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

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.

Keywords: BrlR; EMSA; PA3177; biofilm drug tolerance; biofilm susceptibility; c-di-GMP; diguanylate cyclase; immunoblot.

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).

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.

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.

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.

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.

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.

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.