Cyclic AMP-Independent Control of Twitching Motility in Pseudomonas aeruginosa

Abstract

FimV is a Pseudomonas aeruginosa inner membrane hub protein that modulates levels of the second messenger, cyclic AMP (cAMP), through the activation of adenylate cyclase CyaB. Although type IVa pilus (T4aP)-dependent twitching motility is modulated by cAMP levels, mutants lacking FimV are twitching impaired, even when exogenous cAMP is provided. Here we further define FimV's cAMP-dependent and -independent regulation of twitching. We confirmed that the response regulator of the T4aP-associated Chp chemotaxis system, PilG, requires both FimV and the CyaB regulator, FimL, to activate CyaB. However, in cAMP-replete backgrounds-lacking the cAMP phosphodiesterase CpdA or the CheY-like protein PilH or expressing constitutively active CyaB-pilG and fimV mutants failed to twitch. Both cytoplasmic and periplasmic domains of FimV were important for its cAMP-dependent and -independent roles, while its septal peptidoglycan-targeting LysM motif was required only for twitching motility. Polar localization of the sensor kinase PilS, a key regulator of transcription of the major pilin, was FimV dependent. However, unlike its homologues in other species that localize flagellar system components, FimV was not required for swimming motility. These data provide further evidence to support FimV's role as a key hub protein that coordinates the polar localization and function of multiple structural and regulatory proteins involved in P. aeruginosa twitching motility.IMPORTANCEPseudomonas aeruginosa is a serious opportunistic pathogen. Type IVa pili (T4aP) are important for its virulence, because they mediate dissemination and invasion via twitching motility and are involved in surface sensing, which modulates pathogenicity via changes in cAMP levels. Here we show that the hub protein FimV and the response regulator of the Chp system, PilG, regulate twitching independently of their roles in the modulation of cAMP synthesis. These functions do not require the putative scaffold protein FimL, proposed to link PilG with FimV. PilG may regulate asymmetric functioning of the T4aP system to allow for directional movement, while FimV appears to localize both structural and regulatory elements-including the PilSR two-component system-to cell poles for optimal function.

Keywords: Pseudomonas aeruginosa; subcellular localization; twitching motility; type IV pili.

FIG 1

FimV and PilG have separate cAMP-independent roles in T4aP function. (A) Loss of fimL reduces the levels of PilU (a proxy for intracellular cAMP levels) (25), surface piliation, and twitching motility, and all phenotypes are restored to wild-type levels in a cdpA fimL double mutant. In contrast, the fimV and pilG mutants lack surface piliation and motility even in a cpdA background, even though PilU levels are restored to wild-type or greater levels. Negative controls (−) were pilU or pilA mutant bacteria, respectively. Scale bar = 1 cm. (B) Quantification of relative PilU levels and twitching zone diameters compared to those of wild-type (wt) PAK (set to 100%); measurements are the average of three experiments. Data were analyzed by repeated-measures ANOVA by using the single mutants as the control data set versus double mutants. Matching letters indicate statistically significant differences (P < 0.05).

FIG 2

FimV and its LysM motif are essential for twitching motility. (A) Schematic of the FimV fragments used in this work. FimV's periplasmic and cytoplasmic domains are connected by a single transmembrane segment. The N-terminal periplasmic domain contains a PG-binding LysM motif (green square), which is deleted in frame in the fimV_LysM mutant (white square). (B) The levels of PilU (a proxy for intracellular cAMP levels) (25) and assembly machinery components PilMNOPQ were assessed by Western blotting in various fimV mutant backgrounds complemented with the empty vector or with full-length fimV; a representative blot is shown. Surface piliation and twitching motility are shown in the bottom two rows. Scale bar = 1 cm. (C) Quantification of protein levels and motility in three independent experiments. Note that PilQ multimer levels cannot be accurately quantified by Western blot analysis. Statistically significant changes were identified with a paired t test. In the absence of fimV, levels of PilU, PilN, and PilO and PilQ multimers were reduced and surface piliation and motility were lost. These phenotypes were restored by complementation with fimV in trans. The fimV₁₁₉₄ mutant expressing only the cytoplasmic domain phenocopied the fimV deletion mutant. When FimV's LysM motif was deleted, the levels of the proteins of interest and piliation were similar to those of the wild type but motility remained severely impaired. CyaB requires FimV for its activity, but in the absence of fimV, it can be constitutively activated by an R456L mutation (36). However, the cells have very few surface pili and are nonmotile. Mutants lacking pilU, pilM, pilN, pilO, pilP, pilQ, and pilA were used as negative controls (−) for relevant immunoblot assays.

FIG 3

CyaB activation—but not twitching motility—requires that FimV's cytoplasmic and periplasmic domains be connected. (A) Representative Western blot analysis of whole-cell lysates with anti-PilU antiserum and quantification of PilU levels (a proxy for cAMP levels) by densitometry (n = 3). The reduced levels of PilU in the fimV and fimV₁₁₉₄ mutants are not restored to wild-type (wt) levels by in-trans expression of the N-terminal domain (FimV₅₀₇). A mutant lacking pilU was used as the negative control. (B) Quantification of twitching motility. As reported previously (29), expression of pFimV₅₀₇ in trans complements motility in the fimV₁₁₉₄ mutant, which expresses only the cytoplasmic domain of FimV. It cannot complement the motility of a fimV deletion mutant, even when cAMP levels are restored by expressing constitutively active CyaB-R456L. Data were analyzed by paired t test with GraphPad Prism.

FIG 4

PilS localization depends on FimV. (A) Bright-field microscopy and fluorescence microscopy were used to image PilS-YFP localization in the wild type, fimV mutants, or a pilG mutant, and the average pixel intensity of the YFP signal along the long axis of the cell was quantified with MicrobeJ (51). The number of cells averaged for each population is shown on each graph. PilS-YFP is localized to the poles in wild-type cells, but fluorescence becomes circumferential in the fimV and fimV₁₁₉₄ mutant backgrounds. In a mutant expressing FimV with an in-frame deletion of its LysM PG-binding motif, fluorescence is polar but patchy circumferential fluorescence is also visible. PilS-YFP remains polar in a pilG mutant. Scale bar = 5 μm. (B) Representative Western blot analysis of whole-cell lysates with an anti-YFP antibody. Levels of PilS-YFP were equivalent in all strains. A cross-reactive protein was used as the loading control. The values to the left are molecular sizes in kilodaltons.

FIG 5

Swimming and swarming motilities of wild-type (wt) and mutant PAK strains. (A) Representative assays of swimming motility in 0.3% LB agar. Scale bar = 1 cm. (B) Quantification of swimming zones (average of three experiments). Lowercase letters indicate paired samples that are statistically significantly different from one another as measured by ANOVA by using the single mutants as the control data set when comparing the double mutants in the appropriate background. Loss of fimV does not impact swimming motility. PAK is the parent strain; the fliC mutant that lacks flagellin was included for comparison. (C, D) Representative Western blot assays of whole-cell lysates from strains grown in liquid medium with anti-PilU antiserum (C) and of sheared surface protein preparations with an antibody raised against sheared proteins (recognizes both pilins and flagellins) (D). The values to the left are molecular sizes in kilodaltons. Piliation levels are similar to those of samples prepared from solid medium (Fig. 1). (E) Representative swarming zones. (F) Quantification of swarming zones (average of three experiments).