![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Microbiology and Immunology, Dartmouth Medical School, Hanover, New Hampshire 03755
Received 16 April 2007/ Accepted 11 June 2007
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
sadC mutant was additionally found to be deficient in EPS production and display altered reversal behavior while swimming in high-viscosity medium, two behaviors proposed to influence biofilm formation and swarming motility. Epistasis analysis suggests that the sadC gene is part of a genetic pathway that allows for the concomitant regulation of these aspects of P. aeruginosa surface behavior. We propose that SadC and the phosphodiesterase BifA (S. L. Kuchma et al., J. Bacteriol. 189:8165-8178, 2007), via modulating levels of the signaling molecule cyclic-di-GMP, coregulate swarming motility and biofilm formation as P. aeruginosa transitions from a planktonic to a surface-associated lifestyle. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
P. aeruginosa engages in swarming motility, allowing it to move across semisolid surfaces. The flagellum and surfactants, or surface wetting agents, are required for this process in P. aeruginosa (5, 20). The importance of swarming motility is not well understood; however, a recent report has suggested that it may play a role in determining the ultimate structure of biofilms formed by this microbe (48).
Biofilms have long been known as a predominant feature of the lifestyles of many bacteria (8). These microbial communities are common in a wide variety of both environmental and medical settings, where they can be especially problematic since bacteria within a biofilm have a reduced susceptibility to antimicrobial agents (27). The gram-negative organism P. aeruginosa has served as an important model system for understanding the formation of these microbial communities (21).
Biofilm formation is a stepwise process commencing when planktonic cells encounter a surface. Cells enter a transitional state of reversible surface attachment and, if biofilm formation is to proceed, these cells must stabilize this interaction with the substratum. To date, several factors, including the sadB, pelA, and motAB genes (3, 4, 52), have been identified that contribute to the progression toward irreversible attachment.
As the biofilm matures, cells aggregate into microcolonies and larger macrocolonies (19, 43), which are characteristically encased in an extracellular matrix. This material is thought to function in organizing and structuring the bacterial community (2). The P. aeruginosa biofilm matrix is comprised of multiple constituents, including proteinaceous materials (11, 54), membrane vesicles (44), DNA (56), and exopolysaccharides (EPS) (11, 12, 16, 29), although the relative contributions of each of these components to biofilm integrity are not yet fully understood. It is further becoming apparent that some components of the extracellular matrix also function to promote the early steps in biofilm formation (3, 16, 55, 56).
Emerging as an important signaling molecule in the control of aspects of the transition between a motile and a biofilm lifestyle is the intracellular molecule cyclic di-GMP (c-di-GMP). This messenger has been found in multiple systems to regulate motility and extracellular matrix production, two cellular outputs that influence biofilm formation. The existing paradigm in the field is that high concentrations of this molecule correlate with a sessile lifestyle (e.g., biofilm formation and EPS production), while its absence favors motility (e.g., twitching and swarming). The levels of c-di-GMP are enzymatically modulated by diguanylate cyclases (DGCs), proteins containing a GGDEF domain (37), and phosphodiesterases containing either an EAL (7, 49) or HD-GYP domain (42).
Here we report the characterization of the sadC gene, which encodes a membrane-localized diguanylate cyclase that is involved in biofilm formation in P. aeruginosa. We find that mutations in this gene influence both motility and EPS production, both of which may contribute to modulation of biofilm formation and swarming motility. Finally, we present genetic studies to place sadC within the context of the currently proposed pathway for reciprocally regulating biofilm formation and swarming motility.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Bacterial strains, plasmids, and primers. All bacterial strains, plasmids, and primers are listed in Table 1. Primers were obtained from Integrated DNA Technologies.
|
For quantitative real-time PCR (qRT-PCR), P. aeruginosa in static planktonic cultures or on plates was grown and harvested as previously described (3). Preparation of RNA and cDNA and analysis by qRT-PCR were performed according to published methods (23).
Biofilm and attachment assays. Static biofilm assays were performed in 96-well microtiter plates as previously described (31), followed by incubation at 37°C for 8 h unless otherwise indicated. To monitor the initial stages of biofilm formation, bacteria from a stationary-phase culture were diluted 1:100 in M63 minimal medium supplemented with glucose (0.2%), Casamino Acids (0.5%), and MgSO4 (1 mM). A 500-µl portion of this solution was pipetted into a sterile 24-well plate, followed by incubation at 37°C for 30 min for initial attachment or 1 h to assay reversible attachment (3). The medium was then gently replaced, and phase-contrast images of the surface-associated bacteria were captured at 60 frames/min at x1,400 magnification using OpenLab software. The images were converted into QuickTime movies and analyzed as reported (15).
Motility assays. Swarm assays were conducted as previously described on 0.5% agar plates (53). The area of the plate surface covered by the swarming bacteria was calculated by using ImageJ software (National Institutes of Health) as previously described (3). Directional reversals during swimming were quantified as described by Caiazza et al. (3) using M63 glucose-Casamino Acids medium as for biofilm assays. Ficoll was added to 15% to provide the desired viscosity in the medium. M63 glucose-Casamino Acids medium was solidified with 0.3% agar for swimming motility plates (35). Twitching motility assays were performed in LB medium solidified with 1.5% agar (34).
Assessment of EPS production. M63 salt solution was supplemented as for biofilm assays with the addition of CR (40 µg/ml) and Coomassie brilliant blue (20 µg/ml). For plate-based assay, the medium was solidified with 1% agar (11). Wells and plates were inoculated from overnight cultures grown in LB broth with antibiotics as appropriate, followed by incubation at 37°C for approximately 24 h or as indicated. CR plates were further incubated for 1 or 2 days at room temperature to improve color development.
Protein localization and detection. Strains were grown in LB medium supplemented with gentamicin (60 µg/ml) and arabinose (0.2%). Cells were harvested and resuspended in lysis buffer containing 50 mM Tris (pH 8), 150 mM NaCl, 5% glycerol, and a complete EDTA-free protease inhibitor cocktail tablet (Roche). Lysis was carried out with the addition of 300 µg of lysozyme/ml, followed by incubation at 4°C for 2 h, and samples were subsequently sonicated to fragment the DNA. Unbroken cells were pelleted by centrifugation at 13,000 x g. Cellular fractionations were performed based on the method of Nunn and Lory (33), as modified by Hinsa and O'Toole (15). Fractions were normalized based on protein concentration and analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis on a gradient (4 to 15%) gel, followed by Western blotting. Proteins were detected by using antibodies that recognize the His epitope (QIAGEN), SadB (4), and SecY (15) to distinguish the SadC protein, and cytoplasmic and inner-membrane fractionation controls, respectively.
Protein purification.
PleD*-His was purified essentially as previously described by using a 5-ml HisTrap FF column (GE Healthcare) (37). To purify the His-tagged SadC protein or His-tagged PleD-SadC chimeras, an overnight culture of wild-type (WT) PA14 carrying the appropriate plasmid was diluted into 1 liter of LB medium with gentamicin and grown until the optical density at 600 nm (OD600) reached 0.4. Arabinose was added to 0.2%, and cells were incubated in a shaking flask at 37°C to an OD600 of 
1.0. Cells were harvested and resuspended in binding buffer (20 mM sodium phosphate [pH 7.4], 0.5 M NaCl, 20 mM imidazole) to which an EDTA-free protease inhibitor (Roche) had been added. Triton X-100 (1%) and lysozyme were also included for the full-length His-tagged SadC construct. The PleD-SadC chimera-containing strains were lysed by passage through a French pressure cell, while the cells containing the full-length sadC construct were enzymatically lysed on a rocker at 4°C for 2 to 4 h. Sonication was used to decrease the viscosity of the lysate prior to clearing by centrifugation at 13,000 x g for 30 min. The supernatant was passed through a 0.22-µm-pore-size filter for clarification and then loaded onto a 5-ml HisTrapFF column (GE Healthcare). Purification was carried out according to the manufacturer's instructions. Fractions containing SadC were pooled, concentrated, and dialyzed as reported elsewhere (32).
Diguanylate cyclase assays. c-di-GMP was radiolabeled in vivo and separated from cellular nucleotides by thin-layer chromatography (TLC) on polyethyleneimine cellulose plates (Selecto Scientific) as described by Hickman et al. (13). Plates were exposed to a phosphor screen and analyzed as previously described (32).
In vitro enzyme assays were conducted essentially as reported by Paul et al. (37) and modified by Monds et al. (32) with incubation times of 2 and 24 h at room temperature. Reactions were stopped with addition of 10 µl of 0.5 M EDTA and an equal volume of running buffer (1:1.5 saturated NH4SO4 and 1.5 M KH2PO4 [pH 3.6]). Reaction products were resolved and analyzed as previously described (32).
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
PA4332 mutant was then tested for its ability to form a biofilm under static conditions as reported (36). When compared to WT P. aeruginosa PA14, the PA4332 mutant showed a defect in colonization of the surface after an 8-h incubation (Fig. 1A). This difference becomes less pronounced, although it is still evident, at 24 h.
|
PA4332 mutant strains, biofilm formation was stimulated in both of these backgrounds (Fig. 1B). In light of these data we have elected to rename the PA4332 open reading frame sadC (for surface attachment defective) to reflect its role in biofilm formation and relationship to previously described genes that when mutated yield similar phenotypes.
The fact that the sadC mutant was defective in the microtiter plate assay suggested that this mutation might disrupt early steps in biofilm formation. Accordingly, we further probed the role of the sadC gene in initial attachment and subsequent reversible or irreversible attachment phenotypes through microscopic analysis of these behaviors under similar static conditions.
When planktonic bacteria first attach to a surface, the association is a relatively unstable polar attachment event designated reversible attachment (43). This initial stage of biofilm formation was monitored by inoculating individual compartments of a 24-well plate and quantifying the attached bacteria after a 30-min incubation. We found that the sadC mutant was present on the substratum at levels indistinguishable from those of the WT, suggesting that it is not an inability to initiate attachment that leads to the observed defect in biofilm formation (Fig. 1C).
Because of its early biofilm defect, we also determined the role of sadC in the transition from reversible to irreversible attachment. Microscopic observation of cell-surface association was performed as previously described (3). Bacteria harboring a deletion in sadC are approximately twice as likely to be reversibly attached to the surface as the WT (P = 0.0007, Fig. 1D). These data indicate that the sadC mutant is defective in forming a stable association with the substratum compared to its parental strain.
The sadC mutant displays an increased swarming phenotype.
Published studies conducted in our laboratory (3, 4, 52) found that strains with mutations in the sadB, pelA, and motAB genes display a defect in the transition from reversible to irreversible attachment during biofilm development. Interestingly, these mutants have also been shown to share another trait: they are hyperswarmers. In light of these recent findings, we sought to determine whether the sadC mutant also had an altered swarming phenotype.
We tested swarming motility as previously described (5) by inoculating strains on 0.5% agar and incubating the swarm agar plates for 16 h. The sadC mutant formed swarms that were substantially larger than those of the WT strain (Fig. 2A and B). The sadC mutant was indistinguishable from WT for swimming and twitching motilities (data not shown).
|
sadC mutant backgrounds, expression of the sadC gene on a high-copy plasmid significantly decreases the extent of swarming (Fig. 2A and B).
Mutations in either the sadB, pilJ, or motAB genes were also previously shown to affect the reversal rates of single cells under conditions of high viscosity (3, 53), conditions that are likely analogous to those encountered during swarming motility (53). We counted the number of direction reversals during swimming in the presence of 15% Ficoll and determined that the sadC mutant exhibits a 50% increase in reversals/cell compared to the WT (P = 0.01, Fig. 2C). Taken together, mutating the sadC gene results in phenotypes similar to those observed for the previously reported sadB, pilJ, and motAB mutants, that is, decreased biofilm formation, increased swarming motility, and increased flagellar reversals in high-viscosity medium.
The sadC gene enhances CR staining in a pel-dependent manner. In addition to the aforementioned suite of phenotypes, it was also demonstrated that mutations in the sadB or pilJ genes led to decreased adsorption of the dye CR. CR adsorption has been shown to positively correlate with the presence of EPS in a number of bacterial species including P. aeruginosa (12, 26), Salmonella enterica serovar Typhimurium, and Escherichia coli (59). Furthermore, we noted that when strains maintaining a multicopy plasmid carrying the sadC gene were spotted onto an agar plate, they developed a wrinkled surface, a phenotype frequently associated with overproduction of EPS (13, 18, 58, 59). We therefore decided to assess whether SadC plays a role in the production of the polysaccharide component of the biofilm extracellular matrix. Strains were inoculated onto CR-containing plates and incubated as described previously (3). We observed that a strain harboring a deletion in the sadC gene absorbed less dye than the WT (Fig. 3A), a finding consistent with a diminution in EPS production in this strain. Strikingly, strains in which sadC is expressed from a multicopy plasmid show greatly enhanced amounts of dye bound (Fig. 3A, right panels). These results are consistent with a role for the sadC gene in enhancing production of a CR-binding component of the extracellular matrix.
|
We next assessed whether the difference in EPS production observed on agar plates was also discernible in the context of biofilm formation. By adding CR to our usual microtiter dish biofilm assay, we found that the presence of a sadC-containing plasmid enhanced CR staining at the air-liquid interface, the site of biofilm formation in this assay (Fig. 3B). This is true either in the WT or in the sadC mutant backgrounds, although there is slightly less total crystal violet (CV) staining in the sadC+psadC strain compared to WT+psadC (Fig. 3B). Although the pelA mutant forms some biofilm in this situation, CV staining is no longer stimulated by addition of psadC, nor is there any CR staining visible with this mutant (left panel), a finding consistent with the lack of EPS production.
SadC does not alter pelA/G mRNA levels. The experiments described above suggest a role for SadC in regulating the production of the pel-dependent polysaccharide matrix. In order to further probe this relationship, qRT-PCR was used to gauge whether transcriptional upregulation of the pelA-G locus might be involved in the apparent alteration in exopolysaccharide production observed in strains either lacking or carrying extra copies of the sadC gene.
We assessed transcript levels from the first and last genes in the pel locus: pelA, which shares similarity to endo-
-1,4-polygalactosaminidase and oligogalacturonide lyase (11), and pelG, a putative member of the PST family of proteins, which are generally involved in the translocation of glycolipid precursors out of the cytoplasm to position them for further processing (55). Mutations in either of these genes cause a loss of CR binding (11) and decreased biofilm formation in nonpiliated strains (55), and the pelA strain has also been shown to be defective for biofilm formation (3, 11, 22). Transcript levels from the pelA and pelG genes were unchanged (P = 0.09) by the deletion or overexpression of sadC compared to the vector control when grown on agar plates (Fig. 3C) or as a static planktonic culture (data not shown). These data suggest that the alterations in pel-dependent CR binding in these strains are not due to changes in pelA or pelG mRNA levels.
The SadC protein is localized to the cytoplasmic membrane. The TMpred prediction program calculates that the sadC open reading frame encodes an integral inner membrane protein with five transmembrane domains (http://www.ch.embnet.org/software/TMPRED_form.html). In order to verify this localization prediction, we performed a cellular fractionation of a strain expressing a C-terminally His-tagged SadC protein to isolate the outer and inner membranes, as well as the cytoplasmic compartment. These fractions were probed for appropriate localization controls, as well as an anti-His antibody to recognize the SadC-His protein.
We found that SadC-His can be detected in the inner membrane but not the cytoplasmic fraction, in agreement with the computer prediction (Fig. 4). Antibodies recognizing the SadB and SecY proteins reacted with bands of the expected sizes predominantly in the cytoplasm and inner membrane, respectively (1, 4, 14). We did, however, see the SecY antibody reacting with a protein in the cytoplasm as well. Since there is no cytoplasmic contamination from the SadC-His protein, the recognition of a cytoplasmic band by the SecY antibody may be due to cross-reactivity of the antibody rather than to substantial contamination of the cytoplasmic fraction with the inner membrane.
|
Due to the presence of a putative GGDEF domain, we next sought to determine whether the SadC protein possessed DGC activity. To begin to evaluate SadC's function, we first examined whether the expression of SadC could affect the cellular pools of c-di-GMP produced in vivo. Briefly, radiolabeled nucleotides were extracted from cells either harboring a vector control or a plasmid carrying the sadC gene and separated by two-dimensional TLC (Fig. 5A). Expression of the sadC gene from a multicopy plasmid coincides with an approximately 10-fold increase in the amount of c-di-GMP detected compared to the vector control (Fig. 5A and B).
|
To begin to examine the functionality of these constructs, both were expressed in the WT or sadC mutant backgrounds. Expression of D/C637 in the sadC mutant background resulted in an increase in CR binding and biofilm formation compared to the D/C700 construct (Fig. 5C). Similar results were obtained when expressing the D/C637 or D/C700 constructs in a WT strain (data not shown). These data provide evidence that the D/C637 protein may indeed contain an active GGDEF domain. In cellular fractionation experiments we found that for both of these constructs, while some of the protein is cytoplasmic as expected, a portion of the D/C637 and D/C700 proteins appears to be retained in the membrane fraction (data not shown).
We further explored the diguanylate cyclase activity of these constructs in vitro. The D/C637 and D/C700 chimeras were purified and assayed under conditions that support DGC activity of a PleD* control. We found that the D/C637 but not the D/C700 chimera supported detectable accumulation of c-di-GMP, as assessed by migration of this product by TLC compared to the PleD* control (Fig. 5D). The accumulation of c-di-GMP in this reaction is detectable at 2 h (top) and is more pronounced after a longer incubation period (lower panel). These data suggest that the GGDEF domain of SadC is capable of catalyzing the synthesis of c-di-GMP. Taken together with the in vivo labeling studies and CR data, this finding suggests that SadC is indeed a functioning DGC.
sadC interacts genetically with other biofilm-related genes. In exploring the phenotypes of the sadC mutant, it rapidly became obvious that the profile of phenotypes displayed was also shared by other mutants that had been studied in the laboratory, most notably in the sadB and pilJ genes. The sadB and pilJ genes were previously suggested to be in the same genetic pathway (3). We were impelled to consider whether these genes might be operating in a genetic pathway with the sadC gene, and thus we performed epistasis studies to explore this question.
We first investigated the relationship between the sadB and sadC genes. Multicopy sadC was unable to stimulate biofilm formation in the sadB mutant at early time points (Fig. 6A), a very different outcome than what we observed when the same plasmid is placed in a WT or sadC mutant background (compare with Fig. 1B). Overexpression of the sadB gene in a sadC mutant background was able to increase biofilm formation in this strain, completely rescuing the defect in biofilm formation (Fig. 6A). Furthermore, in a sadB mutant background, the expression of the sadC gene was prevented from enhancing CR binding at early time points (Fig. 6B), although a similar amount of CR adsorption was eventually attained. Taken together, these experiments suggest that SadB is important for SadC function, although such robust overexpression of a DGC may also affect accessory c-di-GMP responsive systems within the cell.
|
bifA mutant accumulates c-di-GMP and is a hyperbiofilm former. Mutating sadC in the bifA background reduced the biofilm formation of this strain, although not completely to the level of the WT (Fig. 6C). A similar trend was observed in respect to the CR binding and swarming phenotypes (Fig. 6D). In both cases, the phenotypes of the double mutants were intermediate between that of the bifA mutant and the WT. In combination, these data suggest that while BifA acts downstream of SadC, there are likely to be other DGCs contributing to the supply of c-di-GMP influencing these phenotypes.
Taken together, these data and previously published work (3) suggest the following ordered genetic pathway: sadC, bifA, sadB, pelA/pilJ. In addition, work presented by Kuchma et al. demonstrated that the sadB gene is required for the phenotypes of the bifA mutant, thus providing further evidence for the proposed genetic model (Fig. 7).
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Prior to and during the course of the present study, our laboratory identified several genes that appear to impact both swarming and biofilm formation. The similarity of these phenotypes, and subsequent epistasis analyses, suggested the possibility that these factors may operate in a common genetic pathway to regulate both biofilm formation and swarming motility (3, 4, 22). We used a genetic approach in order to better understand these findings and to identify additional components of the pathway regulating these surface-associated behaviors. This led us to identify the sadC gene, which when disrupted leads to alterations in both biofilm formation and swarming motility. Specifically, the observed decrease in biofilm formation and increase in swarming motility in the sadC mutant led us to contemplate the potential interrelation of these two aspects of P. aeruginosa biology. In thinking about this question, we first felt it essential to further dissect these two complex behaviors by assessing other contributing phenotypes. We ascertained that, relative to the WT, the sadC mutant produced a reduced amount of the Pel EPS matrix and underwent an abnormally high number of flagellar reversals during incubation in a high-viscosity medium. Could these two phenotypes assist in our understanding of the decreased biofilm formation and hyperswarming of the sadC mutant?
EPS production has previously been shown to play a role in both early and late stages of biofilm development (11, 16, 26, 55), but might it also act as a negative effector of swarming motility? One could easily imagine that bacteria producing an extracellular meshwork of polysaccharides might be hampered in their movements. In support of this idea, a mutation in the pelA gene, required for the manufacture of a glucose-rich polysaccharide, leads to elevated swarming in this organism (3). Thus, we can envision a role for the Pel polysaccharide matrix in both promoting biofilm formation and hindering swarming motility, although this idea requires further experimental verification.
Likewise, we contemplated the consequences of increased flagellar reversal rates in either swarming motility or biofilm formation. At this point, we can only speculate as to the role (if any) of changing flagellar reversal rates in these two surface behaviors. It has been suggested that in E. coli the modulation of flagellar reversals is indeed important in surface attachment (30), although the precise role of reversals is not yet understood. In an examination of motility in semisolid agar, Wolfe and Berg found a general trend that a greater reversal frequency correlated with an overall greater swarm rate. Those authors observed that direction changes prevent the bacteria from becoming "trapped" in the agar matrix (57). In the case of the Sadc mutant, we postulate that direction reversals help to "unstick" the bacteria from the surface, allowing for maximal forward progress (57). Perhaps the increased reversal behavior that promotes swarming motility hampers the ability of the microbe to make a stable commitment to a surface.
Whatever the mechanism, a clear correlation is emerging between the ability to control flagellar reversals and the surface-associated behavior adopted by P. aeruginosa. For example, mutations in the sadC, sadB, pilJ, and motAB genes all yield hyperswarming strains that are biofilm defective at the transition to irreversible attachment; all of these mutants also show increased flagellar reversal rates in high-viscosity medium (3, 4, 52). In contrast, in the accompanying manuscript, we show that a mutation in the BifA phosphodiesterase suppresses swarming motility, even in the absence of the Pel polysaccharide, and shows a 50% decrease in flagellar reversals compared to the WT (22). A further link between control of flagellar function and biofilm formation is presented in a recent report from our group, showing that cells harboring mutations in one of the flagellar stators, which show no defect in swimming motility, still result in reduced biofilm formation (52). Taken together, our data suggest that control of flagellar function, and in particular the rate of reversals, is key for the control of swarming motility and biofilm formation.
The presence of a GGDEF domain in the sadC gene product immediately leads us to reason that SadC might play a role in signaling rather than as part of the machinery directly functioning in flagellar function or EPS production. Proteins containing this motif have been shown to catalyze the synthesis of c-di-GMP, a molecule that was first identified for its role in allosterically regulating cellulose synthesis in Gluconacetobacter xylinus (formerly Acetobacter xylinum) (40, 41). Mutations in genes whose products synthesize or degrade this nucleotide are often associated with alterations in motility- and sessility-related phenotypes such as biofilm formation (25, 32, 50, 51), twitching motility (17), and production of exopolysaccharides (25). The presence of a GGDEF domain is consistent with a model in which SadC is part of a relay system responsible for transmission of an as-yet-unidentified environmental signal to the control of flagellar and polysaccharide-producing machinery of the cell via modulation of c-di-GMP levels.
To test the above hypothesis, we examined the ability of SadC to impact production of the signaling molecule c-di-GMP. Overexpression of SadC results in an increase in the total intracellular levels of c-di-GMP (Fig. 5A). In an in vitro DGC assay, we also found that a chimera protein containing the GGDEF domain of SadC is able to generate c-di-GMP compared to a second chimera missing the first 21 amino acids of the cognate GGDEF domain of SadC. Taken together, with the fact that sadC encodes an intact GGDEF domain, this is strong evidence that the SadC protein possesses the catalytic activity required to synthesize c-di-GMP. A previous study has also explored the diguanylate cyclase activity of SadC in the context of surveying all of the P. aeruginosa PA14 GGDEF or EAL domain-containing proteins. Kulasakara et al. (24) failed to observe production of c-di-GMP in an extract prepared from a strain carrying a plasmid expressing the sadC gene. Given the low levels of c-di-GMP we observe, this disparity may simply be attributable to differential detection limits of the respective methods utilized.
How do the sadC gene and its product fit into the known factors involved in early biofilm formation? We have shown that the sadC mutant is blocked in the transition from reversible to irreversible attachment during the early events in establishing a surface community. Through a series of epistasis experiments, we have placed sadC in the context of other genes that impact this step in the biofilm pathway. The sadB mutant is also defective in irreversible attachment and displays a hyperswarming phenotype. In a sadB mutant background, adding sadC in multicopy no longer enhances CR adsorption to the same degree as that observed in the WT. Likewise, there is a decreased stimulation of biofilm formation in the sadB mutant compared to what we normally observe in response to overexpression of the sadC gene. These data suggest that sadB is genetically downstream of sadC and that its product perhaps functions in transmission of the c-di-GMP signal to downstream effectors. Although a role for sadB in biofilm formation has been previously established (4), its biochemical activity has remained elusive. In addition, SadB has been shown previously to act upstream of pelA, which is required for production of the Pel polysaccharide, and pilJ, a member of the CheIV chemotaxis cluster, which is consistent with its role in controlling flagellar reversals (3). Deletion of the motAB genes, which comprise one of the two P. aeruginosa flagellar stators, also causes a defect in irreversible attachment, hyperswarming, and an increase in swim reversals, suggesting that these genes may be additional downstream candidates of this pathway (10, 52, 53). Alternatively, it may be that some of the components involved in the early stages of biofilm formation are instead functioning in convergent or parallel genetic pathways. Future studies will be required to obtain biochemical data to explain the underpinnings of the phenotypic relationships observed thus far.
In an accompanying manuscript, we show that BifA (22), which contains an inactive GGDEF domain and an EAL domain that functions as a phosphodiesterase, is capable of hydrolyzing c-di-GMP and that deletion of the bifA gene leads to an accumulation of this nucleotide within the cell, a hyperbiofilm phenotype, and complete loss of swarming motility (22). Deletion of the sadC gene in the bifA mutant results in partial restoration of swarming motility, a decrease in CR binding, and decreased biofilm formation. We interpret these data to mean that SadC contributes some but not all of the c-di-GMP to this system under these conditions. Therefore, we can further infer that c-di-GMP from at least one other DCG is also converging on BifA. A possible candidate for this other DGC is the WspR protein which is required for increased CR binding and attachment in a WspF mutant (13). Finally, it has been shown that the sadB gene is required for the phenotypes of the bifA mutant, supporting the conclusion that these genes are all in the same genetic cascade (22).
Based on the data presented here and in other recent studies, we can construct a model of the pathway by which SadC may be regulating aspects of motility and biofilm formation in P. aeruginosa (Fig. 7). We propose that the diguanylate cyclase SadC produces c-di-GMP in response to some as-yet-unknown signal such as contact with a surface or changes in medium viscosity. The amount of this signaling molecule accumulating can be attenuated by the phosphodiesterase BifA, providing a second control point whereby c-di-GMP pools are regulated. This c-di-GMP signal is then transmitted, perhaps via SadB based on our genetic studies, to the pel genes and/or members of the CheIV chemotaxis-like cluster, resulting in control of exopolysaccharide production and flagellar function. The mechanisms linking c-di-GMP pools to the control of EPS production and flagellar function are still under investigation since there are no known c-di-GMP binding proteins present in the genes we have examined. One tempting candidate for allosteric regulation of the Pel polysaccharide is PelD protein, which contains a putative GAF domain. In other systems GAF domains have been shown to bind the cyclic nucleotide cyclic AMP or cyclic GMP. Although studies of SadC and other members of this pathway (3, 22) are providing us with insight into the cellular functions coregulating biofilm formation and swarming motility as cells transition from a planktonic to a surface-associated lifestyle, there remains much to be elucidated in our understanding of this pathway.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Published ahead of print on 22 June 2007.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
