The GrlR-GrlA regulatory system coordinately controls the expression of flagellar and LEE-encoded type III protein secretion systems in enterohemorrhagic Escherichia coli

Abstract

The gene function of the locus of enterocyte effacement (LEE) is essential for full virulence of enterohemorrhagic Escherichia coli (EHEC). Strict control of LEE gene expression is mediated by the coordinated activities of several regulatory elements. We previously reported that the ClpX/ClpP protease positively controls LEE expression by down-regulating intracellular levels of GrlR, a negative regulator of LEE gene expression. We further revealed that the negative effect of GrlR on LEE expression was mediated through GrlA, a positive regulator of LEE expression. In this study, we found that the FliC protein, a major component of flagellar filament, was overproduced in clpXP mutant EHEC, as previously reported for Salmonella. We further found that FliC expression was reduced in a clpXP grlR double mutant. To determine the mediators of this phenotype, FliC protein levels in wild-type, grlR, grlA, and grlR grlA strains were compared. Steady-state levels of FliC protein were reduced only in the grlR mutant, suggesting that positive regulation of FliC expression by GrlR is mediated by GrlA. Correspondingly, cell motility was also reduced in the grlR mutant, but not in the grlA or grlR grlA mutant. Because overexpression of grlA from a multicopy plasmid strongly represses the FliC level, as well as cell motility, we conclude that GrlA acts as a negative regulator of flagellar-gene expression. The fact that an EHEC strain constitutively expressing FlhD/FlhC cannot adhere to HeLa cells leads us to hypothesize that GrlA-dependent repression of the flagellar regulon is important for efficient cell adhesion of EHEC to host cells.

FIG. 1.

Effects of clpXP and/or grlR deletions on the expression of FliC. (A) Coomassie brilliant blue-stained SDS-PAGE profiles of secreted proteins from cells cultured in LB. An arrowhead indicates the FliC band. Strains used: wild type, SKI-5142; grlR, SKI-5152; clpXP, SKI-5149; grlR clpXP, SKI-5156. (B and C) Western blotting analyses of (B) FliC and (C) EspB in equivalent amounts of whole-cell lysates prepared from the same strains used in panel A. Cross-reacting bands recognized by the polyclonal anti-FliC (H7) serum served as loading controls.

FIG. 2.

(A and B) Repression of FliC expression by overproduced GrlA. Shown are Western blotting analyses of FliC prepared from whole-cell lysates containing an equivalent number of cells from the following strains: wild type, SKI-5142; grlR, SKI-5152; grlA, SKI-5153; grlR grlA, SKI-5154; ler, SKI-5143; and pGEM-self or pGEMGA (indicated by − or +, respectively, in panel B). (C) Western blotting analyses of FliC and EspB prepared from whole-cell lysates containing an equivalent number of cells from the same strains with plasmid: wild type, SKI-5142; ler, SKI-5143; pACYC184 or pACGA (indicated by − or +, respectively).

FIG. 3.

Motility phenotypes of wild-type (WT), grlR, grlA, and grlR grlA strains of EHEC (A) and the wild-type strain with pGEM-self (control vector), pGEMGR (carrying grlR), or pGEMGA (carrying grlA) plasmid (B). (C) Cells grown in 1% Bacto-Trypton broth containing 0.5% NaCl were stained with Victoria blue as described in Materials and Methods. The following strains were used: WT, SKI-5142; grlR, SKI-5152; grlA, SKI-5153; and grlR grlA, SKI-5154 (A and/or C) and pGEM-self, SKI-5142 with pGEM-self; grlR⁺, SKI-5142 with pGEMGR; and grlA⁺, SKI-5142 with pGEMGA (B).

FIG.4.

Effect of constitutively expressed FlhD/FlhC on adhesion of EHEC to HeLa cells. (A) The top row was stained with bisbenzimide to see the formation of microcolonies. The bottom row was stained with rhodamine-phalloidin to see accumulated actin under individual attached bacteria. Microcolonies or localized actin accumulations are indicated by arrowheads. The following strains were used: WT/pGEM-self, SKI-5142 with pGEM-self; WT/pGEM-flhDC+, SKI-5142 with pGEMFHDC; grlR/pGEM-self, SKI-5152 with pGEM-self; grlR/pGEM-flhDC+, SKI-5152 with pGEMFHDC. (B) Quantitative analysis of adhesion phenotypes of the strains used in panel A. The bars indicate the numbers of mean bacterial CFU per well. The error bars indicate standard deviations.

FIG.4.

Effect of constitutively expressed FlhD/FlhC on adhesion of EHEC to HeLa cells. (A) The top row was stained with bisbenzimide to see the formation of microcolonies. The bottom row was stained with rhodamine-phalloidin to see accumulated actin under individual attached bacteria. Microcolonies or localized actin accumulations are indicated by arrowheads. The following strains were used: WT/pGEM-self, SKI-5142 with pGEM-self; WT/pGEM-flhDC+, SKI-5142 with pGEMFHDC; grlR/pGEM-self, SKI-5152 with pGEM-self; grlR/pGEM-flhDC+, SKI-5152 with pGEMFHDC. (B) Quantitative analysis of adhesion phenotypes of the strains used in panel A. The bars indicate the numbers of mean bacterial CFU per well. The error bars indicate standard deviations.

FIG. 5.

Model depicting the coordinate regulation of LEE and flagellar gene expressions by the GrlR-GrlA system in EHEC. GrlA activates LEE expression via activation of ler transcription, whereas it inhibits transcription of the flhD operon in response to GrlR levels in the cell. ClpXP positively regulates LEE expression by controlling changes in GrlR and RpoS levels and directly or indirectly represses the flagellar regulon by altering the stabilities of FlhD/FlhC and GrlR. As described in the text, additional regulatory elements, such as Fis, H-NS, IHF, Hha, GadX, YhiE, YhiF, BipA, EtrA, EivF, GrvA, and quorum sensing, have also been implicated in the regulation of LEE.