Fig 1.
The absence of the V. cholerae flagellum filament elicits a flagellum-dependent biofilm regulatory response.
A) Illustration showing the main components of the polar flagellum in V. cholerae with the proteins forming these components shown in brackets. The flagellum sheath is not depicted in this figure. The structures targeted in this study are color coded. B) Representative images of the smooth colony morphology of the WT strain and the corrugated colony morphology of the ΔflaA strain. Scale bars = 1 mm. C) Bar graph of means and standard deviations of relative luminescent units (RLU) obtained from the transcription of vpsL-luxCDABE in colonies of the WT and ΔflaA strains. D) Bar graph of means and standard deviations of c-di-GMP concentration measured by LC-MS/MS in colonies of the WT and ΔflaA strains. E) Bar graph of means and standard deviations of RLU obtained from the expression of the c-di-GMP biosensor in colonies of the WT and ΔflaA strains. Means obtained from three biological replicates were compared with an unpaired t-test. Mean differences with a P value ≤ 0.05 were deemed significant. **** p ≤ 0.0001.
Fig 2.
The FDBR response in the ΔflaA strain requires the presence of the stator and T ring.
A) Representative images of the colony morphologies of the WT strain and strains lacking genes encoding stator and T-ring components in a WT or ΔflaA genetic background. Scale bars = 1 mm. B) Bar graph of means and standard deviations of RLU obtained from the transcription of vpsL-luxCDABE in colonies of the WT and single mutants lacking stator and T-ring genes. C) Bar graph of means and standard deviations of relative fluorescence intensity (RFI) obtained from the expression of the c-di-GMP biosensor in the WT and single mutants lacking stator and T-ring genes. D) Bar graph of means and standard deviations of RLU obtained from the transcription of vpsL-luxCDABE in colonies of the ΔflaA strain and ΔflaA double mutants lacking stator and T-ring genes. E) Bar graph of means and standard deviations of RFI obtained from the expression of the c-di-GMP biosensor in colonies of the ΔflaA strain and ΔflaA double mutants lacking stator and T-ring genes. Means obtained from 3 biological replicates were compared to WT or ΔflaA with a one-way ANOVA followed by Dunnett’s multiple-comparison test. Adjusted P values ≤ 0.05 were deemed significant. *** p ≤ 0.001; **** p ≤ 0.0001. ns not significant. The color of each bar represents the type of flagellum structure to which each gene product belongs as depicted in Fig 1A.
Fig 3.
In the ΔflaA strain, PomB variants with defects in Na+ binding or absence of Na+-NQR components impact colony corrugation.
A) Representative images of the colony morphologies of strains with mutations in pomB in WT and ΔflaA backgrounds. B) Representative images of the colony morphologies of strains lacking subunits of the Na+-NQR complex or the Na+ symporter SssA in the WT and ΔflaA backgrounds. Experiments were performed on 3 biological replicates. Scale bars = 1 mm.
Fig 4.
The absence of the flagellum filament and/or stator influences biofilm formation.
Representative images of flow cell biofilm competition experiments using 1:1 mixtures of WT-RFP (cyan) and WT-GFP or mutant-GFP (yellow) strains. Images were obtained at 40x magnification at stages typical of initial surface attachment (1 hour, 1H), microcolony development (6 hours, 6H), and mature biofilm (24 hours, 24H). Images were generated using Imaris software. Insets in the upper left corners of 1- and 3-hour images are magnifications of regions from the same image that depict single cells and initial microcolonies. Cross sections of the XZ planes are shown for images taken at 6 and 24 hours. Images are representative of a minimum of 3 biological replicates per strain with three technical replicate images obtained per biological replicate at each time point. Scale bars = 20 μm.
Fig 5.
The absence of the flagellum filament and/or stator alters the dynamics of c-di-GMP accumulation and MSHA-surface abundance.
A) Plot of the median RFI for individual WT, ΔflaA, ΔmotX, and ΔflaA ΔmotX cells attached to the surface inside flow cells. For each time point and strain, the distribution of RFI values were obtained from 2 independent experiments, and the median was calculated from these distributions. Time t = 0 h corresponds to the start of image acquisition after flow started, not inoculation time, for a more unbiased comparison of surface attached cells between strains with and without attachment defects. Error estimates for these RFI values, in the form of 95% confidence intervals, are shown in supplementary S1 Fig. B) Surface MshA levels determined by MSHA-specific hemagglutination (HA) assay. The HA titer is defined as the reciprocal of the lowest dilution at which agglutination of sheep erythrocytes was observed for each strain. Equivalent cell numbers were used for each strain, normalized by OD600. Bar graph of means with standard error of the mean of MSHA-specific HA titer. Data were obtained from 5 biological replicates, with 2 technical replicates for each biological replicate per strain. Each mutant HA titer compared to WT via unpaired two-tailed Student’s t-Test, ΔmshA ***p = 0.0002, ΔflaA ns p = 0.1241, ΔmotX **p = 0.0106, ΔflaAΔmotX *p = 0.0133. C) Analysis of surface-attachment ability. A total of 4 biological replicates were analyzed for each strain, and data is presented as mean with the standard deviation. Each mutant compared to WT via unpaired two-tailed Student’s t-Test, ΔflaA ns p = 0.7842, ΔmotX *p = 0.0141, ΔflaAΔmotX **p = 0.0018.
Fig 6.
Three DGCs are required for the FDBR response in the ΔflaA strain.
A) Representative images of the colony morphologies of indicated strains. Scale bars = 1 mm. B) Bar graph of means and standard deviations of RLU obtained from the transcription of vpsL-luxCDABE in colonies of indicated strains. C) Bar graph of means and standard deviations of RFI obtained from the expression of the c-di-GMP biosensor in indicated strains. D) Bar graph of means and standard deviations of RLU obtained from the transcription of vpsL-luxCADBE in colonies of indicated strains. E) Bar graph of means and standard deviations of RFI obtained from the expression of the c-di-GMP biosensor in indicated strains. F) Plots of the median RFI for ΔcdgALO and ΔflaA ΔcdgALO cells on the surface in flow cells. Data from the WT, ΔflaA, ΔmotX, and ΔflaA ΔmotX strains are the same as in Fig 5. Flow cell experiments from Fig 5 and Fig 6 were done in parallel and separated for clarity. For each time point and strain, the distribution of RFI values were obtained from 2 independent experiments, and the median was calculated from these distributions. Time t = 0 h corresponds to when image acquisition began after flow started, not inoculation time for a more unbiased comparison of surface attached cells between strains with and without attachment defects. Error estimates for these RFI values, in the form of 95% confidence intervals, are shown in supplementary S1 Fig. Means were compared to WT or ΔflaA with a one-way ANOVA followed by Dunnett’s multiple-comparison test. Adjusted P values ≤ 0.05 were deemed significant. * p ≤ 0.05; **** p ≤ 0.0001. Experiments were done on 3 biological replicates.
Fig 7.
Strains lacking flagellum regulators or flagellum components have an FDBR response.
A) Representative images of the colony morphologies of the WT strain and strains lacking a variety of flagellum regulators and flagellum components (some of these images are also presented in S5 Fig). Scale bars = 1 mm. B) Bar graph of means and standard deviations of RLU obtained from the transcription of vpsL-luxCDABE in colonies of the WT and flagellar mutant strains. C) Bar graph of means and standard deviations of RFI obtained from the expression of the c-di-GMP biosensor in colonies of the WT and flagellar mutant strains. Means obtained from 3 biological replicates were compared to WT with a one-way ANOVA followed by Dunnett’s multiple-comparison test. Adjusted P values ≤ 0.05 were deemed significant. ** p ≤ 0.01; **** p ≤ 0.0001. The color of each bar represents the type of flagellum structure to which each gene product belongs as depicted in Fig 1A.
Fig 8.
CdgA, CdgL, and CdgO are required for FDBR responses in flagellar mutants.
A) Representative images of the colony morphologies of the WT strain and strains with null mutations in flagellar genes and the cdgA, cdgL, and cdgO genes. Scale bars = 1 mm. B) Bar graph of means and standard deviations of RFI obtained from the expression of the c-di-GMP biosensor in colonies of the strains indicated. Means obtained from 3 biological replicates were compared with an unpaired t-test. Each flagellar gene mutated is color coded accordingly to its function as indicated in the illustration in Fig 1A.
Fig 9.
VpsRD59E or absence of HapR do not promote colony corrugation or increased vps-II expression in the ΔflaA ΔmotX strain to the levels observed in the ΔflaA strain.
A) Representative images of colony morphologies of indicated genetic backgrounds. B) Bar graph of means and standard deviations of RLU obtained from the transcription of vpsL-luxCDABE in colonies. Means obtained from at least three independent biological replicates were transformed to adjust for unequal standard deviations and compared to the WT strain with a one-way ANOVA and Dunnett’s multiple-comparison test. Adjusted P values ≤ 0.05 were deemed significant. **** p ≤ 0.0001, ns not significant.
Fig 10.
A model for signal transduction during the FDBR response in V. cholerae.
Illustration showing the connection between main components of the polar flagellum and processes regulated by c-di-GMP signaling. Lines ending in arrows indicate positive regulation and lines ending with a perpendicular line indicate negative regulation.