Regulation of a Type III and a Putative Secretion System in Edwardsiella tarda by EsrC Is under the Control of a Two-Component System, EsrA-
http://www.100md.com
感染与免疫杂志 2005年第7期
Department of Biological Sciences, Faculty of Science, National University of Singapore, Singapore 117543, Republic of Singapore
ABSTRACT
Edwardsiella tarda is a gram-negative enteric pathogen that causes hemorrhagic septicemia in fish and gastro- and extraintestinal infections in humans. A type III secretion system (TTSS) and a putative secretion system (EVP) have been found to play important roles in E. tarda pathogenesis. Our previous studies suggested that the TTSS and EVP gene clusters were regulated by a two-component system of EsrA-EsrB. In the present study, we characterized another regulator, EsrC, which showed significant sequence similarity to the AraC family of transcriptional regulators. Mutants with in-frame deletions of esrC increased the 50% lethal doses in blue gourami fish, reduced extracellular protein production, and failed to aggregate. Complementation of esrC restored these three phenotypes. Two-dimensional gel electrophoresis showed that EsrC regulated the expression of secreted proteins encoded by the TTSS (such as EseB and EseD) and EVP (EvpC) gene clusters. The expression of esrC required a functional two-component system of EsrA-EsrB. EsrC in turn regulated the expression of selected genes encoded in TTSS (such as the transcriptional unit of orf29and orf30, but not esaC) and genes encoded in the EVP gene cluster. The present study sheds light on the regulation of these two key virulence-associated secretion systems and provides greater insight into the pathogenic mechanisms of this bacterium.
INTRODUCTION
Edwardsiella tarda is an important cause of hemorrhagic septicemia. It infects many commercially important farmed fish species and has led to extensive losses in both freshwater and marine aquaculture (43). The organism has a broad host range and is also known to cause infections in higher animals, including humans (33). In humans, it causes gastro- and extraintestinal infections such as myonecrosis (37), bacteremia (49), septic arthritis (31), and wound infections (1). The pathogenesis of E. tarda is multifactorial. Several potential virulence factors have been reported, including the abilities to invade epithelial cells (22, 26), resist serum (22, 26) and phagocyte-mediated killings (38), and produce toxins and enzymes such as hemolysins (18), catalases (40), and dermatotoxins (45) for disseminating infection.
Using a combination of comparative proteomics of secreted proteins and TnphoA mutagenesis, we identified a type III secretion system (TTSS) (39, 42) and a putative secretion system (EVP) (for E. tarda virulence proteins) (41) that contributed to E. tarda PPD130/91 pathogenesis. TTSSs are generally used by bacterial pathogens to deliver virulence factors into host cells to subvert normal cell functions (21). The TTSS gene cluster of E. tarda PPD130/91 contains at least 30 open reading frames (ORFs) and is in two DNA regions (Y. P. Tan, J. Zheng, S. L. Tung, I. Rosenshine, and K. Y. Leung, unpublished data). The gene members of E. tarda TTSS are composed of apparatus, chaperones, effectors, and regulators. Some of the gene members of the E. tarda TTSS are homologous to other TTSSs encoded by pathogens such as enteropathogenic Escherichia coli, Salmonella enterica serovar Typhimurium, and Yersinia enterocolitica. The designation of the E. tarda TTSS genes is based on the sequence homologs in Salmonella pathogenicity island 2 (SPI-2) (16). The inactivation of the TTSS in E. tarda led to the increase of 50% lethal doses (LD50) in blue gourami fish (39, 41).
The other gene cluster (EVP) is not unique to E. tarda but is widely distributed in many other animal and plant pathogens and symbionts such as Salmonella, Vibrio, Yersinia, Escherichia, Rhizobium, and Agrobacterium species with putative secretion functions (41). In R. leguminosarum, the impairment of EVP-like cluster affected the secretion of at least one protein based on two-dimensional gel electrophoresis (2-DE) analysis (2). Some proteins encoded in this EVP-like locus of Rhizobium shared homology with proteins encoded by the type III or type IV secretion systems (2). Folkesson et al. (11) also analyzed EVP homolog in S. enterica serovar Typhimurium and found that this gene cluster encoded putative cytoplasmic, periplasmic, and outer membrane localizing proteins. In E. tarda, eight genes (evpA to -H) have been found in the EVP cluster (41). evpA and evpC are under the control of a TTSS regulator esrB, while the secretion of EvpC is dependent on EvpB instead of on a functional TTSS. The deletion of evpB or evpC led to lower replication rates in gourami phagocytes and reduced protein secretion and virulence in blue gourami fish (41). Complementation of evpB and evpC deletion mutants restored the secretion of EvpC, partially increased the replication rates in gourami phogocytes and lowered the LD50 values in gourami fish, indicating that these evp genes are vital for E. tarda pathogenesis (41).
The expressions of type III secretion apparatus and effectors are usually subjected to complicated regulation (12). Our previous studies suggested that EsrB regulated both the TTSS (such as eseB and eseD) and the EVP gene cluster (such as evpA and evpC) (41). In addition, the regulation of these two gene clusters is controlled by other factors such as temperature and the pstSCAB-phoU operon (41). This pstSCAB-phoU operon is a high-affinity phosphate-specific transport (PST) operon belonging to the family of ATP binding cassette (ABC) transporters (46). Our previous studies demonstrated that TnphoA insertions in pstB, pstC, and pstS abolished the expression of TTSS, as well as EVP proteins, and the resulting PST mutants were highly attenuated in blue gourami fish (41). However, it is not clear how the two-component system and these factors regulate the TTSS and EVP locus.
A common mechanism of gene regulation in bacteria is via regulatory proteins of the AraC family. To date, this family contains more than 100 members as identified by sequence homology to the AraC protein of Escherichia coli (15). With a few exceptions, AraC homologs are transcriptional activators. We describe here the identification of EsrC, an AraC homolog encoded inside the TTSS. The markerless in-frame deletion mutation of esrC disrupted the expressions of the secreted proteins of the TTSS and EVP locus. The phenotype of the esrC mutant is similar to the phenotypes of the esrA and esrB mutants. Our studies showed that EsrA-EsrB regulates the expression of the secreted proteins of TTSS and EVP proteins through esrC.
MATERIALS AND METHODS
Bacterial strains and plasmids. The bacterial strains and plasmids used in the present study are shown in Table 1. E. tarda stains were grown in tryptic soy broth (TSB; Difco/Becton Dickinson) or Dulbecco modified Eagle medium (DMEM) at 25°C without shaking while Escherichia coli strains were cultured in Luria broth (LB; Difco/Becton Dickinson) at 37°C. Cultivation of bacteria in DMEM was carried out at 5% (vol/vol) CO2 atmosphere. When required, the appropriate antibiotics were supplemented at the following concentrations: ampicillin (100 μg/ml); kanamycin (50 μg/ml); colistin (12.5 μg/ml); chloramphenicol (30 μg/ml); and tetracycline (10 μg/ml).
Construction of deletion mutants and plasmids. Overlap extension PCR (19) was used to generate in-frame deletion of esrC on the E. tarda PPD130/91 chromosome. For the construction of esrC, two PCR fragments were generated from PPD130/91 genomic DNA with the primer pairs of MutC-for plus MutC-int-rev, and MutC-rev plus Mut-int-for (Table 2). The resulting products generated a 754-bp fragment containing the upstream of esrC and a 751-bp fragment containing the downstream of the esrC, respectively. A 16-bp overlap in the sequences (underlined) permitted amplification of a 1,505-bp product during a second PCR with the primers MutC-for and MutC-rev, both of which were introduced into a KpnI restriction site (boldface text in Table 2), respectively. The resulting PCR product contained a deletion from nucleotides 34 to 660 of esrC. The PCR product was cloned into pGEMT-Easy vector, and DNA sequencing was performed to confirm that the construct was correct. The esrC fragment was excised with KpnI, ligated into suicide vector pRE112 (Cmr) (9), and the resulting plasmid then transformed into SM10 pir. The single-crossover mutants were obtained by conjugal transfer into E. tarda PPD130/91. Double-crossover mutants were obtained by plating onto 10% sucrose-tryptic soy agar plates. The deletion mutants were confirmed by PCR and sequencing. The construction of other esrC deletion mutants followed a similar protocol. The primers used for the construction of esrCN1 are MutC-N-for plus N-rev and MutC-N-rev plus N-for. The primers used for the construction of esrCN2 are MutC-N-for plus N2-rev and MutC-N-rev plus N2-for. The primers used for esrC163 are MutC-HTH-for plus HTH-rev and MutC-HTH-rev plus HTH-for. The primers used for esrC263 are MutC-HTH2-for plus HTH2-rev and MutC-HTH2-rev plus HTH2-for. The corresponding restriction sites (KpnI or SacI) are labeled in boldface in Table 2.
For the construction of pACYC+esrC, the complete esrC gene was obtained by PCR with Pfu Turbo polymerase (Stratagene) with E. tarda PPD130/91 as the template DNA. The PCR products were digested and ligated into ScaI- and EcoRI-digested pACYC184 plasmid. For the construction of pQEesrB and pQEesrC, the complete esrB and esrC genes were amplified by using the forward primer esrB-com-for or esrC-com-for containing the ribosome-binding site (underlined) and EcoRI site (boldface) and the reverse primer esrB-com-rev or esrC-com-rev which supplied the BamHI site (boldface) (Table 2). The PCR products were digested with BamHI and EcoRI and cloned into EcoRI- and BamHI-digested PQE60lacI (44).
For the construction of LacZ reporter plasmids, the putative promoter regions were amplified from E. tarda PPD130/91 genomic DNA and the PCR products were cloned into pGEMT-Easy vector. The resulting plasmids were sequenced, and the inserts were cut with BamHI and EcoRI (or MfeI) and subcloned into BamHI- and EcoRI-digested pRW50 plasmid (27).
LD50 determinations. Healthy naive blue gourami (Trichogaster trichopterus Pallas) of ca. 14 g were obtained from a commercial fish farm and infected with E. tarda as described previously (26). The mortality of the fish was recorded over a period of 7 days after infection. The LD50 values were calculated by the method of Reed and Muench (34).
Phagocyte isolation. Phagocytes were isolated from the head kidney of naive gourami and purified according to the procedure of Secombs (36) and as described by Srinivasa Rao et al. (38). Purified phagocytic cells (4 x 106 to 5 x 106 cells/well) were allowed to adhere to 48-well tissue culture plates (Falcon) in an L-15 medium (Sigma) that was supplemented with 5% fetal bovine serum. After 3 h of incubation at 25°C in a 5% (vol/vol) CO2 atmosphere, the cells were washed twice with Hanks balanced salt solution (HBSS; Sigma) and infected with E. tarda at a multiplicity of infection of 1:1 and incubated at 25°C for 30 min. The cells then were washed twice with HBSS and incubated with 100 μg of gentamicin/ml to kill all of the extracellular bacteria for 90 min. The monolayers were then washed twice with HBSS and lysed with 1% Triton X-100, and viable bacterial counts were enumerated by plate count. Phagocyte replication ratios were calculated by dividing the viable bacterial count after 5 h by the 2-h bacterial count.
2-DE and protein identification. Protein isolation and 2-DE were performed as previously described (41). Protein spots were identified with matrix-assisted laser desorption ionization-time of flight as described earlier (25).
-Galactosidase assays. Bacterial were grown in DMEM (for E. tarda) or LB (for E. coli) overnight at 25°C or 37°C. -Galactosidase activities were determined with cells permeabilized with sodium dodecyl sulfate and chloroform as described by Miller (29).
RESULTS
Sequence analysis of regulators. The TTSS of E. tarda is composed of two regions and was subjected to the regulation of a two-component regulator EsrA-EsrB encoded within region 2 (Fig. 1A) (Tan et al., unpublished). The deduced amino acid sequence of EsrA has a high homology to a number of two-component system sensors, including SpiR (SsrA) of Salmonella enterica serovar Typhimurium (31% identity and 45% similarity to residues 29 to 834 of EsrA), which controls the TTSS of SPI-2 (30), and MmoS of Methylococcus capsulatus (31% identity and 48% similarity to residues 411 to 905 of EsrA), which is involved in the copper-dependent regulation of a soluble methane monooxygenase (5). The response regulator EsrB has a CheY-like receiver domain at the N-terminal half and a helix-turn-helix (HTH) domain at the C-terminal half. The homologs of EsrB include the response regulator SsrB of the two-component system SsrA-SsrB of S. enterica serovar Typhimurium (42% identity and 57% similarity to residues 25 to 214 of EsrB) (3) and the response regulator YvqC of the two-component system YvqE-YvqC of Bacillus subtilis (30% identity and 50% similarity to residues 8 to 208 of EsrB) (47).
The characterization of the regulation of the two TTSS regions of E. tarda is ongoing, and we have found that esrC encoded another regulator. esrC is oriented in the same transcriptional direction as esaG, esaH, esaI, esaJ, orf19, esaK, and esaL (Fig. 1A) and are presumably in the same transcriptional unit since no obvious transcription termination sequences have been found between them. The putative ribosome-binding site (AGGAGC) was identified 6 bp upstream of the postulated translational initiation codon of esrC. A homology search of EsrC retrieved a large number of bacterial transcriptional regulators of the AraC family. Some of them are regulators of TTSSs, including ExsA of Pseudomonas aeruginosa (30% identity and 59% similarity to residues 133 to 229 of EsrC) (13) and VirF of Y. enterocolitica (32% identity and 63% similarity to residues 136 to 221 of EsrC) (4) (Fig. 2). Members of this protein family, as exemplified by AraC, contain the dimerization and effector binding domains in the N-terminal half and two potential HTH DNA-binding and transcriptional activation domains at the C-terminal half. The similarity among these proteins is higher at the C-terminal half, which is supposed to be a DNA-binding domain (Fig. 2). However, no homology was found within the 126 N-terminal amino acids of EsrC in a search of the databases.
Role of EsrC in E. tarda virulence and protein secretion. The homology of EsrC to the TTSS transcriptional activators suggests that EsrC may be required for the virulence of E. tarda. A deletion mutant, esrC, was constructed to remove an internal fragment of 627 bp, and the resulting mutant contained the first 33 bp and the last 30 bp of the esrC gene. The esrC mutant increased its LD50 (2.6 logs higher than that of the wild type) in blue gourami fish and showed a lower replication rate in blue gourami phagocytes (Table 3). esrC was also found to secrete less than 10% of the extracellular proteins (ECPs) and did not show any aggregation in DMEM after 24 h of culturing (Table 3). In general, the phenotypes of the esrC mutant are similar to those of esrA and esrB insertional mutants (Table 3). Complementation of the esrC deletion mutant in trans with a plasmid-borne wild-type copy of esrC (esrC+esrC) restored the wild-type phenotypes, including lower LD50 values, an increase in the replication rate inside the blue gourami phagocytes, autoaggregation, and the production of normal amounts of ECPs (Table 3).
To show the importance of the N-terminal sequence and the two HTHs on the function of EsrC, four additional in-frame deletion mutants were constructed: esrCN1 (deletion of 71 amino acids from the N-terminal end), esrCN2 (deletion of 123 amino acids from the N-terminal end), esrC163 (deletion of the first HTH domain), and esrC263 (deletion of the second HTH domain) (Table 1). All of these mutants significantly reduced the LD50 values, and the values were comparable to those of the esrC deletion mutant (Table 3). All four of these esrC mutants also had low ECP production and failed to aggregate in DMEM culture (Table 3). Our results demonstrate that EsrC plays an important role in E. tarda pathogenesis and that an intact and complete protein is required for it to function.
To find out which genes are regulated by EsrC, the total bacterial proteins and the ECPs of the wild type and the esrC mutant were compared by using 2-DE (Fig. 3A and B). Four protein spots (spots 1 to 4) were absent in the esrC mutant. Spots 1 and 2 were confirmed with matrix-assisted laser desorption ionization-time of flight mass spectrometry as the TTSS secreted proteins EseB and EseD, respectively. Spots 3 and 4 were identified as EvpA and EvpC of the EVP proteins, respectively. The two secreted proteins EseC (spot 5) and E2 (spot 6) reported previously (41) were also absent in the ECP profile of the esrC mutant (Fig. 3C and D).
Functional relationship between two-component system EsrA-EsrB and EsrC. The observation that esrA, esrB, and esrC mutants had similar low ECP productions (which were <10% of the wild-type strain) in DMEM (Table 3) led us to examine their interactions in regulating the TTSS of E. tarda. EsrA-EsrB is a putative two-component system that regulates the expressions of the TTSS and EVP gene clusters. However, the opposite orientation of esrA and esrB in the chromosome makes their transcriptional relationship elusive. To investigate the relationships of esrA, esrB, and esrC, we examined the effect of loss-of-function mutations in each of them on the expression of esrA, esrB, and esrC. Plasmids pRWesrA, pRWesrB, and pRWesrC carrying the lacZ transcriptional fusions to the upstream (putative promoter) regions of esrA, esrB, and esrC, respectively, were constructed (Fig. 1A) and introduced into the wild type and the esrA, esrB, and esrC mutants. High -galactosidase activities were detected in all of the wild-type strains containing one of these pRW50 derivatives, indicating the presence of promoters in front of esrA, esrB, and esrC (Fig. 4). The inactivation of esrA did not reduce the -galactosidase expression of esrB-lacZ, and the inactivation of esrB also did not decrease the expression of the lacZ of esrA-lacZ (Fig. 4). When the plasmid pRWesrC was introduced into esrA and esrB mutants, the expression of -galactosidase was reduced 100-fold compared to that of the wild type and was comparable to the vector control pRW50. Our results therefore suggest that EsrA and EsrB exert their regulatory effect upstream on esrC. The inactivation of esrA, esrB, and esrC in the chromosome did not affect the expression of esrA-lacZ, esrB-lacZ, and esrC-lacZ, respectively, indicating that they may not be self-regulated (Fig. 4). To determine whether EsrB directly activates esrC expression, the effect of providing esrB in trans under the control of lacI on the expression of esrC-lacZ from pRWesrC was tested in E. coli. Upon the induction of IPTG (isopropyl--D-thiogalactopyranoside), pQEesrB was able to activate esrC-lacZ expression in E. coli. Very low LacZ activity was detected without the IPTG induction (Table 4). Therefore, esrB may directly activate the expression of esrC.
Regulation of TTSS apparatus genes and orf29 and orf30. To find out whether esrC regulates TTSS apparatus genes, we examined the expression of esaC in the esrC mutant. EsaC is homologous to SsaC of Salmonella SPI-2, which is a TTSS apparatus protein. The inactivation of esaC abolished the secretion of EseB, EseC, and EseD (data not shown). Plasmid pRWesaC carrying the lacZ fusion with the promoter region of esaC (Fig. 1A) was introduced into the wild-type E. tarda and esrA, esrB, and esrC mutants. High -galactosidase activities were detected in the wild type, confirming the presence of a promoter. However, expression of the lacZ fusion was reduced >20-fold in the esrA, and esrB mutants but not in the esrC mutant (Fig. 5). Our results suggest that the EsrA-EsrB two-component system regulates the expression of esaC and that esrC is not required for esaC expression.
Plasmid pRWorf29 carrying the lacZ fusion with the promoter region of orf29 was also constructed (Fig. 1A) and introduced into esrA, esrB, and esrC mutants. As shown in Fig. 5, expression of lacZ of orf29-lacZ in the esrC mutant was found to be four times lower than that in the wild type, demonstrating regulation by EsrC on the transcriptional unit of orf29 and orf30. Expression of lacZ of orf29-lacZ was also examined in the background of esrA or esrB, and the -galactosidase activities in these two mutants were also reduced (Fig. 5), indicating that EsrA-EsrB also can regulate the transcriptional unit of orf29 and orf30 independent of EsrC. Therefore, it is likely that there are two promoters upstream of orf29: one is esrC dependent, and the other is EsrB dependent. Expressions of orf29 and orf30 were regulated by both EsrB and EsrC and could be activated from either of these promoters.
EsrC regulates the EVP gene cluster. EvpA and EvpC were shown to be regulated by EsrC earlier in the 2-DE studies (Fig. 3). The next question is how esrC regulates the expression of EVP genes and whether EsrC also regulates the other genes encoded in the EVP gene cluster. Plasmid pRWevpA containing the putative promoter region of evpA (Fig. 1B) was introduced into the wild-type E. tarda and the esrC mutant. As shown in Fig. 6, high -galactosidase activities were detected in the wild type, which confirmed the presence of a promoter. Expression of the LacZ reporter decreased over 30 times in the esrC mutant compared to the wild type (Fig. 6). The data further confirmed the regulation of EsrC on the evpA-lacZ and indicated that the regulation was at the transcriptional level. To find out whether there were other possible promoters in this EVP gene cluster, we constructed three more reporter fusions in pRW50 plasmids with the upstream sequences of evpC, evpD, and evpH (Fig. 1B). These constructs were introduced into the E. tarda wild type, and the lacZ expression was monitored. The results showed that the expressions of reporters in all of the wild-type E. tarda with these three constructs had low -galactosidase activities that were comparable to the vector control (pRW50) (data not shown). The results indicate that there may not be other promoters in the regions examined. It is possible that the EVP gene cluster transcript in one single unit and esrC is required for the expression of all of the genes in this EVP gene cluster. To determine whether EsrC directly activates evpA expression, the effect of providing esrC in trans under the control of lacI on the expression of evpA-lacZ was tested in E. coli. Under the induction of IPTG, pQEesrC was able to activate evpA-lacZ in E. coli, while very low LacZ activity was detected without the IPTG induction (Table 4). Therefore, esrC may directly activate the expression of evpA.
Regulation of esrA-esrB and esrC. The expression of TTSS secreted proteins and EVP proteins was temperature dependent (41), and EsrC was speculated to contribute to the receiving of temperature change signals due to the similarity to thermo-induced regulator VirF in yersiniae (Fig. 2). However, there is no evidence to support this hypothesis. To clarify this, plasmids pRWesaC and pRWesrC were transformed into the E. tarda wild type, and the expression level of lacZ was studied at either 37 or 25°C. As shown in Fig. 7, the expression of lacZ of esrC-lacZ and esaC-lacZ at 37°C decreased approximately two times compared to that seen at 25°C. These data indicate that the expression of esrC itself is affected by temperature. Since we had shown earlier that esrC was under the control of esrA and esrB, we wanted to determine whether the effect of temperature on the esrC transcription was the result of the decrease in esrA and esrB expression. The expression of lacZ in the wild type of E. tarda carrying pRWesrA and pRWesrB at 37 and 25°C was examined. We found that the expression of lacZ of the E. tarda wild type carrying esrA-lacZ was similar at either 37 or 25°C. However, the expression of lacZ by the wild type carrying esrB-lacZ was two times lower at 37°C than that at 25°C, demonstrating that the expression of esrB, but not esrA, was regulated by temperature. These results suggest that the effect of temperature on the expressions of eseB, eseD, and evp genes is through regulating the expression of esrB.
The insertion of TnphoA in pst genes has been reported to affect the expression of TTSS and EVP genes (41). On the other hand, a TnphoA mutant (i.e., mutant 306, accession no. AY643479) was found to be highly attenuated when injected into blue gourami fish, and it failed to produce TTSS (EseB and EseD) and EVP proteins (EvpA and EvpC) (data not shown). Sequencing data at the point of transposon insertion of this mutant showed similarity to a putative membrane protein of Y. pestis. Therefore, we were very interested in exploring how these mutants affected this regulatory hierarchy. To this end, the plasmids of pRWesrA, pRWesrB and pRWesrC were transformed into pstC and mutant 306. As shown in Fig. 7, in both pstC and mutant 306, the expressions of lacZ of esrA-lacZ and esrB-lacZ did not decrease although eseB, eseD, evpA, and evpC failed to express in these mutants (data not shown) (41). In sharp contrast, the expression of esrC-lacZ was suppressed significantly. These results indicate that the effect of pstC and mutant 306 on the expression of TTSS and evp genes might not be through regulating the transcription of esrA and esrB. We further examined the expression of the TTSS apparatus gene esaC in pstC and mutant 306. Interestingly, in both of these mutants, the expression of lacZ of esaC-lacZ decreased significantly. Such observation indicates that the insertion of TnphoA in pstC and the putative membrane protein also affects TTSS apparatus gene expression.
DISCUSSION
EsrC is a positive regulator. We report here the characterization of another novel regulatory protein, EsrC, which is encoded within the TTSS of E. tarda. EsrC has a significant sequence similarity to members of the AraC family of the transcriptional regulatory proteins. A nonpolar deletion mutation in esrC rendered the mutant less virulent. The wild-type phenotypes could be restored by the complementation of esrC provided by a plasmid with a constitutive promoter. Our results suggest that esrC is an essential virulence gene for E. tarda pathogenesis, and its role is comparable to esrA and esrB (Table 3). The LacZ reporter fusion analysis indicated that EsrC was downstream of the EsrA-EsrB regulatory pathway (Fig. 8). The regulation of EsrB on esrC was possibly direct since EsrB could regulate the expression of the reporter fusion (esrC-lacZ) in E. coli (Table 4). esrC was possibly in the same transcriptional unit and under the same mode of regulation with the apparatus genes (Fig. 8). esrC was also shown to be not self-regulated (Fig. 4). The TTSS apparatus genes, therefore, may be regulated by EsrA-EsrB instead of EsrC. This hypothesis was further confirmed by the reporter esaC-lacZ, which was expressed in the esrC mutant but not in the esrA or esrB insertional mutants (Fig. 5). Although both EsrC and the two-component system EsrA-EsrB control the TTSS, EsrA-EsrB may directly regulate the expression of TTSS apparatus genes, whereas it is EsrC that regulates the expression of some specific transcriptional units such as the secreted proteins of TTSS (e.g., EseB and EseD). The regulation of EsrA-EsrB on TTSSs is, therefore, different from SsrA-SsrB in Salmonella species, in which SsrA-SsrB controls the expression of genes encoding both secreted proteins and the TTSS apparatus. The difference may be due to the lack of another regulatory protein encoded within TTSS of SPI-2.
EsrC regulates orf29 and orf30 and EVP. The expression of orf29 and orf30, the two novel genes identified at the end of the TTSS, was controlled by both EsrA-EsrB and EsrC, which might indicate that they were part of the TTSS gene cluster (Fig. 1A and 8). The regulation of these two ORFs is different from that of other TTSS apparatus genes (e.g., esaC), which are directly regulated only by the EsrA-EsrB two-component system. orf29 and orf30 may have a special function for TTSSs. The analysis with COILs (http://www.ch.embnet.org/software/coils_form.html) showed that the predicted proteins had a high degree of coils. Pallen et al. (32) reported that many of the secreted proteins of the TTSS shared a common coiled-coil structure feature. orf29 and orf30 share no homology with known structure proteins of the TTSS that are believed to be conserved in different bacterial species. Therefore, it is likely that orf29 and orf30 do not encode for apparatus but instead are effector proteins of E. tarda. This possibility will be examined in a future study.
The EVP gene cluster is encoded outside of the TTSS and mutation in esrB has been reported to affect the expression of EvpA and EvpC (41). However, the secretion of EVP proteins is independent of the TTSS. Mutations in the putative translocon proteins of the TTSS did not affect the secretion of EvpC (41). The EsrA-EsrB two-component system regulates EVP via EsrC (Fig. 8). The LacZ reporter study suggests that EsrC regulated the transcription of the evpA-H transcriptional unit; only one promoter sequence was found (Fig. 1 and 6), and the regulation of EsrC on evpA is possibly direct (Table 4). The homologs of the EVP gene cluster are widely distributed in other pathogens such as Escherichia, Salmonella, Vibrio, and Yersinia species with unknown functions (41), and it is possible that these homologs may also be regulated by the regulators encoded in the respective TTSSs. The direct regulation of the EVP gene cluster by the regulators encoded within the TTSS of E. tarda suggests the functional relationship between the TTSS and the EVP gene cluster. The EVP gene cluster and its homologs encode several conserved core proteins and possibly encode a novel secretion system (2, 11, 41). This is the first report to suggest a communication between the TTSS and the EVP in E. tarda. In Salmonella species, SPI-5 encodes at least five effectors (48), among which SigD/SopB is coordinately regulated with SPI-1 genes (6) and is secreted by the SPI-1 TTSS (20). However, PipB, another effector in SPI-5, is part of the SPI-2 regulon and is translocated by the SPI-2 TTSS to the Salmonella-containing vacuole (23). These studies demonstrate a functional and regulatory cross talk between different secretion systems. E. tarda might use EsrC to coordinate the EVP secretion system with the TTSS in infection or survival in host cells.
Involvements of other regulators. The regulators encoded within the TTSS usually respond to regulation by environmental factors or other global regulators. In Salmonella species, the SsrA-SsrB two-component system was regulated by another two-component system OmpR-EnvZ, of which the response regulator OmpR binds directly to the ssrA and ssrB promoter (10, 24). In E. tarda, the transposon mutant 306, which has an insertion in a putative membrane protein, was also shown to affect the expression of the TTSS. Unlike OmpR-EnvZ in Salmonella species, this mutant did not affect the expressions of esrA or esrB but suppressed the expression of esrC and the TTSS apparatus genes (e.g., esaC) (Fig. 7). Interestingly, transposon insertion in pstC also has a similar effect on the expressions of TTSS and EVP (Fig. 7), suggesting that the PST operon has some control over the expressions of TTSS and EVP. The PST operon is required for phosphate transport (46). In S. enterica serovar Typhimurium, the transposon mutant pstS reduced the expression of the TTSS regulator of hilA and the invasion genes, and this repression was due to the negative control of the PhoR-PhoB two-component system. The pstS mutation led to the accumulation of PhoBP, and PhoBP directly or indirectly repressed hilA and the invasion genes (28). E. tarda might have a similar mechanism: the insertion inactivation of the PST system may lead to the accumulation of some negative regulators of the TTSS that have not been identified, thus repressing the expression of TTSS genes. If this is the case, this negative regulation would not modulate the expressions of esrA and esrB since both of their expressions were not affected in the pstC mutant background (Fig. 7). This negative regulator might directly interact with EsrB and interrupt its function. Another hypothesis is that some proteins encoded in the PST operon function as positive regulators and coordinate with EsrB in regulating the TTSS. However, we have shown that in E. coli, under the induction of IPTG, pQEesrB could activate lacZ expression of esrC-lacZ because of large amounts of EsrB accumulation. In E. tarda, as a regulator gene, the expression of esrB was low and EsrB could not activate the expression of the TTSS without the help of the positive regulator encoded in the PST operon. Future studies will attempt to distinguish between these possibilities.
Effect of high temperature. The expression of the esrB-lacZ fusion decreased substantially at 37°C compared to that seen at 25°C, but there was no change for the esrA-lacZ fusion at different temperatures. Our results suggest that the regulation of temperature on the TTSS is through the modulation of esrB expressions (Fig. 8). It is possible that the primary regulatory event is at the esrB transcription and that the accumulation of EsrB increases the transcription of esrC. This results in the increased production of EsrC activator proteins that bind to sites upstream of the target genes and activate transcription. If so, the mechanism response for temperature sensing is independent of EsrB and EsrC. EsrA functions as a sensor in EsrA-EsrB two-component system, and it is possible that EsrA is the one that responds to temperature changes. However, the expression of esrB-lacZ in an ersA mutant background did not decrease compared to that in the wild-type background at 25°C (Fig. 4). From these data, we hypothesize that there must be proteins other than EsrA responding to temperature changes. Alternatively, temperature changes may affect DNA conformation in the esrB promoter region. Changes in DNA topology have been shown to affect expressions of a variety of genes (7, 8, 14, 17). In E. tarda, the topology of the DNA promoter region of esrB may vary at different temperatures and result in altered esrB expressions. Further experiments must be done to confirm these two hypotheses.
In conclusion, we report that the EsrA-EsrB/EsrC regulatory cascade is the key regulon controlling the expressions of TTSS and EVP gene clusters and plays a vital role in E. tarda pathogenesis. EsrA-EsrB controls most of the transcriptional units of the TTSS to ensure that the TTSS is functional. EsrA-EsrB also controls EsrC, which regulates a subset of transcriptional units. They may be effector proteins or proteins for special functions, such as the transcriptional units of orf29 and orf30. Another main role of EsrC is to control the EVP, possibly another secretion system that is different from the TTSS. The EsrA-EsrB/EsrC regulatory cascade, therefore, can coordinate the expressions of these two secretion systems during infection and host-bacterium interactions. The sequencing of the TTSS and EVP gene clusters is ongoing. More transcriptional units will be identified, and this will help to dissect the precise roles of this EsrA-EsrB/EsrC regulatory cascade. Future work will also be carried out to identify and characterize the factors affected by the insertional disruptions of pstC and the putative membrane protein and to identify factors that communicate temperature changes with the TTSS and EVP gene clusters. This information will be useful in elucidating the functions of TTSS and EVP gene clusters and thus the pathogenesis of E. tarda.
ACKNOWLEDGMENTS
We are grateful to the Biomedical Research Council of Singapore (BMRC), ASTAR, the National University of Singapore, and the Lee Hiok Kwee Donation for providing research grants in support of this study.
We thank S. J. Busby of University of Birmingham for providing the plasmid pRW50 and its sequence information. We also thank I. Rosenshine and P. Tang for critical comments, which helped in improving the manuscript.
REFERENCES
1. Banks, A. S. 1992. A puncture wound complicated by infection with Edwardsiella tarda. J. Am. Pediatr. Med. Assoc. 82:529-531.
2. Bladergroen, M. R., K. Badelt, and H. P. Spaink. 2003. Infection-blocking genes of a symbiotic Rhizobium leguminosarum strain that are involved in temperature-dependent protein secretion. Mol. Plant-Microbe Interact. 16:53-64.
3. Cirillo, D. M., R. H. Valdivia, D. M. Monack, and S. Falkow. 1998. Macrophage-dependent induction of the Salmonella pathogenicity island 2 type III secretion system and its role in intracellular survival. Mol. Microbiol. 30:175-188.
4. Cornelis, G., C. Sluiters, C. L. de Rouvroit, and T. Michiels. 1989. Homology between VirF, the transcriptional activator of the Yersinia virulence regulon, and AraC, the Escherichia coli arabinose operon regulator. J. Bacteriol. 171:254-262.
5. Csaki, R., L. Bodrossy, J. Klem, J. C. Murrell, and K. L. Kovacs. 2003. Genes involved in the copper-dependent regulation of soluble methane monooxygenase of Methylococcus capsulatus (Bath): cloning, sequencing, and mutational analysis. Microbiology 149:1785-1795.
6. Darwin, K. H., and V. L. Miller. 2001. Type III secretion chaperone-dependent regulation: activation of virulence genes by SicA and InvF in Salmonella typhimurium. EMBO. J. 20:1850-1862.
7. Dorman, C. J. 1996. Flexible response: DNA supercoiling, transcription and bacterial adaptation to environmental stress. Trends Microbiol. 4:214-216.
8. Dorman, C. J., G. C. Barr, N. N. Bhriain, and C. F. Higgins. 1988. DNA supercoiling and the anaerobic and growth phase regulation of tonB gene expression. J. Bacteriol. 170:2816-2826.
9. Edwards, R. A., L. H. Keller, and D. M. Schifferli. 1998. Improved allelic exchange vectors and their use to analyze 987P fimbria gene expression. Gene 207:149-157.
10. Feng, X., R. Oropeza, and L. J. Kenney. 2003. Dual regulation by phospho-OmpR of ssrA/B gene expression in Salmonella pathogenicity island 2. Mol. Microbiol. 48:1131-1143.
11. Folkesson, A., S. Lofdahl, and S. Normark. 2002. The Salmonella enterica subspecies I specific centisome 7 genomic island encodes novel protein families present in bacteria living in close contact with eukaryotic cells. Res. Microbiol. 153:537-545.
12. Francis, M. S., H. Wolf-Watz, and A. Forsberg. 2002. Regulation of type III secretion systems. Curr. Opin. Microbiol. 5:166-172.
13. Frank, D. W., and B. H. Iglewski. 1991. Cloning and sequence analysis of a trans-regulatory locus required for exoenzyme S synthesis in Pseudomonas aeruginosa. J. Bacteriol. 173:6460-6468.
14. Galan, J. E., and R. Curtiss III. 1990. Expression of Salmonella typhimurium genes required for invasion is regulated by changes in DNA supercoiling. Infect. Immun. 58:1879-1885.
15. Gallegos, M. T., R. Schleif, A. Bairoch, K. Hofmann, and J. L. Ramos. 1997. AraC/XylS family of transcriptional regulators. Microbiol. Mol. Biol. Rev. 61:393-410.
16. Hensel, M., J. E. Shea, S. R. Waterman, R. Mundy, T. Nikolaus, G. Banks, A. Vazquez-Torres, C. Gleeson, F. C. Fang, and D. W. Holden. 1998. Genes encoding putative effector proteins of the type III secretion system of Salmonella pathogenicity island 2 are required for bacterial virulence and proliferation in macrophages. Mol. Microbiol. 30:163-174.
17. Higgins, C. F., C. J. Dorman, D. A. Stirling, L. Waddell, I. R. Booth, G. May, and E. Bremer. 1988. A physiological role for DNA supercoiling in the osmotic regulation of gene expression in Salmonella typhimurium and Escherichia coli. Cell 52:569-584.
18. Hirono, I., N. Tange, and T. Aoki. 1997. Iron regulated hemolysin gene from Edwardsiella tarda. Mol. Microbiol. 24:851-856.
19. Ho, S. N., H. D. Hunt, R. M. Horton, J. K. Pullen, and L. R. Pease. 1989. Site-directed mutagenesis by overlap extension using the polymerase chain reaction. Gene 77:51-59.
20. Hong, K. H., and V. L. Miller. 1998. Identification of a novel Salmonella invasion locus homologous to Shigella ipgDE. J. Bacteriol. 180:1793-1802.
21. Hueck, C. J. 1998. Type III protein secretion systems in bacterial pathogens of animals and plants. Microbiol. Mol. Biol. Rev. 62:379-433.
22. Janda, J. M., S. L. Abott, S. Kroshe-Bystrom, W. K. W. Cheng, C. Powers, R. P. Kokka, and K. Tamura. 1991. Pathogenic properties of Edwardsiella species. J. Clin. Microbiol. 29:1977-2001.
23. Knodler, L. A., J. Celli, W. D. Hardt, B. A. Vallance, C. Yip, and B. B. Finlay. 2002. Salmonella effectors within a single pathogenicity island are differentially expressed and translocated by separate type III secretion systems. Mol. Microbiol. 43:1089-1103.
24. Lee, A. K., C. S. Detweiler, and S. Falkow. 2000. OmpR regulates the two-component system SsrA-SsrB in Salmonella pathogenicity island 2. J. Bacteriol. 182:771-781.
25. Li, M., I. Rosenshine, S. L. Tung, X. H. Wang, D. Friedberg, C. L. Hew, and K. Y. Leung. 2004. Comparative proteomic analysis of extracellular proteins of enterohemorrhagic and enteropathogenic Escherichia coli strains and their ihf and ler mutants. Appl. Environ. Microbiol. 70:5274-5282.
26. Ling, S. H. M., X. H. Wang, L. Xie, T. M. Lim, and K. Y. Leung. 2000. Use of green fluorescent protein (GFP) to track the invasive pathways of Edwardsiella tarda in the in vivo and in vitro fish models. Microbiology 146:7-19.
27. Lodge, J., J. Fear, S. Busby, P. Gunasekaran, and N. R. Kamini. 1992. Broad host range plasmids carrying the Escherichia coli lactose and galactose operons. FEMS. Microbiol. Lett. 95:271-276.
28. Lucas, R. L., C. P. Lostroh, C. C. DiRusso, M. P. Spector, B. L. Wanner, and C. A. Lee. 2000. Multiple factors independently regulate hilA and invasion gene expression in Salmonella enterica serovar Typhimurium. J. Bacteriol. 182:1872-1882.
29. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
30. Ochman, H., F. C. Soncini, F. Solomon, and E. A. Groisman. 1996. Identification of a pathogenicity island required for Salmonella survival in host cells. Proc. Natl. Acad. Sci. USA 93:7800-7804.
31. Osiri, M., T. Tantawichien, and U. Deesomchock. 1997. Edwardsiella tarda bacteremia and septic arthritis in a patient with diabetes mellitus. Southeast Asian J. Trop. Med. Public Health 28:669-672.
32. Pallen, M. J., G. Dougan, and G. Frankel. 1997. Coiled-coil domains in proteins secreted by type III secretion systems. Mol. Microbiol. 25:423-425.
33. Plumb, J. A. 1993. Edwardsiella septicemia, p. 61-79. In V. Inglis, R. J. Roberts, and N. R. Bromage (ed.), Bacterial diseases of fish. Cambridge University Press, Cambridge, United Kingdom.
34. Reed, L. J., and H. Muench. 1938. A simple method of estimating fifty percent end points. Am. J. Hyg. 27:493-497.
35. Rubires, X., F. Saigí, N. Pique, N. Climent, S. Merino, S. Albertí, J. M. Tomas, and M. Regue. 1997. A gene (wbbL) from Serratia marcescens N28b (O4) complements the rfb-50 mutation of Escherichia coli K-12 derivatives. J. Bacteriol. 179:7581-7586.
36. Secombs, C. J. 1990. Isolation of salmonid macrophages and analysis of their killing activity, p. 137-154. In J. S. Stolen, T. C. Fletcher, D. P. Anderson, B. S. Roberson, and W. B. Van Muiswinkel (ed.), Techniques in fish immunology. SOS Publications, Fair Haven, N.J.
37. Slaven, E. M., F. A. Lopez, S. M. Hart, and C. V. Sanders. 2001. Myonecrosis caused by Edwardsiella tarda: a case report and case series of extraintestinal E. tarda infections. Clin. Infect. Dis. 32:1430-1433.
38. Srinivasa Rao, P. S., T. M. Lim, and K. Y. Leung. 2001. Opsonized virulent Edwardsiella tarda strains are able to adhere to and survive and replicate within fish phagocytes but fail to stimulate reactive oxygen intermediates. Infect. Immun. 69:5689-5697.
39. Srinivasa Rao, P. S., T. M. Lim, and K. Y. Leung. 2003. Functional genomics approach to the identification of virulence genes involved in Edwardsiella tarda pathogenesis. Infect. Immun. 71:1343-1351.
40. Srinivasa Rao, P. S., Y. Yamada, and K. Y. Leung. 2003. A major catalase (KatB) that is required for H2O2 and phagocyte-mediated killing in Edwardsiella tarda. Microbiology 149:2635-2644.
41. Srinivasa Rao, P. S., Y. Yamada, Y. P. Tan, and K. Y. Leung. 2004. Use of proteomics to identify novel virulence determinants that are required for Edwardsiella tarda pathogenesis. Mol. Microbiol. 53:573-586.
42. Tan, Y. P., Q. Lin, X. H. Wang, S. Joshi, C. L. Hew, and K. Y. Leung. 2002. Comparative proteomic analysis of extracellular proteins of Edwardsiella tarda. Infect. Immun. 70:6475-6480.
43. Thune, R. L., L. A. Stanley, and R. K. Cooper. 1993. Pathogenesis of gram-negative bacterial infections in warm water fish. Annu. Rev. Fish Dis. 3:37-68.
44. Tu, X., I. Nisan, C. Yona, E. Hanski, and I. Rosenshine. 2003. EspH, a new cytoskeleton-modulating effector of enterohaemorrhagic and enteropathogenic Escherichia coli. Mol. Microbiol. 47:595-606.
45. Ullah, M. A., and T. Arai. 1983. Pathological activities of the naturally occurring strains of Edwardsiella tarda. Fish Pathol. 18:65-70.
46. Webb, D. C., H. Rosenberg, and G. B. Cox. 1992. Mutational analysis of Escherichia coli phosphate-specific transport system, a member of traffic ATPase (or ABC) family of membrane transporters. J. Biol. Chem. 267:24661-24668.
47. Wipat, A., S. C. Brignell, B. J. Guy, M. Rose, P. T. Emmerson, and C. R. Harwood. 1998. The yvsA-yvqA (293 degrees-289 degrees) region of the Bacillus subtilis chromosome containing genes involved in metal ion uptake and a putative sigma factor. Microbiology 144:1593-1600.
48. Wood, M. W., M. A. Jones, P. R. Watson, S. Hedges, T. S. Wallis, and E. E. Galyov. 1998. Identification of a pathogenicity island required for Salmonella enteropathogenicity. Mol. Microbiol. 29:883-891.
49. Yang, C. H., and C. K. Wang. 1999. Edwardsiella tarda bacteraemia complicated by acute pancreatitis and pyomyoma. J. Infect. 38:124-126.(J. Zheng, S. L. Tung, and)
ABSTRACT
Edwardsiella tarda is a gram-negative enteric pathogen that causes hemorrhagic septicemia in fish and gastro- and extraintestinal infections in humans. A type III secretion system (TTSS) and a putative secretion system (EVP) have been found to play important roles in E. tarda pathogenesis. Our previous studies suggested that the TTSS and EVP gene clusters were regulated by a two-component system of EsrA-EsrB. In the present study, we characterized another regulator, EsrC, which showed significant sequence similarity to the AraC family of transcriptional regulators. Mutants with in-frame deletions of esrC increased the 50% lethal doses in blue gourami fish, reduced extracellular protein production, and failed to aggregate. Complementation of esrC restored these three phenotypes. Two-dimensional gel electrophoresis showed that EsrC regulated the expression of secreted proteins encoded by the TTSS (such as EseB and EseD) and EVP (EvpC) gene clusters. The expression of esrC required a functional two-component system of EsrA-EsrB. EsrC in turn regulated the expression of selected genes encoded in TTSS (such as the transcriptional unit of orf29and orf30, but not esaC) and genes encoded in the EVP gene cluster. The present study sheds light on the regulation of these two key virulence-associated secretion systems and provides greater insight into the pathogenic mechanisms of this bacterium.
INTRODUCTION
Edwardsiella tarda is an important cause of hemorrhagic septicemia. It infects many commercially important farmed fish species and has led to extensive losses in both freshwater and marine aquaculture (43). The organism has a broad host range and is also known to cause infections in higher animals, including humans (33). In humans, it causes gastro- and extraintestinal infections such as myonecrosis (37), bacteremia (49), septic arthritis (31), and wound infections (1). The pathogenesis of E. tarda is multifactorial. Several potential virulence factors have been reported, including the abilities to invade epithelial cells (22, 26), resist serum (22, 26) and phagocyte-mediated killings (38), and produce toxins and enzymes such as hemolysins (18), catalases (40), and dermatotoxins (45) for disseminating infection.
Using a combination of comparative proteomics of secreted proteins and TnphoA mutagenesis, we identified a type III secretion system (TTSS) (39, 42) and a putative secretion system (EVP) (for E. tarda virulence proteins) (41) that contributed to E. tarda PPD130/91 pathogenesis. TTSSs are generally used by bacterial pathogens to deliver virulence factors into host cells to subvert normal cell functions (21). The TTSS gene cluster of E. tarda PPD130/91 contains at least 30 open reading frames (ORFs) and is in two DNA regions (Y. P. Tan, J. Zheng, S. L. Tung, I. Rosenshine, and K. Y. Leung, unpublished data). The gene members of E. tarda TTSS are composed of apparatus, chaperones, effectors, and regulators. Some of the gene members of the E. tarda TTSS are homologous to other TTSSs encoded by pathogens such as enteropathogenic Escherichia coli, Salmonella enterica serovar Typhimurium, and Yersinia enterocolitica. The designation of the E. tarda TTSS genes is based on the sequence homologs in Salmonella pathogenicity island 2 (SPI-2) (16). The inactivation of the TTSS in E. tarda led to the increase of 50% lethal doses (LD50) in blue gourami fish (39, 41).
The other gene cluster (EVP) is not unique to E. tarda but is widely distributed in many other animal and plant pathogens and symbionts such as Salmonella, Vibrio, Yersinia, Escherichia, Rhizobium, and Agrobacterium species with putative secretion functions (41). In R. leguminosarum, the impairment of EVP-like cluster affected the secretion of at least one protein based on two-dimensional gel electrophoresis (2-DE) analysis (2). Some proteins encoded in this EVP-like locus of Rhizobium shared homology with proteins encoded by the type III or type IV secretion systems (2). Folkesson et al. (11) also analyzed EVP homolog in S. enterica serovar Typhimurium and found that this gene cluster encoded putative cytoplasmic, periplasmic, and outer membrane localizing proteins. In E. tarda, eight genes (evpA to -H) have been found in the EVP cluster (41). evpA and evpC are under the control of a TTSS regulator esrB, while the secretion of EvpC is dependent on EvpB instead of on a functional TTSS. The deletion of evpB or evpC led to lower replication rates in gourami phagocytes and reduced protein secretion and virulence in blue gourami fish (41). Complementation of evpB and evpC deletion mutants restored the secretion of EvpC, partially increased the replication rates in gourami phogocytes and lowered the LD50 values in gourami fish, indicating that these evp genes are vital for E. tarda pathogenesis (41).
The expressions of type III secretion apparatus and effectors are usually subjected to complicated regulation (12). Our previous studies suggested that EsrB regulated both the TTSS (such as eseB and eseD) and the EVP gene cluster (such as evpA and evpC) (41). In addition, the regulation of these two gene clusters is controlled by other factors such as temperature and the pstSCAB-phoU operon (41). This pstSCAB-phoU operon is a high-affinity phosphate-specific transport (PST) operon belonging to the family of ATP binding cassette (ABC) transporters (46). Our previous studies demonstrated that TnphoA insertions in pstB, pstC, and pstS abolished the expression of TTSS, as well as EVP proteins, and the resulting PST mutants were highly attenuated in blue gourami fish (41). However, it is not clear how the two-component system and these factors regulate the TTSS and EVP locus.
A common mechanism of gene regulation in bacteria is via regulatory proteins of the AraC family. To date, this family contains more than 100 members as identified by sequence homology to the AraC protein of Escherichia coli (15). With a few exceptions, AraC homologs are transcriptional activators. We describe here the identification of EsrC, an AraC homolog encoded inside the TTSS. The markerless in-frame deletion mutation of esrC disrupted the expressions of the secreted proteins of the TTSS and EVP locus. The phenotype of the esrC mutant is similar to the phenotypes of the esrA and esrB mutants. Our studies showed that EsrA-EsrB regulates the expression of the secreted proteins of TTSS and EVP proteins through esrC.
MATERIALS AND METHODS
Bacterial strains and plasmids. The bacterial strains and plasmids used in the present study are shown in Table 1. E. tarda stains were grown in tryptic soy broth (TSB; Difco/Becton Dickinson) or Dulbecco modified Eagle medium (DMEM) at 25°C without shaking while Escherichia coli strains were cultured in Luria broth (LB; Difco/Becton Dickinson) at 37°C. Cultivation of bacteria in DMEM was carried out at 5% (vol/vol) CO2 atmosphere. When required, the appropriate antibiotics were supplemented at the following concentrations: ampicillin (100 μg/ml); kanamycin (50 μg/ml); colistin (12.5 μg/ml); chloramphenicol (30 μg/ml); and tetracycline (10 μg/ml).
Construction of deletion mutants and plasmids. Overlap extension PCR (19) was used to generate in-frame deletion of esrC on the E. tarda PPD130/91 chromosome. For the construction of esrC, two PCR fragments were generated from PPD130/91 genomic DNA with the primer pairs of MutC-for plus MutC-int-rev, and MutC-rev plus Mut-int-for (Table 2). The resulting products generated a 754-bp fragment containing the upstream of esrC and a 751-bp fragment containing the downstream of the esrC, respectively. A 16-bp overlap in the sequences (underlined) permitted amplification of a 1,505-bp product during a second PCR with the primers MutC-for and MutC-rev, both of which were introduced into a KpnI restriction site (boldface text in Table 2), respectively. The resulting PCR product contained a deletion from nucleotides 34 to 660 of esrC. The PCR product was cloned into pGEMT-Easy vector, and DNA sequencing was performed to confirm that the construct was correct. The esrC fragment was excised with KpnI, ligated into suicide vector pRE112 (Cmr) (9), and the resulting plasmid then transformed into SM10 pir. The single-crossover mutants were obtained by conjugal transfer into E. tarda PPD130/91. Double-crossover mutants were obtained by plating onto 10% sucrose-tryptic soy agar plates. The deletion mutants were confirmed by PCR and sequencing. The construction of other esrC deletion mutants followed a similar protocol. The primers used for the construction of esrCN1 are MutC-N-for plus N-rev and MutC-N-rev plus N-for. The primers used for the construction of esrCN2 are MutC-N-for plus N2-rev and MutC-N-rev plus N2-for. The primers used for esrC163 are MutC-HTH-for plus HTH-rev and MutC-HTH-rev plus HTH-for. The primers used for esrC263 are MutC-HTH2-for plus HTH2-rev and MutC-HTH2-rev plus HTH2-for. The corresponding restriction sites (KpnI or SacI) are labeled in boldface in Table 2.
For the construction of pACYC+esrC, the complete esrC gene was obtained by PCR with Pfu Turbo polymerase (Stratagene) with E. tarda PPD130/91 as the template DNA. The PCR products were digested and ligated into ScaI- and EcoRI-digested pACYC184 plasmid. For the construction of pQEesrB and pQEesrC, the complete esrB and esrC genes were amplified by using the forward primer esrB-com-for or esrC-com-for containing the ribosome-binding site (underlined) and EcoRI site (boldface) and the reverse primer esrB-com-rev or esrC-com-rev which supplied the BamHI site (boldface) (Table 2). The PCR products were digested with BamHI and EcoRI and cloned into EcoRI- and BamHI-digested PQE60lacI (44).
For the construction of LacZ reporter plasmids, the putative promoter regions were amplified from E. tarda PPD130/91 genomic DNA and the PCR products were cloned into pGEMT-Easy vector. The resulting plasmids were sequenced, and the inserts were cut with BamHI and EcoRI (or MfeI) and subcloned into BamHI- and EcoRI-digested pRW50 plasmid (27).
LD50 determinations. Healthy naive blue gourami (Trichogaster trichopterus Pallas) of ca. 14 g were obtained from a commercial fish farm and infected with E. tarda as described previously (26). The mortality of the fish was recorded over a period of 7 days after infection. The LD50 values were calculated by the method of Reed and Muench (34).
Phagocyte isolation. Phagocytes were isolated from the head kidney of naive gourami and purified according to the procedure of Secombs (36) and as described by Srinivasa Rao et al. (38). Purified phagocytic cells (4 x 106 to 5 x 106 cells/well) were allowed to adhere to 48-well tissue culture plates (Falcon) in an L-15 medium (Sigma) that was supplemented with 5% fetal bovine serum. After 3 h of incubation at 25°C in a 5% (vol/vol) CO2 atmosphere, the cells were washed twice with Hanks balanced salt solution (HBSS; Sigma) and infected with E. tarda at a multiplicity of infection of 1:1 and incubated at 25°C for 30 min. The cells then were washed twice with HBSS and incubated with 100 μg of gentamicin/ml to kill all of the extracellular bacteria for 90 min. The monolayers were then washed twice with HBSS and lysed with 1% Triton X-100, and viable bacterial counts were enumerated by plate count. Phagocyte replication ratios were calculated by dividing the viable bacterial count after 5 h by the 2-h bacterial count.
2-DE and protein identification. Protein isolation and 2-DE were performed as previously described (41). Protein spots were identified with matrix-assisted laser desorption ionization-time of flight as described earlier (25).
-Galactosidase assays. Bacterial were grown in DMEM (for E. tarda) or LB (for E. coli) overnight at 25°C or 37°C. -Galactosidase activities were determined with cells permeabilized with sodium dodecyl sulfate and chloroform as described by Miller (29).
RESULTS
Sequence analysis of regulators. The TTSS of E. tarda is composed of two regions and was subjected to the regulation of a two-component regulator EsrA-EsrB encoded within region 2 (Fig. 1A) (Tan et al., unpublished). The deduced amino acid sequence of EsrA has a high homology to a number of two-component system sensors, including SpiR (SsrA) of Salmonella enterica serovar Typhimurium (31% identity and 45% similarity to residues 29 to 834 of EsrA), which controls the TTSS of SPI-2 (30), and MmoS of Methylococcus capsulatus (31% identity and 48% similarity to residues 411 to 905 of EsrA), which is involved in the copper-dependent regulation of a soluble methane monooxygenase (5). The response regulator EsrB has a CheY-like receiver domain at the N-terminal half and a helix-turn-helix (HTH) domain at the C-terminal half. The homologs of EsrB include the response regulator SsrB of the two-component system SsrA-SsrB of S. enterica serovar Typhimurium (42% identity and 57% similarity to residues 25 to 214 of EsrB) (3) and the response regulator YvqC of the two-component system YvqE-YvqC of Bacillus subtilis (30% identity and 50% similarity to residues 8 to 208 of EsrB) (47).
The characterization of the regulation of the two TTSS regions of E. tarda is ongoing, and we have found that esrC encoded another regulator. esrC is oriented in the same transcriptional direction as esaG, esaH, esaI, esaJ, orf19, esaK, and esaL (Fig. 1A) and are presumably in the same transcriptional unit since no obvious transcription termination sequences have been found between them. The putative ribosome-binding site (AGGAGC) was identified 6 bp upstream of the postulated translational initiation codon of esrC. A homology search of EsrC retrieved a large number of bacterial transcriptional regulators of the AraC family. Some of them are regulators of TTSSs, including ExsA of Pseudomonas aeruginosa (30% identity and 59% similarity to residues 133 to 229 of EsrC) (13) and VirF of Y. enterocolitica (32% identity and 63% similarity to residues 136 to 221 of EsrC) (4) (Fig. 2). Members of this protein family, as exemplified by AraC, contain the dimerization and effector binding domains in the N-terminal half and two potential HTH DNA-binding and transcriptional activation domains at the C-terminal half. The similarity among these proteins is higher at the C-terminal half, which is supposed to be a DNA-binding domain (Fig. 2). However, no homology was found within the 126 N-terminal amino acids of EsrC in a search of the databases.
Role of EsrC in E. tarda virulence and protein secretion. The homology of EsrC to the TTSS transcriptional activators suggests that EsrC may be required for the virulence of E. tarda. A deletion mutant, esrC, was constructed to remove an internal fragment of 627 bp, and the resulting mutant contained the first 33 bp and the last 30 bp of the esrC gene. The esrC mutant increased its LD50 (2.6 logs higher than that of the wild type) in blue gourami fish and showed a lower replication rate in blue gourami phagocytes (Table 3). esrC was also found to secrete less than 10% of the extracellular proteins (ECPs) and did not show any aggregation in DMEM after 24 h of culturing (Table 3). In general, the phenotypes of the esrC mutant are similar to those of esrA and esrB insertional mutants (Table 3). Complementation of the esrC deletion mutant in trans with a plasmid-borne wild-type copy of esrC (esrC+esrC) restored the wild-type phenotypes, including lower LD50 values, an increase in the replication rate inside the blue gourami phagocytes, autoaggregation, and the production of normal amounts of ECPs (Table 3).
To show the importance of the N-terminal sequence and the two HTHs on the function of EsrC, four additional in-frame deletion mutants were constructed: esrCN1 (deletion of 71 amino acids from the N-terminal end), esrCN2 (deletion of 123 amino acids from the N-terminal end), esrC163 (deletion of the first HTH domain), and esrC263 (deletion of the second HTH domain) (Table 1). All of these mutants significantly reduced the LD50 values, and the values were comparable to those of the esrC deletion mutant (Table 3). All four of these esrC mutants also had low ECP production and failed to aggregate in DMEM culture (Table 3). Our results demonstrate that EsrC plays an important role in E. tarda pathogenesis and that an intact and complete protein is required for it to function.
To find out which genes are regulated by EsrC, the total bacterial proteins and the ECPs of the wild type and the esrC mutant were compared by using 2-DE (Fig. 3A and B). Four protein spots (spots 1 to 4) were absent in the esrC mutant. Spots 1 and 2 were confirmed with matrix-assisted laser desorption ionization-time of flight mass spectrometry as the TTSS secreted proteins EseB and EseD, respectively. Spots 3 and 4 were identified as EvpA and EvpC of the EVP proteins, respectively. The two secreted proteins EseC (spot 5) and E2 (spot 6) reported previously (41) were also absent in the ECP profile of the esrC mutant (Fig. 3C and D).
Functional relationship between two-component system EsrA-EsrB and EsrC. The observation that esrA, esrB, and esrC mutants had similar low ECP productions (which were <10% of the wild-type strain) in DMEM (Table 3) led us to examine their interactions in regulating the TTSS of E. tarda. EsrA-EsrB is a putative two-component system that regulates the expressions of the TTSS and EVP gene clusters. However, the opposite orientation of esrA and esrB in the chromosome makes their transcriptional relationship elusive. To investigate the relationships of esrA, esrB, and esrC, we examined the effect of loss-of-function mutations in each of them on the expression of esrA, esrB, and esrC. Plasmids pRWesrA, pRWesrB, and pRWesrC carrying the lacZ transcriptional fusions to the upstream (putative promoter) regions of esrA, esrB, and esrC, respectively, were constructed (Fig. 1A) and introduced into the wild type and the esrA, esrB, and esrC mutants. High -galactosidase activities were detected in all of the wild-type strains containing one of these pRW50 derivatives, indicating the presence of promoters in front of esrA, esrB, and esrC (Fig. 4). The inactivation of esrA did not reduce the -galactosidase expression of esrB-lacZ, and the inactivation of esrB also did not decrease the expression of the lacZ of esrA-lacZ (Fig. 4). When the plasmid pRWesrC was introduced into esrA and esrB mutants, the expression of -galactosidase was reduced 100-fold compared to that of the wild type and was comparable to the vector control pRW50. Our results therefore suggest that EsrA and EsrB exert their regulatory effect upstream on esrC. The inactivation of esrA, esrB, and esrC in the chromosome did not affect the expression of esrA-lacZ, esrB-lacZ, and esrC-lacZ, respectively, indicating that they may not be self-regulated (Fig. 4). To determine whether EsrB directly activates esrC expression, the effect of providing esrB in trans under the control of lacI on the expression of esrC-lacZ from pRWesrC was tested in E. coli. Upon the induction of IPTG (isopropyl--D-thiogalactopyranoside), pQEesrB was able to activate esrC-lacZ expression in E. coli. Very low LacZ activity was detected without the IPTG induction (Table 4). Therefore, esrB may directly activate the expression of esrC.
Regulation of TTSS apparatus genes and orf29 and orf30. To find out whether esrC regulates TTSS apparatus genes, we examined the expression of esaC in the esrC mutant. EsaC is homologous to SsaC of Salmonella SPI-2, which is a TTSS apparatus protein. The inactivation of esaC abolished the secretion of EseB, EseC, and EseD (data not shown). Plasmid pRWesaC carrying the lacZ fusion with the promoter region of esaC (Fig. 1A) was introduced into the wild-type E. tarda and esrA, esrB, and esrC mutants. High -galactosidase activities were detected in the wild type, confirming the presence of a promoter. However, expression of the lacZ fusion was reduced >20-fold in the esrA, and esrB mutants but not in the esrC mutant (Fig. 5). Our results suggest that the EsrA-EsrB two-component system regulates the expression of esaC and that esrC is not required for esaC expression.
Plasmid pRWorf29 carrying the lacZ fusion with the promoter region of orf29 was also constructed (Fig. 1A) and introduced into esrA, esrB, and esrC mutants. As shown in Fig. 5, expression of lacZ of orf29-lacZ in the esrC mutant was found to be four times lower than that in the wild type, demonstrating regulation by EsrC on the transcriptional unit of orf29 and orf30. Expression of lacZ of orf29-lacZ was also examined in the background of esrA or esrB, and the -galactosidase activities in these two mutants were also reduced (Fig. 5), indicating that EsrA-EsrB also can regulate the transcriptional unit of orf29 and orf30 independent of EsrC. Therefore, it is likely that there are two promoters upstream of orf29: one is esrC dependent, and the other is EsrB dependent. Expressions of orf29 and orf30 were regulated by both EsrB and EsrC and could be activated from either of these promoters.
EsrC regulates the EVP gene cluster. EvpA and EvpC were shown to be regulated by EsrC earlier in the 2-DE studies (Fig. 3). The next question is how esrC regulates the expression of EVP genes and whether EsrC also regulates the other genes encoded in the EVP gene cluster. Plasmid pRWevpA containing the putative promoter region of evpA (Fig. 1B) was introduced into the wild-type E. tarda and the esrC mutant. As shown in Fig. 6, high -galactosidase activities were detected in the wild type, which confirmed the presence of a promoter. Expression of the LacZ reporter decreased over 30 times in the esrC mutant compared to the wild type (Fig. 6). The data further confirmed the regulation of EsrC on the evpA-lacZ and indicated that the regulation was at the transcriptional level. To find out whether there were other possible promoters in this EVP gene cluster, we constructed three more reporter fusions in pRW50 plasmids with the upstream sequences of evpC, evpD, and evpH (Fig. 1B). These constructs were introduced into the E. tarda wild type, and the lacZ expression was monitored. The results showed that the expressions of reporters in all of the wild-type E. tarda with these three constructs had low -galactosidase activities that were comparable to the vector control (pRW50) (data not shown). The results indicate that there may not be other promoters in the regions examined. It is possible that the EVP gene cluster transcript in one single unit and esrC is required for the expression of all of the genes in this EVP gene cluster. To determine whether EsrC directly activates evpA expression, the effect of providing esrC in trans under the control of lacI on the expression of evpA-lacZ was tested in E. coli. Under the induction of IPTG, pQEesrC was able to activate evpA-lacZ in E. coli, while very low LacZ activity was detected without the IPTG induction (Table 4). Therefore, esrC may directly activate the expression of evpA.
Regulation of esrA-esrB and esrC. The expression of TTSS secreted proteins and EVP proteins was temperature dependent (41), and EsrC was speculated to contribute to the receiving of temperature change signals due to the similarity to thermo-induced regulator VirF in yersiniae (Fig. 2). However, there is no evidence to support this hypothesis. To clarify this, plasmids pRWesaC and pRWesrC were transformed into the E. tarda wild type, and the expression level of lacZ was studied at either 37 or 25°C. As shown in Fig. 7, the expression of lacZ of esrC-lacZ and esaC-lacZ at 37°C decreased approximately two times compared to that seen at 25°C. These data indicate that the expression of esrC itself is affected by temperature. Since we had shown earlier that esrC was under the control of esrA and esrB, we wanted to determine whether the effect of temperature on the esrC transcription was the result of the decrease in esrA and esrB expression. The expression of lacZ in the wild type of E. tarda carrying pRWesrA and pRWesrB at 37 and 25°C was examined. We found that the expression of lacZ of the E. tarda wild type carrying esrA-lacZ was similar at either 37 or 25°C. However, the expression of lacZ by the wild type carrying esrB-lacZ was two times lower at 37°C than that at 25°C, demonstrating that the expression of esrB, but not esrA, was regulated by temperature. These results suggest that the effect of temperature on the expressions of eseB, eseD, and evp genes is through regulating the expression of esrB.
The insertion of TnphoA in pst genes has been reported to affect the expression of TTSS and EVP genes (41). On the other hand, a TnphoA mutant (i.e., mutant 306, accession no. AY643479) was found to be highly attenuated when injected into blue gourami fish, and it failed to produce TTSS (EseB and EseD) and EVP proteins (EvpA and EvpC) (data not shown). Sequencing data at the point of transposon insertion of this mutant showed similarity to a putative membrane protein of Y. pestis. Therefore, we were very interested in exploring how these mutants affected this regulatory hierarchy. To this end, the plasmids of pRWesrA, pRWesrB and pRWesrC were transformed into pstC and mutant 306. As shown in Fig. 7, in both pstC and mutant 306, the expressions of lacZ of esrA-lacZ and esrB-lacZ did not decrease although eseB, eseD, evpA, and evpC failed to express in these mutants (data not shown) (41). In sharp contrast, the expression of esrC-lacZ was suppressed significantly. These results indicate that the effect of pstC and mutant 306 on the expression of TTSS and evp genes might not be through regulating the transcription of esrA and esrB. We further examined the expression of the TTSS apparatus gene esaC in pstC and mutant 306. Interestingly, in both of these mutants, the expression of lacZ of esaC-lacZ decreased significantly. Such observation indicates that the insertion of TnphoA in pstC and the putative membrane protein also affects TTSS apparatus gene expression.
DISCUSSION
EsrC is a positive regulator. We report here the characterization of another novel regulatory protein, EsrC, which is encoded within the TTSS of E. tarda. EsrC has a significant sequence similarity to members of the AraC family of the transcriptional regulatory proteins. A nonpolar deletion mutation in esrC rendered the mutant less virulent. The wild-type phenotypes could be restored by the complementation of esrC provided by a plasmid with a constitutive promoter. Our results suggest that esrC is an essential virulence gene for E. tarda pathogenesis, and its role is comparable to esrA and esrB (Table 3). The LacZ reporter fusion analysis indicated that EsrC was downstream of the EsrA-EsrB regulatory pathway (Fig. 8). The regulation of EsrB on esrC was possibly direct since EsrB could regulate the expression of the reporter fusion (esrC-lacZ) in E. coli (Table 4). esrC was possibly in the same transcriptional unit and under the same mode of regulation with the apparatus genes (Fig. 8). esrC was also shown to be not self-regulated (Fig. 4). The TTSS apparatus genes, therefore, may be regulated by EsrA-EsrB instead of EsrC. This hypothesis was further confirmed by the reporter esaC-lacZ, which was expressed in the esrC mutant but not in the esrA or esrB insertional mutants (Fig. 5). Although both EsrC and the two-component system EsrA-EsrB control the TTSS, EsrA-EsrB may directly regulate the expression of TTSS apparatus genes, whereas it is EsrC that regulates the expression of some specific transcriptional units such as the secreted proteins of TTSS (e.g., EseB and EseD). The regulation of EsrA-EsrB on TTSSs is, therefore, different from SsrA-SsrB in Salmonella species, in which SsrA-SsrB controls the expression of genes encoding both secreted proteins and the TTSS apparatus. The difference may be due to the lack of another regulatory protein encoded within TTSS of SPI-2.
EsrC regulates orf29 and orf30 and EVP. The expression of orf29 and orf30, the two novel genes identified at the end of the TTSS, was controlled by both EsrA-EsrB and EsrC, which might indicate that they were part of the TTSS gene cluster (Fig. 1A and 8). The regulation of these two ORFs is different from that of other TTSS apparatus genes (e.g., esaC), which are directly regulated only by the EsrA-EsrB two-component system. orf29 and orf30 may have a special function for TTSSs. The analysis with COILs (http://www.ch.embnet.org/software/coils_form.html) showed that the predicted proteins had a high degree of coils. Pallen et al. (32) reported that many of the secreted proteins of the TTSS shared a common coiled-coil structure feature. orf29 and orf30 share no homology with known structure proteins of the TTSS that are believed to be conserved in different bacterial species. Therefore, it is likely that orf29 and orf30 do not encode for apparatus but instead are effector proteins of E. tarda. This possibility will be examined in a future study.
The EVP gene cluster is encoded outside of the TTSS and mutation in esrB has been reported to affect the expression of EvpA and EvpC (41). However, the secretion of EVP proteins is independent of the TTSS. Mutations in the putative translocon proteins of the TTSS did not affect the secretion of EvpC (41). The EsrA-EsrB two-component system regulates EVP via EsrC (Fig. 8). The LacZ reporter study suggests that EsrC regulated the transcription of the evpA-H transcriptional unit; only one promoter sequence was found (Fig. 1 and 6), and the regulation of EsrC on evpA is possibly direct (Table 4). The homologs of the EVP gene cluster are widely distributed in other pathogens such as Escherichia, Salmonella, Vibrio, and Yersinia species with unknown functions (41), and it is possible that these homologs may also be regulated by the regulators encoded in the respective TTSSs. The direct regulation of the EVP gene cluster by the regulators encoded within the TTSS of E. tarda suggests the functional relationship between the TTSS and the EVP gene cluster. The EVP gene cluster and its homologs encode several conserved core proteins and possibly encode a novel secretion system (2, 11, 41). This is the first report to suggest a communication between the TTSS and the EVP in E. tarda. In Salmonella species, SPI-5 encodes at least five effectors (48), among which SigD/SopB is coordinately regulated with SPI-1 genes (6) and is secreted by the SPI-1 TTSS (20). However, PipB, another effector in SPI-5, is part of the SPI-2 regulon and is translocated by the SPI-2 TTSS to the Salmonella-containing vacuole (23). These studies demonstrate a functional and regulatory cross talk between different secretion systems. E. tarda might use EsrC to coordinate the EVP secretion system with the TTSS in infection or survival in host cells.
Involvements of other regulators. The regulators encoded within the TTSS usually respond to regulation by environmental factors or other global regulators. In Salmonella species, the SsrA-SsrB two-component system was regulated by another two-component system OmpR-EnvZ, of which the response regulator OmpR binds directly to the ssrA and ssrB promoter (10, 24). In E. tarda, the transposon mutant 306, which has an insertion in a putative membrane protein, was also shown to affect the expression of the TTSS. Unlike OmpR-EnvZ in Salmonella species, this mutant did not affect the expressions of esrA or esrB but suppressed the expression of esrC and the TTSS apparatus genes (e.g., esaC) (Fig. 7). Interestingly, transposon insertion in pstC also has a similar effect on the expressions of TTSS and EVP (Fig. 7), suggesting that the PST operon has some control over the expressions of TTSS and EVP. The PST operon is required for phosphate transport (46). In S. enterica serovar Typhimurium, the transposon mutant pstS reduced the expression of the TTSS regulator of hilA and the invasion genes, and this repression was due to the negative control of the PhoR-PhoB two-component system. The pstS mutation led to the accumulation of PhoBP, and PhoBP directly or indirectly repressed hilA and the invasion genes (28). E. tarda might have a similar mechanism: the insertion inactivation of the PST system may lead to the accumulation of some negative regulators of the TTSS that have not been identified, thus repressing the expression of TTSS genes. If this is the case, this negative regulation would not modulate the expressions of esrA and esrB since both of their expressions were not affected in the pstC mutant background (Fig. 7). This negative regulator might directly interact with EsrB and interrupt its function. Another hypothesis is that some proteins encoded in the PST operon function as positive regulators and coordinate with EsrB in regulating the TTSS. However, we have shown that in E. coli, under the induction of IPTG, pQEesrB could activate lacZ expression of esrC-lacZ because of large amounts of EsrB accumulation. In E. tarda, as a regulator gene, the expression of esrB was low and EsrB could not activate the expression of the TTSS without the help of the positive regulator encoded in the PST operon. Future studies will attempt to distinguish between these possibilities.
Effect of high temperature. The expression of the esrB-lacZ fusion decreased substantially at 37°C compared to that seen at 25°C, but there was no change for the esrA-lacZ fusion at different temperatures. Our results suggest that the regulation of temperature on the TTSS is through the modulation of esrB expressions (Fig. 8). It is possible that the primary regulatory event is at the esrB transcription and that the accumulation of EsrB increases the transcription of esrC. This results in the increased production of EsrC activator proteins that bind to sites upstream of the target genes and activate transcription. If so, the mechanism response for temperature sensing is independent of EsrB and EsrC. EsrA functions as a sensor in EsrA-EsrB two-component system, and it is possible that EsrA is the one that responds to temperature changes. However, the expression of esrB-lacZ in an ersA mutant background did not decrease compared to that in the wild-type background at 25°C (Fig. 4). From these data, we hypothesize that there must be proteins other than EsrA responding to temperature changes. Alternatively, temperature changes may affect DNA conformation in the esrB promoter region. Changes in DNA topology have been shown to affect expressions of a variety of genes (7, 8, 14, 17). In E. tarda, the topology of the DNA promoter region of esrB may vary at different temperatures and result in altered esrB expressions. Further experiments must be done to confirm these two hypotheses.
In conclusion, we report that the EsrA-EsrB/EsrC regulatory cascade is the key regulon controlling the expressions of TTSS and EVP gene clusters and plays a vital role in E. tarda pathogenesis. EsrA-EsrB controls most of the transcriptional units of the TTSS to ensure that the TTSS is functional. EsrA-EsrB also controls EsrC, which regulates a subset of transcriptional units. They may be effector proteins or proteins for special functions, such as the transcriptional units of orf29 and orf30. Another main role of EsrC is to control the EVP, possibly another secretion system that is different from the TTSS. The EsrA-EsrB/EsrC regulatory cascade, therefore, can coordinate the expressions of these two secretion systems during infection and host-bacterium interactions. The sequencing of the TTSS and EVP gene clusters is ongoing. More transcriptional units will be identified, and this will help to dissect the precise roles of this EsrA-EsrB/EsrC regulatory cascade. Future work will also be carried out to identify and characterize the factors affected by the insertional disruptions of pstC and the putative membrane protein and to identify factors that communicate temperature changes with the TTSS and EVP gene clusters. This information will be useful in elucidating the functions of TTSS and EVP gene clusters and thus the pathogenesis of E. tarda.
ACKNOWLEDGMENTS
We are grateful to the Biomedical Research Council of Singapore (BMRC), ASTAR, the National University of Singapore, and the Lee Hiok Kwee Donation for providing research grants in support of this study.
We thank S. J. Busby of University of Birmingham for providing the plasmid pRW50 and its sequence information. We also thank I. Rosenshine and P. Tang for critical comments, which helped in improving the manuscript.
REFERENCES
1. Banks, A. S. 1992. A puncture wound complicated by infection with Edwardsiella tarda. J. Am. Pediatr. Med. Assoc. 82:529-531.
2. Bladergroen, M. R., K. Badelt, and H. P. Spaink. 2003. Infection-blocking genes of a symbiotic Rhizobium leguminosarum strain that are involved in temperature-dependent protein secretion. Mol. Plant-Microbe Interact. 16:53-64.
3. Cirillo, D. M., R. H. Valdivia, D. M. Monack, and S. Falkow. 1998. Macrophage-dependent induction of the Salmonella pathogenicity island 2 type III secretion system and its role in intracellular survival. Mol. Microbiol. 30:175-188.
4. Cornelis, G., C. Sluiters, C. L. de Rouvroit, and T. Michiels. 1989. Homology between VirF, the transcriptional activator of the Yersinia virulence regulon, and AraC, the Escherichia coli arabinose operon regulator. J. Bacteriol. 171:254-262.
5. Csaki, R., L. Bodrossy, J. Klem, J. C. Murrell, and K. L. Kovacs. 2003. Genes involved in the copper-dependent regulation of soluble methane monooxygenase of Methylococcus capsulatus (Bath): cloning, sequencing, and mutational analysis. Microbiology 149:1785-1795.
6. Darwin, K. H., and V. L. Miller. 2001. Type III secretion chaperone-dependent regulation: activation of virulence genes by SicA and InvF in Salmonella typhimurium. EMBO. J. 20:1850-1862.
7. Dorman, C. J. 1996. Flexible response: DNA supercoiling, transcription and bacterial adaptation to environmental stress. Trends Microbiol. 4:214-216.
8. Dorman, C. J., G. C. Barr, N. N. Bhriain, and C. F. Higgins. 1988. DNA supercoiling and the anaerobic and growth phase regulation of tonB gene expression. J. Bacteriol. 170:2816-2826.
9. Edwards, R. A., L. H. Keller, and D. M. Schifferli. 1998. Improved allelic exchange vectors and their use to analyze 987P fimbria gene expression. Gene 207:149-157.
10. Feng, X., R. Oropeza, and L. J. Kenney. 2003. Dual regulation by phospho-OmpR of ssrA/B gene expression in Salmonella pathogenicity island 2. Mol. Microbiol. 48:1131-1143.
11. Folkesson, A., S. Lofdahl, and S. Normark. 2002. The Salmonella enterica subspecies I specific centisome 7 genomic island encodes novel protein families present in bacteria living in close contact with eukaryotic cells. Res. Microbiol. 153:537-545.
12. Francis, M. S., H. Wolf-Watz, and A. Forsberg. 2002. Regulation of type III secretion systems. Curr. Opin. Microbiol. 5:166-172.
13. Frank, D. W., and B. H. Iglewski. 1991. Cloning and sequence analysis of a trans-regulatory locus required for exoenzyme S synthesis in Pseudomonas aeruginosa. J. Bacteriol. 173:6460-6468.
14. Galan, J. E., and R. Curtiss III. 1990. Expression of Salmonella typhimurium genes required for invasion is regulated by changes in DNA supercoiling. Infect. Immun. 58:1879-1885.
15. Gallegos, M. T., R. Schleif, A. Bairoch, K. Hofmann, and J. L. Ramos. 1997. AraC/XylS family of transcriptional regulators. Microbiol. Mol. Biol. Rev. 61:393-410.
16. Hensel, M., J. E. Shea, S. R. Waterman, R. Mundy, T. Nikolaus, G. Banks, A. Vazquez-Torres, C. Gleeson, F. C. Fang, and D. W. Holden. 1998. Genes encoding putative effector proteins of the type III secretion system of Salmonella pathogenicity island 2 are required for bacterial virulence and proliferation in macrophages. Mol. Microbiol. 30:163-174.
17. Higgins, C. F., C. J. Dorman, D. A. Stirling, L. Waddell, I. R. Booth, G. May, and E. Bremer. 1988. A physiological role for DNA supercoiling in the osmotic regulation of gene expression in Salmonella typhimurium and Escherichia coli. Cell 52:569-584.
18. Hirono, I., N. Tange, and T. Aoki. 1997. Iron regulated hemolysin gene from Edwardsiella tarda. Mol. Microbiol. 24:851-856.
19. Ho, S. N., H. D. Hunt, R. M. Horton, J. K. Pullen, and L. R. Pease. 1989. Site-directed mutagenesis by overlap extension using the polymerase chain reaction. Gene 77:51-59.
20. Hong, K. H., and V. L. Miller. 1998. Identification of a novel Salmonella invasion locus homologous to Shigella ipgDE. J. Bacteriol. 180:1793-1802.
21. Hueck, C. J. 1998. Type III protein secretion systems in bacterial pathogens of animals and plants. Microbiol. Mol. Biol. Rev. 62:379-433.
22. Janda, J. M., S. L. Abott, S. Kroshe-Bystrom, W. K. W. Cheng, C. Powers, R. P. Kokka, and K. Tamura. 1991. Pathogenic properties of Edwardsiella species. J. Clin. Microbiol. 29:1977-2001.
23. Knodler, L. A., J. Celli, W. D. Hardt, B. A. Vallance, C. Yip, and B. B. Finlay. 2002. Salmonella effectors within a single pathogenicity island are differentially expressed and translocated by separate type III secretion systems. Mol. Microbiol. 43:1089-1103.
24. Lee, A. K., C. S. Detweiler, and S. Falkow. 2000. OmpR regulates the two-component system SsrA-SsrB in Salmonella pathogenicity island 2. J. Bacteriol. 182:771-781.
25. Li, M., I. Rosenshine, S. L. Tung, X. H. Wang, D. Friedberg, C. L. Hew, and K. Y. Leung. 2004. Comparative proteomic analysis of extracellular proteins of enterohemorrhagic and enteropathogenic Escherichia coli strains and their ihf and ler mutants. Appl. Environ. Microbiol. 70:5274-5282.
26. Ling, S. H. M., X. H. Wang, L. Xie, T. M. Lim, and K. Y. Leung. 2000. Use of green fluorescent protein (GFP) to track the invasive pathways of Edwardsiella tarda in the in vivo and in vitro fish models. Microbiology 146:7-19.
27. Lodge, J., J. Fear, S. Busby, P. Gunasekaran, and N. R. Kamini. 1992. Broad host range plasmids carrying the Escherichia coli lactose and galactose operons. FEMS. Microbiol. Lett. 95:271-276.
28. Lucas, R. L., C. P. Lostroh, C. C. DiRusso, M. P. Spector, B. L. Wanner, and C. A. Lee. 2000. Multiple factors independently regulate hilA and invasion gene expression in Salmonella enterica serovar Typhimurium. J. Bacteriol. 182:1872-1882.
29. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
30. Ochman, H., F. C. Soncini, F. Solomon, and E. A. Groisman. 1996. Identification of a pathogenicity island required for Salmonella survival in host cells. Proc. Natl. Acad. Sci. USA 93:7800-7804.
31. Osiri, M., T. Tantawichien, and U. Deesomchock. 1997. Edwardsiella tarda bacteremia and septic arthritis in a patient with diabetes mellitus. Southeast Asian J. Trop. Med. Public Health 28:669-672.
32. Pallen, M. J., G. Dougan, and G. Frankel. 1997. Coiled-coil domains in proteins secreted by type III secretion systems. Mol. Microbiol. 25:423-425.
33. Plumb, J. A. 1993. Edwardsiella septicemia, p. 61-79. In V. Inglis, R. J. Roberts, and N. R. Bromage (ed.), Bacterial diseases of fish. Cambridge University Press, Cambridge, United Kingdom.
34. Reed, L. J., and H. Muench. 1938. A simple method of estimating fifty percent end points. Am. J. Hyg. 27:493-497.
35. Rubires, X., F. Saigí, N. Pique, N. Climent, S. Merino, S. Albertí, J. M. Tomas, and M. Regue. 1997. A gene (wbbL) from Serratia marcescens N28b (O4) complements the rfb-50 mutation of Escherichia coli K-12 derivatives. J. Bacteriol. 179:7581-7586.
36. Secombs, C. J. 1990. Isolation of salmonid macrophages and analysis of their killing activity, p. 137-154. In J. S. Stolen, T. C. Fletcher, D. P. Anderson, B. S. Roberson, and W. B. Van Muiswinkel (ed.), Techniques in fish immunology. SOS Publications, Fair Haven, N.J.
37. Slaven, E. M., F. A. Lopez, S. M. Hart, and C. V. Sanders. 2001. Myonecrosis caused by Edwardsiella tarda: a case report and case series of extraintestinal E. tarda infections. Clin. Infect. Dis. 32:1430-1433.
38. Srinivasa Rao, P. S., T. M. Lim, and K. Y. Leung. 2001. Opsonized virulent Edwardsiella tarda strains are able to adhere to and survive and replicate within fish phagocytes but fail to stimulate reactive oxygen intermediates. Infect. Immun. 69:5689-5697.
39. Srinivasa Rao, P. S., T. M. Lim, and K. Y. Leung. 2003. Functional genomics approach to the identification of virulence genes involved in Edwardsiella tarda pathogenesis. Infect. Immun. 71:1343-1351.
40. Srinivasa Rao, P. S., Y. Yamada, and K. Y. Leung. 2003. A major catalase (KatB) that is required for H2O2 and phagocyte-mediated killing in Edwardsiella tarda. Microbiology 149:2635-2644.
41. Srinivasa Rao, P. S., Y. Yamada, Y. P. Tan, and K. Y. Leung. 2004. Use of proteomics to identify novel virulence determinants that are required for Edwardsiella tarda pathogenesis. Mol. Microbiol. 53:573-586.
42. Tan, Y. P., Q. Lin, X. H. Wang, S. Joshi, C. L. Hew, and K. Y. Leung. 2002. Comparative proteomic analysis of extracellular proteins of Edwardsiella tarda. Infect. Immun. 70:6475-6480.
43. Thune, R. L., L. A. Stanley, and R. K. Cooper. 1993. Pathogenesis of gram-negative bacterial infections in warm water fish. Annu. Rev. Fish Dis. 3:37-68.
44. Tu, X., I. Nisan, C. Yona, E. Hanski, and I. Rosenshine. 2003. EspH, a new cytoskeleton-modulating effector of enterohaemorrhagic and enteropathogenic Escherichia coli. Mol. Microbiol. 47:595-606.
45. Ullah, M. A., and T. Arai. 1983. Pathological activities of the naturally occurring strains of Edwardsiella tarda. Fish Pathol. 18:65-70.
46. Webb, D. C., H. Rosenberg, and G. B. Cox. 1992. Mutational analysis of Escherichia coli phosphate-specific transport system, a member of traffic ATPase (or ABC) family of membrane transporters. J. Biol. Chem. 267:24661-24668.
47. Wipat, A., S. C. Brignell, B. J. Guy, M. Rose, P. T. Emmerson, and C. R. Harwood. 1998. The yvsA-yvqA (293 degrees-289 degrees) region of the Bacillus subtilis chromosome containing genes involved in metal ion uptake and a putative sigma factor. Microbiology 144:1593-1600.
48. Wood, M. W., M. A. Jones, P. R. Watson, S. Hedges, T. S. Wallis, and E. E. Galyov. 1998. Identification of a pathogenicity island required for Salmonella enteropathogenicity. Mol. Microbiol. 29:883-891.
49. Yang, C. H., and C. K. Wang. 1999. Edwardsiella tarda bacteraemia complicated by acute pancreatitis and pyomyoma. J. Infect. 38:124-126.(J. Zheng, S. L. Tung, and)