ABSTRACT
Cyclic di-GMP is a conserved signaling molecule regulating the transitions between motile and sessile modes of growth in a variety of bacterial species. Recent evidence suggests that Pseudomonas species harbor separate intracellular pools of c-di-GMP to control different phenotypic outputs associated with motility, attachment, and biofilm formation, with multiple diguanylate cyclases (DGCs) playing distinct roles in these processes, yet little is known about the potential conservation of functional DGCs across Pseudomonas species. In the present study, we demonstrate that the P. aeruginosa homolog of the P. fluorescens DGC GcbA involved in promoting biofilm formation via regulation of swimming motility likewise synthesizes c-di-GMP to regulate surface attachment via modulation of motility, however, without affecting subsequent biofilm formation. P. aeruginosa GcbA was found to regulate flagellum-driven motility by suppressing flagellar reversal rates in a manner independent of viscosity, surface hardness, and polysaccharide production. P. fluorescens GcbA was found to be functional in P. aeruginosa and was capable of restoring phenotypes associated with inactivation of gcbA in P. aeruginosa to wild-type levels. Motility and attachment of a gcbA mutant strain could be restored to wild-type levels via overexpression of the small regulatory RNA RsmZ. Furthermore, epistasis analysis revealed that while both contribute to the regulation of initial surface attachment and flagellum-driven motility, GcbA and the phosphodiesterase DipA act within different signaling networks to regulate these processes. Our findings expand the complexity of c-di-GMP signaling in the regulation of the motile-sessile switch by providing yet another potential link to the Gac/Rsm network and suggesting that distinct c-di-GMP-modulating signaling pathways can regulate a single phenotypic output.
INTRODUCTION
The secondary messenger molecule cyclic diguanylic monophosphate (c-di-GMP) has emerged in recent years as a key bacterial regulator of various processes, including virulence, differentiation, and biofilm formation (1–4). Its highly conserved role in facilitation of biofilm formation is associated with such processes as production of extracellular polymeric substance (EPS) matrices and proper localization of adhesins, as well as regulation of motility to enable transition to surface-associated growth. While high levels of c-di-GMP favor biofilm growth by promoting EPS production and suppressing motility, reductions in the level of the molecule favor free-floating, planktonic growth and dispersion of established biofilms (5–7). Cellular levels of c-di-GMP are modulated by the opposing activities of diguanylate cyclases (DGCs) and phosphodiesterases (PDEs) synthesizing and degrading c-di-GMP, respectively, with genomes of various bacterial species containing multiple predicted DGCs and PDEs (3, 8, 9). Although the positive correlation between sessile growth and c-di-GMP levels has been firmly established, it is becoming increasingly clear that c-di-GMP regulation is far more complex than previously thought. Mounting evidence has suggested that dedicated DGCs and PDEs modulate specific phenotypic outputs. Among such findings are the observations that inactivation of various DGCs results in similar reductions in total cellular c-di-GMP levels but correlates with distinct phenotypic manifestations (9, 10). Other findings have also suggested the importance of overall cellular c-di-GMP levels, with various c-di-GMP-modulating enzymes demonstrating the ability to impact the same cellular processes (11).
The finding that each of various Pseudomonas species harbors over 30 genes coding for proteins containing conserved DGC motifs suggests different roles for the putative c-di-GMP-synthesizing enzymes (8, 9). In Pseudomonas aeruginosa, at least five DGCs have been described to control the transition to surface-associated growth: WspR, SadC, RoeA, SiaD, and YfiN or TpbB (10, 12–16). WspR is the response regulator of the Wsp chemosensory system, which functions to produce c-di-GMP as its effector output. The Wsp system includes a predicted membrane-bound methyl-accepting chemotaxis protein (MCP)-like receptor (WspA), CheW-like scaffolding proteins (WspB and WspD), a CheA/Y hybrid histidine sensor kinase with a receiver domain (WspE), a methyltransferase CheR homologue (WspC), a methylesterase CheB homologue (WspF), and the DGC WspR (12, 17). In response to surface contact, signals are relayed through the P. aeruginosa PAO1 Wsp system to phosphorylate and activate WspR, which subsequently produces c-di-GMP to regulate cell aggregation and biofilm formation, at least partially through the regulation of expression of the EPS gene operons pel and psl (12, 17). Analysis of this system provided support for the notion of spatial c-di-GMP regulation, with phosphorylated active WspR recently shown to form clusters within P. aeruginosa cells (18). Two additional DGCs that regulate the transition from the motile to the sessile lifestyle in P. aeruginosa are SadC and RoeA. Specifically, SadC modulates the transition from reversible to irreversible attachment via regulation of flagellar reversal rates and swarming motility, while RoeA regulates Pel polysaccharide production (10, 13). Although both proteins are associated with the inner membrane, their distinct roles correlate with different distributions throughout the cell: RoeA was found to have a patchy distribution throughout the cell, while SadC formed foci around the cell periphery (10). In addition to the DGCs promoting surface attachment and biofilm formation, a fourth P. aeruginosa DGC, SiaD, was found to promote cellular aggregation in response to detergent-associated stress (19). More recently, SiaD and SadC were implicated in a positive feedback regulatory circuit between Psl polysaccharide and c-di-GMP production, with the two DGCs responsible for increases in c-di-GMP levels in response to extracellular Psl (14). Another P. aeruginosa cyclase, named either YfiN (16, 20) or TpbB (15), has been found to contribute to the regulation of Pel polysaccharide production and consequently biofilm formation, small colony variant formation, and persistence.
Similarly to P. aeruginosa, Pseudomonas fluorescens harbors over 20 genes encoding putative DGCs, with 4 confirmed to be involved in the regulation of surface attachment and/or biofilm formation. P. fluorescens harbors a homolog of the Wsp system, with WspR, associated with the wrinkly spreader phenotype, functioning in the regulation of production of acetylated cellulose and of cell attachment (21, 22). Systematic analysis of P. fluorescens DGCs revealed that inactivation of only four genes encoding putative DGCs, including wspR, resulted in reduced biofilm formation (9), with these DGCs regulating reversible-to-irreversible attachment transitions, adhesins, and motility. In addition to WspR, three novel DGCs were found to have distinct regulatory roles in promoting biofilm formation: GcbA mainly controlled swimming motility, GcbB preferentially affected localization of the LapA adhesin, and GcbC affected both LapA and motility (9). P. fluorescens uses c-di-GMP to regulate cell surface localization of the large adhesin protein LapA, which helps promote the transition from reversible to irreversible attachment and is exported out of the cell via an ATP-binding cassette (ABC) transporter composed of LapE, LapB, and LapC (23). Localization of the adhesin LapA is regulated by an inside out signaling mechanism, in which the inner membrane protein LapD, upon binding c-di-GMP, sequesters the periplasmic protease LapG, thus preventing the cleavage of and securing the cell surface localization of LapA. Newell and colleagues (9) demonstrated that the DGCs GcbB and GcbC are likely responsible for modulating the c-di-GMP bound by LapD to promote surface localization of LapA.
The inverse regulation of motility and EPS production, as well as distinct functions of multiple DGCs, during initiation of surface attachment and biofilm formation appears to be widely conserved, yet, with the exception of WspR, little work has been performed to study the potential homology or conservation of function of DGCs across Pseudomonas species. To date, no functional homolog of the LapA adhesin has been identified in P. aeruginosa. It is thus not surprising that the computational predictions from the Pseudomonas Genome database (www.pseudomonas.com) do not reveal orthologs of the LapA-controlling GcbB and GcbC DGCs in P. aeruginosa. However, P. aeruginosa does harbor a homolog of GcbA, the P. fluorescens DGC involved in the regulation of swimming motility, but not of LapA localization. This homolog of GcbA is conserved in a variety of P. aeruginosa strains, including PAO1, PA14, and PAK. Similarly to the P. fluorescens GcbA, this P. aeruginosa protein, encoded by PA4843, is a PleD-like response regulator consisting of a receiver domain and a DGC domain, harboring a conserved active site GGEEF motif. In a manner similar to gcbA, PA4843 is also located adjacent to a gene coding for the putative chemotaxis transducer CtpL (9, 24). Given the homology of PA4843 to P. fluorescens GcbA, as well as our recent findings suggesting that it may function as an active DGC (5, 25), we presently asked whether PA4843 is likewise involved in promoting biofilm formation by affecting swimming motility via regulation of intracellular c-di-GMP. For reasons discussed below and because P. fluorescens GcbA is able to complement the P. aeruginosa PA4843 mutant phenotypes to wild type, we named PA4843 GcbA.
MATERIALS AND METHODS
Bacterial strains, plasmids, media, and culture conditions.All bacterial strains and plasmids used in this study are listed in Table S1 in the supplemental material. P. aeruginosa strain PAO1 was used as the parental strain. All planktonic strains were grown in Lennox broth (LB) (BD Biosciences) or Vogel-Bonner minimal medium (VBMM) (26, 27) in shake flasks at 220 rpm in the absence or presence of 0.1 to 1.0% arabinose. Antibiotics were used at the following concentrations: 50 to 75 μg/ml gentamicin and 60 μg/ml tetracycline for P. aeruginosa and 100 μg/ml ampicillin, 20 μg/ml gentamicin, and 20 μg/ml tetracycline for Escherichia coli.
Strain construction.Isogenic mutants were constructed by allelic replacement using sucrose counterselection as previously described with the gene replacement vector pEX18Gm (28). Complementation and overexpression were accomplished by placing the respective genes under the control of an arabinose-inducible promoter in the pJN105 vector (29). Additionally, single-copy chromosomal complementation of the ΔgcbA mutation was accomplished by introducing gcbA under the control of its native promoter (−500 bp) into pMini-CTX (30). C-terminal V5/6×His tagging was accomplished by subcloning into pET101D (Life Technologies). The tagged constructs were then introduced into pJN105 or pMini-CTX. The identity of vector inserts was confirmed by PCR and sequencing. Plasmids were introduced into P. aeruginosa via conjugation. Primers used for strain construction are listed in Table S2 in the supplemental material.
Assessment of initial attachment.Initial attachment to a polystyrene surface was measured using the polystyrene microtiter dish assay system with crystal violet staining following 6 h of growth in LB medium, with the plates incubated at 220 rpm to ensure proper aeration and prevent cell lysis (31). An inoculum having an optical density at 600 nm (OD600) of ∼0.03 was used. Initial attachment to glass was visualized following 6 h of growth under flowing conditions in flow cell reactors (BioSurface Technologies) as previously described (32, 33). Flow cells were inoculated using 4 ml of overnight-grown cultures as previously described (34). The cells were viewed by transmitted light using an Olympus BX60 microscope and Olympus UPLNFLN 20× and 40× objectives. Images were captured using a ProgRes CF camera (Jenoptik, Jena, Thuringia) and processed via ProgRes CapturePro 2.7.7 software. The total number of cells attached to the glass surface of a flow cell covering an area of 400 μm2 was quantified using ImageJ software with the cell counter function. For quantitation of polar attachment rates, reversibly attached cells were defined as being attached to the surface via their poles, as evidenced by bacterial cells exhibiting rotation along the central axis, while irreversibly attached cells were defined as cells attached to the surface longitudinally along their entire length and exhibiting little to no motion.
Reversal rate measurements.Reversal rates, measured as previously described in references 32 and 35, report the frequency at which a motile cell changes its direction. Briefly, overnight cultures grown in LB medium were diluted 1:100 into M63 minimal salt medium supplemented with 1 mM magnesium sulfate and 0.2% glucose and containing 3% (low-viscosity, swimming conditions) or 15% (high-viscosity, swarming conditions) Ficoll. Following incubation for 2 h at 37°C, 300-μl aliquots were added to the wells of 96-well plates. The cells were observed by transmitted light using an Olympus BX60 microscope with an Olympus UPLNFLN 40× objective, and real-time videos were captured using a ProgRes CF camera and ProgRes CapturePro 2.7.7 software. The videos were subsequently used to monitor individual cells within the field of view for reversal of motility direction. Only cells remaining within the field of view for the duration of the video were considered. Reversal rates are reported as number of changes in motility direction per bacterial cell per minute.
Biofilm formation.Biofilms were grown under flowing conditions for a period of up to 6 days in flow cell reactors (BioSurface Technologies). Biofilms were grown at 22°C in 20-fold-diluted LB medium. Biofilm architecture was assessed via confocal laser scanning microscopy (CLSM) using a Leica TCS SP5 confocal microscope and the BacLight LIVE/DEAD viability kit (Life Technologies). The CLSM images were processed using LAS AF software v2.4.1.
Motility assays.Swarming motility was determined using M8 medium supplemented with 0.5% Casamino acids, 0.2% glucose, 1 mM magnesium sulfate, and 0.45%, 0.5%, or 0.55% agar (35, 36), while swimming motility was assessed using medium containing 0.1% tryptone, 0.5% sodium chloride, and 0.3% agar as previously described (10, 37). When necessary, the medium contained 75 μg/ml gentamicin and 1% arabinose for plasmid maintenance and induction of gene expression from the PBAD promoter, respectively. Stationary-phase cultures of the indicated strains grown in LB were stab-inoculated into the middle of petri dishes with the respective media and subsequently incubated upright for 24 and 48 h, at which time points, motility zone diameters were measured.
qRT-PCR.Quantitative real-time reverse-transcription PCR (qRT-PCR) was used to determine gene expression levels using 1 μg of total RNA isolated from the P. aeruginosa wild-type and indicated mutant strains grown planktonically in LB medium for 6 h. An initial inoculum having an OD600 of ∼0.03 was used. Isolation of mRNA and cDNA synthesis were carried out as previously described (37–40). qRT-PCR was performed using the Bio-Rad CFX Connect real-time PCR detection system and SsoAdvanced SYBR green supermix (Bio-Rad) with the oligonucleotides listed in Table S2 in the supplemental material. mreB was used as a control. Relative transcript quantitation was accomplished with CFX Manager software (Bio-Rad), by first normalizing transcript abundance (based on the threshold cycle [CT] value) to mreB followed by determining transcript abundance ratios. Melting curve analyses were employed to verify specific single-product amplification.
RT-PCR was used to determine whether PA4843 (gcbA) and PA4844 (ctpL) are cotranscribed. Following extraction from planktonic cultures, RNA was subjected to cDNA synthesis as described above, with the modification that a negative-control sample in the absence of reverse transcriptase was processed as well. Following cDNA synthesis, the samples were subjected to 35 cycles of PCR using the primer pair PA4843delF2/PA4844delR1 (see Table S2 in the supplemental material), with P. aeruginosa PAO1 genomic DNA used as a positive control.
In vivo quantification of c-di-GMP from P. aeruginosa.c-di-GMP was extracted in triplicate from the indicated strains grown planktonically for 6 h or as biofilms for 6 days using heat and ethanol precipitation (6) and quantitated essentially as previously described (5). Cells for initial extraction were adjusted to an equivalent of 1 ml of culture with an OD600 of 2.0 per extraction. Following the repeated three heat and ethanol extractions, supernatants were combined, dried using a Speed-Vac, and resuspended in 200 μl of 10 mM ammonium acetate. Samples (20 μl) were analyzed using an Agilent 1100 high-performance liquid chromatography (HPLC) device equipped with an autosampler, degasser, and detector set to 253 nm and separated using a reverse-phase C18 Targa column (2.1 by 40 mm; 5 μm) at a flow rate of 0.2 ml/min with the following gradient: 0 to 9 min, 1% B; 9 to 14 min, 15% B; 14 to 19 min, 25% B; 19 to 26 min, 90% B; 26 to 40 min, 1% B (buffer A, 10 mM ammonium acetate; buffer B, methanol plus 10 mM ammonium acetate). Commercially available cyclic di-GMP was used as a reference for the identification and quantification of cyclic di-GMP in cell extracts. Intracellular c-di-GMP content was further normalized to the total cell protein remaining after extraction, as determined using the modified Lowry protein assay kit (Thermo Scientific).
Diguanylate cyclase assays.Diguanylate cyclase assays were adapted from procedures previously described in references 9 and 41. Overnight cultures of E. coli DH5α harboring gcbA under the control of the arabinose-inducible PBAD promoter in the pJN105 vector or the empty pJN105 vector control were inoculated into fresh LB with 20 μg/ml gentamicin and grown for 3 h at 37°C, followed by addition to the medium of 1% arabinose and an additional 3-h incubation at 37°C. An initial inoculum having an OD600 of ∼0.05 was used. Following collection of cells via centrifugation at 16,000 × g for 5 min, total cell extracts were obtained by sonication as previously described (34) followed by centrifugation for 5 min at 21,200 × g to pellet unbroken cells. A total of 1 μg cell protein extract, as determined by the modified Lowry protein assay, was added to the reaction mixtures with 75 mM Tris-HCl at pH 7.8, 250 mM NaCl, 25 mM KCl, and 10 mM MgCl2 in a 200-μl total reaction volume. Reactions were started by the addition GTP at a final concentration of 25 μM. The reaction mixtures were incubated for 0, 60, and 120 min at 37°C, terminated by heating to 95°C, filtered through a 0.5-μm-pore syringe filter (Upchurch Scientific), and subsequently analyzed by HPLC as described above.
Statistical analysis.All statistical analyses were performed in Microsoft Excel using a two-tailed, Student's t test assuming equal variance or single-factor analysis of variance (ANOVA).
RESULTS
P. aeruginosa GcbA does not contribute to biofilm formation.GcbA is a PleD-like response regulator protein consisting of a response regulator receiver domain (Rec) and a diguanylate cyclase domain harboring a conserved active-site GGEEF motif (Fig. 1A). Our earlier studies indicated that overexpression of the gene encoding GcbA in P. aeruginosa results in a hyperaggregative phenotype (5, 25). Together with the predicted domain organization of the protein, these results suggested a possible role for GcbA in the regulation of c-di-GMP levels and/or surface-associated growth. As the P. fluorescens homolog GcbA (for the purpose of clarity, referred to as GcbA-Pfl below), exhibiting 74.12% sequence identity to the P. aeruginosa GcbA (E value, 0.0) (Fig. 1A; see Fig. S1A in the supplemental material), was found to regulate biofilm formation by affecting motility via regulation of the intracellular c-di-GMP pool (9), we asked whether GcbA plays a similar role in P. aeruginosa biofilm formation. Biofilms of the wild-type strain PAO1 and the isogenic gcbA mutant strain were cultivated under continuously flowing conditions over the course of 6 days, and subsequently biofilm formation by the two strains was analyzed by confocal microscopy. Inactivation of gcbA had no effect on biofilm development, as biofilms formed by the ΔgcbA strain following 144 h of growth under continuously flowing conditions did not exhibit any formation defects compared to PAO1. As determined by confocal microscopy and subsequent COMSTAT analysis, the mutant biofilms were characterized by microcolony formation, biomass accumulation, and overall biofilm architecture similar to those observed for the wild type (Fig. 1B to D). The findings of GcbA not affecting biofilm formation prompted us to assess the pattern of gcbA expression. Interestingly, when biofilms were grown for 144 h, the same time period used for confocal microscopy analysis, gcbA expression was reduced 7.59 ± 1.47-fold relative to that of cells grown planktonically, reinforcing the finding of an absence of GcbA effects during biofilm growth.
P. aeruginosa DGC impacts initial attachment to surfaces but not biofilm formation. (A) GcbA, a homolog of the previously characterized P. fluorescens DGC GcbA, is a PleD-like response regulator harboring conserved receiver (Rec) and DGC (GGDEF) domains. Numbers indicate amino acids corresponding to the start and end of the proteins and respective domains. (B) Confocal laser scanning microscopy (CLSM) images of PAO1 wild-type and ΔgcbA and ΔctpL mutant biofilm cells grown for 6 days under flowing conditions and stained with the LIVE/DEAD BacLight viability stain. Size bars, 100 μm. The CLSM was followed by COMSTAT analysis to evaluate biofilm biomass accumulation (C) and average and maximum thickness (D). (E) Evaluation of attachment to a polystyrene surface as determined by crystal violet (CV) staining following 6 or 24 h of growth under shaking conditions. Attachment is expressed as a percentage relative to that of the wild-type PAO1 control. (F) Growth curve of wild-type PAO1 and the isogenic ΔgcbA mutant strain. All assays were repeated at least three times, with at least 6 images acquired per CSLM analysis and at least 8 replicates used per attachment or growth curve assay. *, significantly different from PAO1 (P < 0.01).
GcbA plays a role in initial attachment to polystyrene surfaces.Our results suggested that despite a high degree of sequence homology, GcbA did not promote P. aeruginosa biofilm formation in the manner observed for P. fluorescens, a difference that may be attributed to different experimental conditions or distinct biofilm formation mechanisms utilized by the two Pseudomonas strains. Newell and colleagues (9) used a static microtiter plate assay to assess the effect of GcbA-Pfl on biofilm formation. Thus, we also assessed the effect of GcbA on surface adherence of P. aeruginosa cells in a microtiter plate system, with biomass accumulation assessed following 6 and 24 h of growth, corresponding to the stages of initial attachment to a surface and transition to biofilm formation postattachment, respectively. Inactivation of gcbA in P. aeruginosa PAO1, while not impacting growth rates in liquid (Fig. 1F), resulted in a significant 3-fold reduction in attachment to polystyrene relative to the wild type following 6 h of growth (Fig. 1E). Complementation (ΔgcbA/PgcbA-gcbA) restored attachment to wild-type levels, while overexpression (PAO1/pJN-gcbA) enhanced attachment more than 3-fold compared to that of the vector control (Fig. 1E). However, in contrast to the cells grown for 6 h, no difference in the attached biofilm biomass was noted in the gcbA mutant and overexpressing strains grown for 24 h relative to the respective wild-type and vector control strains (Fig. 1C). This observation was consistent with the finding that inactivation of gcbA does not affect biofilm development (Fig. 1B). Instead, our findings suggested GcbA is involved in the regulation of initial stages of P. aeruginosa attachment to surfaces.
The chemotaxis transducer CtpL is not required for GcbA effects on attachment.Similarly to gcbA-Pfl, the P. aeruginosa gcbA (gcbA-Paer) is located adjacent to a gene encoding the putative chemotaxis transducer CtpL (Fig. 2A), potentially involved in chemotaxis toward inorganic phosphate (9, 24). The proteins encoded by these loci, Pfl01_0622 in P. fluorescens and PA4844 in P. aeruginosa, are predicted to be homologs and share 65.6% sequence identity (E value, 0.0) (see Fig. S1 in the supplemental material). Using cDNA obtained from P. aeruginosa PAO1 cells grown to the exponential phase in LB, gcbA and ctpL were found to be cotranscribed (Fig. 2A). This prompted us to ask whether CtpL is involved in the regulation of initial attachment to surfaces in conjunction with GcbA. Inactivation of ctpL significantly reduced attachment by P. aeruginosa to polystyrene by approximately 40% relative to the wild type, while overexpression of cptL did not enhance attachment (Fig. 2B). In contrast to the ΔgcbA strain, however, biofilms formed by the ΔctpL mutant strain were characterized by significantly reduced biomass accumulation (Fig. 1B and C). Moreover, CtpL was found not to be required for GcbA to enhance attachment, as multicopy expression of gcbA in a ΔctpL mutant background resulted in over 2-fold-elevated attachment to polystyrene relative to the vector control (Fig. 2B). These findings suggested that while CtpL may contribute to the regulation of initial attachment to surfaces and subsequent biofilm formation, it is not essential for the GcbA-dependent enhanced-attachment phenotype.
gcbA is cotranscribed with ctpL, a chemotaxis transducer affecting initial surface attachment. (A) Genomic organization of gcbA and ctpL and RT-PCR results indicating cotranscription of gcbA and ctpL. P1 and P2 indicate the positions of primers PA4843delF2 and PA4844delR1, respectively (see Table S2 in the supplemental material), used to test cotranscription of gcbA and ctpL. cDNA, complementary DNA after reverse transcription; gDNA, genomic DNA used as a positive control. The absence (−) or presence (+) of reverse transcriptase is indicated. (B) Evaluation of attachment to a polystyrene surface by crystal violet (CV) staining following 6 h of growth under shaking conditions. Attachment is expressed as percentage relative to that of the wild-type PAO1 control. All assays were repeated at least three times. *, significantly different from PAO1 (P < 0.01).
GcbA regulates flagellar motility and flagellar reversal rates in a viscosity-independent manner.Initial attachment to surfaces and biofilm formation are driven by the inverse regulation of EPS production and motility (32, 42). As GcbA-Pfl was found to regulate flagellum-driven swimming motility without significantly affecting the LapA adhesin (9), we therefore asked whether the P. aeruginosa GcbA likewise contributes to the regulation of flagellar motility behaviors. Inactivation of gcbA resulted in small, but significant increases in both swimming and swarming motilities, while overexpression of gcbA suppressed both forms of flagellum-driven motility (Fig. 3A and C). When the ΔgcbA strain was tested for the expression of FleQ-regulated flagellum genes (fliE and fleS), no increase in transcript abundance was noted (Table 1). Interestingly, these genes appeared to be reduced in expression by approximately 2-fold in the ΔgcbA strain, suggesting that transcriptional regulation of these genes could not account for the observed swimming and swarming phenotypes.
GcbA suppresses swimming and swarming motilities in a manner independent of surface hardness via regulation of flagellar reversal rates. (A to C) The swarming and swimming motilities of the indicated strains were analyzed using 0.5% and 0.3% agar, respectively, following 24 h of growth. Additionally, the surface hardness dependence of GcbA effects on swarming was assessed by testing the motilities of the PAO1 wild-type and ΔgcbA mutant strains on 0.45, 0.5, and 0.55% agar. Motility assays were repeated at least three times, with a minimum of four plates used per strain-agar concentration combination. (D) Flagellar reversal rates in M63 medium with 3 or 15% Ficoll, representing swimming and swarming conditions, respectively, were measured as changes in direction of movement of cells. Rates are expressed as reversals per cell per minute. The assays were repeated at least three times, with five videos acquired per sample and 10 cells analyzed per video, for a total of 150 cells analyzed per strain-viscosity combination. *, significantly different from PAO1 or the respective vector control PAO1/pJN105 (P < 0.01).
Results from qRT-PCR analysis
Additional factors contributing to the regulation of swimming and swarming motilities include the flagellar stators MotAB and MotCD (35, 43), the signaling protein SadB (44), and chemotaxis cluster IV (32). However, their contribution to motility is limited to conditions of higher viscosity and increased surface hardness (32, 35, 43). To determine whether the effects of GcbA on attachment are also dependent on surface hardness and are thus potentially linked to these systems, swarming motility assays using increasing concentrations of agar (0.45, 0.50, and 0.55%) were performed. Under the conditions tested, the ΔgcbA strain demonstrated significantly elevated swarming motility relative to the wild-type strain PAO1 regardless of the agar concentration (Fig. 3B and C), suggesting that GcbA regulates swarming motility in a manner independent of surface hardness.
Regulation of the flagellum-dependent swimming and swarming modes of motility has furthermore been linked to the modulation of flagellar reversal rates, with dependency of surface hardness effects on motility correlating with dependency of viscosity effects on flagellum reversals (32, 43). Consequently, the effect of GcbA on reversal rates (i.e., the frequency at which a motile cell changes its movement direction) was tested in media with different viscosities mimicking swimming (3% Ficoll) and swarming (15% Ficoll) conditions (32). In agreement with GcbA affecting both swimming and swarming motilities and functioning in a viscosity-independent manner, inactivation of gcbA resulted in significantly elevated frequency of changes in cellular movement direction under both conditions tested compared to those of the wild-type strain (Fig. 3D).
GcbA promotes the transition from polar to longitudinal attachment.Regulation of flagellar reversal rates has been linked to the process of reversibly adhering cells transitioning to irreversible attachment to a surface (32, 35, 43). Transition from reversible to irreversible attachment has been associated with P. aeruginosa cells switching from polar to longitudinal attachment. Cells adhering only via their poles are capable of spinning along their axis and rapidly detaching from the surface. Longitudinally attached cells interact with the surface along their entire length and while still capable of exhibiting motility along the surface, are less likely to detach and return to the bulk fluid (32, 45–47). Considering that GcbA was found to affect flagellar reversal rates and that inactivation of gcbA only temporarily reduced attachment, the effect of GcbA on the switch from polar attachment to longitudinal attachment was assessed via bright-field microscopy. In accord with the microtiter plate attachment assays, significantly fewer attached ΔgcbA cells (30.9 ± 12.4 cells per 400 μm2) compared to the wild-type strain PAO1 (92.3 ± 34.0 cells per 400 μm2) were detected following 6 h of growth under flowing conditions (Fig. 4A and B). Quantitative analysis indicated that a significantly higher fraction of ΔgcbA cells compared to wild-type cells attached in a polar manner (Fig. 4C). While only a quarter of surface-associated PAO1 cells were attached via their poles, over 35% of ΔgcbA cells were polarly attached. Multicopy expression of gcbA from the pJN-gcbA vector significantly enhanced attachment of the ΔgcbA strain, with the numbers of cells attached to the surface following 1 or 6 h of incubation being too high to count (Fig. 4A). Therefore, the single-copy chromosomal complementation strain (ΔgcbA/PgcbA-gcbA) was tested for the polar-longitudinal attachment transition phenotype. Single-copy expression of gcbA complemented ΔgcbA, as indicated by the finding of the ΔgcbA/PgcbA-gcbA strain having rates of polar attachment comparable to those observed for the wild type (Fig. 4C). The finding suggesting that GcbA promotes the transition to more permanent surface contact was consistent with our observation of GcbA playing only a temporal role in attachment, with no effects observed following 24 h of growth.
GcbA mediates the transition from polar to longitudinal attachment. (A) Attachment of the PAO1 wild-type and ΔgcbA mutant strains to a glass surface following 6 h of growth under flowing conditions and of the ΔgcbA/pJN-gcbA strain following 1 or 6 h of growth, was assessed via bright-field microscopy, with images captured by using ProgRes CapturePro software. An example of reversibly or polarly attached cells is indicated with a dashed arrow, while that of irreversibly attached cells (adhered longitudinally along their entire length) is indicated with a solid arrow. (B) Total numbers of attached cells per 400 μm2 were quantified using ImageJ software with the cell counter tool. (C) Numbers of polarly, reversibly attached cells and irreversibly attached cells were quantified using ImageJ software with the cell counter tool. The rate of polar attachment is reported as a percentage of total attached cells. The assays were repeated at least in triplicate, with a minimum of 10 images acquired per strain per replicate. *, significantly different from PAO1 (P < 0.01).
P. fluorescens gcbA complements the P. aeruginosa ΔgcbA mutant strain.Given the sequence homology and similarity in the phenotypic outputs between GcbA and P. fluorescens GcbA (GcbA-Pfl), we next asked whether GcbA-Pfl is functional in P. aeruginosa and will rescue the P. aeruginosa ΔgcbA mutant phenotypes. When gcbA-Pfl under the control of the arabinose-inducible PBAD promoter within the pMQ72 vector was introduced into the P. aeruginosa gcbA mutant strain (referred to as the ΔgcbA-Paer strain in Fig. 5 and this section), swimming and swarming motilities were restored to wild-type levels (Fig. 5A and B). Multicopy expression of gcbA-Pfl in the wild-type PAO1 background, however, did not result in statistically significant differences relative to the vector control (Fig. 5A and B). Furthermore, multicopy expression of gcbA-Pfl also rescued initial attachment by ΔgcbA-Paer cells (Fig. 5C) and reduced the rate of polar surface attachment to wild-type levels (Fig. 5D). The findings indicated GcbA-Pfl to be a functional homolog of P. aeruginosa GcbA.
P. fluorescens gcbA rescues the P. aeruginosa gcbA mutant phenotypes. In order to test whether the P. fluorescens GcbA (GcbA-Pfl) is functional in P. aeruginosa and will rescue the P. aeruginosa ΔgcbA (ΔgcbA-Paer) phenotypes, gcbA-Pfl under the control of the arabinose-inducible PBAD promoter within the pMQ72 vector (pGcbA-Pfl) or the pMQ72 empty vector control was introduced into the P. aeruginosa PAO1 and the gcbA-Paer mutant strains. The respective strains were tested for swarming (A) and swimming (B) motilities, attachment to glass under flowing conditions (C), and rates of polar attachment (D). All assays were repeated at least in triplicate. *, significantly different from PAO1 (P < 0.01).
GcbA regulation of motility is independent of EPS production effects.Fine-tuning of early surface attachment has been linked to the inverse regulation of motility and EPS production. Therefore, we asked whether GcbA also affects EPS production and whether its regulation of motility is partially mediated via EPS effects. However, the reduction in attachment of the ΔgcbA strain following 6 h of growth (Fig. 1C) did not correlate with a reduction in Congo red binding (not shown). In accord with this, qRT-PCR analysis revealed no significant differences between the wild-type PAO1 strain and ΔgcbA mutant with respect to expression of the psl operon (Table 1) encoding the biosynthetic machinery for the predominant EPS polysaccharide of PAO1. Expression of the pel operon also did not appear to be significantly affected by GcbA (Table 1). Overexpression of gcbA similarly had no effect on psl gene expression, suggesting that GcbA does not regulate Psl abundance in PAO1 at the level of transcription.
Given these findings, we hypothesized that the effects of GcbA on motility are independent of EPS, with suppression of motility upon gcbA multicopy expression not resulting from elevated EPS production and enhanced cell-cell interactions. In order to test this, we assessed the effect of gcbA overexpression on swimming and swarming motilities in the absence of Psl and/or Pel polysaccharides (i.e., Δpsl, Δpel, and Δpel Δpsl mutant strains). Multicopy expression of gcbA suppressed both swimming and swarming motilities in all three polysaccharide mutants to a degree comparable to that observed for the wild-type PAO1 strain (Fig. 6). This finding indicated that reduction of motility following gcbA overexpression is not the result of hyperaggregation and enhanced cell-cell interactions.
GcbA-mediated suppression of motility is independent of EPS production. Effects of gcbA overexpression in the wild-type PAO1 and Δpel, Δpsl, and Δpel Δpsl EPS mutant strains on swarming (A and C) and swimming (B and D) motilities were analyzed using 0.5% and 0.3% agar, respectively, following 24 h of growth. All assays were repeated at least in triplicate. *, significantly different from the respective strain not overexpressing gcbA (P < 0.01).
GcbA is an active diguanylate cyclase that modulates c-di-GMP levels during planktonic growth.Our results indicated that GcbA regulates the transition from polar to longitudinal surface attachment by regulating flagellar reversal rates and flagellum-dependent motility in a viscosity-independent manner. We next sought to determine how GcbA exerts its regulatory effects. As P. fluorescens GcbA was found to regulate swimming motility by acting as a functional DGC, we asked whether the P. aeruginosa GcbA similarly plays a regulatory role by synthesizing c-di-GMP. In order to determine whether GcbA is an active DGC, cyclase assays were performed using protein extracts of E. coli strains harboring pJN-gcbA or the vector control plasmid pJN105. HPLC analysis revealed conversion of GTP to c-di-GMP in the presence of GcbA in E. coli extracts following incubation at 37°C for 60 and 120 min. c-di-GMP production exhibited by GcbA was comparable to that observed for the previously described DGC PleD (Fig. 7A). Overall, the specific activity of the extracts containing GcbA was found to be 448.9 ± 48.6 pmol/mg · min−1. In contrast, no c-di-GMP detected in the extracts of the vector control (Fig. 7A). Consistent with GcbA functioning as an active DGC and our previous observations (5, 25), overexpression of gcbA in P. aeruginosa resulted in a hyperaggregative phenotype in liquid culture (Fig. 7B). The hyperaggregative phenotype correlated with elevated c-di-GMP levels: PAO1/pJN-gcbA cultures on average contained 228 pmol of c-di-GMP per mg of total cell protein, compared to less than 35 pmol/mg c-di-GMP observed for the wild type and the vector control (Fig. 7C).
GcbA is an active diguanylate cyclase. (A) Elution profiles of enzymatically produced c-di-GMP. Diguanylate cyclase assays were performed using 25 μM GTP and 1 μg of the total cell protein of indicated E. coli strains harboring either gcbA under the control of the arabinose-inducible PBAD promoter in the pJN105 vector or the empty pN105 vector control. E. coli expressing the DGC PleD was used as a positive control. The reaction mixtures were analyzed by HPLC for the presence of c-di-GMP 0, 60, and 120 min postinitiation of DGC assays. Representative peaks corresponding to c-di-GMP are shown. (B) Wild-type PAO1 exhibits a hyperaggregative phenotype in liquid culture upon overexpression of gcbA. c-di-GMP levels in extracts obtained from cells of indicated P. aeruginosa strains grown planktonically (C) or as biofilms (D) were determined using HPLC analysis. c-di-GMP levels are shown as pmol per mg of total cell protein. The wild-type PAO1 strain harboring the empty vector pJN105 was used as a vector control. Error bars denote standard deviations. All assays were repeated at least in triplicate. *, significantly different from PAO1 biofilm cells and the vector control PAO1/pJN105 (P < 0.01).
While confirming that GcbA functions as an active DGC, the finding of elevated c-di-GMP levels in PAO1/pJN-gcbA did not answer the question of whether GcbA affects the intracellular c-di-GMP pool in vivo under physiological conditions. Consequently, we compared c-di-GMP levels in wild-type strain PAO1 and an isogenic gcbA mutant. Inactivation of gcbA reduced total cellular c-di-GMP levels 2-fold under planktonic growth conditions, with ΔgcbA cells on average containing 17.8 pmol/mg compared to the 34.1 pmol/mg detected in PAO1 cells (Fig. 7C). When biofilm-grown cells were analyzed, however, no differences in c-di-GMP levels between the gcbA mutant and the wild type were observed (Fig. 7D). These findings indicated that GcbA contributes to the cellular levels of c-di-GMP in planktonically growing cells but not in biofilm cells postattachment.
GcbA and the phosphodiesterase DipA inversely regulate attachment and motility in planktonic cells by converging but distinct mechanisms.Having established GcbA as an active DGC mediating initial surface attachment via regulation of motility, we asked whether it functions within a network of other proteins known to modulate c-di-GMP levels during planktonic growth. The PDE DipA, which contributes significantly to the overall PDE activity of planktonic cells, has recently been linked to GcbA, with dipA overexpression capable of suppressing the hyperaggregative and nonswarming phenotypes of the gcbA overexpresser strain (5). Presently, inactivation of dipA was found to affect swimming and swarming motilities independently of surface hardness (Fig. 8A). We therefore asked whether DipA and GcbA act in an opposite manner within the same regulatory pathway. Thus, a ΔdipA ΔgcbA double mutant was tested, assuming that if both proteins participate in the same signaling network, the double mutant should exhibit wild-type phenotypes. Interestingly, however, a secondary mutation in gcbA failed to rescue the motility phenotype of the dipA mutant to wild-type levels (Fig. 8A). The ΔdipA ΔgcbA strain exhibited reduced swimming and swarming motility levels similar to those observed for the dipA mutant alone. Moreover, while inactivation of dipA resulted in significantly enhanced initial attachment, with a 3-fold increase in total surface-adhered cells relative to the wild type observed following 6 h of growth, a secondary mutation in gcbA, however, reduced attachment of a dipA mutant below that of the wild type, to levels observed for the gcbA mutant alone (Fig. 8B). These findings suggested that GcbA and DipA are unlikely to be part of the same regulatory pathway and that DipA predominates in the regulation of motility on semisolid surfaces, while GcbA exerts a greater effect than DipA on the regulation of initial attachment to surfaces.
GcbA and DipA regulate attachment and motility by distinct yet converging pathways. (A) Swarming (0.45, 0.50, and 0.55% agar) and swimming (0.3% agar) motilities of the wild-type PAO1 and ΔgcbA, ΔdipA, and ΔdipA ΔgcbA mutant strains were assessed following 24 h of growth. (B) Total numbers of attached cells per 400 μm2 were quantified using ImageJ software with the cell counter tool. (C) Numbers of polarly attached cells were quantified using ImageJ software, and the rate of polar attachment is reported as the percentage of total attached cells. (D) Flagellar reversal rates in M63 medium with 3% or 15% Ficoll, representing swimming and swarming conditions, respectively, were measured as changes in direction of movement of cells. Rates are expressed as reversals per cell per minute. (E) c-di-GMP levels extracts obtained from cells of indicated P. aeruginosa strains grown planktonically were determined by HPLC analysis. c-di-GMP levels are shown as pmol per mg total cell protein. All assays were repeated at least in triplicate. *, significantly different from PAO1 (P < 0.01).
We subsequently asked which protein predominates in the regulation of the polar-longitudinal attachment transition. Interestingly, the dipA mutant demonstrated an increase in the percentage of reversibly attached cells relative to the wild type similar to that observed for ΔgcbA (Fig. 8C; see Fig. S2 in the supplemental material). This finding was surprising, as a significant increase in cellular attachment was anticipated to correlate with an increased percentage of irreversibly attached cells and thus a reduction in the fraction of polarly attached cells (Fig. 8C). Rather, this observation indicated that DipA might regulate the initial transition from free-swimming growth to surface contact, with the elevated numbers of polarly attached cells reflecting increased instances of surface contact. In order to further explore the hierarchical relationship between DipA and GcbA, as well as to assess whether the elevated rate of polar attachment in the dipA mutant is associated with alterations in flagellar reversals, reversal rates in the single and double gcbA and dipA mutants were compared. Interestingly, inactivation of dipA had no significant effects on flagellar reversal rates at either high or low viscosity (Fig. 8D). In contrast, inactivation of gcbA elevated flagellar reversal rates regardless of the presence of DipA, with both the ΔgcbA and ΔdipA ΔgcbA mutants demonstrating significant increases relative to the wild type in reversal rates under conditions of both low and high viscosity (Fig. 8D).
These findings suggested that DipA and GcbA are not part of the same regulatory pathway. To further determine DipA and GcbA to function in separate pathways, total levels of c-di-GMP present in planktonic cells of the ΔdipA, ΔgcbA, and ΔdipA ΔgcbA mutant strains were determined. We reasoned that if DipA and GcbA were not part of the same regulatory network, inactivation of both respective genes would be expected not to return total cellular c-di-GMP levels to those observed in the wild-type strain, PAO1. However, restoration of wild-type c-di-GMP levels would be expected if DipA and GcbA work in concert. In accord with previous findings (5), inactivation of dipA resulted in elevated c-di-GMP levels (Fig. 8E). However, a secondary mutation in gcbA did not rescue the dipA mutant phenotype, as the double ΔdipA ΔgcbA mutant exhibited c-di-GMP levels above those observed for the wild type and similar to those observed for the dipA mutant alone (Fig. 8E). This funding suggested that DipA makes a more significant contribution to the modulation of the overall cellular c-di-GMP levels and underscored the conclusion that DipA and GcbA likely act within distinct regulatory networks.
Regulation of attachment and motility by GcbA is dependent on RsmZ.The switch from planktonic to surface-associated growth is controlled by two major regulatory mechanisms in P. aeruginosa—modulation of c-di-GMP levels and the Gac/Rsm signal transduction network—the functions of which have recently been linked to each other (48). The small RNA (sRNA) RsmZ, while playing a role in initial attachment, was found to be suppressed within 24 h of surface-associated growth in order to allow biofilm formation postattachment (49). As GcbA appears to be active only during initial attachment and not following 24 h of surface-associated growth, we asked whether GcbA function is related to RsmZ. qRT-PCR indicated a positive correlation between gcbA and rsmZ transcript abundance (Table 1), with gcbA inactivation and overexpression resulting in a 3-fold reduction and a 4-fold increase in RsmZ levels, respectively. In agreement with GcbA and DipA regulating attachment via different mechanisms, inactivation of dipA did not significantly affect RsmZ levels.
We therefore asked whether the observed gcbA mutant phenotypes are linked to reduced RsmZ levels and can be rescued by elevating levels of this sRNA. Multicopy expression of rsmZ in the ΔgcbA mutant restored attachment to wild-levels, as revealed by crystal violet staining and microscopy (Fig. 9A and B). Moreover, the percentages of polarly attached cells were restored to wild-type levels in the ΔgcbA/pJN-rsmZ strain (Fig. 9C), suggesting that rsmZ expression facilitated the transition from polar to longitudinal attachment in gcbA mutant cells. Overexpression of rsmZ in the gcbA mutant, furthermore, suppressed both swimming and swarming motilities back to wild-type levels (Fig. 9D). It is of interest to note that no significant differences in initial surface attachment, transition to longitudinal attachment, or motility were observed between PAO1/pJN-rsmZ and the vector control (Fig. 9). These findings indicated that the GcbA phenotypes are, at least partially, dependent on RsmZ levels.
Multicopy expression of rsmZ restores attachment and motility of ΔgcbA cells to wild-type levels. (A) Evaluation of attachment to a polystyrene surface by crystal violet staining following 6 h of growth under shaking conditions. Attachment is expressed as a percentage relative to the wild-type PAO1 control. (B) Total numbers of attached cells per 400 μm2 were quantified using ImageJ software with the cell counter tool. (C) Numbers of reversibly attached cells were quantified using ImageJ software, and the rate of polar attachment is reported as a percentage of total attached cells. (D) The swarming and swimming motilities of the indicated strains were analyzed using 0.5% and 0.3% agar, respectively, following 24 h of growth. All assays were repeated at least in triplicate. *, significantly different from PAO1 (P < 0.01).
DISCUSSION
The c-di-GMP-dependent regulation of initial surface attachment and biofilm formation in P aeruginosa is a complex multisystem process involving the inverse regulation of EPS production and motility, with c-di-GMP synthesis for these processes dependent on at least five DGCs (WspR, SadC, RoeA, SiaD, and YfiN/TpbB) (10, 12–16). Similarly, four DGCs (WspR, GcbA, GcbB, and GcbC) have been identified to regulate biofilm formation in P. fluorescens via inverse regulation of motility and cell adhesin localization (9, 21, 22). While the Wsp chemotactic systems and WspR DGCs are homologous in P. fluorescens and P. aeruginosa, information about additional homology in c-di-GMP-modulating systems remains limited. Here, we demonstrate that the P. aeruginosa homolog of the P. fluorescens DGC GcbA-Pfl, involved in promoting biofilm formation via regulation of swimming motility, likewise synthesizes c-di-GMP (Fig. 7) to regulate surface attachment (Fig. 1) and that the respective P. aeruginosa mutant can be complemented by gcbA-Pfl. We furthermore, demonstrate that the P. aeruginosa GcbA functions to modulate an intracellular pool of c-di-GMP to suppress flagellar reversal rates and flagellum-associated motility and to promote more permanent cell-surface interactions following initial attachment in a manner independent of the expression of EPS-biosynthetic genes.
The P. aeruginosa GcbA was found to promote initial surface adhesion but not biofilm formation postattachment (Fig. 1). Our findings indicated that while facilitating the reversible-to-irreversible attachment switch, GcbA is not required for this transition, as evidenced by the fact that a gcbA mutant is capable of forming wild type-like biofilms once the cells overcome the hurdle of reversible attachment. These findings suggested that GcbA likely acts in concert with other DGCs, such as WspR, SadC, and RoeA, which in the absence of GcbA are capable of promoting more permanent interactions with the surface at a reduced rate. The four P. fluorescens DGCs described by Newell and colleagues (9) similarly appear to act in concert to contribute to c-di-GMP regulation, as combinatorial analysis of multiple-DGC mutants revealed additive effects on biofilm formation.
In agreement with the temporal role of GcbA in regulating attachment is the finding of growth mode-specific effects of GcbA on cellular c-di-GMP levels (Fig. 7C). GcbA appears to produce c-di-GMP to suppress flagellum-dependent motility to promote the transition to surface-associated growth. However, following the initiation of subsequent biofilm formation, the contribution of GcbA to the cellular levels of c-di-GMP becomes reduced or minimal (Fig. 7C), as indicated by the significant reduction in gcbA transcript levels within biofilm cells. However, while we noted that the activity and contribution of GcbA to cellular c-di-GMP levels are rendered undetectable within the context of the overall high DGC activity level of biofilm cells, we cannot preclude that GcbA is still active in biofilm cells. Support for GcbA functioning in biofilm cells stems from a recent report by Jones et al. (50) indicating a contribution of GcbA to the regulation of biofilm formation in the absence of the transcriptional regulator AmrZ. More specifically, Jones et al. (50) identified gcbA (named adcA for AmrZ-dependent cyclase A) as being repressed by AmrZ, with inactivation of amrZ, correlating with elevated gcbA levels, and enhanced biofilm biomass accumulation, which could be reduced by a secondary mutation in gcbA (50). Yet, the full contribution of GcbA to c-di-GMP turnover and regulation within wild-type biofilms remains to be investigated.
At present, GcbA has been found to modulate attachment by regulating motility and flagellar reversal rates in a manner independent of surface hardness or viscosity, suggesting a regulatory function distinct from those of a number of previously characterized systems, including flagellar stators MotAB and MotCD (35, 43), the signaling protein SadB (44), chemotaxis cluster IV (32), the DGC SadC (13), and the PDE BifA (51, 52). The contribution of these factors is limited to swarming motility at higher viscosity and increased surface hardness, conditions that are experienced by cells already attached to a surface, rather than by free-swimming planktonic cells (32, 35, 43). Thus, the high-viscosity-dependent factors appear to facilitate the transition to more permanent surface interactions and biofilm formation by suppressing the swarming motility of initially attached cells. In contrast, GcbA was found to affect both swimming and swarming motilities and to impact the rate of flagellar reversals independently of medium viscosity, suggesting this DGC to function both in free-swimming and initially attached cells. These findings implied that while promoting the polar-longitudinal attachment transition and thus likely promoting the transition from the reversible to irreversible attachment stage, GcbA might also facilitate initial surface contact as well. It is possible that GcbA-Pfl functions in a similar manner to promote P. fluorescens biofilm formation, as it was likewise found to regulate swimming motility (9). However, the effect of GcbA-Pfl on swarming has not been tested. Recent findings caution that the motilities of P. fluorescens and P. aeruginosa may be differentially regulated by homologous mechanisms: while P. aeruginosa SadB and SadC were found to exclusively regulate swarming motility (32, 44), P. fluorescens SadB and SadC were found to affect both swimming and swarming (53, 54).
In addition to being driven by increasing c-di-GMP levels, initial attachment to surfaces by P. aeruginosa and P. fluorescens is regulated by the Gac/Rsm signal transduction network (53, 55–57). Recent evidence has established a link between c-di-GMP signaling and the Gac/Rsm network in both species. The regulatory functions of the P. fluorescens Gac/Rsm network and the protein SadB have been shown converge to regulate expression of genes encoding the transcriptional regulator FleQ and the sigma factor AlgU to ultimately modulate motility and surface attachment (53, 57) The link between the P. aeruginosa LadS/RetS/Gac/Rsm network and c-di-GMP regulation has been established through evidence demonstrating that the acute-to-chronic switch modulated by LadS requires the presence of the DGC WspR. In turn, the WspR-dependent c-di-GMP regulation of the switch was found to require the sRNAs RsmY and RsmZ (48). Currently, we present evidence of a link between GcbA and RsmZ in the regulation of the motile-sessile switch. The temporal role of GcbA in regulating attachment is reminiscent of that recently reported for RsmZ: while playing a role in initial attachment, RsmZ was found to be suppressed within 24 h of surface-associated growth in order to allow biofilm formation postattachment (49). Furthermore, RsmZ has previously been implicated in the regulation of flagellar motility (58, 59). Here, GcbA levels were found to positively correlate with RsmZ levels, with gcbA inactivation and overexpression reducing and elevating RsmZ abundance, respectively (Table 1). The changes in RsmZ levels, however, were not the result of a general perturbation in the cellular c-di-GMP levels, as inactivation of dipA, which elevates c-di-GMP abundance by significantly lowering cellular PDE activity (5), did not elevate RsmZ levels (Table 1). Furthermore, the ΔgcbA attachment and motility phenotypes were rescued to wild-type levels upon multicopy expression of rsmZ (Fig. 9). These findings indicated that the GcbA phenotypes are at least partially dependent on RsmZ, with c-di-GMP synthesized by GcbA potentially contributing to the regulation of RsmZ abundance.
Regulation of intracellular levels of c-di-GMP involves the opposing action of DGCs and PDEs. For instance, recent findings have suggested that regulation of P. fluorescens biofilm formation is dependent on the opposing activities of the DGC WspR and PDE BifA in a manner dependent on the novel c-di-GMP-sensing protein FlgZ (54). Furthermore, the PDE RapA participates as part of the inside out signaling mechanism to regulate localization of the LapA adhesin, acting in an opposing manner to the DGCs GcbB and GcbC under conditions of phosphate limitation (60, 61). In P. aeruginosa PA14, the PDE BifA acts downstream of the DGCs SadC and RoeA to degrade the c-di-GMP produced by these proteins to regulate swarming motility and EPS production (10). Here, we demonstrate that the DGC GcbA and PDE DipA function through different, but converging, mechanisms to regulate motility and attachment. Specifically, DipA superseded GcbA in the regulation of swimming and swarming motilities on semisolid surfaces regardless of surface hardness (Fig. 8A). GcbA, on the other hand, predominated in the regulation of initial cellular attachment to surfaces, as well as that of flagellar reversal rates (Fig. 8B and D). Interestingly, inactivation of either GcbA or DipA resulted in similar elevated percentages of polarly attached cells (Fig. 8C). While the increased fraction of polarly attached ΔgcbA cells is likely the result of a failure to transition to more permanent interactions with the surface, an elevated percentage of dipA mutant cells exhibiting polar attachment is likely indicative of enhanced surface sensing and/or increased instances of surface contact. Furthermore, the present findings indicated that DipA likely promotes motile growth by regulating motility through mechanisms that are distinct from those regulating flagellar reversal rates (Fig. 8). Thus, DipA is likely part of a regulatory network distinct from SadC or GcbA, both of which are associated with the control of flagellar reversal rates.
While recent evidence is available to suggest the importance of both the general cellular c-di-GMP level and the spatial separation of c-di-GMP pools, our present findings support the notion of different c-di-GMP-modulating enzymes regulating distinct phenotypic outputs. Comparisons between the P. aeruginosa GcbA and P. fluorescens GcbA-Pfl suggest strong conservation among c-di-GMP-regulating mechanisms, yet underscore the existence of subtle adaptations to the environment or niche specifics of c-di-GMP signaling modulating biofilm formation. Previous studies of Pseudomonas species have provided evidence that distinct c-di-GMP-modulating enzymes separately regulate motility and EPS production. Our present findings indicate an additional level of complexity, with the results suggesting that distinct c-di-GMP-modulating pathways can be involved in the regulation of a single phenotypic output.
ACKNOWLEDGMENTS
We thank Daniel Wozniak at the Ohio State University, Dieter Haas and Cornelia Reimmann at the University of Lausanne, and George O'Toole at Dartmouth Medical School for providing strains used in this study.
This work was supported by a grant from the NIH (1RO1 A107525701A2).
FOOTNOTES
- Received 2 March 2014.
- Accepted 22 May 2014.
- Accepted manuscript posted online 2 June 2014.
- Address correspondence to Karin Sauer, ksauer{at}binghamton.edu.
Supplemental material for this article may be found at http://dx.doi.org/10.1128/JB.01628-14.
REFERENCES
- Copyright © 2014, American Society for Microbiology. All Rights Reserved.
