• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of jbacterPermissionsJournals.ASM.orgJournalJB ArticleJournal InfoAuthorsReviewers
J Bacteriol. Jul 2004; 186(13): 4199–4208.
PMCID: PMC421588

Identification of Inducers of the Yersinia enterocolitica Phage Shock Protein System and Comparison to the Regulation of the RpoE and Cpx Extracytoplasmic Stress Responses


Known inducers of the phage shock protein (Psp) system suggest that it is an extracytoplasmic stress response, as are the well-studied RpoE and Cpx systems. However, a random approach to identify conditions and proteins that induce the Psp system has not been attempted. It is also unknown whether the proteins or mutations that induce Psp are specific or if they also activate the RpoE and Cpx systems. This study addressed these issues for the Yersinia enterocolitica Psp system. Random transposon mutagenesis identified null mutations and overexpression mutations that increase Φ(pspA-lacZ) operon fusion expression. The results suggest that Psp may respond exclusively to extracytoplasmic stress. Null mutations affected glucosamine-6-phosphate synthetase (glmS), which plays a role in cell envelope biosynthesis, and the F0F1 ATPase (atp operon). The screen also revealed that in addition to several secretins, the overexpression of three novel putative inner membrane proteins (IMPs) induced the Psp response. We also compared induction of the Y. enterocolitica Psp, RpoE, and Cpx responses. Overexpression of secretins or the three IMPs or the presence of an atpB null mutation only induced the Psp response. Similarly, known inducers of the RpoE and Cpx responses did not significantly induce the Psp response. Only the glmS null mutation induced all three responses. Therefore, Psp is induced distinctly from the RpoE and Cpx systems. The specific IMP inducers may be valuable tools to probe specific signal transduction events of the Psp response in future studies.

Misfolding and/or mislocalization of envelope proteins induce extracytoplasmic stress responses in bacteria. The RpoE and Cpx systems of Escherichia coli and its relatives are well-studied examples (reviewed in references 41 and 42). These systems control many genes, with some overlap between their regulons, which encode proteases, envelope protein folding factors, and several proteins of unknown function (10, 41, 43). Mounting a response to extracytoplasmic stress is extremely important. rpoE is an essential gene in E. coli (14) and Yersinia enterocolitica (21), although this is apparently not the case in the closely related Salmonella genus (23).

The RpoE response is important for virulence in Salmonella enterica serovar Typhimurium (23, 45) and Vibrio cholerae (30). In addition, the RpoE and Cpx responses are induced by the overproduction of P pilus subunits from uropathogenic E. coli (25), and the Cpx system affects assembly and expression of the P pilus (24). The Cpx system is also important for the attachment of E. coli to surfaces (37), which is a critical step during biofilm formation.

The phage shock protein (Psp) system may be another example of an extracytoplasmic stress response. pspA operon expression, studied most extensively in E. coli K-12, is induced by the mislocalization of secretin proteins and by environmental conditions that induce the RpoE response (reviewed in reference 35). The precise inducing signal of the Psp response is unknown, as is its physiological role. However, the PspA protein helps maintain the proton motive force (PMF) when a mutant PhoE porin is overproduced in E. coli (29). Dissipation of the PMF could be a signal that induces the Psp response. Besides PspA, the other proteins encoded by the psp locus are only known to play roles in regulating expression of the pspA operon itself (35).

A working model for induction of the Psp response is derived from published studies of E. coli (reviewed in reference 35) and Y. enterocolitica (12). pspA promoter activity depends on the enhancer binding protein PspF (27). PspF activity is negatively controlled by an interaction with the peripheral inner membrane protein PspA (4, 17). The current hypothesis is that when the Psp response is active, the inner membrane proteins PspB and PspC detect an inducing signal and interact with PspA. This prevents PspA from interfering with PspF, and PspF-dependent genes are induced (the Psp response). Under noninducing conditions, PspA is free to interfere with PspF, and PspF-dependent genes are not expressed. This model awaits direct experimental investigation, although interactions among the PspA, PspB, and PspC proteins do occur (1).

The Psp response has been shown to play an important role in a virulent organism, the gastrointestinal pathogen Y. enterocolitica (11, 12). Different psp null mutations have various effects on the virulence of this organism, with some (e.g., pspC) causing complete attenuation (12). Y. enterocolitica pspA operon expression is induced when the Ysc type III secretion system is produced, and this is apparently due to mislocalization of the YscC secretin (12). Furthermore, YscC secretin mislocalization inhibits the growth of some psp null mutants.

Although published data suggest that the Psp system may only respond to extracytoplasmic stress, this hypothesis has not been addressed in an unbiased manner. Therefore, in this study we randomly identified genes that induce the Y. enterocolitica Psp response when they are overexpressed or disrupted. If the extracytoplasmic stress response hypothesis is correct, then the overexpressed genes would be expected to encode cell envelope components or proteins involved in controlling their expression or processing. Null mutations that induce the Psp response would be expected to affect the cell envelope or some of its protein components. A second aim of this study was to investigate whether inducers of the Psp response are specific or whether they also induce the RpoE and Cpx extracytoplasmic stress responses.


Bacterial strains, plasmids, and routine growth conditions.

The bacterial strains and plasmids used in this study are shown in Table Table1.1. For routine plasmid manipulations, the E. coli host strain was DH5α, CC118 λpir, S17-1 λpir, or SM10 λpir. E. coli strains were grown at 37°C, and Y. enterocolitica strains were grown at 26 or 37°C as noted. Strains were routinely grown in Luria-Bertani (LB) broth or on LB agar plates (33). The antibiotics used were ampicillin (200 μg ml−1), streptomycin (50 μg ml−1), spectinomycin (50 μg ml−1), nalidixic acid (20 μg ml−1), trimethoprim (100 μg ml−1), kanamycin (75 μg ml−1 for E. coli, 100 μg ml−1 for Y. enterocolitica), and chloramphenicol (25 μg ml−1 for E. coli, 12.5 μg ml−1 for Y. enterocolitica).

Strains and plasmids used in this study

Transposon construction, mutagenesis, and mutant characterization.

An ~1.5-kb fragment from plasmid pMAL-p2, encoding lacIq and tacp, was amplified by PCR with the lacI Up SpeI and tacp Dwn SpeI primers (Table (Table2).2). This fragment was cloned into the unique SpeI site of pTnMod-RKm′, and the desired orientation was confirmed by restriction digest analysis. The resulting plasmid was named pAJD428.

Primers used in this study

For transposon mutagenesis, pAJD428 was transferred from E. coli S17-1 λpir to Y. enterocolitica strain YVM576 by conjugation as described previously (11). Transposon insertion mutants with increased Φ(pspA-lacZ) expression were identified as described in Results.

A Southern blot assay with the kanamycin resistance gene of the transposon as a probe was done to ensure that each mutant contained a single transposon insertion and to determine the size of the EcoRI fragment containing the transposon and flanking DNA (data not shown). The conditional R6K ori of the transposon was used to recover transposon-chromosome junctions as replicating plasmids. Briefly, chromosomal DNA was digested with EcoRI and then treated with T4 DNA ligase. The ligation mixture was used to transform E. coli CC118 λpir to kanamycin resistance. Restriction digest analysis was done to confirm that each plasmid contained the transposon fragment predicted by the Southern blot assay. The DNA sequence of the transposon-chromosome junction was determined with primer tacp Tn P7.

β-Galactosidase assays.

Cultures were routinely grown in LB broth buffered with 100 mM 3-(N-morpholino)propanesulfonic acid (MOPS; pH 7). In some experiments, the medium was buffered with 100 mM 2-(N-morpholino)ethanesulfonic acid (MES; pH 5.5 or 6) or 100 mM N-tris(hydroxymethyl)methyl-3-aminopropanesulfonic acid (TAPS; pH 8 or 9), as indicated later in the text or figure legends. In most experiments, saturated cultures were diluted into 5 ml of medium in 18-mm-diameter test tubes, either with or without 1 mM IPTG. The initial optical density (600 nm) was approximately 0.05. The cultures were grown on a roller drum at 26 or 37°C until an optical density of ~0.6 was reached (approximately 4 to 8 h; mid-exponential phase). In some cases, cultures were grown for 24 h (late stationary phase). When ready, cultures were immediately chilled on ice and cells were collected by centrifugation.

Cultures used to determine the effect of overexpressing genes cloned into plasmid pVLT35 were grown differently (12) owing to toxic effects when some of the genes were present in multiple copies. Saturated cultures were diluted into 5 ml of medium in 18-mm-diameter test tubes. The initial optical density (600 nm) was approximately 0.05. The cultures were grown on a roller drum at 37°C for 2 h. A 0.2 mM final concentration of IPTG was then added to induce tacp promoter expression. Growth continued at 37°C for 2 h more prior to harvest.

β-Galactosidase enzyme activity was determined at room temperature (approximately 22°C) in permeabilized cells as described previously (31). Activities are expressed in arbitrary Miller units (33). Individual cultures were assayed in duplicate, and values were averaged from at least three independent cultures, the standard error of which was not more than 15%.

λ red recombinase mutagenesis.

The λ red recombinase system (13) was adapted for Y. enterocolitica, which is naturally ampicillin resistant. The λ red recombinase expression plasmid pKD46 (13) encodes ampicillin resistance and cannot be used. Therefore, a fragment encoding trimethoprim resistance was inserted into the unique PstI site of pKD46. The resulting plasmid, pAJD434, was transferred into Y. enterocolitica by electroporation.

The gene replacement system was used to insert a cassette encoding kanamycin resistance, lacIq, and an outward-facing tac promoter at specific chromosomal locations. A 2.9-kb kan lacIq tacp fragment was amplified from pAJD428 by PCR. Primer pairs (Table (Table2)2) were designed with 42 nucleotides at their 5′ ends that were homologous to regions flanking the desired chromosomal insertion site. The PCR product was introduced into Y. enterocolitica strain YVM576, containing pAJD434, by electroporation. Mutant colonies were isolated on LB agar containing kanamycin. The strains were cured of plasmid pAJD434, and the insertion was confirmed by Southern hybridization analysis (data not shown).

The λ red recombinase system was also used to construct ΔglmS and ΔatpB in-frame deletions. The kanamycin resistance gene of pKD13 was amplified by PCR with primers that included 42 nucleotides homologous to the 5′ or 3′ end of glmS or atpB (Table (Table2).2). The PCR product was introduced into Y. enterocolitica strain JB580v, containing pAJD434, by electroporation. To remove the chromosomally inserted kanamycin resistance gene and leave an in-frame “scar,” plasmid pLH29, encoding IPTG-inducible FLP recombinase, was introduced into each mutant by electroporation. Following FLP-mediated excision of the kan gene, the strains were cured of pLH29 and checked by Southern blot analysis (data not shown). The deleted region was amplified by PCR, and the sequence of the product was determined to confirm an in-frame deletion.

Construction of plasmid pVLT33/35 derivatives.

Genes were amplified from chromosomal DNA by PCR with primer pairs that annealed immediately upstream and downstream of each gene and incorporated unique restriction sites, and in some cases a C-terminal six-His tag (Table (Table2).2). The fragments were cloned into plasmid pVLT33 or pVLT35, and the DNA sequence of each fragment was confirmed.

Construction of Φ(cpxP-lacZ) and Φ(rpoE-lacZ) operon fusion strains.

cpxR′-cpxP′ and nadC′-rpoE′ fragments were amplified from chromosomal DNA by PCR and cloned into the pFUSE derivative pKN8 (Table (Table1).1). The sequence of each fragment was checked. These fusions were integrated onto the Y. enterocolitica chromosome by homologous recombination. Following integration, the merodiploid strains encode intact cpxP and rpoE genes under the control of their native promoters, in addition to the Φ(cpxP-lacZ) and Φ(rpoE-lacZ) operon fusions. Correct integration was checked by Southern hybridization analysis (data not shown).

Western blot assays.

Total cell proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on gels containing 12.5% polyacrylamide. Proteins were transferred to nitrocellulose by electroblotting. Chemiluminescence detection followed sequential incubation with penta-His monoclonal primary antibody (QIAGEN) used at a 1-in-2,000 dilution and then goat anti-mouse immunoglobulin G-horseradish peroxidase conjugate (Bio-Rad) used at a 1-in-5,000 dilution.


Screen for induction of the Psp response.

To identify null mutations and overexpressed genes that induce Φ(pspA-lacZ) expression, we constructed a transposon encoding the E. coli lac repressor, an outward-facing tac promoter, and a conditional origin of replication. This transposon causes null mutations by insertional activation and/or IPTG-dependent overexpression of downstream genes. A similar strategy was pioneered with a Tn5 derivative that lacked the conditional replication origin (7). Transposon insertion mutants of a Φ(pspA-lacZ) operon fusion strain were identified after growth in the presence of IPTG at either 26 or 37°C on MacConkey-lactose or LB-5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside (X-Gal) indicator agar. Mutants with increased Φ(pspA-lacZ) expression were identified as red (MacConkey) or dark blue (X-Gal) colonies, which was confirmed by β-galactosidase enzyme assay. The screen was done with a virulence plasmid-free strain to avoid induction of Φ(pspA-lacZ) expression by the Ysc type III secretion system (12).

We anticipated that many of the insertions would be in the psp locus. The transposon tac promoter could be inserted upstream of the Φ(pspA-lacZ) operon fusion. Alternatively, increased expression of some psp genes, or inactivation of pspA, increases Φ(pspA-lacZ) expression (12). Therefore, we used multiplex PCR to amplify the psp locus from each transposon mutant as three separate fragments. Alteration of the sizes of these fragments indicated disruption of the psp locus, and these mutants were eliminated from further analysis (data not shown).

Approximately 100,000 random TnMod-RKm′-lacIqtacp insertion mutants were screened on indicator plates. One hundred sixty-two mutants had increased Φ(pspA-lacZ) activity in β-galactosidase enzyme assays. Multiplex PCR analysis indicated that 54 of these did not have insertions within the psp locus, and their transposon-chromosome junctions were isolated as plasmids. In a few cases, the plasmids did not have the expected restriction patterns or the cloning failed. For the remaining mutants, the DNA sequence adjacent to the transposon insertion was determined and used to locate the insertion sites in the complete Y. enterocolitica genome sequence.

Overexpression of different secretins induces Φ(pspA-lacZ) expression.

Most mutants had an IPTG-dependent increase in Φ(pspA-lacZ) expression, indicating that tacp-dependent overexpression of a gene(s) downstream of the transposon was responsible. These insertions were mapped to seven loci (Fig. (Fig.1).1). During preliminary characterization, we noticed that in some cases the level of Φ(pspA-lacZ) expression was significantly affected by the growth temperature. We do not understand the reason for this, but β-galactosidase assay data for one mutant for each locus, grown at 26 or 37°C, are presented in Table Table33.

FIG. 1.
Overexpression mutations that induce the Psp response. A flag shows the approximate location of each transposon insertion, and the arrow shows the orientation of the tac promoter. Closed flags indicate the mutants used to determine the data shown in Table ...
TnMod-RKm′-lacIq tacp mutants with IPTG-dependent induction of Φ(pspA-lacZ) expression

The virulence plasmid-encoded YscC secretin was previously shown to induce Φ(pspA-lacZ) expression (12). However, it was a relatively weak inducer (fivefold), even when it was overexpressed from a multicopy plasmid. Considering that the inducers identified here were overexpressed from the chromosome, many of them were more efficient inducers than YscC. This was confirmed when the new inducers were overexpressed from multicopy plasmids (see below). Even so, the overexpression inducers varied in the ability to increase Φ(pspA-lacZ) expression from 2- to 94-fold (Table (Table33).

Four of the loci had a secretin gene downstream of the transposon insertion (Fig. (Fig.1).1). It is reasonable to assume that overexpression of the secretin gene is responsible for the phenotype. The virulence plasmid YscC secretin induces Φ(pspA-lacZ) expression in Y. enterocolitica (12), and several secretins induce PspA synthesis in E. coli (35).

Three Psp response overexpression inducers are putative inner membrane proteins.

Three of the loci identified in mutants with an IPTG-dependent increase in Φ(pspA-lacZ) expression do not encode secretins (the lower three loci in Fig. Fig.1).1). In each case, there were multiple genes downstream of the transposon tacp promoter.

A combination of two procedures was used to identify the downstream gene responsible for inducing Φ(pspA-lacZ) expression. The λ red recombinase system was used to insert a lacIqtacp cassette immediately upstream of specific downstream genes (see Materials and Methods). Alternatively, individual or multiple downstream genes were cloned into tacp expression plasmid pVLT33. This analysis revealed that for each of the three loci a single gene was responsible for inducing the Psp response. These were the ampE, yggT, and YPO0432 ortholog genes (Fig. (Fig.1).1). Each of these genes is predicted to encode a conserved inner membrane protein of unknown function.

Psp system overexpression inducers do not induce other extracytoplasmic stress responses.

Many previous Psp studies did not determine whether the inducing proteins, mutations, or conditions also induced other extracytoplasmic stress responses such as the RpoE and Cpx systems. Therefore, we compared activation of the Y. enterocolitica Psp, RpoE, and Cpx systems in response to overexpression of some of the genes identified in the screen.

Y. enterocolitica strains were constructed with single-copy Φ(rpoE-lacZ) and Φ(cpxP-lacZ) operon fusions to monitor induction of the RpoE and Cpx responses, respectively. The RpoE and Cpx responses of E. coli are specifically induced by overexpression of the ompX and cutF (nlpE) genes, respectively (32, 44). Therefore, the Y. enterocolitica orthologs of these genes were used as positive controls. The ysaC, yggT, ampE, YPO0432 ortholog, ompX, and cutF genes were cloned into the tac promoter expression plasmid pVLT35. We then tested their effects on Φ(pspA-lacZ), Φ(cpxP-lacZ), and Φ(rpoE-lacZ) expression.

Each reporter fusion behaved as expected. The best inducers of Φ(pspA-lacZ) expression were ysaC, yggT, ampE, and the YPO0432 ortholog, which induced expression by approximately 30-fold to more than 100-fold (Fig. (Fig.2).2). This was higher induction than that which occurred in the original transposon insertion mutants (Table (Table3)3) because the genes were expressed from multicopy plasmids. CutF was the most efficient inducer of Φ(cpxP-lacZ) expression (more than 40-fold), and OmpX induced Φ(rpoE-lacZ) expression by approximately 3-fold. The fold induction of Φ(rpoE-lacZ) expression by ompX was small but reproducible. This is probably due at least in part to the fact that the basal level of expression from this fusion was high (approximately 2,000 Miller units; data not shown). Overexpression of ompX also has relatively small effects on some RpoE-dependent promoters in E. coli (e.g., see references 8 and 32). Therefore, although the sensitivity may be limited, the Φ(rpoE-lacZ) fusion is a suitable reporter for induction of the RpoE system.

FIG. 2.
Overexpression inducers are specific for the Psp, Cpx, and RpoE stress responses. Plasmid pVLT35 derivatives encoding the indicated genes expressed from the tac promoter were transferred into Φ(pspA-lacZ), Φ(cpxP-lacZ), and Φ( ...

The four Psp inducers did not affect Φ(rpoE-lacZ) expression (Fig. (Fig.2).2). Similarly, yggT and the YPO0432 ortholog did not induce Φ(cpxP-lacZ) expression. With IPTG, the tacp-ysaC plasmid induced Φ(pspA-lacZ) expression by 114-fold but also had a minor inducing effect on Φ(cpxP-lacZ) expression (7-fold). However, in the absence of IPTG the tacp-ysaC plasmid still induced Φ(pspA-lacZ) expression by 46-fold but no longer induced Φ(cpxP-lacZ) expression (Fig. (Fig.2).2). With 0.2 mM IPTG, the tacp-ampE plasmid induced Φ(pspA-lacZ) expression by 31-fold but also induced Φ(cpxP-lacZ) expression by 6-fold. However, reducing the IPTG concentration to 0.01 mM did not significantly reduce the fold induction of Φ(pspA-lacZ) expression (approximately 26-fold; data not shown) but reduced induction of Φ(cpxP-lacZ) expression by 50% (from sixfold to threefold; data not shown).

The RpoE inducer OmpX had no effect on the Psp response (Fig. (Fig.2).2). There was some induction of Φ(pspA-lacZ) expression by the Cpx inducer CutF when 0.2 mM IPTG was used for induction (Fig. (Fig.2).2). However, with 0.04 mM IPTG the tacp-cutF plasmid no longer induced Φ(pspA-lacZ) expression but still induced Φ(cpxP-lacZ) expression by 44-fold (data not shown).

Taken together, these results indicate that there is no overlap between these proteinaceous inducers of the Psp response and those of the RpoE and Cpx responses. Each system appears to be exquisitely sensitive to the production of specific cell envelope proteins.

Detection of overexpressed proteins.

We considered the possibility that the inner membrane protein inducers indirectly affect the outer membrane protein profile. We were also interested to know if the overexpression plasmids led to massive protein overproduction. Therefore, strains containing the pVLT35 vector or the ysaC, ampE, yggT, or YPO0432 ortholog overexpression plasmids were grown as for the β-galactosidase assay experiments in Fig. Fig.2.2. Cells were separated into cytoplasm, periplasm, inner membrane, and outer membrane fractions. None of the proteins were overexpressed sufficiently to allow their identification by Coomassie brilliant blue-stained SDS-PAGE (data not shown). Furthermore, none of the overexpression plasmids had a global effect on protein profiles in any subcellular fraction (data not shown).

The inability to detect Psp-inducing proteins by Coomassie brilliant blue-stained SDS-PAGE made it important to confirm that the overexpression plasmids increased the levels of these proteins. Therefore, we constructed expression plasmids carrying ysaC, ampE, yggT, or the YPO0432 ortholog with six-His tags at their C termini. These plasmids were still able to induce Φ(pspA-lacZ) expression (data not shown). Strains containing these plasmids were grown as for the β-galactosidase assay experiments in Fig. Fig.2,2, and total cell proteins were separated by SDS-PAGE and detected following Western blotting with anti-penta-His antibody (Fig. (Fig.3).3). Each of the proteins was detected in the presence of IPTG. Overexposure of the film also revealed a low level of each protein in the absence of IPTG (data not shown).

FIG. 3.
Detection of Psp response-inducing proteins. Strains containing plasmid pVLT35 (vector) or derivatives encoding the indicated genes were grown as described in the Fig. Fig.22 legend. Total cell proteins were separated by SDS-12.5% PAGE, transferred ...

Only secretins specifically inhibit the growth of a pspC null mutant.

In addition to inducing Φ(pspA-lacZ) expression, overexpression of the YscC secretin inhibits the growth of a pspC null mutant (12). Therefore, secretin overexpression may cause a stress that cannot be tolerated in the pspC null mutant. The novel overexpression inducers identified here may cause the same stress as secretins, or they may induce by a different mechanism.

To begin to investigate this, we determined the effect of overexpressing the ysaC secretin, yggT, ampE, and the YPO0432 ortholog on the growth of pspC+ and ΔpspC mutant strains as described previously (12). Expression of the ysaC secretin gene completely inhibited growth of the ΔpspC mutant without significantly affecting the pspC+ strain (data not shown). However, none of the other expression plasmids specifically inhibited the growth of the ΔpspC mutant (data not shown). These experiments were done with various concentrations of IPTG to control the expression level of the inducers, but the conclusion was the same (data not shown).

These experiments suggest that overexpression of either ampE, yggT, or the YPO0432 ortholog does not cause a stress that specifically affects a pspC null mutant, despite the fact that they induce the Psp response. This may indicate that the mechanism by which they induce Φ(pspA-lacZ) expression is different from that of secretin proteins such as YsaC and YscC.

Disruption of the F0F1 ATPase specifically induces the Psp response during growth at alkaline pH.

A minority of the mutants identified in the screen demonstrated IPTG-independent induction of Φ(pspA-lacZ) expression (data not shown). Most of these mutants had insertions in the atpA or atpB gene (Fig. (Fig.4),4), which encode components of the F0F1 proton-translocating ATPase that interconverts PMF and ATP.

FIG. 4.
Null mutations that induce the Psp response. A flag represents the approximate location of each transposon insertion, and the arrow shows the orientation of the tac promoter. Open reading frames are shown as horizontal arrows in the direction of their ...

The pH of unbuffered LB increases during growth of Y. enterocolitica. This causes an approximately twofold increase in Φ(pspA-lacZ) expression (data not shown). The β-galactosidase assays described so far were done with LB medium buffered to pH 7. In preliminary experiments with unbuffered medium, we noticed that the atp null mutations increased Φ(pspA-lacZ) expression significantly. However, they did not do so if the medium was buffered to pH 7 (the pH did not significantly affect other inducers identified in this screen). Therefore, we investigated the effect of the external pH on Φ(pspA-lacZ) expression in atp+ and atp null strains.

The atpA and atpB insertion mutations significantly increased Φ(pspA-lacZ) expression at pH 8. After 8 h (mid-exponential phase), Φ(pspA-lacZ) expression was induced approximately five- and eightfold in the atpA and atpB mutants, respectively (data not shown). Induction was much higher in late stationary phase (24 h; 10-fold for the atpA mutant and 22-fold for the atpB mutant; data not shown). At pH 7, the atpA and atpB null mutations had a less-than-twofold effect on Φ(pspA-lacZ) expression (data not shown).

We also investigated whether an atp null mutation induced the RpoE and Cpx responses. An atpB in-frame deletion mutation was combined with either the Φ(pspA-lacZ), the Φ(cpxP-lacZ), or the Φ(rpoE-lacZ) fusion. The strains were grown at 26°C into mid-exponential phase (8 h) or late stationary phase (24 h) at pH 7 or 8, and β-galactosidase activity was determined (Fig. (Fig.5).5). The ΔatpB mutation had an effect on Φ(pspA-lacZ) expression similar to that of the original transposon insertion mutation (Fig. (Fig.55 and data not shown). In contrast, expression of Φ(cpxP-lacZ) and Φ(rpoE-lacZ) was indistinguishable in atpB+ and ΔatpB strains. However, we observed that Φ(cpxP-lacZ) expression was significantly induced at pH 8 compared to pH 7 (Fig. (Fig.5).5). An inducing effect of elevated pH has also been reported for the cpxP gene of E. coli (9), which offers a further indication that the Cpx responses of Y. enterocolitica and E. coli are regulated similarly. Although alkaline pH alone was sufficient to induce Φ(cpxP-lacZ) expression, induction of Φ(pspA-lacZ) required both alkaline pH and an atp null mutation. This suggests different mechanisms underlying the effects of alkaline pH on the Psp and Cpx responses.

FIG. 5.
Effects of an atpB null mutation and alkaline pH on the Psp, Cpx, and RpoE systems. Φ(pspA-lacZ), Φ(cpxP-lacZ), and Φ(rpoE-lacZ) operon fusion strains YVM576, AJD243, and AJD242, respectively (WT [wild type]), or their Δ ...

We also determined the effect of an expanded range of pH values on the ability of an atpB null mutation to induce the Psp response. The results showed that an atpB mutation did not significantly induce Φ(pspA-lacZ) expression at pH 5.5, 6, or 7 but did induce it at pHs 8 and 9 (data not shown). There was no significant difference in the fold induction at pHs 8 and 9, although at pH 9 there was a toxic effect on the growth of atpB+ and atpB null strains. These experiments confirm that an atpB null mutation only induces the Psp response during growth at alkaline pH.

A glmS null mutation causes general induction of extracytoplasmic stress responses including Psp.

One of the IPTG-independent mutants did not grow on MacConkey agar. It grew poorly on LB agar and reached a low cell density in liquid culture before undergoing an apparent lysis event (not shown). The transposon had inserted into a homolog of the E. coli glmS gene (Fig. (Fig.4),4), which encodes glucosamine-6-phosphate synthetase. This mutation caused an approximately 14-fold induction of Φ(pspA-lacZ) expression (data not shown). GlmS converts fructose 6-phosphate, produced by glycolysis, into glucosamine 6-phosphate. Glucosamine 6-phosphate is a precursor for the biosynthesis of several cell envelope components (lipid A, O-antigen, enterobacterial common antigen, and peptidoglycan; 40). Therefore, a cell wall biosynthesis defect of the glmS mutant may be responsible for inducing Φ(pspA-lacZ) expression.

The growth defect of an E. coli glmS mutant is relieved by addition of d-glucosamine to the growth medium (47). Furthermore, a cell wall biosynthesis defect might be expected to induce other extracytoplasmic stress responses, in addition to Psp. Therefore, we constructed an in-frame deletion mutation of glmS and determined its effect on Φ(pspA-lacZ), Φ(cpxP-lacZ), and Φ(rpoE-lacZ) expression in the presence or absence of d-glucosamine.

The ΔglmS mutation induced Φ(pspA-lacZ) expression by approximately 20-fold in the absence of d-glucosamine but did not cause any induction in the presence of d-glucosamine (data not shown). In the absence of d-glucosamine, the ΔglmS mutation also induced Φ(cpxP-lacZ) and Φ(rpoE-lacZ) expression by 29- and 4-fold, respectively (data not shown). Once again, the induction did not occur in the presence of d-glucosamine.

These results demonstrate that the glmS null mutation causes general induction of extracytoplasmic stress responses, rather than being a specific inducer of the Psp system. It is likely that major disruption of the cell envelope has many effects on the cell that trigger the precise inducing signals of multiple stress response systems.

Conservation of inducing signals between the Y. enterocolitica and E. coli Psp responses.

To rule out strain-specific factors, we determined whether the specific Psp inducers identified here also induce the E. coli Psp response. The tacp expression plasmids were introduced into E. coli strain MC3, which has a single-copy Φ(pspA-lacZ) operon fusion (3). The strains were grown at 37°C in the presence of 0.2 mM IPTG, and β-galactosidase activities were determined. The ysaC, ampE, yggT, and YPO0432 ortholog expression plasmids induced E. coli Φ(pspA-lacZ) expression by 75-, 36-, 51-, and 21-fold, respectively (data not shown).

The Y. enterocolitica and E. coli PspA proteins are 78% identical (12). Therefore, we also used a Y. enterocolitica PspA polyclonal antiserum to detect Y. enterocolitica and E. coli PspA proteins in atpB+ and atpB null strains (Table (Table1)1) grown for 24 h at pH 7 or 8. In both Y. enterocolitica and E. coli, the PspA protein was most abundant in atpB null mutants grown at pH 8 (data not shown).


Observations, primarily from E. coli studies, suggest that the Psp system responds to stress that occurs in the cell envelope (reviewed in reference 35). Like the well-studied RpoE and Cpx systems, Psp appears to consist of inner membrane proteins (PspB and PspC) that communicate the presence of an extracytoplasmic stress to cytoplasmic proteins (PspA and PspF). Here we addressed the questions of whether Psp is likely to be exclusively a response to extracytoplasmic stress and if there is overlap among induction of the Psp, RpoE, and Cpx responses.

The Psp system responds to extracytoplasmic stress.

The screen identified seven protein inducers of the Y. enterocolitica Psp response (Table (Table3).3). Four are predicted to be secretins, which adds to the YscC secretin that was known to induce Φ(pspA-lacZ) expression (12). Primary sequence analysis of the other three inducers predicts that they are likely to be associated with the inner membrane (data not shown). This would be a novel finding because all previously known inducers of the Psp response are outer membrane proteins (secretins and mutant PhoE and LamB porins of E. coli). The identification of secretins and putative inner membrane protein inducers, coupled with the failure to identify any cytoplasmic inducers, strongly suggests that the Psp system responds exclusively to extracytoplasmic stress.

We also isolated one mutation that is predicted to affect the integrity of the cell envelope, an insertion in the glmS gene. It was not surprising to find that a glmS null mutation also induces the Cpx and RpoE extracytoplasmic stress responses. The glmS mutation serves to support the characterization of Psp as an extracytoplasmic stress response, but its future analysis is unlikely to tell us anything about the precise nature of the inducing signal.

atp null mutations induce the Psp response.

We isolated several insertions in the atpA and atpB genes, which encode subunits of the F0F1 ATPase. AtpA is subunit α of the F1 sector, and AtpB is subunit a of the proton-translocating F0 sector (19). This reversible enzyme complex interchanges the two energy currencies of the cell, ATP and the PMF. In E. coli, expression of a localization-defective PhoE protein induces the Psp response and also reduces the PMF in a pspA null strain (29). The authors of that study hypothesized that one function of PspA is to help maintain the PMF under stress conditions and that PMF dissipation is the inducing signal. As our study was under way, it was shown that depletion of the E. coli YidC protein induces the Psp response (46) and this was also confirmed in a later study (26). YidC depletion leads to instability of the F0F1 ATPase and cytochrome o oxidase and causes depletion of the PMF (46). However, YidC depletion has pleiotropic effects on many E. coli membrane proteins (6), rather than specifically affecting the F0F1 ATPase. In contrast, our data show a specific link between loss of the F0F1 ATPase and Psp induction.

atp null mutations only induced Φ(pspA-lacZ) expression in alkaline growth media (Fig. (Fig.5).5). The PMF is made up of the sum of the pH gradient across the membrane (ΔpH) and the membrane potential (ΔΨ; negative charge inside, positive outside). The E. coli cytoplasmic pH is maintained at 7.6 to 7.8 over a wide range of external pH values (38, 39). Something similar probably occurs in Y. enterocolitica. Therefore, with neutral or acidic medium there would always be a positive contribution of ΔpH to the PMF. In basic growth medium, the ΔpH would have a negative contribution to the PMF, which might make the role of the F0F1 ATPase important to prevent induction of the Psp response. However, we stress that we do not yet have any data that link the atp null mutations to changes in the PMF under any growth conditions.

Overexpression inducers of the Psp response.

The screen identified seven genes that induce the Psp response when they are overexpressed. Four encode outer membrane secretin proteins, including the Yts1D and YsaC secretins, which were the most potent inducers (Table (Table3).3). The other three are all predicted to encode inner membrane proteins, which is a class of proteins not previously implicated in inducing psp gene expression (except for the inner membrane pspC gene). All seven of these overexpression inducers were specific for the Psp response. However, we do not understand the mechanism(s) by which they cause induction. This is confounded by the fact that the three inner membrane proteins have no known function in any bacterial species. However, the Vibrio alginolyticus YggT homolog was at least implicated in the transport of K+ ions and has sequence similarity to other K+ transporters and the E. coli Na+/H+ antiporter NhaB (36). Perhaps overexpression of Y. enterocolitica YggT causes aberrant ion transport, which could adversely affect the membrane potential.

One possible unifying hypothesis for all of the overexpression inducers is that their overexpression depletes the PMF. It seems that secretin mislocalization causes severe stress in Y. enterocolitica because it inhibits the growth of a pspC null mutant (data not shown and reference 12). However, overexpression of ampE, yggT, and the YPO0432 ortholog did not specifically affect the growth of a pspC null mutant (data not shown). Perhaps there is a fundamental difference in how secretins and the putative inner membrane proteins (IMPs) induce the Psp response.


This study provides support for the hypothesis that the Psp system is likely to respond exclusively to extracytoplasmic stress and that there is little overlap between induction of the Psp, RpoE, and Cpx responses. The demonstration that atp null mutations induce the Psp response is consistent with a role for PMF in induction but does not prove it. The novel IMP Psp inducers should be valuable tools in future studies to investigate Psp signal transduction events. A major goal of future work will be to address the role of these inducers on maintenance of the PMF and/or in transducing a signal to the Psp system.


We thank Virginia Miller, Michael Cox, Friederike Turnowsky, and the E. coli Genetic Stock Center for providing strains and plasmids. We are grateful to Heran Darwin, Virginia Miller, and Valley Stewart for critical review of the manuscript. Y. enterocolitica genome sequence data were produced by the Y. enterocolitica Sequencing Group at the Sanger Institute and can be obtained from http://www.sanger.ac.uk/Projects/Y_enterocolitica/.

This study was supported by Public Health Service grant AI-052148 from the National Institute of Allergy and Infectious Diseases and by a grant from the Speaker's Fund for Biomedical Research: toward the Science of Patient Care, awarded by the City of New York.


1. Adams, H., W. Teertstra, J. Demmers, R. Boesten, and J. Tommassen. 2003. Interactions between phage-shock proteins in Escherichia coli. J. Bacteriol. 185:1174-1180. [PMC free article] [PubMed]
2. Bäumler, A. J., R. M. Tsolis, A. W. M. van der Velden, I. Stojiljkovic, S. Anic, and F. Heffron. 1996. Identification of a new iron regulated locus of Salmonella typhi. Gene 183:207-213. [PubMed]
3. Bergler, H., D. Abraham, H. Aschauer, and F. Turnowsky. 1994. Inhibition of lipid biosynthesis induces the expression of the pspA gene. Microbiology 140:1937-1944. [PubMed]
4. Bordes, P., S. R. Wigneshweraraj, J. Schumacher, X. Zhang, M. Chaney, and M. Buck. 2003. The ATP hydrolyzing transcription activator phage shock protein F of Escherichia coli: identifying a surface that binds sigma 54. Proc. Natl. Acad. Sci. USA 100:2278-2783. [PMC free article] [PubMed]
5. Butlin, J. D., G. B. Cox, and F. Gibson. 1971. Oxidative phosphorylation in Escherichia coli K12: mutations affecting magnesium ion- or calcium ion-stimulated adenosine triphosphatase. Biochem. J. 124:75-81. [PMC free article] [PubMed]
6. Chen, M., K. Xie, F. Jiang, L. Yi, and R. E. Dalbey. 2002. YidC, a newly defined evolutionarily conserved protein, mediates membrane protein assembly in bacteria. Biol. Chem. 383:1565-1572. [PubMed]
7. Chow, W. Y., and D. E. Berg. 1988. Tn5tac1, a derivative of transposon Tn5 that generates conditional mutations. Proc. Natl. Acad. Sci. USA 85:6468-6472. [PMC free article] [PubMed]
8. Danese, P. N., and T. H. Silhavy. 1997. The σE and the Cpx signal transduction systems control the synthesis of periplasmic protein-folding enzymes in Escherichia coli. Genes Dev. 11:1183-1193. [PubMed]
9. Danese, P. N., and T. J. Silhavy. 1998. CpxP, a stress-combative member of the Cpx regulon. J. Bacteriol. 180:831-839. [PMC free article] [PubMed]
10. Dartigalongue, C., D. Missiakas, and S. Raina. 2001. Characterization of the Escherichia coli sigma E regulon. J. Biol. Chem. 276:20866-20875. [PubMed]
11. Darwin, A. J., and V. L. Miller. 1999. Identification of Yersinia enterocolitica genes affecting survival in an animal host using signature-tagged transposon mutagenesis. Mol. Microbiol. 32:51-62. [PubMed]
12. Darwin, A. J., and V. L. Miller. 2001. The psp locus of Yersinia enterocolitica is required for virulence and for growth in vitro when the Ysc type III secretion system is produced. Mol. Microbiol. 39:429-444. [PubMed]
13. Datsenko, K. A., and B. L. Wanner. 2000. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci. USA 97:6640-6645. [PMC free article] [PubMed]
14. De Las Penas, A., L. Connolly, and C. A. Gross. 1997. σE is an essential sigma factor in Escherichia coli. J. Bacteriol. 179:6862-6864. [PMC free article] [PubMed]
15. de Lorenzo, V., L. Eltis, B. Kessler, and K. N. Timmis. 1993. Analysis of Pseudomonas gene products using lacIq/Ptrp-lac plasmids and transposons that confer conditional phenotypes. Gene 123:17-24. [PubMed]
16. Dennis, J. J., and G. J. Zylstra. 1998. Plasposons: modular self-cloning minitransposon derivatives for rapid genetic analysis of gram-negative bacterial genomes. Appl. Environ. Microbiol. 64:2710-2715. [PMC free article] [PubMed]
17. Dworkin, J., G. Jovanovic, and P. Model. 2000. The PspA protein of Escherichia coli is a negative regulator of σ54-dependent transcription. J. Bacteriol. 182:311-319. [PMC free article] [PubMed]
18. Ellison, D. W., B. Young, K. Nelson, and V. L. Miller. 2003. YmoA negatively regulates expression of invasin from Yersinia enterocolitica. J. Bacteriol. 185:7153-7159. [PMC free article] [PubMed]
19. Harold, F. M., and P. C. Maloney. 1996. Energy transduction by ion currents, p. 283-306. In F. C. Neidhardt, R. Curtis III, J. L. Ingraham, E. C. C. Lin, K. Brooks Low, B. Magasanik, W. S. Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella: cellular and molecular biology. ASM Press, Washington, D.C.
20. Herrero, M., V. de Lorenzo, and K. N. Timmis. 1990. Transposon vectors containing non-antibiotic resistance selection markers for cloning and stable chromosomal insertion of foreign genes in gram-negative bacteria. J. Bacteriol. 172:6557-6567. [PMC free article] [PubMed]
21. Heusipp, G., M. A. Schmidt, and V. L. Miller. 2003. Identification of rpoE and nadB as host responsive elements of Yersinia enterocolitica. FEMS Microbiol. Lett. 226:291-298. [PubMed]
22. Huang, L. 1997. Convenient and reversible site-specific targeting of exogenous DNA into a bacterial chromosome by use of the FLP recombinase: the FLIRT system. J. Bacteriol. 179:6076-6083. [PMC free article] [PubMed]
23. Humphreys, S., A. Stevenson, A. Bacon, A. B. Weinhardt, and M. Roberts. 1999. The alternative sigma factor, σE, is critically important for the virulence of Salmonella typhimurium. Infect. Immun. 67:1560-1568. [PMC free article] [PubMed]
24. Hung, D. L., T. L. Raivio, C. H. Jones, T. J. Silhavy, and S. J. Hultgren. 2001. Cpx signaling pathway monitors biogenesis and affects assembly and expression of P pili. EMBO J. 20:1508-1518. [PMC free article] [PubMed]
25. Jones, C. H., P. N. Danese, J. S. Pinkner, T. J. Silhavy, and S. J. Hultgren. 1997. The chaperone-assisted membrane release and folding pathway is sensed by two signal transduction systems. EMBO J. 16:6394-6406. [PMC free article] [PubMed]
26. Jones, S. E., L. J. Lloyd, K. K. Tan, and M. Buck. 2003. Secretion defects that activate the phage shock response of Escherichia coli. J. Bacteriol. 185:6707-6711. [PMC free article] [PubMed]
27. Jovanovic, G., L. Weiner, and P. Model. 1996. Identification, nucleotide sequence, and characterization of PspF, the transcriptional activator of the Escherichia coli stress-induced psp operon. J. Bacteriol. 178:1936-1945. [PMC free article] [PubMed]
28. Kinder, S. A., J. L. Badger, G. O. Bryant, J. C. Pepe, and V. L. Miller. 1993. Cloning of the YenI restriction endonuclease and methyltransferase from Yersinia enterocolitica serotype O:8 and construction of a transformable RM+ mutant. Gene 136:271-275. [PubMed]
29. Kleerebezem, M., W. Crielaard, and J. Tommassen. 1996. Involvement of stress protein PspA (phage shock protein A) of Escherichia coli in maintenance of the protonmotive force under stress conditions. EMBO J. 15:162-171. [PMC free article] [PubMed]
30. Kovacikova, G., and K. Skorupski. 2002. The alternative sigma factor σE plays an important role in intestinal survival and virulence in Vibrio cholerae. Infect. Immun. 70:5355-5362. [PMC free article] [PubMed]
31. Maloy, S. R., V. J. Stewart, and R. K. Taylor. 1996. Genetic analysis of pathogenic bacteria. Cold Spring Harbor Laboratory Press, Plainview, N.Y.
32. Mecsas, J., P. E. Rouviere, J. E. Erickson, T. J. Donohue, and C. A. Gross. 1993. The activity of σE, an Escherichia coli heat-inducible σ-factor, is modulated by expression of outer membrane proteins. Genes Dev. 7:2618-2628. [PubMed]
33. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
34. Miller, V. L., and J. J. Mekalanos. 1988. A novel suicide vector and its use in construction of insertion mutations: osmoregulation of outer membrane proteins and virulence determinants in Vibrio cholerae requires toxR. J. Bacteriol. 170:2575-2583. [PMC free article] [PubMed]
35. Model, P., G. Jovanovic, and J. Dworkin. 1997. The Escherichia coli phage-shock-protein (psp) operon. Mol. Microbiol. 24:255-261. [PubMed]
36. Nakamura, T., Y. Katoh, Y. Shimizu, Y. Matsuba, and T. Unemoto. 1996. Cloning and sequencing of novel genes from Vibrio alginolyticus that support the growth of K+ uptake-deficient mutant of Escherichia coli. Biochim. Biophys. Acta 1277:201-208. [PubMed]
37. Otto, K., and T. J. Silhavy. 2002. Surface sensing and adhesion of Escherichia coli controlled by the Cpx-signaling pathway. Proc. Natl. Acad. Sci. USA 99:2287-2292. [PMC free article] [PubMed]
38. Padan, E., and S. Schuldiner. 1987. Intracellular pH and membrane potential as regulators in the prokaryotic cell. J. Membr. Biol. 95:189-198. [PubMed]
39. Padan, E., D. Zilberstein, and H. Rottenberg. 1976. The proton electrochemical gradient in Escherichia coli cells. Eur. J. Biochem. 63:533-541. [PubMed]
40. Raetz, C. R. H. 1996. Bacterial lipopolysaccharides: a remarkable family of bioactive macroamphiphiles, p. 1035-1063. In F. C. Neidhardt, R. Curtis III, J. L. Ingraham, E. C. C. Lin, K. Brooks Low, B. Magasanik, W. S. Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella: cellular and molecular biology. ASM Press, Washington, D.C.
41. Raivio, T. L., and T. J. Silhavy. 2001. Periplasmic stress and ECF sigma factors. Annu. Rev. Microbiol. 55:591-624. [PubMed]
42. Raivio, T. L., and T. J. Silhavy. 1999. The σE and Cpx regulatory pathways: overlapping but distinct envelope stress responses. Curr. Opin. Microbiol. 2:159-165. [PubMed]
43. Rezuchova, B., H. Miticka, D. Homerova, M. Roberts, and J. Kormanec. 2003. New members of the Escherichia coli σE regulon identified by a two-plasmid system. FEMS Microbiol. Lett. 225:1-7. [PubMed]
44. Snyder, W. B., L. J. B. Davis, P. N. Danese, C. L. Cosma, and T. J. Silhavy. 1995. Overproduction of NlpE, a new outer membrane lipoprotein, suppresses the toxicity of periplasmic LacZ by activation of the Cpx signal transduction pathway. J. Bacteriol. 177:4216-4223. [PMC free article] [PubMed]
45. Testerman, T. L., A. Vazquez-Torres, Y. Xu, J. Jones-Carson, S. J. Libby, and F. C. Fang. 2002. The alternative sigma factor σE controls antioxidant defences required for Salmonella virulence and stationary-phase survival. Mol. Microbiol. 43:771-782. [PubMed]
46. van der Laan, M., M. L. Urbanus, C. M. Ten Hagen-Jongman, N. Nouwen, B. Oudega, N. Harms, A. J. Driessen, and J. Luirink. 2003. A conserved function of YidC in the biogenesis of respiratory chain complexes. Proc. Natl. Acad. Sci. USA 100:5801-5806. [PMC free article] [PubMed]
47. Vogler, A. P., S. Trentmann, and J. W. Lengeler. 1989. Alternative route for biosynthesis of amino sugars in Escherichia coli K-12 mutants by means of a catabolic isomerase. J. Bacteriol. 171:6586-6592. [PMC free article] [PubMed]

Articles from Journal of Bacteriology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...