Fig 1.
FlhF affects surface behaviors of P. aeruginosa.
(A) ΔflhF cells are deficient in biofilm formation. Biomass of static biofilms was measured by crystal violet staining after 24 (solid) and 48 hours (open bars). Bars show mean ± S.D. (n = 4). Both ΔflhF and ΔfliC differ from wild-type PAK (***, p < 0.001), but not from each other (2 way ANOVA with Bonferroni post-test). (B) ΔflhF cells are more likely to rotate after surface tethering. Percentage of rotating vs. non-rotating cells was determined 5 min after tethering to anti-FliC coated slides for PAK (n = 151), ΔflhF (n = 239) and ΔpilA (n = 135) bacteria. Each symbol represents an independent experiment; lines indicate means for each condition. The percentage of rotating ΔflhF cells, but not of ΔpilA, differs significantly from PAK (***, p < 0.001; 2 way ANOVA with Bonferroni post-test). (C) ΔflhF rotation persists over time. Rotating bacteria were identified 5 minutes after tethering to anti-FliC coated slides (t = 0) and observed for 45 min. The proportion of rotating cells (red), attached cells (black), and detached cells (white) was determined in 4–8 independent experiments for PAK, ΔflhF and ΔpilA. Survival curves of rotating cells were analyzed with a Mantel-Cox test. The percentage of rotating ΔflhF cells differed from PAK or ΔpilA (***, p < 0.0001); the percentage of rotating PAK v ΔpilA cells did not differ significantly (p = 0.36).
Fig 2.
The C ring component FliG interacts with FlhF.
FlhF interactions with flagellar proteins were assayed by bacterial two-hybrid assay. ω or Zif fusions were tested as indicated, with interactions resulting in beta-galactosidase expression and activity (reported in Miller units). Bars show mean ± S.D. (n = 3) for a representative experiment. FlhF (WT), which forms a homodimer, served as a positive control (black bar). FliG (G) interacted with wild-type FlhF (WT) (red bars), but gave no signal when co-expressed with either the ω or Zif domain alone (white bars). Other tested rotor components (red bars: FliF (F), FliM (M), or FliN (N)) and stator proteins (blue bars: MotA (A), MotB (B), MotC (C), or MotD (D)) did not show interactions with FlhF.
Fig 3.
FimV(L7P) is a suppressor of FlhF(R251G).
(A) FimV constructs. The predicted signal sequence (SS, red), LysM domain (white), periplasmic region (grey), transmembrane helix (TM, black) and tetratricopeptide repeats (TPR) (blue) are diagrammed. ω or Zif fusions for bacterial 2-hybrid screen are also shown (B2H, green). (B) Loss of FimV function suppresses FlhF(R251G). Swimming zone diameters of PAK (black), PAK fimV(L7P) (red), PA14 (blue) and PA14 fimV::Tn (green) harboring either pUCP (vector control (VC)) or pFlhF(R251G) (open) are shown; each symbol is a biological replicate (line indicates mean). Disruption of fimV significantly alters swimming behavior of FlhF(R251G)-overexpressing bacteria (***, p < 0.001; 2 way ANOVA followed by Bonferroni posttest). (C) Pil/Chp system mutants associated with low intracellular [cAMP] do not suppress FlhF(R251G). Wild-type and transposon mutant bacteria harboring either empty vector (pUCP; solid) or pFlhF(R251G) (open) were assayed for swimming motility. Each symbol is a biological replicate; line indicates mean. Only disruption of fimV suppresses the dominant-negative phenotype associated with FlhF(R251G) overexpression (***, p < 0.001; 2 way ANOVA with Bonferroni posttest).
Fig 4.
The carboxy terminus of FimV mediates interactions with FlhF and is required to stop rotation of tethered bacteria.
(A) FlhF interacts with the carboxy terminus of FimV. FlhF interactions with FimV and Vfr were assayed by bacterial two-hybrid. ω or Zif fusions were constructed as indicated, with interactions resulting in beta-galactosidase expression and activity (reported in Miller units). Bars show mean ± S.D. (n = 3) for a representative experiment. FlhF (FWT), which forms a homodimer, served as a positive control (black bar). FimV (VC) interacted with both wild-type (FWT) and R251G alleles (FR) of FlhF (red bars), but gave no signal when co-expressed with either the ω or Zif domain alone (white bars). Neither allele of FlhF interacted with a FimV construct lacking the final 150 aa (VCΔC, blue bars) nor with Vfr (green bars). (B) Surface-tethered bacteria expressing fimV(L7P) or fimV(ΔC) exhibit persistent rotation. Rotating bacteria identified 5 minutes after tethering on anti-FliC coated slides (t = 0) were observed for 45 min. The proportion of rotating cells (red), attached cells (black), and detached cells (white) was determined over 3–8 independent experiments. Survival curves of rotating fimV(L7P) and fimV(ΔC) differed significantly from PAK (p < 0.0001), but not from ΔflhF (p = 0.53 and p = 0.52, respectively; Mantel-Cox test). (C) fimV(L7P) and fimV(ΔC) assemble a unipolar flagellum. Cells were labelled with anti-FliC antibodies conjugated to Alexa Fluor 488 and visualized by phase and fluorescence microscopy.
Fig 5.
Deletion of both P. aeruginosa adenylate cyclases is required to suppress FlhF(R251G)-mediated inhibition of swimming.
PAK or isogenic mutants lacking one or both of the P. aeruginosa adenylate cyclases were transformed with a plasmid driving constitutive expression of FlhF(R251G) or the empty vector, pUCP. Overnight cultures grown in VBM plus carbenicillin (200 μg/mL) were adjusted to an OD600 of ~0.3 and spotted to 0.3% LB agar plates supplemented with 250 μg/mL carbenicillin. Bacteria were incubated at 30°C for 20h before swimming zones were measured. The wild-type, cyaA, cyaB and cyaAB strains are isogenic.
Fig 6.
Exogenous cAMP reverses the persistent rotation phenotype of surface-tethered flhF, fimV and vfr mutants.
Bacteria were incubated with anti-FliC coated slides for 5 min in the presence or absence of 20 mM cAMP. Rotating bacteria were identified (t = 0) and observed for 45 min. The proportion of rotating (red), attached (black), and detached cells (white) was determined in 3–8 independent experiments. Survival curves were analyzed with the Mantel-Cox test to determine whether exogenous cAMP had a significant effect on persistence of rotation. This was the case for ΔflhF, fimV(L7P), NG01(vfr(G73D)) and ΔflhF pilA (p < 0.0001), but not for the parental strain PAK (p = 0.64).
Fig 7.
Increased flagellar load in ΔfleN bacteria increases cAMP in a FimV-dependent manner.
(A) ΔfleN bacteria assemble increased numbers of polar flagella. Wild-type or ΔfleN cells in which endogenous fliC was replaced with the fliC(T394C) allele were labelled with maleimide-AlexFluor488 and visualized by phase and fluorescence microscopy. (B) cAMP levels measured by EIA. Intracellular cAMP levels were measured by EIA. Bars show mean ± S.D. of triplicate samples from a representative experiment of 3–5 independent assays.
Fig 8.
FlhF-FimV interaction is upstream of a cAMP signal in surface-tethered bacteria.
A. The P. aeruginosa flagellar motor C-ring (outlined in blue) interacts with FlhF (blue solid oval) by bacterial 2-hybrid. A GDP-locked allele of FlhF, FlhF(R251G), has stronger and/or more persistent interaction with FimV. This causes an inappropriate cAMP increase in liquid-grown bacteria that inhibits flagellar rotation. B. When bacteria tether at a surface, an undefined “signal” causes a conformational change in FlhF that allows it to interact with FimV. This leads to an increase in intracellular cAMP that stops flagellar rotation and promotes bacterial cell body attachment to a surface. If either FlhF or FimV are absent, the interaction does not occur, cAMP levels do not rise, and bacterial rotation persists in surface-tethered cells.
Fig 9.
ΔmotAB and ΔmotCD bacteria show distinct distributions of velocity (A) and reversal frequency (B) during liquid swimming.
Cells of each genotype were imaged under dark-field microscopy at 30 fps. Image analysis was carried out as described in Methods using a custom MATLAB code. In (B), boxplots span 1st and 3rd quartiles with a line at the median. The number of cells tracked is shown above each sample in panel B; data from these tracked cells was also used to generate the histogram in panel A, in which the y-axis shows the probability of observing a given velocity.
Fig 10.
The MotAB stator is required for tethered bacteria to stop rotating.
Bacteria were incubated with anti-flagellin antibody coated slides for 5 min in the presence or absence of 20 mM cAMP. (A) Behavior of tethered bacteria over time. Videomicroscopy was used to identify bacteria rotating at t = 0, and the behavior of these cells was followed for 45 min. The proportion of rotating cells (red), attached cells (black), and detached cells (white) was determined in 4–8 independent experiments. To compare the percentage of tethered cells that were rotating in presence or absence of cAMP, survival curves were analyzed with a Mantel-Cox test. Exogenous cAMP had a significant effect on PAK ΔmotAB (***, p = 0.0005), but not on PAK ΔflhF ΔmotAB (ns, p = 0.0916). (B) Reversal frequency of tethered bacteria. Movies of tethered bacteria imaged at 100 fps for 60s were analyzed by a custom Matlab script to determine reversal frequency. Each symbol indicates reversals made by an individual cell over 60s (n = 21–23); lines indicate medians. cAMP increases the reversal frequency of ΔflhF, but decreases reversal frequency of the ΔflhF motAB mutant (Kruskal-Wallis with Dunn’s multiple comparison test; *, p <0.05; **, p < 0.01). (C) Rotation speeds of tethered bacteria. CW and CCW rotation speeds were extracted from tethered bacteria analyzed in panel (B).
Fig 11.
A model for modulating flagellar rotation in response to surface tethering.
1) Swimming bacteria can make contact with a surface via their flagellum. 2) Continued flagellar rotation of these attached bacteria increases the load on the flagellar motor. MotCD must be present for tethered bacteria to rotate, suggesting that it is recruited to the motor in this setting. 3) FlhF interacts with FimV in tethered bacteria, which we propose leads to an increase in intracellular cAMP. cAMP is a sufficient signal to modulate flagellar switching frequency if MotAB is present in the cell. 4) Horizontal attachment of the cell body to a surface follows flagellar surface-tethering if FlhF and FimV are present in cells.