrio cholerae. Dev Cell. 2003;5:647–656. doi: 10.1016/s1534-5807(03)00295-8. [DOI] [PubMed] [Google Scholar]
  45. Zhu J, Miller MB, Vance RE, Dziejman M, Bassler BL, Mekalanos JJ. Quorum-sensing regulators control virulence gene expression in Vibrio cholerae. Proc Natl Acad Sci. 2002;99:3129–3134. doi: 10.1073/pnas.052694299. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Zogaj X, Nimtz M, Rohde M, Bokranz W, Romling U. The multicellular morphotypes of Salmonella typhimurium and Escherichia coli produce cellulose as the second component of the extracellular matrix. Mol Microbiol. 2001;39:1452–1463. doi: 10.1046/j.1365-2958.2001.02337.x. [DOI] [PubMed] [Google Scholar]

RESOURCES