The PA3177 Gene Encodes an Active Diguanylate Cyclase That Contributes to Biofilm Antimicrobial Tolerance but Not Biofilm Formation by Pseudomonas aeruginosa

A hallmark of biofilms is their heightened resistance to antimicrobial agents. Recent findings suggested a role for bis-(3′-5′)-cyclic dimeric GMP (c-di-GMP) in the susceptibility of bacteria to antimicrobial agents; however, no c-di-GMP modulating enzyme(s) contributing to the drug tolerance phenotype of biofilms has been identified.

ABSTRACT

A hallmark of biofilms is their heightened resistance to antimicrobial agents. Recent findings suggested a role for bis-(3′-5′)-cyclic dimeric GMP (c-di-GMP) in the susceptibility of bacteria to antimicrobial agents; however, no c-di-GMP modulating enzyme(s) contributing to the drug tolerance phenotype of biofilms has been identified. The goal of this study was to determine whether c-di-GMP modulating enzyme(s) specifically contributes to the biofilm drug tolerance of Pseudomonas aeruginosa. Using transcriptome sequencing combined with biofilm susceptibility assays, we identified PA3177 encoding a probable diguanylate cyclase. PA3177 was confirmed to be an active diguanylate cyclase, with overexpression affecting swimming and swarming motility, and inactivation affecting cellular c-di-GMP levels of biofilm but not planktonic cells. Inactivation of PA3177 rendered P. aeruginosa PAO1 biofilms susceptible to tobramycin and hydrogen peroxide. Inactivation of PA3177 also eliminated the recalcitrance of biofilms to killing by tobramycin, with multicopy expression of PA3177 but not PA3177_GGAAF harboring substitutions in the active site, restoring tolerance to wild-type levels. Susceptibility was linked to BrlR, a previously described transcriptional regulator contributing to biofilm tolerance, with inactivation of PA3177 negatively impacting BrlR levels and BrlR-DNA binding. While PA3177 contributed to biofilm drug tolerance, inactivation of PA3177 had no effect on attachment and biofilm formation. Our findings demonstrate for the first time that biofilm drug tolerance by P. aeruginosa is linked to a specific c-di-GMP modulating enzyme, PA3177, with the pool of PA3177-generated c-di-GMP only contributing to biofilm drug tolerance but not to biofilm formation.

INTRODUCTION

Bis-(3′-5′)-cyclic dimeric GMP (c-di-GMP) is an intracellular signaling molecule that is produced from two molecules of GTP by diguanylate cyclases (DGCs) and broken down to 5′-phosphoguanylyl-(3′-5′)-guanosine and subsequently to two GMP molecules by phosphodiesterases (1). Known since the late 1980s, this “second messenger extraordinaire” (2) is now recognized as a near-ubiquitous second messenger that coordinates diverse aspects of bacterial growth and behavior, including motility, virulence, cell cycle progression, and biofilm formation, in which c-di-GMP aids in the transition of planktonic to sessile lifestyle (3). The switch to sessile lifestyle coincides with increased levels of this second messenger. For instance, while Pseudomonas aeruginosa planktonic cells have been described to contain less than 30 pmol/mg, mature P. aeruginosa biofilm cells contain on average 75 to 110 pmol of c-di-GMP per mg total cell extract (4, 5). Elevated c-di-GMP levels, in turn, result in increased production of extracellular matrix components, including exopolysaccharides, adhesive pili, nonfimbrial adhesins, extracellular DNA, and other biofilm-associated characteristics such as increased adhesiveness and autoaggregation, but repressed motility (3, 6–11).

Recent findings further suggest a role of c-di-GMP in the susceptibility of bacteria to antimicrobial agents. This is supported by clinical isolates from the cystic fibrosis lung that display aggregative phenotypes in liquid or small colony variants (SCVs) on agar (phenotypes indicative of high c-di-GMP levels), displaying 2- to 8-fold-higher MIC values to a broad range of anti-Pseudomonas agents than revertants of SCVs (12–16). Likewise, Gupta et al. (17) demonstrated that planktonic cells were rendered more resistant to antimicrobial agents by increasing intracellular c-di-GMP levels to those more commonly found in biofilm cells. Specifically, Gupta et al. (17) demonstrated that planktonic cells were rendered resistant when c-di-GMP levels reached or exceeded 55 pmol/mg. In addition, differences in the intracellular level of the second messenger molecule c-di-GMP present in biofilm and free-floating, dispersed cells have been shown to correlate with differences in the production of proteins involved in antimicrobial peptide resistance and resistance toward colistin (18) or tobramycin (19).

Biofilm cells have been shown to be 10- to 1,000-fold less susceptible to various antimicrobial agents than their planktonic counterparts. While the nature of this biofilm tolerance has been attributed to a combination of different factors, including slow growth, reduced metabolic rates, increased stress tolerance, changes in membrane permeability resulting in reduced uptake, and increased sequestration (18, 20–32), recent findings furthermore suggested that the level of c-di-GMP plays a role in the heightened tolerance of biofilm cells to killing by antimicrobial agents. By analyzing biofilms by P. aeruginosa mutant strains characterized by various intracellular levels of c-di-GMP and strains overexpressing diguanylate cyclases and phosphodiesterases, Gupta et al. (17) demonstrated that biofilm cells remain resistant to antimicrobial agents at c-di-GMP levels above 62 pmol/mg but not at or below 40 pmol/mg. These findings suggested a switch in susceptibility in a manner dependent on the c-di-GMP level but independent of the mode of growth or biofilm biomass accumulation (17, 33). Moreover, the findings suggested a disconnect between the biofilm architecture and biofilm tolerance (33). Additional evidence of c-di-GMP contributing specifically to biofilm drug tolerance stems from two P. aeruginosa proteins required for biofilm drug tolerance: the two-component sensory protein SagS and the transcriptional regulator BrlR. Inactivation of sagS rendered biofilms as susceptible as planktonic cells to various antimicrobial agents, including tobramycin, norfloxacin, and hydrogen peroxide (34). SagS contribution to biofilm tolerance was dependent on c-di-GMP, with ΔsagS mutant biofilms demonstrating cellular c-di-GMP levels reminiscent of those present in planktonic cells (17, 33, 34). Moreover, ΔsagS mutant biofilms were characterized by significantly reduced brlR transcript and BrlR protein levels (17, 33, 34). BrlR is a c-di-GMP-responsive transcriptional regulator that in turn activates the expression of several multidrug efflux pumps and ABC transporters (35–38), with the ABC transport system PA1874-77 directly contributing to the drug tolerance of biofilms (38).

Despite the mounting evidence of resistance to antimicrobial agents being linked to the level of the intracellular c-di-GMP, no c-di-GMP modulating enzyme(s) contributing to the drug tolerance phenotype of biofilms has been identified. We therefore sought to determine whether a pool of c-di-GMP modulating enzyme(s) exists that specifically contributes to the biofilm drug tolerance of P. aeruginosa. To address this question, we first made use of transcriptome sequencing (RNA-Seq) to determine genes encoding c-di-GMP modulating enzymes that are increased in transcript abundance upon biofilm growth.

RESULTS

Identification of genes encoding c-di-GMP modulating enzymes that are induced upon biofilm formation by P. aeruginosa.

We first assessed the transcriptomes of Pseudomonas aeruginosa PAO1 cells grown planktonically and as biofilms using RNA-Seq (39) for genes encoding known and predicted c-di-GMP modulating enzymes (Fig. 1A). The analysis revealed several genes demonstrating a >3-fold difference in transcript abundance when the two different growth conditions were compared. These included probable and confirmed diguanylate cyclases (only harbor GGDEF domain) PA0169 (SiaD), PA1851, PA2870, PA3177, and PA3343 (HsbD), genes encoding probable and confirmed phosphodiesterases (EAL harboring) PA5017 (DipA) and PA2818 (arr), and PA2072 harboring both GGDEF and EAL domains (40). Among these genes, the transcript abundance of PA1851, PA2818, and PA3343 was found to be reduced in biofilms, with PA1851 demonstrating the largest reduction in transcript abundance (11.4-fold) (Fig. 1A). In contrast, the transcript abundance of PA0169, PA2870, PA3177, and PA5017 was found to be increased in biofilm cells relative to planktonic cells, while the largest increase in transcript abundance being noted for PA2870 (10.6-fold) (Fig. 1A).

FIG 1.

Transcript abundance of genes encoding c-di-GMP modulating enzymes and susceptibility phenotype of select P. aeruginosa biofilms to tobramycin. (A) Fold change in transcript abundance of genes encoding c-di-GMP modulating enzymes, as determined using RNA-Seq. RNA-Seq was performed in triplicate using biological replicates. Fold changes were calculated using EdgeR. P values were calculated using EdgeR, with P < 0.001 for the data shown. Genes highlighted in orange were chosen for subsequent biofilm susceptibility assays. (B) Susceptibility phenotype of P. aeruginosa biofilms by PAO1 and mutant strains PA0169::IS (siaD), PA1851::IS, PA2818::IS (arr), PA2870::IS, PA3177::IS, PA3343::IS (hsbD), PA4843 (gcbA), and PA5017 (dipA) to tobramycin (150 μg/ml). Viability was determined from CFU counts. Susceptibility is expressed as the log₁₀ reduction in viability. *, significantly different compared to PAO1 (<0.005), as determined using analysis of variance (ANOVA).

Screening biofilms by mutants inactivated in genes encoding probable and confirmed diguanylate cyclases and phosphodiesterases for their susceptibility to tobramycin.

To determine whether there is a pool of c-di-GMP modulating enzyme(s) that specifically contributes to the biofilm drug tolerance of P. aeruginosa, we next focused on genes that were found by RNA-Seq to be differentially expressed in biofilms relative to planktonic cells (Fig. 1A). Specifically, we focused on PA5017, PA0169, PA1851, PA2818, PA2870, PA3177, and PA3343. To determine whether the respective c-di-GMP modulating enzymes contribute to the susceptibility phenotype of P. aeruginosa biofilms, mutant strains inactivated in PA5017 (dipA) and strains harboring transposon insertions in PA0169, PA1851, PA2818, PA2870, PA3177, and PA3343 were subjected to biofilm susceptibility assays using the aminoglycoside tobramycin. The antibiotic was chosen since it is frequently used in the treatment of chronic airway infection with Pseudomonas aeruginosa (41). Strains inactivated in sagS and brlR were used as controls. PA4843 encoding the active diguanylate cyclase GcbA has previously been demonstrated to modulate c-di-GMP levels during planktonic growth but not under biofilm growth conditions. In agreement with previous findings, the transcript abundance of PA4843 was found to be decreased in biofilms (Fig. 1A). We therefore included the mutant ΔgcbA strain as a negative control. Under the conditions tested, biofilms by the ΔgcbA strain were as resistant as wild-type PAO1 biofilms (Fig. 1B). Likewise, insertional inactivation of PA0169, PA1851, PA2870, PA3343, and PA5017 had no effect on the susceptibility of P. aeruginosa biofilms to tobramycin. In contrast, insertional inactivation of PA2818 and PA3177 rendered biofilm cells more susceptible to tobramycin, as indicated by 2.7-log and 3.5-log reductions in viability, respectively (Fig. 1B). The observed decrease in viability by biofilm cells of strain PA3177::IS was similar to that noted for ΔbrlR and ΔsagS mutant biofilms (Fig. 1B).

PA3177 encodes an active diguanylate cyclase.

Our findings indicated biofilms by strains PA2818::IS and PA3177::IS to be as susceptible to tobramycin as biofilms by ΔsagS and ΔbrlR strains. However, because the transcript abundance of PA2818 was reduced in biofilms relative to planktonic cells but the transcript abundance of PA3177 was found to be increased, we chose to only further investigate PA3177. Quantitative reverse transcription-PCR (qRT-PCR) confirmed the increase in transcript abundance of PA3177 to be (6.5 ± 0.9)-fold in biofilm cells relative to planktonic cells. Based on pseudomonas.com data, PA3177 is annotated as a hypothetical protein, composed of 307 amino acids, and predicted to harbor a N-terminal GAF-like domain (named after some of the proteins in which it is found, including cGMP-specific phosphodiesterases, adenylyl cyclases, and FhlA) and a C-terminal GGDEF domain (40). Conserved domain search confirmed the presence of a GGDEF domain located near the C-terminal end of the PA3177; however, the search revealed no other domains. Sequence homology to other diguanylate cyclases by P. aeruginosa, including WspR and SadC, further suggested that PA3177 likely encodes a diguanylate cyclase. Similar to SadC and WspR, PA3177 is characterized by the presence of conserved GGEEF motif present within the GGDEF domain, where glutamate is replaced for aspartate, and an I site characterized by an RXXD motif (“X” is any residue) that is separated from the GGEEF motif by a linker composed of 5 amino acids (Fig. 2) (42, 43). Additional support for PA3177 encoding a diguanylate cyclase stems from PA3177 having been identified in pulldown assays via its affinity to immobilized c-di-GMP and subsequent mass spectrometry in a manner similar to WspR (44). Moreover, overexpression of PA3177 in Escherichia coli BL21 has been shown to result in increased cellular c-di-GMP levels (45), while the overexpression of PA3177 in P. aeruginosa PA14 coincided with reduced swimming and swarming motility, and an aggregative phenotype when grown in liquid (45).

FIG 2.

PA3177 is characterized by a putative GGDEF domain. The sequence alignment of PA3177 with other known diguanylate cyclases, including SadC and WspR from P. aeruginosa, depicting homology is depicted. The inhibitory site (I-site) is characterized by RxxD, where “X” is any residue. The active site (A-site) is characterized by GGDEF domain outlined in the diagram at the top of the figure by a rectangular box. *, identical residues. The terms “:” and “.” beneath the sequences indicate conserved and semiconserved substitutions, respectively.

Since diguanylate cyclases contribute to increased intracellular c-di-GMP levels which in turn, negatively affects motility, we first determined the role of PA3177 in swimming and swarming motility to determine whether PA3177 encodes a diguanylate cyclase. We therefore generated a mutant strain in which PA3177 is inactivated. The resulting mutant ΔPA3177 was then subjected to swimming and swarming assays. No difference in swimming motility was noted when strain ΔPA3177 was compared to the isogenic parental strain PAO1 (Fig. 3A), suggesting that strain ΔPA3177 is not hypermotile. Although inactivation of PA3177 had no effect on swimming, multicopy expression of PA3177 in strain ΔPA3177 coincided with reduced swimming motility, suggesting that PA3177 likely encodes a diguanylate cyclase. Diguanylate cyclase activity is associated with the conserved GGDEF domain, where the GGDEF (or GGEEF) motif corresponds to the active site (A site) of a diguanylate cyclase (46–49). Any mutation in the motif, except for aspartate (D) to glutamate (E), inhibits the enzymatic activity (47).To further confirm PA3177 to be an active diguanylate cyclase, we substituted the two glutamate residues present in the GGEEF motif to alanine. The resulting construct PA3177_GGAAF was cloned into pJN105 under the control of an arabinose inducible P_BAD promoter, and transferred into ΔPA3177, generating the strain ΔPA3177/pJN-PA3177_GGAAF. Multicopy expression of the variant gene PA3177_GGAAF had no effect on swimming motility.

FIG 3.

Overexpression of PA3177 affects swarming and swimming motility, while PA3177_GGAAF harboring substitution in active-site residues has no effect. (A) Swimming motility of the indicated strains after 24 and 48 h of growth on 0.3% agar medium by the indicated strains. The photographs above the graph show representative images of swimming behavior. (B) Swarming motility after 24 and 48 h growth on M8 medium containing 0.4% agar by indicated strains. The photographs above the graph show representative images of swarming behavior. Error bars represent standard deviations. All experiments were performed in triplicates. *, significantly different (P < 0.005) from PAO1, as indicated by ANOVA.

Inactivation of PA3177 also had no effect on swarming motility (Fig. 3B). In contrast, overexpression of PA3177 resulted in reduced swarming, while multicopy expression of variant gene PA3177_GGAAF harboring a substitution of the active-site motif had no effect on swarming (Fig. 3B). Our findings suggested that PA3177 contributes to swimming and swarming motility in a manner dependent on the active-site GGEEF. Moreover, given that PA3177 only affected motility when overexpressed, our findings suggested that PA3177 is not expressed under planktonic growth conditions.

Inactivation of PA3177 correlates with reduced c-di-GMP levels in biofilm but not planktonic cells.

We next sought to determine whether PA3177 affects the intracellular c-di-GMP pool in vivo under physiological conditions. Consequently, we compared c-di-GMP levels in wild-type PAO1 and an isogenic ΔPA3177 mutant when grown planktonically to exponential phase and as 3-day-old biofilms. c-di-GMP was extracted by heat and ethanol precipitation and subsequently analyzed by high-pressure liquid chromatography (HPLC) to quantitate intracellular c-di-GMP levels. Inactivation of PA3177 reduced total cellular c-di-GMP levels 5-fold under biofilm growth conditions, with ΔPA3177 cells on average containing 22.4 pmol/mg compared to the 97.6 pmol/mg detected in wild-type PAO1 biofilm cells (Fig. 4A). Multicopy expression of PA3177 significantly increased c-di-GMP levels of PA3177 mutant biofilm cells to an average of 137.9 pmol/mg. It is of interest to note that multicopy expression of PA3177 coincided with a 35 ± 4-fold increase in the PA3177 transcript abundance. In contrast, multicopy expression of the variant gene PA3177_GGAAF harboring a substitution of the active-site motif, had no effect on the c-di-GMP level of PA3177 mutant biofilms (Fig. 4A). While inactivation of PA3177 affected the intracellular c-di-GMP levels of biofilms, PA3177 was found to play no role in the intracellular c-di-GMP levels of planktonic cells. No differences in c-di-GMP levels between the PA3177 mutant and the wild type were observed (Fig. 4B). These findings indicated that PA3177 contributes to the cellular levels of c-di-GMP in biofilms, but not in planktonically growing cells.

FIG 4.

Inactivation of PA3177 coincides with reduced intracellular c-di-GMP levels in biofilm but not planktonic cells. (A) Intracellular c-di-GMP levels present in biofilm cells by the indicated strains, grown under continuous-flow conditions for 3 days. (B) Intracellular c-di-GMP levels present in planktonic cells by wild-type PAO1 and mutant strains ΔPA3177. c-di-GMP was quantified by HPLC. The “pmol/mg” refers to the c-di-GMP levels (pmol) per total cell protein (in mg). Error bars represent standard deviations, and experiments were performed in triplicates. *, significantly different compared to PAO1 (<0.005), as determined using ANOVA.

Inactivation of PA3177 has no effect on attachment and the formation of mature biofilms characterized by a three-dimensional biofilm architecture.

Several DGCs in P. aeruginosa, including WspR, SadC, RoeA, SiaD, and YfiN/Tpb, have been shown to contribute to c-di-GMP-dependent formation of biofilms (11, 23, 39, 41–43). Considering that expression of PA3177 was found to be biofilm specific, with PA3177 contributing to the c-di-GMP pool present in biofilms, we next sought to determine whether PA3177 plays any role in the formation of biofilms. We therefore evaluated attachment and the formation of biofilms following 5 days of growth. Attachment capabilities were quantitated following 24 h of growth in 96-well microtiter plates using crystal violet (CV) (50, 51). No significant difference in attachment upon inactivation or multicopy expression of PA3177 relative to wild-type PAO1 cells was observed (Fig. 5A).

FIG 5.

Inactivation of PA3177 has no effect on biofilm formation. (A) Attachment by indicated strains after 24 h, as indicated by CV staining. (B) Representative images of flow-cell-grown biofilms by the indicated strains after 2, 4, and 5 days of growth under continuous-flow conditions. Biofilm cells were visualized by live/dead staining. (C) Quantitative analysis of the biofilm biomass by the indicated strains after 2, 4, and 5 days of growth under continuous-flow conditions. (D) Representative images of flow-cell-grown biofilms by the indicated strains after 5 days of growth under continuous-flow conditions. (E and F) Quantitative analysis of the biofilm biomass (E) and the biofilm average and maximum height (F) by the indicated strains after 5 days of growth using COMSTAT. (G) Average number of viable cells present in biofilm tube reactor grown biofilms, as determined using CFU count. Error bars represent standard deviations. All experiments were performed in triplicates. Each attachment assay was composed of eight technical replicates. COMSTAT analysis was based on eight images per biofilm experiment, while CFU counts were performed using two technical duplicates per experiment. *, significantly different compared to PAO1 (<0.005) as determined using ANOVA.

We next assessed the biofilm architecture of 5-day-old biofilms inactivated or overexpressing PA3177, with the biofilm architecture being a measure of the biofilm biomass (52, 53). Considering that inactivation of PA3177 coincided with significantly reduced intracellular c-di-GMP levels under biofilm growth conditions (Fig. 4A), we anticipated ΔPA3177 to form biofilms with aberrant biofilm structure relative to wild-type biofilms and coincide with reduced biofilm biomass accumulation. However, biofilms by ΔPA3177 were indistinguishable from P. aeruginosa wild-type biofilms, which formed biofilms characterized by large microcolonies, as evidenced by average and maximum biofilm heights of 8 and 57 μm, respectively (Fig. 5B, C, E, and F). The similarity in the biofilm architecture was not limited to the 5-day time point and instead extended to earlier time points as well, with biofilms by ΔPA3177 and P. aeruginosa PAO1 featuring similar biofilm architectures following 2 and 4 days of growth (Fig. 5B and C). Likewise, no significant difference in the biofilm architecture at the 5-day time point was observed upon multicopy expression of PA3177 or PA3177_GGAAF in ΔPA3177 (Fig. 5D to F). To further confirm that ΔPA3177 and wild-type PAO1 forming biofilms were comprised of similar biomasses, the number of viable cells was determined from biofilms grown under flowing conditions in biofilm tube reactors for 5 days. No difference in the number of viable cells present in PAO1 and ΔPA3177 was noted (Fig. 5G).

Inactivation of PA3177 renders biofilms susceptible to tobramycin and hydrogen peroxide.

PA3177 was initially identified as a biofilm-induced gene encoding a probable diguanylate cyclase which contributes to the susceptibility phenotype of P. aeruginosa to tobramycin (Fig. 1). However, biofilm susceptibility was determined using a mutant harboring a transposon insertion in PA3177. To ensure that the PA3177 indeed contributes to the susceptibility phenotype of biofilms when grown under flowing conditions for 3 days, we first determined the susceptibility of biofilms by P. aeruginosa PAO1 and the isogenic mutant strain inactivated in PA3177 to tobramycin. Treatment of 3-day-old ΔPA3177 mutant biofilm cells with tobramycin (150 μg/ml) for 1 h resulted in a 3.1-log reduction in viability compared to a 1-log reduction in the viability of the PAO1 wild-type biofilms (Fig. 6A). Multicopy expression of PA3177 restored the susceptibility phenotype of ΔPA3177 mutant biofilm cells to wild-type levels. However, no restoration was observed when PA3177_GGAAF was overexpressed in ΔPA3177 mutant biofilm cells (Fig. 6A). Having confirmed a role of PA3177 in the susceptibility of PAO1 biofilms to tobramycin, we next sought to determine whether inactivation of PA3177 also plays a role in the susceptibility of P. aeruginosa PAO1 biofilms to hydrogen peroxide. Using identical growth conditions as for tobramycin treatment, exposure of 3-day-old wild-type biofilms with 0.6% hydrogen peroxide coincided with a 0.8-log reduction, while biofilms formed by ΔPA3177 demonstrated a 3-log reduction (Fig. 6B). Multicopy expression of PA3177 restored the susceptibility phenotype of ΔPA3177 mutant biofilm cells to wild-type levels, while multicopy expression of PA3177_GGAAF failed to do so (Fig. 6B).

FIG 6.

PA3177 contributes to the susceptibility and tolerance of P. aeruginosa biofilms to antimicrobial agents. (A and B) Susceptibility phenotype of P. aeruginosa PAO1 and PA3177 harboring an empty vector or overexpressing PA3177 to tobramycin (150 μg/ml) (A) and 0.6% hydrogen peroxide (B). Viability was determined from CFU counts. Susceptibility is expressed as the log₁₀ reduction in viability. *, significantly different compared to PAO1 (<0.005), as determined using ANOVA. (C and D) Biofilm-MBC assay results. Biofilms of indicated strains were grown for 3 days and subsequently treated for 24 h with increasing concentrations of tobramycin (75 to 300 μg/ml) under continuous-flow conditions before recovering and enumerating the surviving cells. Viability was determined by CFU counts (biofilm CFU, obtained from biofilm tube reactors having an inner surface area of 25 cm²). (C) Biofilms by P. aeruginosa PAO1, ΔPA3177, and ΔbrlR. #, no viable bacteria were detected. (D) Biofilms by ΔPA3177 expressing PA3177 or PA3177_GGAAF. #, no viable bacteria were detected. Error bars denote standard deviations. Experiments were carried out in triplicate, with two technical replicates per experiment.

Considering that in P. aeruginosa PA14, the gene expression of PA3177 was previously found to be inducible by the oxidative stress agent hypochlorite (45), we determined whether PA3177 is likewise induced by tobramycin. We therefore subjected P. aeruginosa PAO1 to tobramycin and analyzed PA3177 gene expression by qRT-PCR. Unlike hypochlorite, no significant difference in gene expression was noted in response to tobramycin, since the difference in PA3177 transcript abundance in the absence and presence of tobramycin was only 0.7 ± 1.2.

Inactivation of PA3177 eliminates the recalcitrance of biofilms to killing by tobramycin.

Given the increased susceptibility phenotype of ΔPA3177 biofilms relative to wild-type biofilms, with PA3177 gene expression being independent of tobramycin exposure and multicopy expression of PA3177 but not PA3177_GGAAF coinciding with restoration of the biofilm susceptibility phenotype of ΔPA3177 biofilms to wild-type levels, we surmised that PA3177 may contribute to the drug tolerance of P. aeruginosa biofilms. This drug tolerance is distinct from biofilm susceptibility and refers to the recalcitrance of biofilms to killing by cidal antimicrobial agents. Tolerance and susceptibility being distinct is supported by the findings that the two efflux pumps MexAB-OprM and MexEF-OprN contribute to biofilm susceptibility but play no role in the recalcitrance phenotype of biofilms to killing by bactericidal antimicrobials (36). We therefore determined whether PA3177 contributes to the drug tolerance of P. aeruginosa biofilm cells to antimicrobial agents, using biofilm minimum bactericidal concentration (biofilm-MBC) assays. The biofilm-MBC has been defined by Monzon et al. (54), Villain-Guillot et al. (55), and Moriarty et al. (56) as the concentration at which no further increase in log reduction following antimicrobial treatment is observed. We previously demonstrated that for P. aeruginosa wild-type biofilms, no further increase in log reduction was observed at concentrations higher than ∼75 μg/ml tobramycin following 24 h of treatment, with higher concentrations resulting in neither increased log reduction nor in complete killing of P. aeruginosa wild-type biofilms (37). In contrast, inactivation of brlR eliminated the tolerance of biofilms to tobramycin, which was apparent from the complete killing of ΔbrlR mutant biofilms by 75 μg/ml tobramycin (37). We therefore used biofilms by PAO1 and the ΔbrlR mutant as controls. In agreement with previous findings, exposure to tobramycin at concentrations exceeding 75 μg/ml failed to eradicate wild-type biofilms, while no viable cells were recovered from biofilms by brlR mutant strains (Fig. 6C). Similar to ΔbrlR mutant biofilms, no viable cells were recovered from PA3177 mutant biofilms after 24 h exposure to tobramycin (Fig. 6C). However, while ΔbrlR biofilms were eradicated at 75 μg/ml tobramycin, concentrations exceeding 150 μg/ml were needed to eradicate ΔPA3177 biofilms (Fig. 6C).

We furthermore surmised that if the tolerance of P. aeruginosa biofilms to tobramycin is indeed dependent on the diguanylate cyclase activity of PA3177, the expression of intact PA3177 but not PA3177_GGAAF would restore the tolerance phenotype of biofilms by ΔPA3177 to wild-type levels. We therefore evaluated biofilms by the ΔPA3177 mutant strains expressing intact PA3177 and the variant gene PA3177_GGAAF using biofilm-MBC assays. As anticipated, biofilms produced by strain ΔPA3177/pMJT-PA3177 were as tolerant to tobramycin as P. aeruginosa wild-type biofilms (Fig. 6C and D), indicating that tolerance to tobramycin can be restored upon expression of PA3177. However, the expression of PA3177_GGAAF failed to restore the drug tolerance phenotype to wild-type levels, as evidenced by ΔPA3177/pMJT-PA3177_GGAAF biofilms being as susceptible to tobramycin as biofilms by ΔPA3177 (Fig. 6C and D). The findings further suggested a contribution of PA3177 to the recalcitrance of biofilms to tobramycin. Moreover, our findings indicated that PA3177 contributes to the drug tolerance of biofilms in a manner dependent on its diguanylate cyclase activity.

Inactivation of PA3177 coincides with BrlR abundance being reduced in biofilms and unable to bind to P_mexE.

Drug tolerance by P. aeruginosa biofilms has previously been linked to the expression of brlR encoding the biofilm resistance regulator BrlR (36, 37). In addition, the two-component sensor hybrid SagS was found to contribute to drug tolerance by P. aeruginosa biofilms by functioning as a switch to enable the transition of biofilm cells from an antimicrobial susceptible to a highly tolerant state (34–37, 57, 58), with transition to the tolerant state coinciding with brlR expression and activation of BrlR. BrlR, in turn, activates the expression of several multidrug efflux pumps and ABC transporters (35–38), with the ABC transport system PA1874-77 directly contributing to the drug tolerance of biofilms (38).

Given the contribution of PA3177 in biofilm susceptibility and drug tolerance, we reasoned that inactivation of PA3177 coincides with reduced BrlR protein abundance, in a manner comparable to that noted for ΔsagS mutants (17, 34). We therefore made use of strains expressing a chromosomally located V5/His₆-tagged BrlR under the control of its own promoter (P_brlR-brlR-V5/His₆). This construct ensured the expression of brlR at native levels. Using immunoblot analysis, BrlR was detected in cell extracts obtained from wild-type biofilms (Fig. 7A). In contrast, little to no BrlR was detected in cell extracts obtained from PA3177 mutant biofilm cells, while multicopy expression restored BrlR to detectable levels (Fig. 7A). Our findings suggested that PA3177 likely contributes to the recalcitrance of biofilm to tobramycin by interfering with BrlR.

FIG 7.

Inactivation of PA3177 coincides with reduced BrlR abundance and BrlR unable to bind to the promoter P_mexE. (A) Detection of BrlR by immunoblot analysis. Total cell extracts (TCE) obtained from 3-day-old biofilms by PAO1, ΔPA3177, and ΔPA3177/pJN-PA3177 expressing a chromosomally located V5/His₆-tagged BrlR under the control of its own promoter (P_brlR-brlR-V5/His₆) were probed for the presence of BrlR by immunoblot analysis with anti-V5 antibodies. A total of 15 μg of total cell extract (TCE) was loaded. The corresponding SDS-PAGE gel image obtained after transfer demonstrates equal loading. (B) Streptavidin magnetic bead DNA-binding assay using cell extracts obtained from 3-day-old biofilms by strains PAO1 and ΔPA3177 harboring pMJT-brlR-V5/His₆ in the presence or absence of plasmid pJN-PA3177. Binding assays were carried out with total cell extracts harboring the equivalent of 5 pmol of BrlR-V5/His₆ protein and 1 pmol of biotinylated P_mexE in wild-type and ΔPA3177 strains. Representative images are shown. All experiments were carried out in triplicate.

BrlR is a c-di-GMP responsive transcriptional regulator (35). Previous findings demonstrated BrlR to be absent or detectable at only low levels in total extracts characterized by low c-di-GMP levels, and unable to bind to its target promoters P_brlR, P_mexA, and P_mexE, resulting in significantly increased susceptibility but reduced brlR, mexA, and mexE transcript levels (17, 33–36). However, BrlR levels and BrlR-DNA binding was restored upon or overexpression of a diguanylate cyclase or the addition of c-di-GMP to the binding assay, respectively (17, 35). Given that ΔPA3177 biofilm cells are characterized by increased susceptibility to antimicrobial agents, low c-di-GMP levels, and significantly reduced BrlR levels, we next sought to determine whether BrlR produced by PA3177 mutant biofilm cells is capable of binding to DNA. To analyze the DNA binding capabilities of BrlR, we made use of a recombinant BrlR harboring a His₆V5 sequence at the C terminus of BrlR (referred to as His₆/V5-tagged BrlR). The gene encoding the recombinant protein, cloned into pMJT1 under the control of the arabinose inducible P_BAD promoter, was transferred into P. aeruginosa wild-type and ΔPA3177 strains. The resulting strains (PAO1/pMJT-brlR-His₆/V5 and ΔPA3177/pMJT-brlR-His₆/V5) were grown as biofilms for 3 days, and then the biofilm cells were harvested and lysed to obtain total cell extracts. The resulting cell extracts harboring tagged BrlR protein were used in DNA binding assays with the BrlR-target promoter DNA P_mexE. While wild-type produced BrlR did bind to the P_mexE target promoter DNA, very little or no binding was noted for BrlR obtained from total cell extracts of ΔPA3177 biofilm cells (Fig. 7B).

Considering that overexpression of PA3177 restored not only the c-di-GMP levels and the susceptibility phenotype but also BrlR levels of biofilms to wild-type levels (Fig. 4, 6, and 7A), we next determined whether expression of PA3177 could restore BrlR function by using the overexpression vector pJN-PA3177. As shown in Fig. 7B, multicopy expression of PA3177 coincided with restoration of DNA binding by His₆/V5-tagged BrlR obtained from ΔPA3177/pJN-PA3177/pMJT-brlR biofilm extracts to P_mexE.

DISCUSSION

The goal of the present study was to determine whether there is a pool of c-di-GMP modulating enzyme(s) that contributes to the biofilm drug tolerance of P. aeruginosa. Our transcriptome analysis revealed an increase in the transcript abundance of several genes encoding c-di-GMP modulating enzymes in biofilm relative to planktonic cells. Among the genes demonstrating increased transcript abundance in the sessile mode of growth was PA3177. PA3177 encodes a diguanylate cyclase that, like WspR and SadC, harbors a conserved GGDEF domain, where the GGEEF motif corresponds to the active site (A site) (46–49). Given that PA3177 was found to affect motility and intracellular c-di-GMP levels in biofilms by P. aeruginosa PAO1 and PA14, with the GGAAF variant of PA3177 generated in this study, being inactive, our findings support PA3177 encoding an active diguanylate cyclase.

With the exception of PA2818 encoding a phosphodiesterase, the diguanylate cyclase PA3177 was the only factor among the candidate genes tested to contribute to the susceptibility phenotype of P. aeruginosa biofilms to tobramycin and hydrogen peroxide. Moreover, inactivation of PA3177 coincided with P. aeruginosa biofilm cells being eradicated upon exposure to tobramycin, while wild-type biofilm cells persisted. The drug tolerance phenotype of PA3177 mutant biofilms was restored upon overexpression of intact PA3177 but not PA3177_GGAAF harboring substitutions in the active site. Our findings are in agreement with previous reports of the role of PA3177 affecting the susceptibility of P. aeruginosa PA14 biofilms to hypochlorite without affecting the MIC to hypochlorite under planktonic growth conditions (45). However, while the authors of that study demonstrated that PA3177 contributes to the biofilm susceptibility by PA3177 being induced upon exposure to hypochlorite (45), we found here that PA3177 contributes to drug tolerance in a manner dependent on the levels of c-di-GMP and BrlR. More specifically, we demonstrate that the restoration of drug tolerance coincided with the restoration of BrlR to wild-type levels. The contribution of c-di-GMP to biofilm tolerance is in agreement with previous findings demonstrating that restoration of c-di-GMP levels to wild-type biofilm-like levels restored BrlR production and DNA binding by BrlR to wild-type levels (17, 34, 35). Thus, our findings clearly indicate resistance to antimicrobial agents being linked to the level of the intracellular c-di-GMP, with PA3177 being the c-di-GMP modulating enzyme contributing to the drug tolerance phenotype of biofilms.

More important, however, is the finding that PA3177 plays no role in the formation of P. aeruginosa biofilms. This is supported by the finding of biofilms by ΔPA3177 being comparable in architecture to wild-type biofilms. Our findings suggest that while PA3177 is expressed during biofilm growth, the c-di-GMP produced by PA3177 does not appear to contribute to biofilm-associated characteristics typically associated with diguanylate cyclase (3, 6–11). Our findings are in agreement with reports by Strempel et al. (45) indicating that biofilms produced by the ΔPA3177 mutant and the isogenic parental strain P. aeruginosa PA14 are indistinguishable in architecture. Instead, our findings suggest PA3177 to only contribute to the drug tolerance phenotype of biofilms. Diguanylate cyclase PA3177 contributing only to biofilm drug tolerance is in agreement with recent findings by Petrova et al. (33) suggesting the presence of two different pools of c-di-GMP, one contributing to biofilm formation while the second contributes to biofilm drug tolerance by P. aeruginosa.

For PA3177-generated c-di-GMP to only contribute to biofilm drug tolerance but not to biofilm formation, c-di-GMP must be functionally separated, sequestrated, or locally confined within the cell. There is increasing evidence that c-di-GMP signaling occurs in “microcompartments” (reviewed in reference 1). Such microcompartments are likely created by multiprotein complexes that comprise a specific diguanylate cyclase and/or phosphodiesterase and have in common that they are controlled by distinct input signals, as well as specific effector and target components, which associate by specific protein-protein or protein-DNA interactions (1). For example, the HD-GYP-type phosphodiesterase domain of RpfG in Xanthomonas axonopodis pv. citri was found to interact directly with several GGDEF domain proteins (59). Likewise, specific interactions were detected between the diguanylate cyclase HmsT, the phosphodiesterase HmsP, the putative glycosyltransferase HmsR, and its accessory factor HmsS of Yersinia pestis, all of which are attached to the inner membrane and are required for biofilm-associated synthesis and excretion of exopolysaccharide matrix substances (60). Similarly, the membrane-bound diguanylate cyclase NicD by P. aeruginosa has been reported to be closely associated with the chemosensory protein BdlA and two phosphodiesterases (61). The multiprotein complex was not only found to be attached via NicD to the inner membrane but to result upon perception of exogenous dispersion cues in a short-lived burst of c-di-GMP (61).

The findings suggest close association of relevant protein components to likely coincide with c-di-GMP being locally available and/or be prevented from diffusing away. While it is not known at this time whether PA3177 is functional within a multiprotein complex, our results support the notion of biofilm drug tolerance in P. aeruginosa to coincide with a locally confined pool of c-di-GMP. Moreover, while c-di-GMP is generally believed to contribute to the formation of biofilms (3), our data on PA3177-dependent c-di-GMP contributing to the drug tolerance of biofilms but not the the formation of biofilms adds a new dimension to the role of c-di-GMP in biofilms.

MATERIALS AND METHODS

Bacterial strains, plasmids, media, and culture conditions.

All bacterial strains and plasmids used in this study are listed in Table 1. P. aeruginosa strain PAO1 was used as the parental strain. All planktonic cultures were grown at 37°C and 220 rpm in Erlenmeyer flasks for 16 to 18 h in Lennox broth (LB; BD Biosciences). For plasmid maintenance, antibiotics were used at the following concentrations: 50 to 75 μg/ml gentamicin and 200 to 250 μg/ml carbenicillin for P. aeruginosa and 20 μg/ml gentamicin and 50 μg/ml ampicillin for E. coli.

TABLE 1.

Strains and plasmids

Strain or plasmid	Relevant genotype or descriptiona	Source or reference
Strains
Escherichia coli
DH5α	F− ϕ80lacZΔM15 Δ(lacZYA-argF)U169 recA1 endA1 hsdR17(rK− mK+) phoA supE44 thi-1 gyrA96 relA1 tonA	Life Technologies
Pseudomonas aeruginosa
PAO1	Wild type	B. H. Holloway
PA0169::IS	PAO1, PA0169::ISphoA; Tetr	77
PA1851::IS	PAO1, PA1851::ISphoA; Tetr	77
PA2818::IS	PAO1, PA2818::ISphoA; Tetr	77
PA2870::IS	PAO1, PA2870::ISphoA; Tetr	77
PA3177::IS	PAO1, PA3177::ISphoA; Tetr	77
ΔPA3177	PAO1, ΔPA3177	This study
PA3343::IS	PAO1, PA3343::ISphoA; Tetr	77
ΔPA4843	PAO1, ΔPA4843 (gcbA)	62
ΔPA5017	PAO1, ΔPA5017 (dipA)	3
ΔsagS	PAO1, ΔsagS (PA2824)	57
Plasmids
pET101D	Vector for directional cloning and V5/6XHis fusion protein expression; Ampr	Life Technologies
pEX18Gm	Allelic replacement suicide vector; pUC18 MCS; oriT+; sacB+; Gmr	63
pRK2013	Helper plasmid for triparental mating; mob; tra; Kmr	78
pJN105	Arabinose-inducible gene expression vector; pBRR-1 MCS; araC-PBAD; Gmr	65
pMJT-1	Arabinose-inducible gene expression vector; pUCP18 MCS; araC-PBAD; Amp/Carbr	64
pJN-PA3177	C-terminal V5-tagged PA3177 cloned into pJN105 at NheI/SacI	This study
pJN-PA3177_GGAAF	C-terminal V5-tagged PA3177 cloned into pJN-105 with mutation GGAAF; Gmr	This study
pMJT-brlR-V5/His6	C-terminal V5/His6-tagged brlR cloned into pMJT1 at NheI/XbaI; Amp/Carbr	37
PbrlR120-brlR-V5/His6	C-terminal V5/His6-tagged brlR with native promoter cloned into mini-CTX-lacZ at XbaI/XmaI; Tetr	17

^aCarb^r, carbenicillin resistance; Gm^r, gentamicin resistance; Km^r, kanamycin resistance; Tet^r, tetracycline resistance; Ap^r, ampicillin resistance.

Strain construction.

Isogenic mutant for PA3177 was constructed by allelic replacement using sucrose counterselection as previously described (37, 62) using the gene replacement vector pEX18Gm (63). C-terminal tagging was achieved by introducing the sequence of a V5 tag into the indicated genes via PCR using a primer (Table 2). The tagged constructs were cloned into pJN105 or pMJT-1, and the vectors were subsequently introduced into P. aeruginosa. Site-directed mutagenesis of GGEEF domain to GGAAF domain was achieved by utilizing Q5 site-directed mutagenesis kit (NEB) according to the manufacturer's protocol. The identity of vector inserts was confirmed by PCR and sequencing. Complementation and overexpression was achieved by placing the respective gene under the control of an arabinose-inducible P_BAD promoter in the pJN105 or pMJT-1 vectors (64, 65). The primers used for strain construction are listed in Table 2.

TABLE 2.

Oligonucleotides

Procedure and oligonucleotide	Sequence (5′–3′)a
PCR
PA3177_F	TCGCCTTAGCCCTACTGTCG
PA3177_R	CAGCCATAAGAGGCAGGACC
PA3177_Del_chk_F	GGTACTGATCGATCTCGTCGC
PA3177_Del_chk_R	GATCCTGGCTATCGCGCTC
qRT-PCR
PA3177_F	CTCCCAAGCAGTCCAATA
PA3177_R	GAAGAACAGGCCGAGAATC
mreB_F	CTTCATCAACAAGGTCCACGA
mreB_R	GCTCTTCGATCAGGAACACC
Cloning
PA3177_NheI_F	GCGCGCGCgctagcATGGCTCCTCCCAAGCAGTC
PA3177_V5_SacI_R	GCGCGCGCgagctcTCACGTAGAATCGAGACCGAGGAGAGGGTTAGGGATAGGCTTACCGGCGCAGCGCAGGCGGTTG
PA3177-_Del_F1_EcoRI	GCGCGCGCgaattcCCTGGTTCATCAACGCCTTC
PA3177-_Del_R1_BamHI	GCGCGCGCggatccCGTCCAGCGGGATCAGTT
PA3177-_Del_F2_BamHI	GCGCGCGCggatccTCGTCCTGCTGTCCAGCAC
PA3177-_Del_R2_HindIII	GCGCGCGCaagcttGACTTCATCACTGTCTACAAGCC
PA3177_GGAAF_F	CTTCGGCGGCgcggcaTTCCTCGTCC
PA3177_GGAAF_R	CGGAACACCATGTCGACATTGC
Vector specific
pJN105_MCS_F	TAGCGGATCCTACCTGACGC
pJN105_MCS_R	CCATTCGCCATTCAGGCTG
pMJT1_MCS_F	GACCGCGAATGGTGAG
pMJT1_MCS_R	GAGCTGATACCGCTCG
Streptavidin bead binding assay
PmexE_F	GGATCAGCATGTTCATCG
PmexE_Rb	CTGTTCCATGCTTGACTC

^aRestriction sites are denoted in lowercase letters.

^bBiotinylated.

qRT-PCR.

Isolation of mRNA and cDNA synthesis was carried out as previously described (66–68). qRT-PCR was performed using a Bio-Rad CFX Connect Real-Time PCR detection system and SsoAdvanced Universal SYBR green Supermix (Bio-Rad) with the oligonucleotides listed in Table 2. mreB was used as the control. The differential expression analysis was performed using CFX Manager software (Bio-Rad) by first normalizing transcript abundance (based on the threshold cycle value (C_T) to mreB, followed by determining the transcript abundance ratios. Melting-curve analysis was used to detect single product amplification. Increased multicopy gene expression of PA3177 was evaluated by comparing PA3177 transcript abundance in strains by ΔPA3177/pMJT-1 relative to ΔPA3177/pMJT-PA3177. Tobramycin-inducible gene expression of PA3177 was evaluated in P. aeruginosa PAO1 and PAO1 exposed to tobramycin (20 μg/ml) for 30 min.

Biofilm growth conditions.

For biofilm antibiotic susceptibility, biofilm-MBC assays, and RNA extraction, biofilms were grown in a once-through continuous flow reactor system with size 13 (inner surface area, 25 cm²) Masterflex silicon tubing (Cole Parmer) for 3 days as previously described (68–70). For the analysis of the biofilm architecture, the biofilms were grown in flow cells (BioSurface Technologies) for 5 days as previously described (68–70). All biofilms were grown at 22°C in 20-fold-diluted LB medium. To maintain plasmid expression, gentamicin at 2 μg/ml and carbenicillin at 10 μg/ml were added to the growth medium. Quantitative analysis of the confocal laser scanning microscope images of the flow cell grown biofilms was performed using COMSTAT (52). The wild-type PAO1 and respective mutant strains harboring the empty plasmids pJN105 or pMJT-1 were used as vector controls to demonstrate that the presence of the empty plasmid has no effect.

Attachment.

Attachment capabilities of bacterial cells were tested using CV microtiter plate assay as previously described (50). Briefly, overnight cultures were adjusted to an optical density at 600 nm of 0.05 and grown in fresh LB medium in 96-well microtiter plate. After 24 h of growth at 37°C with shaking at 220 rpm, the cultures were stained with 0.1% crystal violet for 15 min at room temperature, followed by a washing with distilled water. The stain was then solubilized for 10 min with 95% ethanol and quantified in terms of the absorbance at 570 nm.

Biofilm antibiotic susceptibility testing.

The susceptibility of biofilm cells to antibiotic treatment was determined by growing biofilms for 3 days using a continuous-flow reactor. Biofilm cells in the continuous-flow reactor were then exposed to tobramycin (150 μg/ml) and hydrogen peroxide (0.6%) for 1 h under flowing conditions. Upon completion of the treatment, the cells were harvested and homogenized to disrupt any aggregates and cell clusters. The homogenized cells were serially diluted, and up to 7 dilutions were drop plated onto LB agar. After overnight incubation at 37°C, the CFU were determined. Viability was determined via CFU counts, and susceptibility was expressed as a log reduction. Under the conditions tested, a total of 1.1 × 10¹⁰ CFU/biofilm were obtained on average following 3 days of biofilm growth.

Biofilm-MBC.

The MBC is the concentration required to prevent the growth of bacterial cells in antibiotic-free media after exposure to antibiotic (22). Biofilm-MBC is defined as the concentration at which no further increase in log reduction is observed (54–56). To determine the biofilm-MBC, biofilm cells were grown for 3 days, after which the medium was switched to one containing antibiotics at concentrations ranging from 75 to 300 μg/ml tobramycin. The chosen concentration range has previously been shown to demonstrate recalcitrance to killing by wild-type biofilm cells, while complete eradication was achieved for ΔbrlR mutant biofilm cells (37). The cells were treated for 24 h and subsequently harvested, homogenized, and plated to measure viability.

Streptavidin magnetic bead DNA binding assay.

BrlR binding to the promoter region of mexE operon was determined using the streptavidin magnetic bead DNA binding assay as previously described (34, 35, 71). A reaction mixture was prepared containing 50 ng/μl poly(dI-dC) as nonspecific competitor DNA, 10× ligase buffer (Invitrogen), 200 mM EDTA, 1 pmol of biotinylated DNA, nonbiotinylated DNA competitor, and 5 pmol of His₆/V5-tagged BrlR protein extract. The binding reaction mixture was incubated for 30 min at room temperature. Streptavidin magnetic beads (100 μg; Thermo Scientific) were used to capture biotinylated DNA. After four washes, the proteins copurifying with the biotinylated DNA were separated by SDS–11% PAGE and assessed by immunoblot analysis for the presence of BrlR using anti-V5 antibodies (Invitrogen). An aliquot prior to the addition of streptavidin magnetic beads was used to determine total BrlR present in each DNA binding assay (loading control).

Motility assays.

Swimming motility was assessed at an agar concentration of 0.3% with 0.1% tryptone and swarming motility at 0.4% agar concentration using 5× concentration M8 medium, as previously described (72–74).

c-di-GMP extraction.

Intracellular c-di-GMP levels were quantitated by extracting c-di-GMP in triplicate using ethanol-heat precipitation (75) and quantitated essentially as previously described (57, 76). The supernatants harboring c-di-GMP were dried using a Speed-Vac and stored at −80°C until further analysis. The precipitate was dissolved in 200 μl water and 20 μl was analyzed using an Agilent 1100 high-performance liquid chromatography (HPLC) system equipped with an autosampler, degasser, and detector set to 253 nm and were separated using a reverse-phase C₁₈ Targa column at a flow rate of 0.2 ml/min to quantitate c-di-GMP levels. For separation, 10 mM Ammonium acetate dissolved in water was used as solvent A, and 10 mM ammonium acetate dissolved in methanol was used as solvent B. The following HPLC solvent gradient was used to elute c-di-GMP: 0 to 9 min, 1% B (= 1% solvent B and 99% solvent A); 9 to 14 min, 15% B; 14 to 19 min, 25% B; 19 to 26 min, 90% B; and 26 to 40 min, 1% B.

Data availability.

Strain, plasmid, and other data will be provided upon request.