The c-di-GMP phosphodiesterase BifA regulates biofilm development in Pseudomonas putida

Authors

  • Alicia Jiménez-Fernández,

    1. Centro Andaluz de Biología del Desarrollo, Universidad Pablo de Olavide/Consejo Superior de Investigaciones Científicas/Junta de Andalucía, Sevilla, Spain
    2. Departamento de Biología Molecular e Ingeniería Bioquímica, Universidad Pablo de Olavide, Sevilla, Spain
    Search for more papers by this author
  • Aroa López-Sánchez,

    1. Centro Andaluz de Biología del Desarrollo, Universidad Pablo de Olavide/Consejo Superior de Investigaciones Científicas/Junta de Andalucía, Sevilla, Spain
    2. Departamento de Biología Molecular e Ingeniería Bioquímica, Universidad Pablo de Olavide, Sevilla, Spain
    Search for more papers by this author
  • Patricia Calero,

    1. Centro Andaluz de Biología del Desarrollo, Universidad Pablo de Olavide/Consejo Superior de Investigaciones Científicas/Junta de Andalucía, Sevilla, Spain
    2. Departamento de Biología Molecular e Ingeniería Bioquímica, Universidad Pablo de Olavide, Sevilla, Spain
    Current affiliation:
    1. Novo Nordisk Foundation Center for Biosustainability, Hørsholm, Denmark
    Search for more papers by this author
  • Fernando Govantes

    Corresponding author
    1. Centro Andaluz de Biología del Desarrollo, Universidad Pablo de Olavide/Consejo Superior de Investigaciones Científicas/Junta de Andalucía, Sevilla, Spain
    2. Departamento de Biología Molecular e Ingeniería Bioquímica, Universidad Pablo de Olavide, Sevilla, Spain
    • For correspondence. E-mail fgovrom@upo.es; Tel. +34-954977877; Fax +34-954349376.

    Search for more papers by this author

Summary

We previously showed the isolation of biofilm-persistent Pseudomonas putida mutants that fail to undergo biofilm dispersal upon entry in stationary phase. Two such mutants were found to bear insertions in PP0914, encoding a GGDEF/EAL domain protein with high similarity to Pseudomonas aeruginosa BifA. Here we show the phenotypic characterization of a ΔbifA mutant in P. putida KT2442. This mutant displayed increased biofilm and pellicle formation, cell aggregation in liquid medium and decreased starvation-induced biofilm dispersal relative to the wild type. Unlike its P. aeruginosa counterpart, P. putida BifA did not affect swarming motility. The hyperadherent phenotype of the ΔbifA mutant correlates with a general increase in cyclic diguanylate (c-di-GMP) levels, Congo Red-binding exopolysaccharide production and transcription of the adhesin-encoding lapA gene. Integrity of the EAL motif and a modified GGDEF motif (altered to GGDQF) were crucial for BifA activity, and c-di-GMP depletion by overexpression of a heterologous c-di-GMP phosphodiesterase in the ΔbifA mutant restored wild-type biofilm dispersal and lapA expression. Our results indicate that BifA is a phosphodiesterase involved in the regulation of the c-di-GMP pool and required for the generation of the low c-di-GMP signal that triggers starvation-induced biofilm dispersal.

Introduction

The alternation between individual free-swimming planktonic cells and the formation of structured, highly cooperative polymer-encased sessile communities, named biofilms, is a staple of bacterial life in the environment (Costerton et al., 1995). Biofilm formation is a form of coordinated collective behaviour that has been regarded as an evolutionary precursor of developmental processes. Biofilm formation proceeds through stages of adhesion, proliferation and microcolony formation and maturation, and is terminated by programmed biofilm dispersal (O'Toole et al., 2000; Monds and O'Toole, 2009). Transition through these stages requires the timely production of different factors in response to environmental and physiological traits occurring during the biofilm developmental cycle, and involves a variety of signal transduction and regulatory pathways to connect such traits to the adequate physiological responses (Monds and O'Toole, 2009).

Several factors relevant to biofilm development in the Gram-negative soil bacterium Pseudomonas putida have been identified (Reviewed by Klausen et al., 2006). The high molecular weight adhesin proteins LapA and LapF are important determinants for cell–surface and cell–cell interactions in early and mature biofilms respectively (Martínez-Gil et al., 2010). Flagella have been shown to contribute to initial surface attachment and the transition from microcolonies to the mature biofilm, but their relevance has not been unequivocally determined (Klausen et al., 2006). The extracellular matrix of P. putida biofilms has been shown to contain extracellular DNA (Steinberger and Holden, 2005) and a mixture of exopolysaccharides (EPS). Four gene clusters involved in EPS synthesis and export are present in P. putida and the contribution of different types of EPS to the extracellular biofilm matrix and biofilm stability has been determined (Camesano and Abu-Lail, 2002; Chang and Halverson, 2003; Chang et al., 2007; Nilsson et al., 2011).

P. putida biofilms undergo rapid dispersal in response to nutritional stress (Gjermansen et al., 2005). Studies performed in P. putida and the related species Pseudomonas fluorescens have determined that dispersal is effected by proteolytic cleavage of LapA by the periplasmic protease LapG. Dispersal conditions are signalled by a decrease in the intracellular cyclic diguanilate (c-di-GMP) concentration, which is sensed by the membrane-bound signal transduction protein LapD, in turn regulating the protease activity of LapG (Gjermansen et al., 2009; Newell et al., 2009; reviewed by Boyd and O'Toole, 2012). Despite the detailed knowledge of the effector system for biofilm dispersal, the pathways that regulate c-di-GMP synthesis and degradation in response to nutrient availability leading to activation of the dispersal response remain unknown.

The nucleotide c-di-GMP is a second messenger that is ubiquitously used in bacteria for signalling the transition between planktonic and sessile lifestyles. c-di-GMP is synthesized from GTP by diguanylate cyclase (DGC) activities, present in proteins with GGDEF domains and degraded by specific phosphodiesterase (PDE) activities, present in proteins with EAL or HD-GYP domains. Many bacterial genomes encode multiple proteins containing one or more of these domains, implying that c-di-GMP signalling is likely complex. Multiple pathways have been shown to regulate the c-di-GMP levels in different bacteria, and there is growing evidence of the arrangement of DGCs and PDEs along with c-di-GMP-sensing proteins in dedicated temporally or spatially isolated modules to control particular functions. The biology of c-di-GMP signalling has been extensively reviewed in recent years (Hengge, 2009; Mills et al., 2011; Boyd and O'Toole, 2012; Römling, 2012). The genome of P. putida KT2440 encodes 43 polypeptides potentially involved in c-di-GMP turnover (Ulrich and Zhulin, 2007). However, it was only recently that Matilla and colleagues (2011) characterized the product of locus PP4959, a DGC whose synthesis is induced in response to stationary phase, microaerobiosis and the presence of root exudates, and Österberg and colleagues (2013) characterized the product of locus PP2258, a dual activity DGC/PDE involved in the regulation of flagellar motility. To the best of our knowledge, no DGC or PDE has been unambiguously linked to the regulation of the biofilm developmental cycle in this organism.

BifA was characterized in Pseudomonas aeruginosa as a c-di-GMP PDE bearing an EAL and a modified GGDEF domain involved in the control of the intracellular concentration of c-di-GMP and the reciprocal regulation of biofilm formation and swarming motility (Kuchma et al., 2007). Recently, we described the isolation of two biofilm-persistent mutants bearing transposon insertions in the P. putida KT2442 bifA gene, encoding ortholog of P. aeruginosa BifA (López-Sánchez et al., 2013). Here we describe the phenotypic characterization of a ΔbifA mutant. Our results unveil a major role of BifA in c-di-GMP metabolism and the regulation of biofilm formation and dispersal.

Results and discussion

Phenotypic characterization of a P. putida ΔbifA mutant

We used allelic replacement to construct an unmarked deletion of bifA in the chromosome of P. putida KT2442. Initial characterization of the ΔbifA mutant MRB32 was performed by means of a serial dilution-based growth curve method we described recently (López-Sánchez et al., 2013), in which a dilution series is used to recapitulate the time course of planktonic and biofilm growth in microtiter plate wells using Luria–Bertani (LB) and K10T-1 as growth media (Fig. 1). When compared with the wild-type strain, the ΔbifA mutant growth curve showed two salient features: the biofilm dispersal response was not observed upon entry into stationary phase, and ∼ 2-fold-higher biofilm levels were observed in K10T-1 medium. Complementation analysis using a miniTn7 derivative expressing P. putida KT2442 bifA from its own promoter region confirmed that the biofilm dispersal defect and biofilm overproduction phenotypes in MRB32 are due to the absence of a functional bifA gene product (Supporting Information Fig. S1). These results are consistent with our previous observation that BifA inactivation prevents the onset of the starvation-induced biofilm dispersal response. In addition, BifA downregulates biofilm formation in K10T-1 medium.

Figure 1.

Dilution series-based growth curves of KT2442 and MRB32. Planktonic (left axes, open symbols) or biofilm growth (right axes, closed symbols) is plotted against the initial A600 of each dilution. Circles represent the wild-type KT2442 strain and squares represent the ΔbifA mutant MRB32.

A. Cells grown on 96-well microtiter plates containing LB medium or;

B. K10T-1 medium.

Plots show one representative experiment of at least three biological replicates. Error bars represent the standard deviation of six technical replicates. Full methods available as supporting information.

CV, crystal violet.

Pellicle formation (i.e., colonization of the interphase between the aqueous growth medium and the air) by the wild-type strain and the ΔbifA mutant was tested in shaking culture tubes containing 1% tryptone broth (Fig. 2A). While the wild-type strain showed little or no pellicle biomass, a large amount of pellicle material was evident in the ΔbifA mutant. In addition, turbidity of the growth medium was greatly diminished and a sediment was observed in the ΔbifA mutant culture, suggesting that most of the biomass was in the form of glass-attached pellicle and cell aggregates. Again, complementation analysis confirmed the involvement of BifA in this phenotype (Fig. 2B). These results indicate that BifA negatively regulates pellicle formation and aggregation.

Figure 2.

Pellicle and aggregation phenotypes of the bifA mutants.

A. Photographs of planktonic cultures of the wild-type, bifA:miniTn5-Km and ΔbifA mutants showing pellicle formation and culture clarification due to aggregation.

B. Complementation analysis, showing the wild-type and ΔbifA strains bearing the empty miniTn7BB-Gm transposon (Ø) or the miniTn7BB-bifA transposon expressing bifA (+BifA). Full methods available as supporting information.

P. aeruginosa BifA was shown to inversely regulate biofilm formation and swarming motility (Kuchma et al., 2007). However, swimming and swarming assays failed to reveal significant differences between the wild-type and ΔbifA P. putida strains (Supporting Information Fig. S2), indicating that BifA is not involved in the regulation of swarming motility in P. putida.

BifA downregulates EPS production and the synthesis of the large adhesin LapA

The biofilm overproduction, pellicle formation and cell aggregation phenotypes of the ΔbifA mutant are all suggestive of an increase in cell adhesiveness in this background. P. putida was previously shown to produce Congo red (CR)-binding EPS in response to elevated c-di-GMP levels (Gjermansen et al., 2006). To test whether BifA has an influence on the production of this matrix component, we exposed KT2442 and MRB32 cells to a CR solution, and quantified the amount of dye retained (Fig. 3A). In this assay, the ΔbifA mutant bound twice as much CR as the wild-type, suggesting that BifA negatively regulates the production of CR-binding EPS.

Figure 3.

Effect of the ΔbifA mutation on P. putida biofilm matrix components.

A. Normalized CR adsorption by stationary phase (overnight) wild-type (closed bar) and ΔbifA (open bar) cells.

B. Expression of the PlapA promoter assessed as β-galactosidase activity from stationary phase (24 h) cultures of the wild-type (closed bar) and ΔbifA (open bar) strains bearing the PlapA-gfp-lacZ fusion plasmid pMRB67. Bars represent the average and standard deviation of at least three independent assays. Full methods available as supporting information.

The outer membrane protein LapA is a critical determinant of irreversible attachment to surfaces as well as an essential component of the P. putida biofilm matrix (Espinosa-Urgel et al., 2000; Hinsa et al., 2003; Gjermansen et al., 2009). To test whether BifA may also influence the synthesis of LapA, expression of a transcriptional fusion of the lapA promoter region to the reporters gfp-mut3 and lacZ in plasmid pMRB67 was assessed from stationary phase cultures by means of β-galactosidase assays (Fig. 3B). Expression was increased twofold in the ΔbifA mutant relative to the wild-type, suggesting that BifA directly or indirectly contributes to negative regulation of lapA transcription. The fact that BifA negatively regulates the production of EPS and LapA, two matrix components critical for adhesion, provides a suitable rationale to the hyperadherent phenotype of the ΔbifA mutant.

P. putida BifA is a c-di-GMP PDE that regulates the intracellular c-di-GMP concentration

The P. putida BifA sequence displays 83% identity and 93% similarity to that of its P. aeruginosa counterpart, a c-di-GMP PDE that regulates the size of the intracellular c-di-GMP pool (Kuchma et al., 2007). Both proteins bear a highly conserved EAL domain and a less conserved GGDEF domain, with its signature sequence altered to GGDQF (Supporting Information Fig. S3). The PDE activity of the P. aeruginosa protein requires the integrity of both the EAL and GGDQF motifs (Kuchma et al., 2007). To test the relevance of these sequences to P. putida BifA function, its EAL and GGDQF motifs were modified to AAL and AAAQF or GEDQF, respectively, by site-directed mutagenesis, and the ability of these bifA derivatives to complement the dispersal defect of the ΔbifA strain MRB32 was tested by means of an end-point biofilm formation/dispersal assay (Supporting Information Fig. S4). The lack of complementation observed with all three mutants strongly suggests that P. putida BifA requires both the EAL and GGDQF motifs for activity.

To test a possible role of P. putida BifA in the control of the intracellular c-di-GMP concentration, we used plasmid pCdrA::gfpC (Rybtke et al., 2012), a recently developed fluorescent reporter of c-di-GMP levels containing a fusion of the c-di-GMP-induced P. aeruginosa PcdrA promoter to the table gfpmut3 allele (Borlee et al., 2010). Normalized GFP fluorescence was threefold elevated in the ΔbifA strain bearing pCdrA::gfpC relative to the wild-type (Fig. 4A), suggesting that inactivation of bifA provokes an increase in the size of the c-di-GMP pool. To test this hypothesis further, c-di-GMP was depleted by overproducing the Escherichia coli PDE YhjH in P. putida, as previously described (Gjermansen et al., 2006; Österberg et al., 2013). For this purpose, we produced YhjH from the salicylate-inducible promoter Psal in plasmid pMRB89 (Fig. 4B). PcdrA expression from the induced ΔbifA culture was decreased to levels similar to those in the wild-type, suggesting that elevated fluorescence of the reporter fusion in the ΔbifA strain is indeed due to increased c-di-GMP levels in this background. Taken together, our results strongly suggest that, similar to its P. aeruginosa counterpart, P. putida BifA is a c-di-GMP PDE involved in the regulation of the intracellular c-di-GMP levels.

Figure 4.

The interaction of c-di-GMP with BifA-dependent regulation.

A. Effect of BifA on the intracellular c-di-GMP levels. GFP fluorescence measurements from stationary phase (24 h) LB cultures of wild-type (open bar) and ΔbifA (closed bar) P. putida strains bearing pCdrA::gfpC.

B. Effect of overexpression of the PDE YhjH on the intracellular c-di-GMP levels. GFP fluorescence measurements from stationary phase LB cultures of wild-type and ΔbifA P. putida strains bearing pCdrA::gfpC and the empty vector pKT230 (Ø) or the Psal-yhjH plasmid pMRB89 (+YhjH) in the absence (open bars) or in the presence of the Psal promoter inducer, salicylate (closed bars).

C. Effect of c-di-GMP depletion on biofilm dispersal. Eight-hour biofilms of the wild-type biofilms of the wild-type, ΔbifA and lapG- strains bearing the empty vector pKT230 (Ø) or the Psal-yhjH plasmid pMRB89 (+YhjH) were treated with salicylate, and biofilm growth was determined after a final 24 h incubation. The plot represents biofilm biomass prior to induction (open bars) and after 24 h induction with 2 mM salicylate (closed bars).

D. Effect of c-di-GMP depletion on PlapA expression. Expression was assessed as β-galactosidase activity from stationary phase (24 h) cultures of the wild-type and ΔbifA strains bearing the PlapA-gfp-lacZ fusion plasmid pMRB67 and the empty vector pKT230 (Ø) or the Psal-yhjH plasmid pMRB89 (+YhjH) in the absence (open bars) or in the presence of the Psal promoter inducer, salicylate (closed bars). Bars represent the average and standard deviation of at least three independent assays. Full methods available as supporting information.

BifA-dependent changes in c-di-GMP levels regulate biofilm dispersal and LapA synthesis

Since the starvation-induced dispersal response is triggered by a decrease in the intracellular c-di-GMP concentration (Gjermansen et al., 2005), the dispersal defect of the ΔbifA mutant may be a consequence of its inability to generate this signal in response to nutrient limitation. To test this hypothesis, YhjH synthesis was induced in grown biofilms of the wild-type and ΔbifA strains, and biofilm biomass was monitored after induction (Fig. 4C). Biofilm cultures of the wild-type strain showed four to sixfold dispersal regardless of the presence of the YhjH-producing plasmid pMRB89, likely reflecting natural starvation-induced dispersal in these conditions. Biofilm dispersal was not observed in the ΔbifA mutant bearing the empty control plasmid. However, YhjH production in this background elicited a threefold decrease in biofilm biomass, consistent with induction of biofilm dispersal in response to YhjH-mediated c-di-GMP depletion. These results strongly suggest that BifA is required to decrease the intracellular c-di-GMP levels in response to nutritional stress during the dispersal response.

YhjH-mediated c-di-GMP depletion was also used to investigate the role of the intracellular c-di-GMP levels in BifA-mediated regulation of lapA transcription. Induction of YhjH synthesis resulted in decreased lapA expression in both the wild-type and ΔbifA backgrounds (Fig. 4D). These results strongly suggest that PlapA transcription is sensitive to the intracellular levels of c-di-GMP, and therefore the effect of BifA on PlapA expression is likely due to its ability to decrease the size of the intracellular pool of this second messenger.

BifA and the regulation of the biofilm developmental cycle

The present work provides evidence of the involvement of the P. putida protein BifA in the regulation of the biofilm developmental cycle. BifA was initially described in P. aeruginosa as a PDE that controls the size of the c-di-GMP pool (Kuchma et al., 2007). By showing that (i) P. putida BifA activity requires the same conserved motifs as P. aeruginosa BifA (Supporting Information Fig. S3); (ii) a P. putida ΔbifA mutant displays increased c-di-GMP levels (Fig. 4A); and (iii) c-di-GMP depletion rescues the phenotypes of a ΔbifA mutant (Fig. 4C and D), we provide evidence that P. putida BifA is likely a c-di-GMP PDE functionally similar to its P. aeruginosa counterpart. However, the set of functions regulated by both proteins appears to be different. P. aeruginosa BifA reciprocally regulates biofilm formation and swarming motility and no role in biofilm dispersal has been described (Kuchma et al., 2007). In contrast, our data support a dual role of P. putida BifA in the regulation of both biofilm formation and dispersal (Figs 1, 2 and 4), but do not provide evidence of its involvement in flagella-mediated motility phenomena (Supporting Information Fig. S2).

In P. putida, biofilm dispersal is triggered by a decrease in the intracellular c-di-GMP levels in response to nutrient limitation (Gjermansen et al., 2005; 2009). Several of our results provide evidence that BifA is required for the generation of such signal. First, a ΔbifA mutant fails to disperse the biofilm upon entry into stationary phase (Fig. 1). Second, the P. putida ΔbifA mutant shows elevated c-di-GMP levels, suggesting that BifA contributes to lower the c-di-GMP concentration (Fig. 3). Finally, ectopic expression of the heterologous c-di-GMP PDE YhjH in the ΔbifA mutant decreases the c-di-GMP levels and restores starvation-induced dispersal, suggesting a causal relationship between the elevated c-di-GMP concentration and the inability to disperse the biofilm by the ΔbifA strain (Fig. 4).

In addition to its function in biofilm dispersal, several phenotypic traits suggest an additional role of BifA in the early stages of biofilm development. The ΔbifA mutant displayed elevated biofilm levels in K10-T medium (Fig. 1B), high pellicle formation in the medium–air interface and an aggregative phenotype (Fig. 2). The cause of these phenotypes may be traced to increased levels of at least two biofilm matrix components, the large adhesin LapA and CR-binding EPS. By decreasing the intracellular c-di-GMP concentrations, BifA may restrict the amount of LapA exposed on the cell surface by two mechanisms. First, BifA counteracts positive c-di-GMP-dependent regulation of lapA transcription (Fig. 3B). Second, BifA may provoke partial release of LapD-mediated inhibition of LapG, thus promoting LapA proteolysis (Gjermansen et al., 2009). We have also shown that BifA prevents the accumulation of CR-binding EPS (Fig. 3A). Increased retention of CR in response to an elevation of the c-di-GMP levels was previously shown in P. putida, but only as a consequence of overexpression of a DGC activity (Gjermansen et al., 2006; Matilla et al., 2011). CR has been shown to bind strongly to cellulose and cellulose-like polysaccharides containing β-(1→4)-linked D-glucopyranosyl residues (Teather and Wood, 1982). Interestingly, Gjermansen and colleagues (2009) proposed that LapA interacts with a cellulase-sensitive polysaccharide to form a scaffold for the biofilm matrix. The fact that LapA and EPS are released from the cell surface in response to nutrient starvation in other P. putida and P. fluorescens strains lends further support to this notion (Kachlany et al., 2001; Gjermansen et al., 2009; Newell et al., 2009). We hope that future research will provide a better understanding of the mechanisms underlying these regulatory phenomena.

Acknowledgements

We wish to thank Tim Tolker-Nielsen (University of Copenhagen, Denmark) for providing plasmids, Søren Molin (Technical University of Denmark) and all members of the Govantes and Santero laboratories at CABD for providing materials and critical discussion. This work was funded by grant BIO2010–17853 of the Spanish Ministerio de Ciencia e Innovación, and a JAE-CSIC fellowship awarded to A.J.-F.

Ancillary

Advertisement