SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

RovA is a MarR-type transcriptional regulator that controls transcription of rovA, the expression of the primary invasive factor invasin and other virulence genes of Yersinia pseudotuberculosis in response to environmental signals. Using a genetic approach to identify regulatory components that negatively influence rovA expression, we identified a new LysR-type regulatory protein, designated RovM, which exhibits homology to the virulence regulator PecT/HexA of plant pathogenic Erwinia species. DNA-binding studies revealed that RovM interacts specifically with a short binding site between promoters P1 and P2 within the rovA regulatory region and negatively modulates rovA transcription in cooperation with the histone-like protein H-NS. The rovM gene itself is under positive autoregulatory control and is significantly induced during growth in minimal media as shown in regulation studies. Disruption of the rovM gene leads to a significant increase of RovA and invasin synthesis and enhances internalization of Y. pseudotuberculosis into host cells. Finally, we show that a Y. pseudotuberculosis rovM mutant is more virulent than wild type and higher numbers of the bacteria are detectable in gut-associated lymphatic tissues and organs in the mouse infection model system. In contrast, elevated levels of the RovM protein, which exert a positive effect on flagellar motility, severely attenuate the ability of Y. pseudotuberculosis to disseminate to deeper tissues. Together, our data show, that RovM is a key regulator implicated in the environmental control of virulence factors, which are crucial for the initiation of a Yersinia infection.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Yersinia pseudotuberculosis is a Gram-negative food-borne pathogen that causes a variety of intestinal and extraintestinal syndromes (yersiniosis), including self-limiting enteritis, diarrhoea, mesenteric lymphadenitis and autoimmune disorders (Bottone, 1997; Smego et al., 1999). The ability of the bacteria to encounter the host mucosal surfaces overlaying the intestinal cells and the subsequent penetration through the M cells of the epithelial cell layer into the Peyer's patches is one of the key features of the disease process (Grutzkau et al., 1990). The outer membrane protein invasin is the most efficient factor that promotes cellular uptake to transfer enteric Yersinia through epithelial layers (Pepe and Miller, 1993; Marra and Isberg, 1997). The transcription of this invasive factor is tightly regulated in response to certain external cues and requires the activator protein RovA, a member of the MarR family of winged-helix transcription factors (Nagel et al., 2001; Ellison and Miller, 2006). The RovA protein forms stable dimers with two central winged-helix DNA-binding motifs and interacts directly with two AT-rich regions within the inv promoter, which both contribute to inv activation (Heroven et al., 2004; Tran et al., 2005).

As previously shown, RovA of enteropathogenic Yersinia is required for efficient invasion of tissue culture cells and colonization of gut-associated lymphatic tissues and organs in a mouse model system. A Yersinia enterocolitica rovA mutant strain was reported to be significantly attenuated in virulence after oral infection, and showed a considerable increase in the LD50 compared with wild type and an inv mutant (Revell and Miller, 2000; Nagel et al., 2001; Dube et al., 2003). RovA analogous proteins (SlyA, Hor, Rap) found in other pathogenic Enterobacteriaceae have also been shown to regulate virulence. For example, SlyA in Salmonella typhimurium is required for survival in macrophages and virulence in mice (Libby et al., 1994; Spory et al., 2002; Stapleton et al., 2002). Recent analysis of the SlyA transcriptome and proteome has identified more than 40 loci that were positively or negatively affected by SlyA, including some that have been previously shown to be important for environmental adaptation and virulence (Spory et al., 2002; Navarre et al., 2005). SlyA of Escherichia coli induces the expression of a cryptic haemolysin (ClyA) and genes required for resistance to heat and acid stress, suggesting that SlyA of E. coli is also crucial for surviving stress conditions in the host (Oscarsson et al., 1999; Spory et al., 2002). Other RovA homologues, such as Rap in Serratia marcescens and Hor in the phytopathogen Erwinia carotovora regulate the production of secondary metabolites, including antibiotics, pigments and exoenzymes implicated in plant pathogenicity (Thomson et al., 1997). This indicated that the RovA/SlyA regulators comprise a subset of the MarR-like family of transcriptional regulators, which control a variety of physiological processes in bacterial pathogens, that are involved in environmental and host-associated stress adaptation and virulence.

The rovA gene in Y. pseudotuberculosis is transcribed by two different promoters P1 and P2, which are located 76 nt and 343 nt upstream of the translational start site (Heroven et al., 2004). Expression studies further revealed that rovA transcription is autoregulated and subject to silencing by the nucleoid-associated H-NS protein (Nagel et al., 2001; Heroven et al., 2004). RovA induces the transcription of its own gene in response to moderate temperature (20–28°C), stationary phase or nutrient rich growth medium by binding to an extended AT-rich sequence located upstream of promoter P2 (Nagel et al., 2001). Under these conditions RovA appears to disrupt the H-NS silencer complex, through the displacement of H-NS from an extended AT-rich promoter segment located upstream of rovA promoter P2 and stimulates RNA polymerase directly, most likely via its C-terminal domain (Heroven et al., 2004; Tran et al., 2005). However, autoactivation at high intracellular RovA levels is blocked by the interaction of RovA with a lower affinity site located downstream of P1, where binding would interfere with RNA polymerase. This supports a model in which RovA tightly controls the activation of its own gene using a concentration-dependent mechanism (Heroven et al., 2004). The identity of the signals or the control factors involved in the upregulation of rovA expression upon the appearance of the appropriate environmental signals is still unknown. In order to decipher the regulatory network, a genetic screen for molecular components of the rovA signalling cascade was performed. A result of this search was the identification of RovM, a LysR-type regulator that acts as a direct repressor of rovA expression in Y. pseudotuberculosis.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Identification of RovM, a LysR-like transcriptional repressor of rovA expression

In order to identify regulatory factors that control rovA expression in response to environmental signals, a plasmid-born gene library from Y. pseudotuberculosis strain YPIII was introduced into the merodiploid rovA–lacZ reporter strain YP38. Approximately 2·104 clones were screened on agar plates containing the indicator X-Gal. Three candidates were isolated which formed lighter blue colonies and exhibited a significantly reduced rovA–lacZ expression level (< 25%) at 25°C. One of the candidates harboured a gene bank plasmid encoding the H-NS protein of Y. pseudotuberculosis, which was previously shown to repress rovA transcription (Heroven et al., 2004), demonstrating that our selection procedure was effective. The other independent clones carried a single, common open reading frame (ORF), encoding a protein of 310 amino acids that bears considerable amino acid identity to transcriptional regulator proteins of the LysR family of Enterobacteriaceae (Schell, 1993). The new LysR family member was tentatively named RovM for modulator of rovA expression. Its monocistronic coding sequence was located between the nuoA-N locus, encoding the NADH dehydrogenase and the aat gene of a putative amino transferase, and is conserved in Yersinia pestis, Y. enterocolitica and other related bacteria. The predicted RovM protein exhibited substantial identity to HexA from E. carotovora (65% identity), and PecT from Erwinia chrysanthemi (66% identity), which control biofilm formation and act as repressors of virulence factors, such as plant cell wall-degrading exoenzymes, in order to cause soft rot disease (Surgey et al., 1996; Harris et al., 1998; Mukherjee et al., 2000). RovM also shared 65% identity with LrhA from E. coli, a key regulator of motility genes, type 1 fimbriae synthesis and biofilm formation, and exhibits 64% identity with a HexA homologue of Photorhabdus temperata, which appears to repress symbiosis with entomophagous nematodes, but induces pathogenicity to insect larvae (Fig. S1) (Lehnen et al., 2002; Joyce and Clarke, 2003; Blumer et al., 2005). To ensure that the LysR-type protein is solely responsible for rovA repression, the rovM gene was subcloned, and the resulting plasmid pAKH42 was introduced into Y. pseudotuberculosis YPIII pAKH47 harbouring a rovA–lacZ fusion. Introduction of the rovM+ plasmid considerably reduced the expression of the rovA–lacZ fusion at a growth temperature of 25°C, at which rovA transcription is normally induced (Fig. 1A). Furthermore, we found that under these conditions no or very little of the endogenous RovA protein was detectable within the whole-cell extract of Y. pseudotuberculosis YPIII harbouring the rovM+ plasmid (Fig. 1B, lane 4), similar to YPIII grown at 37°C, when RovA synthesis is usually repressed (Fig. 1) (Nagel et al., 2001). These data support a model in which RovM encodes a regulatory component that represses rovA expression under normally inducing growth temperatures.

image

Figure 1. rovA expression in Y. pseudotuberculosis wild type and the rovM mutant strain YP41 in the presence or absence of the rovM+ plasmid pAKH42. A. Strains YPIII and YP41 (rovM), harbouring a rovA–lacZ fusion with the empty vector (pAKH85) or a rovM+ plasmid (pAKH42), were grown overnight in LB medium at 25°C or 37°C. β-Galactosidase activity from the overnight cultures was determined and is given in μmol min−1 mg−1 for comparison. The data represent the average ± SD from at least three different experiments each performed in duplicate. B. Whole-cell extracts from overnight cultures of Y. pseudotuberculosis wild-type YPIII and mutant strain YP41 (rovM) transformed with pACYC184 (V) or pAKH42 (rovM+) grown at 25°C or 37°C were prepared, and analysed by Western blotting with a polyclonal antibody directed against RovA. A prestained molecular weight marker (M) is loaded on the left.

Download figure to PowerPoint

Construction and characterization of a rovM mutant strain

In order to evaluate the activity of RovM on RovA synthesis and determine its influence on cell attachment, invasion and virulence, we constructed a rovM mutant strain. To do so, a polymerase chain reaction (PCR)-generated fragment encoding a kanamycin resistance gene with flanking regions of the 5′ and 3′ non-coding sequence of rovM was introduced into Y. pseudotuberculosis YPIII pKOBEG-sacB expressing the phage lambda RED recombinase. A recombinant strain (YP41) with a rovM::kan insertion on the chromosome was cured of pKOBEG-sacB (for details, see Experimental procedures).

To test whether the loss of the RovM protein had an effect on RovA synthesis, we compared rovA–lacZ expression and RovA protein levels in YPIII and the isogenic rovM mutant strain. The absence of the transcriptional repressor led to a significant increase in rovA–lacZ expression (Fig. 1A), and considerably higher amounts of the endogenous RovA protein were detected in the rovM mutant strain grown at 25°C (Fig. 1B, lane 6). This derepression of the rovA gene could be fully overcome by in trans complementation with a rovM+ plasmid (Fig. 1). Interestingly, RovA synthesis was still repressed in the rovM derivative at 37°C, indicating that rovA expression in the absence of RovM is still under temperature control.

As rovA is also regulated in response to growth phase and nutrient availability (Nagel et al., 2001), we further tested the influence of RovM on rovA expression in Y. pseudotuberculosis during different growth stages in complex and minimal media. As shown previously (Nagel et al., 2001), rovA expression in the wild type was significantly repressed in the exponential growth phase (Fig. 2A and B). Expression of the rovA–lacZ fusion in the rovM mutant was also reduced (Fig. 2A) and no RovA protein was detectable during log phase (Fig. 2B). In addition, a strong repression of rovA transcription was found when Y. pseudotuberculosis was grown to stationary phase in minimal media (Fig. 2A and B; (Nagel et al., 2001)). In contrast, the absence of RovM fully eliminated repression of rovA–lacZ transcription under these conditions (Fig. 2A), and identical amounts of RovA were found in the rovM mutant grown in minimal medium and in the wild type grown in Luria–Bertani (LB) (Fig. 2B). In summary, we conclude that the RovM regulatory protein is important for the regulation of rovA expression in response to different growth media, but does not appear to be implicated in temperature and growth phase or growth rate control.

image

Figure 2. Expression of rovA in the Y. pseudotuberculosis rovM mutant in response to different growth phases and growth media. A. Y. pseudotuberculosis YPIII harbouring a rovA–lacZ fusion was grown at 25°C in complex medium (LB) or minimal medium (MMA). Cells were harvested in the exponential phase (A600 = 0.6) and stationary phase (A600 = 4.5), β-galactosidase activity of the cells was determined and is given in μmol min−1 mg−1 for comparison. The data represent the average ± SD from at least three different experiments each performed in duplicate. B. Y. pseudotuberculosis wild type and mutant strain YP41 (rovM) were grown in LB medium or minimal medium (MMA) at 25°C. The bacteria were harvested in the exponential phase (A600 = 0.6) and the stationary phase (A600 = 4.5). Whole-cell extracts were prepared with equal cell numbers, separated on a 15% SDS-polyacrylamide gel and analysed by Western blotting with a polyclonal antibody directed against RovA. A molecular weight marker (M) is loaded on the left. C. Whole-cell extracts from overnight cultures of Y. pseudotuberculosis YPIII grown in LB, MMA, M9 or RPMI medium and YP41 grown in LB at 25°C were prepared and analysed by Western blotting with a polyclonal antibody directed against RovM. A prestained molecular weight marker is loaded on the left.

Download figure to PowerPoint

Environmental control of rovM expression

As the loss of rovM had the most dramatic effect on RovA levels in minimal medium, we examined whether rovA expression in response to different media compositions is mediated by regulation of rovM. To do so, endogenous RovM protein levels were analysed in Y. pseudotuberculosis grown in a complex and different minimal media (Fig. 2C). Considerably more RovM protein was detectable in Yersinia whole-cell extracts upon growth in MMA, RPMI and M9 compared with that in LB medium, indicating that rovA repression in minimal media results due to an increase of RovM. In contrast, no significant increase of RovM synthesis was found under other non-inducing conditions of rovA, such as 37°C and exponential phase growth (data not shown), indicating that temperature- and growth phase-dependent rovA expression does not occur through RovM, but requires additional regulatory factors.

Interaction of RovM with the rovA promoter region

To investigate whether RovM-mediated repression of the rovA promoter is direct or requires other regulatory factors of Y. pseudotuberculosis, we first analysed the expression of a rovA–lacZ fusion in E. coli K-12 strain MC4100 pAKH47 in the presence or absence of a low-copy rovM+ plasmid (pAKH42). We found that the rovA promoter activity was reduced about threefold in E. coli in the presence of RovM, suggesting that RovM influence on the rovA gene in Y. pseudotuberculosis is direct (data not shown).

In order to provide evidence for the interaction of the RovM protein with the rovA promoter region we incubated PCR fragments encompassing different portions of the rovA promoter region with increasing concentrations of purified His-tagged RovM protein and assayed for protein-DNA complex formation (for details, see Experimental procedures). As shown in Fig. 3, a retarded DNA band with decreased motility representing the RovM-DNA complex was obtained with all tested DNA fragments harbouring the rovA promoter sequence from position −68 to −47 with respect to the transcription initiation site of rovA promoter P1. In contrast, no complex formation was found with control fragments (data not shown) and rovA promoter segments including sequences upstream of position −68 or downstream of position −47 (Fig. 3). This demonstrated DNA binding of RovM to the rovA promoter region and suggested a specific RovM binding site closely upstream of the rovA promoter P1.

image

Figure 3. Interaction of RovM with the rovA regulatory region. Individual DNA fragments comprising different portions of the rovA upstream region used for DNA bandshift assays are illustrated. The numbers indicate the nucleotides with respect to the transcriptional initiation site of P1 of rovA. The grey box indicates the potential RovM binding region. The bandshift analysis of chosen fragments is shown below. The rovA promoter fragments were incubated without or with increasing amounts of the purified RovM protein. The DNA-RovM complexes were separated on a 4% polyacrylamide gel. The corresponding molecular weights are indicated on the left. The positions of the rovA fragments are indicated, an arrow shows the higher molecular weight RovM-DNA complex.

Download figure to PowerPoint

To obtain precise information about the RovM binding sites, we performed DNase I footprinting experiments with increasing concentrations of purified RovM bound to DNA fragments encoding the entire rovA regulatory region (Fig. 4). The analysis of both strands revealed approximately the same RovM binding site within the rovA promoter extending from position −80 to −47 and −74 to −47 respectively. Most nucleotides in this region were protected from DNase I digestion, but single nucleotides became hypersensitive towards DNase I in the presence of RovM, suggesting that RovM binding to the rovA promoter is accompanied by a drastic conformational change. Taken together, these results confirmed our gel retardation data and further demonstrated that the recognition site of RovM is distinct and does not overlap with the superimposed RovA and H-NS binding sites within the rovA regulatory region (Fig. 4) (Heroven et al., 2004).

image

Figure 4. Analysis of the RovM binding site within the rovA regulatory region. The Digoxigenin-labelled fragments represent the non-coding (A) or coding (B) strand of the rovA regulatory region incubated with increasing amounts of purified RovM. Reaction mixtures were migrated alongside sequencing reactions of the labelled fragments (lanes C and G). RovM binding sites are indicated by solid lines, and the sites hypersensitive to DNase I are shown by short arrows. The positions of the protected areas are given relative to the transcriptional initiation site of P1rovA. An overview of the rovA regulatory region is shown below. The transcriptional start sites are indicated by broken horizontal arrows. The putative −35 and −10 regions of the rovA promoters are underlined. The numbers indicate the distance in bp from the transcriptional initiation site with respect to P1rovA. The dark grey box below the sequence shows the RovA binding region, the light grey box illustrates the preferential H-NS binding site and the black box indicates the identified RovM binding site.

Download figure to PowerPoint

RovM and H-NS are both required for silencing of the rovA gene

Besides RovM, also binding of the nucleoid-associated protein H-NS to the extended AT-rich region upstream of rovA promoter P2 was shown to reduce rovA transcription in Y. pseudotuberculosis (Heroven et al., 2004). To determine whether H-NS is involved in the regulatory network implicating the RovM repressor, we analysed rovA–lacZ expression in Y. pseudotuberculosis YPIII pAKH47 in the presence and/or absence of functional H-NS and RovM. Because a deletion of the hns gene appears to be deleterious to Y. pseudotuberculosis, we expressed a dominant negative N-terminal H-NS fragment, which was previously shown to abolish H-NS-mediated rovA silencing (Heroven et al., 2004). As expected, rovA–lacZ expression in Y. pseudotuberculosis YPIII was significantly repressed when rovM or hns was expressed from a low-copy plasmid (Fig. 5A). However, H-NS-mediated silencing of the rovA–lacZ fusion was eliminated in the absence of rovM, and vice versa, nearly no repression by RovM was detectable in Y. pseudotuberculosis harbouring the dominant negative hns′ allele. To confirm this result, we performed Western blot analysis to investigate endogenous RovA levels in the different hns- and rovM-positive and -negative backgrounds. In agreement with our previous results, we found that RovA production was considerably increased in the rovM or the hns′ expressing strain, and was completely repressed in the presence of rovM+ and hns+ plasmids (Fig. 5B). In contrast, introduction of the hns+ plasmid was not sufficient to prevent RovA synthesis in the absence of rovM, and plasmid-mediated RovM overproduction was not able to suppress rovA expression in the absence of functional H-NS. These results demonstrated that rovM and hns are both required for efficient silencing of the rovA gene in Y. pseudotuberculosis. In order to provide additional experimental proof, we also performed DNA binding studies with H-NS and RovM, and found that both regulatory proteins are able to bind simultaneously to the rovA promoter (Figs S2 and S3).

image

Figure 5. RovM and H-NS are both required for silencing of the rovA gene. Y. pseudotuberculosis strains YPIII and YP41 (rovM) harbouring a rovA–lacZ fusion with or without an additional rovM+ plasmid pAKH42, hns+ plasmid pAKH74 or hns′ plasmid pAKH30 were grown overnight at 25°C in LB medium. A. β-Galactosidase activity of overnight cultures grown in LB at 25°C was determined and is given in μmol min−1 mg−1 for comparison. The data represent the average ± SD from at least three different experiments each performed in duplicate. B. Whole-cell extracts of the Y. pseudotuberculosis strains were prepared and the RovA protein in the extracts was detected by Western blotting using a polyclonal antibody against RovA. A prestained molecular weight marker is loaded on the left.

Download figure to PowerPoint

Autoregulation of rovM expression

Many LysR-type regulators tightly control their own synthesis (Schell, 1993). To assess whether rovM is subject to an autoregulatory mechanism, the response of rovM expression to the presence of RovM was tested using the rovM–lacZ gene fusion in rovM-positive and -negative backgrounds (Fig. 6). The expression of rovM in the wild type was rather low, but the elimination of the rovM gene further reduced the expression of the rovM–lacZ fusion to about 30%. In contrast, when the concentration of RovM was raised in the strains by introducing a plasmid-encoded copy of the rovM gene, rovM transcription was significantly increased and exceeded that of the wild type by a factor of 2. RovM-mediated autoinduction was observed under all previously tested growth conditions (data not shown), indicating that rovM autoregulation per se is not under environmental control.

image

Figure 6. Positive autoregulation of rovM expression. E. coli K-12, Y. pseudotuberculosis YPIII and YP41 (rovM) harbouring the rovM–lacZ fusion plasmid pAKH63 with the empty vector (pAKH85) or a rovM+ plasmid (pAKH42) were grown overnight in LB medium at 25°C. β-Galactosidase activity from overnight cultures of the different E. coli and Y. pseudotuberculosis strains was determined and is given in μmol min−1 mg−1 for comparison. The data represent the average ± SD from at least three different experiments each performed in duplicate.

Download figure to PowerPoint

Binding of RovM to its promoter region was also tested by gel retardation, using DNA fragments encoding different portions of the entire regulatory and of the coding region of rovM. However, no specific interaction of RovM with the rovM promoter was detected with the same conditions as for rovA (data not shown), indicating that autoregulation requires additional binding components. This suggestion is supported by the fact that expression of a plasmid-encoded rovM–lacZ fusion in E. coli is generally very low and cannot be significantly induced by RovM even when higher amounts of the protein are present in the bacterial cell (Fig. 6).

Effect of rovM on host cell invasion

The RovA protein, which is negatively affected by RovM, was shown to activate the synthesis of invasin, the primary internalization factor of enteropathogenic Yersinia (Revell and Miller, 2000; Nagel et al., 2001). To test the effect of RovM on invasin synthesis and host cell invasion, we compared the amount of the invasin protein in the Y. pseudotuberculosis wild type and the rovM mutant strain and investigated the efficiency of these bacteria to enter human cells. High levels of the invasin protein were detectable in the rovM mutant, whereas overexpression of rovM reduced the amount of invasin to concentrations similar to those of a rovA mutant (Fig. 7A). The rovA deficient strain showed a fourfold lower uptake rate than the parental strain, but no or only a slight reduction of cellular invasion was detectable with the strain harbouring the rovM+ plasmid (Fig. 7B). This indicated that rovM might affect other mechanisms that overcome the negative effect on RovA-mediated invasin expression.

image

Figure 7. Invasin synthesis and entry efficiency of Y. pseudotuberculosis strains YPIII, YP41 (rovM) and YP3 (rovA). A. Overnight cultures of the different Y. pseudotuberculosis strains used for the cell infection experiments were also used for the preparation of cell extracts and analysed by Western blotting with a monoclonal antibody directed against invasin. A molecular weight marker is loaded on the left. B. About 106 bacteria of an overnight culture were used to challenge 5 × 104 HEp-2 cells, which were incubated for 30 min at 37°C to determine cell uptake efficiency as described in Experimental procedures. Each bar represents the average of three independent experiments relative to the invasion rate promoted by the Y. pseudotuberculosis wild type, defined as 1.0.

Download figure to PowerPoint

Impact of rovM on the motility of Y. pseudotuberculosis

Motility was demonstrated to be essential for virulence of a number of pathogens (Josenhans and Suerbaum, 2002; Ramos et al., 2004), and was shown to be required for the initiation of Yersinia host cell invasion, as it provides the migration capability to contact host cells (Young et al., 2000). Because RovM has considerable homology to LrhA, a LysR-type regulator of flagellation, motility and chemotaxis from E. coli (Lehnen et al., 2002), we also investigated the effect of RovM on the swimming behaviour of Y. pseudotuberculosis. As shown in Fig. 8, the motility of the rovM mutant was slightly decreased on trypthone swarm agar plates compared with wild type. In contrast, both wild type and the rovM mutant strain, harbouring a low-copy rovM+ plasmid were hypermotile and formed large concentric swarm rings with significantly increased diameters around the point of inoculation (Fig. 8), indicating that an increase in rovM expression improves motility. RovM-mediated stimulation of flagellar motility is independent of RovA, because a rovA mutant is unaffected in its swimming behaviour (data not shown), and this may also explain why a hypermotile RovM-overexpressing strain was more invasive than a rovA mutant, although the amount of invasin was similarly reduced in both strains (Fig. 7A).

image

Figure 8. Effect of RovM on Y. pseudotuberculosis motility. An equal number of suspended Y. pseudotuberculosis wild-type YPIII, the rovM mutant YP41 and the rovM+-overexpressing strains YPIII pAKH42 and YP41 pAKH42 was spotted onto tryptone swarm soft agar plates and incubated at 25°C for 16 h.

Download figure to PowerPoint

Effect of rovM on pathogenesis

Previous studies revealed that a lack of RovA in Y. enterocolitica attenuated virulence and resulted in a significant increase in the LD50 compared with wild type in the mouse infection model (Revell and Miller, 2000). BALB/c mice infection experiments with a Y. pseudotuberculosis wild type (YPIII) and a rovA mutant strain (YP3) showed a similar strong impact of RovA on the pathogenesis of a Y. pseudotuberculosis infection (Fig. 9A). Although the rovA mutant is still capable of establishing an infection and replicating in Peyer's patches in the mice, fewer bacteria were recovered from the Peyer's patches compared with the wild type. Furthermore, the rovA deficient strain was severely attenuated in its ability to reach and/or replicate in the deeper tissues and organs, e.g. liver and spleen. In order to determine whether a change in the levels of the negative modulator RovM also alters rovA expression in vivo and thus affects pathogenesis of a Y. pseudotuberculosis infection, we orally infected BALB/c mice with isogenic Y. pseudotuberculosis strains, expressing wild-type levels (YPIII), no (YP41) or elevated levels of RovM (YPIII pAKH42). The results showed that 3- to 15-fold higher numbers of the rovM mutant bacteria were detected in all dissected tissues and organs compared with the wild type (Fig. 9B). In contrast, the rovM overexpressing strain was highly attenuated and showed a strong reduction in its infection potential (Fig. 9C). About 100-fold fewer bacteria were found in the Peyer's patches and in the mesenteric lymph nodes. The difference in the dissemination of the bacteria was even more striking. No bacteria were able to reach the deeper tissues, although the rovA mutant was occasionally detectable in the liver and spleen. This may indicate that rovM controls virulence-associated genes in addition to rovA. However, neither the expression of the major adhesin YadA nor the synthesis and secretion of the antiphagocytic Yop effector proteins that are encoded on the Yersinia virulence plasmid pYV appear to be affected by RovM (Fig. S4).

image

Figure 9. Oral infection of BALB/c mice with a rovA, a rovM and a rovM+-overexpressing Y. pseudotuberculosis strain. For oral infection, groups of six BALB/c mice were infected via the orogastric route with 5 × 109 bacteria of the Y. pseudotuberculosis wild-type strain YPIII and YP3 (rovA) (A), YP41 (rovM) (B) or YPIII pAKH42 (rovM+) (C). One day post infection, the mice were sacrificed and the numbers of surviving bacteria in the liver (LI), kidney (KI), spleen (SP), mesenterial lymph nodes (MLN), and Peyer's patches (PP) were determined as described in Experimental procedures. Data are presented as a scatter plot of numbers of cfu per gram of organ as determined by counts of viable bacteria on plates. Each spot represents the cfu count, in the indicated tissue samples from one mouse. The levels of statistical significance for differences between test groups were determined by Student's t-test. The cfu of YP3, YP41 and YPIII pAKH42 in PP, MLN and LI were significantly different from the cfu of the wild type (P < 0.05).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Enteropathogenic Yersinia species need to monitor, adapt and respond to a large variety of environmental conditions and host-mediated assaults in order to establish a successful infection. One way in which this is accomplished is through transcriptional regulation of virulence genes that are required for a certain stage in the infection process. The transcription regulator RovA has been shown to be important for the regulation of virulence genes and efficient colonization of host tissues by yersiniae, but little is known about the molecular mechanisms that control RovA synthesis in response to multiple environmental signals and metabolic stimuli (Nagel et al., 2001). We present here the identification and characterization of a novel LysR-type regulatory protein RovM in Y. pseudotuberculosis, which represses the expression of rovA and affects cellular invasion, virulence and motility of this pathogen. Removal of the rovM gene activates RovA-dependent expression of the internalization factor invasin and increases invasion of Y. pseudotuberculosis into human epithelial cells. We further show that the elimination of the rovM gene increases the virulence of Y. pseudotuberculosis to BALB/c mice and leads to higher numbers of the bacteria in the Peyer's patches, the mesenterial lymph nodes and the organs of orally infected mice. In contrast, RovM overproduction results in a strong attenuation of virulence in this mouse infection model. Fewer bacteria are found in the lymphatic tissue and no bacteria are detectable in the organs of the mice. This phenotype is even more striking if one considers that rovM overexpression increases the motility of Yersinia, which was shown to stimulate invasion into host cells (Young et al., 2000).

Sequence analysis revealed that RovM undoubtedly belongs to the LysR-type transcriptional regulator family. The LysR family is composed of a large number of individual regulatory proteins which apparently evolved from a distant ancestor into subfamilies with diverse functions found in various prokaryotic genera. They control a wide range of biological processes in order to permit optimal bacterial survival and environmental adaptation (Schell, 1993), and some LysR-like regulators, including CrgA of Neisseria gonorrhoae, AphB of Vibrio cholerae and SpvR of S. typhimurium have been shown to be important for regulating cell adhesion and/or virulence of human pathogens (Kovacikova and Skorupski, 1999; Deghmane et al., 2000). The RovM protein is most closely related to the virulence gene regulators PecT and HexA which have been characterized from the phytopathogenic bacteria E. chrysanthemi and E. carotovora, and from Photorhabdus temperata, a bacterium that is mutuallistically associated with nematodes and pathogenic to insect larvae (Surgey et al., 1996; Harris et al., 1998; Mukherjee et al., 2000; Joyce and Clarke, 2003). They comprise a subset of the LysR family of transcriptional regulators, which have pleiotropic effects on gene expression and were shown to regulate exoenzymes, flagella and biofilm synthesis genes involved in host interaction and pathogenicity. RovM represents the first member of this subfamily known to affect virulence of an animal and human pathogen.

LysR proteins are coinducer-responsive regulators with sizes ranging from 250 to 350 residues and their N-terminus contains the helix-turn-helix DNA-binding motif, which is the most conserved of the LysR-like regulators, including RovM. Most LysR-type transcription factors are positive transcriptional regulators, which are transcribed divergently from the genes they control (Schell, 1993). However, the rovM gene is not closely linked to other genes, and it represses the virulence regulator gene rovA. DNA binding studies revealed that RovM interacts specifically with DNA fragments, including a 30 bp region closely upstream of the rovA promoter P1 (−80 to −47) and led to the formation of a single RovM-DNA complex. LysR-type regulators commonly recognize a ∼15 bp partially dyadic sequence centred near −65, which has the conserved T-N11-A core motif proven critical for binding of the LysR-type proteins GcvA, NodD, NahR and MetR (Schell, 1993; Heil et al., 2002). The predicted RovM binding region within the rovA promoter includes two palindromic sequences (−63 to −53: ATcaTTT-N5-AAAgaAT; −62 to −55: TCaTT-N6-AAaGA) with similarity to other LysR-type binding sites, which could serve this function. Several hyper-reactive bases, which became more exposed to DNase I action on RovM binding, were also identified within this promoter region. This suggests that the DNA undergoes significant bending or looping upon RovM binding, a phenomenon that has also been described for LysR-type regulators Nod, CrgA and AphB (Fisher and Long, 1989; Kovacikova and Skorupski, 2002; Deghmane et al., 2004). The structural alterations might modify rovA promoter geometry in a way that hinders proper recognition by the RNA polymerase or prevents the transition from a closed transcription initiation complex to an open form of this complex. Furthermore, RovM interaction with the rovA regulatory region at positions −80 to −47 would impede rovA transcription from P2rovA at position −267.

Transcription of rovA is also subject to silencing by the histone-like protein H-NS, which preferentially interacts with extended AT-rich sequences at positions −422 to −526 located upstream of P2rovA (Heroven et al., 2004). The distant positions of the RovM and H-NS binding sites suggest that the two negative regulators operate independently of each other. However, rovA expression cannot be repressed by RovM when the amount of functional H-NS is reduced, and H-NS has no repressor capabilities in the absence of RovM, even when higher quantities of the silencer protein are present. This strongly suggests that RovM and H-NS are both required for silencing of the rovA promoters in Y. pseudotuberculosis and implies an intriguing mechanism that may be unique among LysR-like protein regulated genes. In this and our previous study (Heroven et al., 2004) we have shown that both H-NS and RovM are able to interact independently with the rovA regulatory region in vitro. However, it is possible that the interaction of both repressors on the DNA and/or structural alterations of the rovA regulatory sequences (e.g. DNA looping), which are accompanied by H-NS/RovM-DNA binding, are required for the formation of a stable repressor complex which cannot easily be disrupted by the RovA activator protein.

It is obvious from this study that RovM is not only a rovA specific regulatory factor. It also controls genes involved in flagellar motility and activates the expression of its own gene. The latter is different from most LysR-type regulators, which are commonly under negative autoregulation (Schell, 1993). The molecular mechanism of rovM autoinduction is not yet clear, because no specific RovM binding to the rovM promoter region was found as for rovA. Similarly, positive autoregulation has been shown for the rovM homologue hexA, but an interaction of HexA with the hexA promoter was also not detectable (Harris et al., 1998). It is possible that positive autoregulation of the rovM and hexA genes requires additional factors or different binding conditions (e.g. special DNA topology). In fact, it has been shown that the affinity of LysR regulators for specific binding sites on DNA is altered by other regulatory factors (Heil et al., 2002; Kovacikova et al., 2005). Furthermore, it is well known, that most LysR-type regulators interact with small specific signal molecules that are typically metabolites, catabolites or intermediates of a biochemical pathway they regulate (Schell, 1993). However, despite several efforts, a coeffector of the RovM/HexA/PecT subfamily of LysR proteins has not yet been identified.

Nonetheless, certain chemical compounds seem to be important for RovM-dependent gene regulation in Y. pseudotuberculosis. Comparative analysis of rovA and rovM in this study revealed a reciprocal regulation pattern of the regulator genes in response to different growth media. The expression of rovM in Y. pseudotuberculosis is considerably induced during growth in minimal media, when intracellular RovA levels are reduced. Yet, rovA repression during growth in different minimal media is completely eliminated in a rovM mutant strain and equals rovA expression of the wild type during growth in complex medium. This strongly suggests that the regulation of rovA expression in response to nutrient availability is mediated through RovM. In contrast, RovM levels are not elevated at 37°C and during exponential growth phase, indicating that additional regulatory mechanisms are required for temperature- and growth phase-dependent regulation of rovA. The identity of the regulatory mechanisms involved in these environmental control processes and the regulatory network of all individual components is subject of our future investigations.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Bacterial strains, cell culture, media and growth conditions

The strains used in this study are listed in Table 1. Overnight cultures of E. coli were routinely grown at 37°C, Yersinia strains were grown at 25°C or 37°C in LB broth if not indicated otherwise. To analyse the effect of minimal medium, MMA, M9 (Miller, 1992) and RPMI1640 (Sigma) containing glucose (0.2%), casamino acids (0.2%) and 1 mM MgSO4 were used to culture Y. pseudotuberculosis as indicated. The antibiotics used for bacterial selection were as follows: ampicillin 100 μg ml−1, chloramphenicol 30 μg ml−1, tetracycline 5 μg ml−1, kanamycin 50 μg ml−1 and gentamicin 50 μg ml−1.

Table 1.  Bacterial strains and plasmids.
Strains/plasmidsDescriptionSource and reference
  • a.

    The number indicates the codon of rovA, in which the resistance cassette has been inserted.

  • b.

    The number indicates the codon of rovApstb fused to lacZ.

  • c.

    The number indicates the codons of the truncated hns allele.

  • d.

    The number indicates the codon of rovMpstb fused to lacZ.

Bacterial strains
 E. coli K-12
  BL21λDE3F-, gal met rmhdsSλlysplacUV5-T7-Gen1 placIqlacIStudier and Moffatt (1986)
  PD145MC4100 (hns205::Tn10)Heroven et al. (2004)
  MC4100FΔ(argF-lac)U169 rpsL150 deoC1 relA1 ptsF25 flbB5501 rbsRCasadaban (1976)
 Y. pseudotuberculosis
  YPIIIpIB1, wild typeBolin et al. (1982)
  YP3pIB1, rovA::Tn10(60)a, CmRNagel et al. (2001)
  YP12pIB1Bolin et al. (1982)
  YP38pIB1, rovA–lacZ(129)b, ApRThis study
  YP41pIB1, ΔrovM, KnRThis study
Plasmids
 pACYC177Cloning vector, p15A, ApR, KanRChang and Cohen (1978)
 pACYC184Cloning vector, p15A, CmR, TetRChang and Cohen (1978)
 pAKH26pPD281, rovA–lacZ(129)b, ApRThis study
 pAKH30pK18, hns′(1–109)c, KnRHeroven et al. (2004)
 pAKH42pACYC184, rovM+, CmRThis study
 pAKH43pET28a, rovM+, KnRThis study
 pAKH47pGP20, rovA–lacZ(23)b, TetRThis study
 pAKH62pBR322, rovM+, ApRThis study
 pAKH63pGP20, rovM–lacZ(41)d, TetRThis study
 pAKH74pACYC184, hns, CmRThis study
 pAKH85pACYC184, CmR, TetSThis study
 pBR322Cloning vector, TetR, ApRBolivar et al. (1977)
 pET28aT7 overexpression vector, KnRNovagen
 pGN36pACYC184, −626 bp rovA upstream region, CmRHeroven et al. (2004)
 pGP20pSC101-based promoter probe vector, lacZ+, TetRPetra Gerlach
 pK18cloning vector, lacPO, KnRPridmore (1987)
 pKOBEG-sacBrecombination vector, sacB+, CmRDerbise et al. (2003)
 pPD281R6K derivative, rovA–lacZ(129)b, ApRNagel et al. (2001)

HEp-2 cells were cultured in RPMI1640 media (Sigma) supplemented with 7.5% newborn calf serum (Invitrogen) and 2 mM glutamine at 37°C in the presence of 5% CO2.

DNA manipulations and sequence analysis

All DNA manipulations, restriction digestions, ligations and transformations were performed using standard genetic and molecular techniques (Sambrook et al., 2001; Miller, 1992). Plasmid DNA was purified using a Qiagen kit. Restriction and DNA-modifying enzymes were obtained from Roche or New England Biolabs. The oligonucleotides used for amplification by PCR, sequencing and primer extension were purchased from Metabion. PCR reactions were performed routinely in a 100 μl mix for 25 cycles using Taq polymerase (New England Biolabs) or Pfu polymerase (Promega) according to the manufacturer's instructions. PCR products were purified with the QIAquick kit (Qiagen) before and after digestion of the amplification product. Double-stranded DNA sequencing of the rovA upstream region was carried out by the dideoxy-chain termination procedure, using a Thermo Sequenase cycle sequencing kit (Amersham Life Science). Other sequencing reactions were performed by Agowa (Berlin, Germany).

Plasmids used in this study are listed in Table 1. pAKH26 was constructed by inserting a PCR fragment harbouring the rovA promoter region into pPD281 cut with SalI. The fragment was amplified from chromosomal DNA of Y. pseudotuberculosis using the upstream primer 5′-GCCGCGGTCGACGGCCGCTGAAAGCAGCGTTCA-3′ and downstream primer 5′-CGCCGCGTCGACCTGATAACACTGCAATTTCATCCG-3′. The sequence and the correct orientation of the fragment were proven by DNA sequencing. The rovM+ fragment of Y. pseudotuberculosis in pAKH42 and pAKH62 was generated by PCR using primers 5′-GGGCCCGGATCCGGTCTTAAAAAGAGAACATTC-3′ and 5′-GCGGCGGTCGACGCAGGATCACCGATACTG-3′, digested with BamHI and SalI, and inserted into pACYC184 and pBR322 respectively. For overexpression of the rovM gene of Y. pseudotuberculosis, a PCR fragment amplified with the upstream primer 5′-GCGGCGGCTAGCATGACAAATGCAAATCGTCCG-3′ and the downstream primer 5′-GGGCGCCTCGAGGACTACTTAATCTTCATCACC-3′ was integrated between the NheI and XhoI sites of pET28a, generating pAKH43. Plasmid pAKH47 was constructed by the insertion of a PCR fragment amplified with primers 5′-GCCGCGGAATTCTGCCGCCTTCCTGCAACTCG-3′ and 5′-GCGGCGGAATTCCGCGCCAAACGCGAACTAATCG-3′ inserted into the EcoRI site of pGP20. Plasmid pAKH63 carries a PCR generated fragment harbouring the rovM promoter region of Y. pseudotuberculosis (upstream primer 5′-GCGGCGCCCGGGGGTCTTAAAAAGAGAACATTC-3′ and downstream primer 5′-CGGCGCCCCGGGTGCTGACTGAGTTCGGCAG-5′) cloned into the SmaI site of pGP20. Plasmid pAKH74 was constructed by inserting a PCR fragment amplified with primers 5′-GCGGCGGTCGACCTCCGATTGAACATTAACC-3′ and 5′-GCGGCGGGATCCCAGCGGCTGTACAGTGGAC-3′ into the SalI and BamHI sites of pACYC184. For the construction of pAKH85, pACYC184 was digested with HincII and religated.

Construction of the Y. pseudotuberculosis gene bank, isolation of sequences influencing the rovA–lacZ expression in Y. pseudotuberculosis

The Y. pseudotuberculosis gene bank was generated as described previously (Nagel et al., 2001). The plasmid library was introduced into Y. pseudotuberculosis YP38 carrying a chromosomal rovA–lacZ fusion. Gene bank plasmids from clones showing a lower rovA expression on plates supplemented with chloramphenicol and X-Gal were isolated and retransformed into YP38 to verify their influence on rovA–lacZ expression.

Construction of the chromosomal rovA–lacZ fusion strain

For the construction of the chromosomal rovA–lacZ translational fusion strain YP38, plasmid pAKH26 was mated from E. coli S17–1λpir (tra+) into Y. pseudotuberculosis YPIII as described (Nagel et al., 2001). Active lacZ+ colonies were selected on Yersinia selective agar (Oxoid) supplemented with Amp, selecting for the antibiotic resistance of the plasmid. The resulting merodiploid transconjugants, which include a wild-type copy of rovA and the rovA–lacZ fusion, were analysed by PCR and DNA sequencing to prove proper integration of the plasmid into the rovA locus.

Construction of the rovM deletion mutant of Y. pseudotuberculosis

For the construction of a rovM deletion mutant a kanamycin cassette was integrated into the rovM locus of Y. pseudotuberculosis using the RED recombinase system as described (Derbise et al., 2003). To do so, a 1266 bp rovM::KanR PCR fragment was generated using pACYC177 as a template with primers 1 and 2 composed of 100 nucleotides which are homologous to the up- or downstream region of the rovM gene followed by 20 nucleotides homologous to the 5′- or 3′-end of the kanamycin resistance gene of pACYC177 (underlined) (Primer 1: 5′-CCCTACTCCAATGACTGGGGTACTGTCATACTTAGCGTTGTCCTTTATTGATAACGCTGATACTTAAAACACCAGGGTAGTTTGTAATTAGAATTCAAAAAAAGGGTGATTTTGAACTTTTGCTTTGC-3′; Primer 2: 5′-CGTTGAATATAAAATTAAGCTGAAATGAAAAAAAAGAGGATTTACTAAAATAAATAAAAAAAAGCCCCTTAAAATACAGAGACTTGTGATGATGAATTAAAAACAGACCCAGTGTTACAACCAATTAACC-3′). This PCR fragment was transformed into Y. pseudotuberculosis YPIII pKOBEG-sacB and chromosomal integration of the fragment, generating a rovM::kanR knock-out mutant was selected by plating on LB kanamycin agar plates. After primary selection, mutants were grown on LB agar plates without NaCl, but containing 10% sucrose; fast growing colonies without pKOBEG-sacB were selected. One strain, YP41, harbouring the rovM::Kan mutation in the rovM locus, as proven by PCR and DNA sequencing was used for further studies.

Expression and purification of the RovM protein

A 1 l culture of BL21λDE3 pAKH43 was grown to an A600 of 0.6 and expression of 6His-RovM was induced by adding IPTG to a final concentration of 2 mM. The cells were grown for additional 2 h, harvested and resuspended in 50 mM NaH2PO4, pH 8.0, 300 mM NaCl, 10 mM imidazole. Subsequently, the cells were lysed with a French press (120 000 psi), and the soluble 6His-RovM extract was separated from insoluble cell material by centrifugation at 25 000 g. The 6His-RovM protein was then purified by affinity chromatography on Ni-NTA agarose (Qiagen). The column was washed with three column volumes of 50 mM NaH2PO4, pH 8.0, 300 mM NaCl, 20 mM imidazole and eluted with 50 mM NaH2PO4, pH 8.0, 300 mM NaCl containing 250 mM imidazole. The purity of the RovM protein was estimated to be > 95%.

DNA retardation assay

For DNA-binding studies, the purified RovM protein was dialysed against the DNA-binding buffer (10 mM Tris-HCl, pH 7.5, 1 mM EDTA, 5 mM dithiothreitol, 5% glycerol, 10 mM NaCl, 1 mM MgCl2, 100 μg ml−1 BSA). Defined PCR fragments, carrying different portions of the rovA regulatory region, were mixed in an equimolar ratio and incubated in a 20 μl reaction mixture containing the DNA binding buffer and increasing amounts of RovM. The reaction mixture was incubated for 20 min at room temperature and subsequently loaded onto a 4% polyacrylamide gel and stained with ethidium bromide. Fragment a of the rovA promoter region was amplified with the upstream primer 5′-TCGTATTTATTGCTATTATCC-3′ and the downstream primer 5′-TTATTACTTAGTTTGTAATTG-3′. Fragments b and c of the rovA promoter region were generated with the upstream primer 5′-GTGCTAACGACAATGAC-3′ and the downstream primer 5′-GAGTCCAGTCTATTTTGGG-3′ for fragment b and 5′-CTCATTACCCAGCATCG-3′ for fragment c.

DNase I footprinting

For DNase I footprinting, different segments of the rovA promoter region were amplified by PCR using a digoxigenin (DIG)-labelled primer and a non-labelled primer. The following primer sets were used: rovA-A (5′-CCGATTTCGCAGTAGATAACG-3′ and 5′-DIG-GAGTCCAGTCTATTTTGGG-3′), rovA-B (5′-DIG-CGTATCACCTGACGGAG-3′), leading to fragments corresponding to nt −246 to +37 and −216 to +37 with respect to the transcriptional initiation site of P1rovA. PCR fragments were purified, incubated with the purified RovM protein in 20 μl of DNA-binding buffer as described for the gel retardation assays. The PCR products were digested with DNase I of an appropriate dilution and the resulting products were separated and visualized as described (Heroven et al., 2004). The protected bands were identified by comparison with a sequence ladder generated with the same DIG-labelled primer used for the amplification of the fragment by PCR. For double-stranded sequencing of the rovA regulatory region, plasmid pGN36 was used as a template.

β-Galactosidase assays

The activity of the chromosomal and plasmid-encoded rovA–lacZ and rovM–lacZ fusions were measured in permeabilized cells as described previously (Nagel et al., 2001). The activity was calculated as follows: β-galactosidase activity OD420 · 6,75 · OD600−1 · Δt (min)−1 · Vol (ml)−1.

Motility assay

A 2 μl portion of an overnight culture was spotted onto semisolid tryptone swarm plates (1% tryptone, 0.5% NaCl, 0.3% Difco Bacto agar) (Tisa and Adler, 1995). The capacity of each strain to spread was monitored after 16 h at 25°C.

Polyclonal antibodies against RovM of Y. pseudotuberculosis

Polyclonal antibodies to RovM were generated in a rabbit (Davids Biotechnologie, Regensburg, Germany) by injecting 500 μg of purified protein. For further purification, the serum was incubated with precipitated cell lysate of YP41. As a negative control, serum was taken from preimmune blood of the rabbit.

Gel electrophoresis, preparation of cell extracts and Western blotting

For immunological detection of the RovA and RovM proteins, cultures of Y. pseudotuberculosis were grown under specific environmental conditions as described. Cell extracts of equal amounts of the bacteria were prepared and separated on a 15% SDS-PAGE (Sambrook, 2001). Subsequently, the samples were transfered onto an Immobilon-P membrane (Millipore) and probed with a polyclonal RovA, RovM or β-galactosidase antibody as described (Heroven et al., 2004). The cell extracts used for Western blotting were also separated by SDS-PAGE and stained with Coomassie blue to ensure that the protein concentrations in the different cell extracts are comparable; about 10 μg protein was applied of each sample.

To compare expression of the YadA protein in the Y. pseudotuberculosis wild type, the rovM mutant and the rovM overexpressing strains, bacteria were grown to an A600 of 0.6 at 37°C in LB medium supplemented with 20 mM Na oxalate and 20 mM MgCl2. The immunological detection of the YadA protein in the whole-cell extracts was performed as described previously (Eitel and Dersch, 2002).

Cell invasion assay

In preparation of the cell adhesion and uptake assay, 5 × 104 HEp-2 cells were seeded and grown overnight in individual wells of 24-well cell culture plates (Nunc). Cell monolayers were washed three times with phosphate-buffered saline and incubated in binding buffer (RPMI 1640 medium supplemented with 20 mM HEPES pH 7.0 and 0.4% bovine serum albumin) before the addition of bacteria. Approximately 106 bacteria were added to the monolayer and incubated at 37°C. Bacterial uptake was assessed 30 min after infection as the percentage of bacteria which survived killing by the addition of the antibiotic gentamicin to the external medium, as described previously (Dersch and Isberg, 1999). For each strain, the relative level of bacterial uptake was determined by calculating the number of colony-forming units (cfu) relative to the total number of bacteria introduced onto monolayers. The experiments were routinely performed in triplicate.

Virulence experiments

Groups of six BALB/c mice (6–8 weeks, female) were orally infected with 5 × 109 bacteria of Y. pseudotuberculosis strains YPIII, YP3, YP41 and YPIII pAKH42 (animal licensing committee permission no. G0194/00). After 24 h of infection, mice were killed, and the organs were homogenized in 10 ml of PBS per gram of organ. The ileum was incubated for 30 min in PBS with 100 μg ml−1 gentamicin and then intensively washed with PBS. The numbers of bacteria were determined by plating three independent serial dilutions of the homogenates on Yersinia selective medium (Oxoid, Germany) and the cfu were counted. When infected with YPIII pAKH42, 200 single colonies of each organ of one mouse were patched on LB chloramphenicol plates to ensure that all bacteria isolated from host tissue still contained the rovM overexpressing plasmid.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

We thank Drs Birgitta Beatrix, Martin Fenner, Sabine Hunke, Barbara Schulz and Barbara Weissenmayer for helpful discussions and critical reading of the manuscript and Ulla Thiesen and Susanne Kneip for technical support. This work was supported by Deutsche Forschungsgemeinschaft Grant DE616/3.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Blumer, C., Kleefeld, A., Lehnen, D., Heintz, M., Dobrindt, U., Nagy, G., et al. (2005) Regulation of type 1 fimbriae synthesis and biofilm formation by the transcriptional regulator LrhA of Escherichia coli. Microbiology 151: 32873298.
  • Bolin, I., Norlander, I., and Wolf-Watz, H. (1982) Temperature-inducible outer membrane protein of Yersinia pseudotuberculosis and Yersinia enterocolitica is associated with the virulence plasmid. Infect Immun 37: 506512.
  • Bolivar, F., Rodriguez, R.L., Greene, P.J., Betlach, M.C., Heyneker, H.L., and Boyer, H.W. (1977) Construction and characterization of new cloning vehicles. II. A multipurpose cloning system. Gene 2: 95113.
  • Bottone, E.J. (1997) Yersinia enterocolitica: the charisma continues. Clin Microbiol Rev 10: 257276.
  • Casadaban, M.J. (1976) Transposition and fusion of the lac genes to selected promoters in Escherichia coli using the bacteriophage lambda and Mu. J Mol Biol 104: 541555.
  • Chang, A.C., and Cohen, S.N. (1978) Construction and characterization of amplifiable multicopy DNA cloning vehicles derived from the P15A cryptic miniplasmid. J Bacteriol 134: 11411156.
  • Deghmane, A.E., Petit, S., Topilko, A., Pereira, Y., Giorgini, D., Larribe, M., and Taha, M.K. (2000) Intimate adhesion of Neisseria meningitidis to human epithelial cells is under the control of the crgA gene, a novel LysR-type transcriptional regulator. EMBO J 19: 10681078.
  • Deghmane, A.E., Giorgini, D., Maigre, L., and Taha, M.K.(2004) Analysis in vitro and in vivo of the transcriptional regulator CrgA of Neisseria meningitidis upon contact with target cells. Mol Microbiol 53: 917927.
  • Derbise, A., Lesic, B., Dacheux, D., Ghigo, J.M., and Carniel, E. (2003) A rapid and simple method for inactivating chromosomal genes in Yersinia. FEMS Immunol Med Microbiol 38: 113116.
  • Dersch, P., and Isberg, R.R. (1999) A region of the Yersinia pseudotuberculosis invasin protein enhances integrin-mediated uptake into mammalian cells and promotes self association. EMBO J 18: 11991213.
  • Dube, P.H., Handley, S.A., Revell, P.A., and Miller, V.L. (2003) The rovA mutant of Yersinia enterocolitica displays differential degrees of virulence depending on the route of infection. Infect Immun 71: 35123520.
  • Eitel, J., and Dersch, P. (2002) The YadA protein of Yersinia pseudotuberculosis mediates high-efficiency uptake into human cells under environmental conditions in which invasin is repressed. Infect Immun 70: 48804891.
  • Ellison, D.W., and Miller, V.L. (2006) Regulation of virulence by members of the MarR/SlyA family. Curr Opin Microbiol 9: 153159.
  • Fisher, R.F., and Long, S.R. (1989) DNA footprint analysis of the transcriptional activator proteins NodD1 and NodD3 on inducible nod gene promoters. J Bacteriol 171: 54925502.
  • Grutzkau, A., Hanski, C., Hahn, H., and Riecken, E.O. (1990) Involvement of M cells in the bacterial invasion of Peyer's patches: a common mechanism shared by Yersinia enterocolitica and other enteroinvasive bacteria. Gut 31: 10111015.
  • Harris, S.J., Shih, Y.L., Bentley, S.D., and Salmond, G.P. (1998) The hexA gene of Erwinia carotovora encodes a LysR homologue and regulates motility and the expression of multiple virulence determinants. Mol Microbiol 28: 705717.
  • Heil, G., Stauffer, L.T., and Stauffer, G.V. (2002) Glycine binds the transcriptional accessory protein GcvR to disrupt a GcvA/GcvR interaction and allow GcvA-mediated activation of the Escherichia coli gcvTHP operon. Microbiology 148: 22032214.
  • Heroven, A.K., Nagel, G., Tran, H.J., Parr, S., and Dersch, P. (2004) RovA is autoregulated and antagonizes H-NS-mediated silencing of invasin and rovA expression in Yersinia pseudotuberculosis. Mol Microbiol 53: 871888.
  • Josenhans, C., and Suerbaum, S. (2002) The role of motility as a virulence factor in bacteria. Int J Med Microbiol 291: 605614.
  • Joyce, S.A., and Clarke, D.J. (2003) A hexA homologue from Photorhabdus regulates pathogenicity, symbiosis and phenotypic variation. Mol Microbiol 47: 14451457.
  • Kovacikova, G., and Skorupski, K. (1999) A Vibrio cholerae LysR homolog, AphB, cooperates with AphA at the tcpPH promoter to activate expression of the ToxR virulence cascade. J Bacteriol 181: 42504256.
  • Kovacikova, G., and Skorupski, K. (2002) Binding site requirements of the virulence gene regulator AphB: differential affinities for the Vibrio cholerae classical and El Tor tcpPH promoters. Mol Microbiol 44: 533547.
  • Kovacikova, G., Lin, W., and Skorupski, K. (2005) Dual regulation of genes involved in acetoin biosynthesis and motility/biofilm formation by the virulence activator AphA and the acetate-responsive LysR-type regulator AlsR in Vibrio cholerae. Mol Microbiol 57: 420433.
  • Lehnen, D., Blumer, C., Polen, T., Wackwitz, B., Wendisch, V.F., and Unden, G. (2002) LrhA as a new transcriptional key regulator of flagella, motility and chemotaxis genes in Escherichia coli. Mol Microbiol 45: 521532.
  • Libby, S.J., Goebel, W., Ludwig, A., Buchmeier, N., Bowe, F., Fang, F.C., et al. (1994) A cytolysin encoded by Salmonella is required for survival within macrophages. Proc Natl Acad Sci USA 91: 489493.
  • Marra, A., and Isberg, R.R. (1997) Invasin-dependent and invasin-independent pathways for translocation of Yersinia pseudotuberculosis across the Peyer's patch intestinal epithelium. Infect Immun 65: 34123421.
  • Miller, J.H. (1992) A Short Course in Bacterial Genetic: a Laboratory Manual and Handbook for Escherichia coli and Related Bacteria. Cold Spring Habor, NY: Cold Spring Harbor Laboratory Press.
  • Mukherjee, A., Cui, Y., Ma, W., Liu, Y., and Chatterjee, A.K. (2000) hexA of Erwinia carotovora ssp. carotovora strain Ecc71 negatively regulates production of RpoS and rsmB RNA, a global regulator of extracellular proteins, plant virulence and the quorum-sensing signal, N-(3-oxohexanoyl)-L-homoserine lactone. Environ Microbiol 2: 203215.
  • Nagel, G., Lahrz, A., and Dersch, P. (2001) Environmental control of invasin expression in Yersinia pseudotuberculosis is mediated by regulation of RovA, a transcriptional activator of the SlyA/Hor family. Mol Microbiol 41: 12491269.
  • Navarre, W.W., Halsey, T.A., Walthers, D., Frye, J., McClelland, M., Potter, J.L., et al. (2005) Co-regulation of Salmonella enterica genes required for virulence and resistance to antimicrobial peptides by SlyA and PhoP/PhoQ. Mol Microbiol 56: 492508.
  • Oscarsson, J., Mizunoe, Y., Li, L., Lai, X.H., Wieslander, A., and Uhlin, B.E. (1999) Molecular analysis of the cytolytic protein ClyA (SheA) from Escherichia coli. Mol Microbiol 32: 12261238.
  • Pepe, J.C., and Miller, V.L. (1993) Yersinia enterocolitica invasin: a primary role in the initiation of infection. Proc Natl Acad Sci USA 90: 64736477.
  • Pridmore, R.D. (1987) New and versatile cloning vectors with kanamycin-resistance marker. Gene 56: 309312.
  • Ramos, H.C., Rumbo, M., and Sirard, J.C. (2004)Bacterial flagellins: mediators of pathogenicity and host immune responses in mucosa. Trends Microbiol 12: 509517.
  • Revell, P.A., and Miller, V.L. (2000) A chromosomally encoded regulator is required for expression of the Yersinia enterocolitica inv gene and for virulence. Mol Microbiol 35: 677685.
  • Sambrook, J. (2001) Molecular Cloning: a Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratories.
  • Schell, M.A. (1993) Molecular biology of the LysR family of transcriptional regulators. Annu Rev Microbiol 47: 597626.
  • Smego, R.A., Frean, J., and Koornhof, H.J. (1999) Yersiniosis I: microbiological and clinicoepidemiological aspects of plague and non-plague Yersinia infections. Eur J Clin Microbiol Infect Dis 18: 115.
  • Spory, A., Bosserhoff, A., Von Rhein, C., Goebel, W., and Ludwig, A. (2002) Differential regulation of multiple proteins of Escherichia coli and Salmonella enterica serovar Typhimurium by the transcriptional regulator SlyA. J Bacteriol 184: 35493559.
  • Stapleton, M.R., Norte, V.A., Read, R.C., and Green, J. (2002) Interaction of the Salmonella typhimurium transcription and virulence factor SlyA with target DNA and identification of members of the SlyA regulon. J Biol Chem 277: 1763017637.
  • Studier, F.W., and Moffatt, B.A. (1986) Use of bacteriophage T7 RNA polymerase to direct selective high-level expression. J Mol Biol 189: 113130.
  • Surgey, N., Robert-Baudouy, J., and Condemine, G. (1996) The Erwinia chrysanthemi pecT gene regulates pectinase gene expression. J Bacteriol 178: 15931599.
  • Thomson, N.R., Cox, A., Bycroft, B.W., Stewart, G.S., Williams, P., and Salmond, G.P. (1997) The Rap and Hor proteins of Erwinia, Serratia and Yersinia: a novel subgroup in a growing superfamily of proteins regulating diverse physiological processes in bacterial pathogens. Mol Microbiol 26: 531544.
  • Tisa, L.S., and Adler, J. (1995) Chemotactic properties of Escherichia coli mutants having abnormal Ca2+ content. J Bacteriol 177: 71127118.
  • Tran, H.J., Heroven, A.K., Winkler, L., Spreter, T., Beatrix, B., and Dersch, P. (2005) Analysis of RovA, a transcriptional regulator of Yersinia pseudotuberculosis virulence that acts through antirepression and direct transcriptional activation. J Biol Chem 280: 4242342432.
  • Young, G.M., Badger, J.L., and Miller, V.L. (2000) Motility is required to initiate host cell invasion by Yersinia enterocolitica. Infect Immun 68: 43234326.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Fig. S1. Multiple amino acid sequence alignment of RovMlike LysR-type regulatory proteins. Fig. S2. DNase I protection assay with RovM and H-NS. Fig. S3. Interaction of H-NS and RovM with the rovA promoter region. Fig. S4. Analysis of yadA expression and Yop protein secretion in wild type (YPIII), in the rovM mutant strain (YP41) and the rovM overexpressing strains (YPIII pAKH42, YP41pAKH42).

FilenameFormatSizeDescription
MMI5458Supplementary_material.pdf1076KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.