SseA is a chaperone for the SseB and SseD translocon components of the Salmonella pathogenicity-island-2-encoded type III secretion system

Javier Ruiz-Albert, Rosanna Mundy, Xiu-Jun Yu, Carmen R. Beuzón{dagger} and David W. Holden

Department of Infectious Diseases, Centre for Molecular Microbiology and Infection, Imperial College School of Medicine, Armstrong Road, London SW7 2AZ, UK

Correspondence
David W. Holden
d.holden{at}ic.ac.uk


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
NOTE ADDED IN PROOF
REFERENCES
 
The type III secretion system (TTSS) encoded by the Salmonella pathogenicity island 2 (SPI-2) is required for bacterial replication inside macrophages and for systemic infection in mice. Many TTSS secreted proteins, including effectors and components of the translocon, require chaperones which promote their stability, prevent their premature interactions or facilitate their secretion. In this study, the function of the first gene (sseA) of one of the SPI-2 operons (sseAG) was investigated. This operon includes genes that encode translocon components (SseB, SseC and SseD), translocated proteins (SseF and SseG) and putative chaperones (SscA and SscB). sseA encodes a 12·5 kDa protein with a C-terminal region with the potential to form a coiled-coil structure, but no sequence similarity to other proteins. Mutation of sseA results in severe virulence attenuation and an intracellular replication defect. It is shown here that SseA is not a secreted protein, but is required for SPI-2-dependent translocation of two effector proteins (SifA and PipB). Furthermore, the translocon components SseB and SseD were not detected in an sseA mutant strain. By using a yeast two-hybrid assay and column binding experiments, it is demonstrated that SseA interacts directly with SseB and SseD. These results indicate that SseA is a chaperone for SseB and SseD. The inability of an sseA mutant to assemble the SPI-2 TTSS translocon accounts for its high level of virulence attenuation in vivo. To the authors' knowledge, this is the first chaperone described for the SPI-2 TTSS.


Abbreviations: CI, competitive index; HA, haemagglutinin; Sif, Salmonella-induced filament; SPI-2, Salmonella pathogenicity island 2; S. typhimurium, Salmonella enterica serovar Typhimurium; TTSS, type III secretion system

{dagger}Present address: Departamento de Biología Celular, Genética y Fisología, Facultad de Ciencias, Universidad de Málaga, Campus Teatinos, 29071, Málaga, Spain.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
NOTE ADDED IN PROOF
REFERENCES
 
Type III secretion systems (TTSSs) are specialized protein delivery systems found in many Gram-negative pathogenic bacteria. They are responsible for the secretion and translocation of effector proteins from bacteria into the host cell cytosol, where they subvert a variety of different cellular processes. Type III secretion typically involves a secreton for exporting proteins from the bacterial cell, a translocon for transferring effector proteins into host cells, various regulators that control gene transcription and protein secretion, and chaperones (Hueck, 1998). Chaperones have been identified for both effector proteins and translocon components. They are necessary to prevent degradation and promote secretion of effector proteins, and to prevent premature degradation of, and interaction between, translocon components. Generally, chaperones of the effectors specifically interact with one substrate, while chaperones of the translocators associate with two substrates (Page & Parsot, 2002). The genes encoding these chaperones are often located in the vicinity of the genes encoding their substrates (Page & Parsot, 2002). Chaperones of the TTSSs have common physiochemical characteristics, including low molecular masses (<15 kDa) and predicted amphipathic helices near their C termini (Page & Parsot, 2002).

Salmonella enterica serovar Typhimurium (S. typhimurium) has two TTSSs, encoded by Salmonella pathogenicity islands 1 and 2 (SPI-1 and SPI-2, respectively). The SPI-1 TTSS is required for cell invasion, and it induces significant intestinal secretory and inflammatory responses (Galan, 2001; Wallis & Galyov, 2000). The SPI-2 TTSS is required for intracellular replication and systemic infection (Cirillo et al., 1998; Hensel et al., 1995, 1998; Ochman et al., 1996; Shea et al., 1996).

On the basis of similarities of sequence and organization with genes of other bacterial pathogens, several genes located in a 9 kb region within SPI-2 (operon sseA–G) were initially proposed to encode secreted proteins (Hensel et al., 1998). SseB, SseC and SseD are encoded by this operon and have been shown to be secreted in vitro onto the bacterial surface, where they remain associated, and to be necessary for translocation of effector proteins into the host cell (Beuzón et al., 1999; Klein & Jones, 2001; Nikolaus et al., 2001). Strains carrying mutations in sseB, sseC or sseD are as attenuated as those that are completely defective for secretion, and display the same intracellular phenotypes (Cirillo et al., 1998; Hensel et al., 1998; Klein & Jones, 2001; Yu et al., 2002). SscA and SscB have been proposed to be chaperones on the basis of their similarities to the SycD and IpgC chaperones from Yersinia spp. and Shigella flexneri, respectively (Cirillo et al., 1998; Hensel et al., 1998). Mutation of sseE does not lead to a noticeable attenuation of virulence (Hensel et al., 1998); SseF and SseG have been shown to be translocated into the host cell (Hansen-Wester et al., 2002; Kuhle & Hensel, 2002). Strains carrying mutations in sseF or sseG are mildly attenuated and partially defective in the formation of tubular membranous structures known as Salmonella-induced filaments (Sifs), which are induced upon SPI-2-mediated translocation of the effector protein SifA into HeLa cells (Beuzón et al., 2000; Brumell et al., 2002; Garcia-del Portillo et al., 1993; Guy et al., 2000; Kuhle & Hensel, 2002; Stein et al., 1996). The first gene of the operon encodes SseA, a small protein that contains a domain with potential to form a coiled-coil structure, but which has no significant sequence similarity to other proteins (Hensel et al., 1998). An sseA mutant strain is strongly attenuated in virulence and severely defective in intracellular replication (Hensel et al., 1998).

In this study, we have investigated the function of SseA. Unlike other proteins encoded within the same SPI-2 operon, SseA does not appear to be secreted, but is required for SPI-2-dependent translocation of two different effector proteins (SifA and PipB). Furthermore, the SPI-2 TTSS translocon components SseB and SseD were not detected in an sseA mutant strain, although the intracellular levels of SseC remained unaffected. We show that SseA binds to SseB and SseD. These results, together with the predicted physiochemical properties of SseA, indicate that this protein is a chaperone for SseB and SseD. Therefore, the inability of an sseA mutant to correctly assemble the SPI-2 TTSS translocon explains its high level of virulence attenuation in vivo.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
NOTE ADDED IN PROOF
REFERENCES
 
Bacterial strains and growth conditions.
The S. typhimurium strains used in this study are listed in Table 1. Bacteria were grown in Luria–Bertani (LB) medium supplemented with carbenicillin (100 µg ml-1), or chloramphenicol (35 µg ml-1), as appropriate. To induce SPI-2 gene expression and SPI-2-dependent secretion, bacteria were grown in magnesium minimal medium MES (MgM/MES), containing 170 mM 2-(N-morpholino)ethanesulphonic acid at pH 5·0, 5 mM KCl, 7·5 mM (NH4)2SO4, 0·5 mM K2SO4, 1 mM KH2PO4, 8 µM MgCl2, 38 mM glycerol and 0·1 % Casamino acids (Beuzón et al., 1999) with the corresponding antibiotics when appropriate. Bacteria were grown at 37 °C overnight with aeration.


View this table:
[in this window]
[in a new window]
 
Table 1. S. typhimurium strains used in this study

 
Plasmids.
The complementing plasmid psseA (Hensel et al., 1998) is a derivative of pACYC184 (Chang & Cohen, 1978) carrying the sseA gene under the control of a constitutive promoter.

The complementing plasmid psseBD is a derivative of the complementing plasmid psseB (Hensel et al., 1998), carrying the sseB and sseD genes under the control of a constitutive promoter. A DNA fragment including the complete ORF of sseD with its corresponding ribosome-binding site was amplified by PCR from 12023 genomic DNA using primers SSED-F (5'-ATAGCATGCATAGTAGCGTTCAGGCGAG-3') and SSED-R (5'-ATAGTCGACTTACCTCGTTAATGCCCGG-3'). The 640 bp PCR product, containing terminal SphI and SalI sites, was digested and ligated into the corresponding sites of psseB, generating psseBD.

Plasmid pACB C-2HA (Knodler et al., 2002), a derivative of pACYC184 (Chang & Cohen, 1978) encoding an haemagglutinin (HA)-tagged version of PipB, was a kind gift from L. Knodler (Rocky Mountain Laboratories, Hamilton, MT, USA). Plasmid pFVP25.1, carrying gfpmut3A under the control of the rpsM constitutive promoter (Valdivia & Falkow, 1996), was introduced into bacterial strains and used for fluorescence visualization where indicated.

Preparation of cell fractions.
Bacterial cell densities were determined by measurement of the OD600 value. To ensure that protein from equal numbers of cells was analysed, in all experiments protein samples were adjusted to OD600 values such that a volume corresponding to 10 ml of a culture of OD600 0·6 was taken up in 100 µl of protein denaturing buffer for gel electrophoresis. Cell cultures were cooled on ice, centrifuged at 35 000 g for 5 min at 4 °C, and the proteins present in the supernatant were collected by trichloroacetic acid precipitation (culture supernatant). After removal of the culture medium, the cell pellets were washed with ice-cold PBS and resuspended in 0·3 ml of PBS. The suspensions were mixed gently with 0·2 ml n-hexadecane for 5 min at room temperature and centrifuged at 10 000 g for 10 min at room temperature (Michiels et al., 1990). The organic layer was discarded, the aqueous layer mixed with 1·2 ml acetone and held at -20 °C to form a precipitate, and the cell pellet was stored at -70 °C before analysis (bacterial pellet). The acetone precipitate was centrifuged, dried and stored at -70 °C before analysis (hexadecane fraction). All fractions were dissolved in protein denaturing buffer before PAGE.

PAGE and Western analysis of proteins.
Protein fractions were dissolved in the appropriate volume of protein denaturing buffer (Beuzón et al., 1999) and held at 100 °C for 5 min. Proteins were immediately separated on 12 % SDS-polyacrylamide gels (Laemmli, 1970). For Western analysis, gels were transferred to Immobilon-P membranes (Millipore) and examined using the ECL detection system under conditions recommended by the manufacturer (Amersham Life Science). Incubation of membranes with primary antibodies (rabbit anti-SseA, anti-SseB, anti-SseC, anti-SseD) was followed by incubation with horseradish peroxidase-conjugated anti-rabbit as secondary antibody.

Recombinant SseA protein purification and generation of anti-SseA polyclonal antibody.
For the expression of recombinant SseA, a DNA fragment including the complete ORF of sseA minus the start codon was amplified by PCR from 12023 genomic DNA, using primers SSEA-T72 (5'-GGATCCATGATAAAGAAAAAGCTGCGT-3') and SSEA-T73 (5'-GGATCCTTACCTTTTTGTTTTTCCTGACGG-3'). The 350 bp PCR product, containing terminal BamHI sites, was digested and ligated into vector pET-15b (Novagen), generating pID834. This plasmid expresses a full-length version of SseA bearing an N-terminal hexahistidine-tag (His6SseA), under the control of a T7 promoter. Plasmid pID834 was introduced into Escherichia coli strain BL21 (DE3), and SseA was expressed following standard protocols (Sambrook & Russell, 2001). For antibody production, the 14·3 kDa recombinant protein was purified under denaturing conditions by metal-chelating chromatography using the QIAexpress Protein Purification System according to the instructions of the manufacturer (Qiagen) and used as an antigen for rabbit immunization (Abcam). Antibody titres were checked by Western blot using total cell fractions of E. coli strain BL21 (DE3) expressing SseA.

Affinity purification of the anti-SseA antibody.
For affinity purification of the anti-SseA antiserum, purified His6SseA (20 mg ml-1) was coupled to water-activated Affi-Gel 10 (Bio-Rad) in the presence of 0·1 M MOPS, pH 7·5, and 0·3 M NaCl. The slurry was then centrifuged at 400 g for 5 min at 4 °C. After removing the supernatant, the remaining reactive esters were blocked by addition of Tris/HCl to a final concentration of 50 mM at pH 7·2. It was then washed twice with 0·1 M MOPS, pH 7·5/0·3 M NaCl, and twice with elution buffer (0·1 M glycine, pH 2·5). The washed slurry was neutralized by succesive washings with 50 mM Tris/HCl (50 mM), pH 7·2, until neutral pH was reached. Anti-SseA serum (1·8 ml) and 0·2 ml of 10x TBS (120 mM Tris/HCl, pH 8·0/1·5 M NaCl) were added to the resin and incubated overnight with gentle shaking at 4 °C. The slurry was then poured into an Econo-Column chromatography column support (Bio-Rad), and both the flow-through and the washes with 10 ml of 1x TBS were collected. The purified antibody was eluted with 0·1 M glycine/HCl, pH 2·5, and neutralized by addition of 1 M Tris/HCl, pH 8·0. The affinity-purified rabbit polyclonal anti-SseA antibody was used for Western analysis at a dilution of 1 : 5000.

Column binding assay.
The ability of Ni2+–NTA column-immobilized His6SseA to bind SseB was determined following the method described by Elliott et al. (1999). Briefly, His6SseA was expressed from pID834 following standard procedures (Sambrook & Russell, 2001). The bacterial pellet was resuspended in buffer containing 300 mM NaCl and 50 mM NaH2PO4 (pH 8·0) (column buffer) supplemented with 10 mM imidazole, and lysed by sonication. The lysate was cleared by centrifugation and incubated at 4 °C for 1 h with 1 ml of Ni2+–NTA beads (Pierce) with gentle mixing. The suspension was then poured into a column, the flow-through was collected and re-applied twice more, followed by three washes with 4 ml of column buffer supplemented with 20 mM imidazole. Wild-type Salmonella-secreted proteins corresponding to the bacterial surface protein fraction (hexadecane fraction, as described above) were concentrated from 100 ml of culture grown in minimal medium at low pH, resuspended in column buffer supplemented with 20 mM imidazole, and passed three times through the packed column. The column was then washed three times with 4 ml of column buffer supplemented with 20 mM imidazole. Bound proteins, including the excess of His6SseA bound to the column and any other proteins interacting with it, were eluted in 0·5 ml aliquots with column buffer supplemented with 250 mM imidazole. The resulting samples were examined by Western blotting.

Yeast two-hybrid system.
The yeast two-hybrid host used in this study was PJ69-4A (MATa trp1-901 leu2-3112 ura3-52 his3-200 gal4D gal80D LYS2 : : GAL1–HIS3 GAL2–ADE1 met2 : : GAL7–lacZ), which has three independent reporter genes under the control of three different GAL promoters (James et al., 1996). A DNA fragment encoding SseA was amplified by PCR from wild-type S. typhimurium strain 12023 genomic DNA using the primers SSEAY-F (5'-ATCGAATTCATAAAGAAAAAGGCTGCG-3') and SSEAY-R (5'-ATCCTCGAGTTACCTTTTTGTTTTTCC-3'). The PCR product, containing terminal EcoRI and XhoI sites, was digested and ligated into the EcoRI and SalI sites of the ADH-1-driven fusion vector pGBT9. Likewise, a DNA fragment encoding SseD was amplified by PCR from 12023 genomic DNA using the primers SSEDY-F (5'-ATCGAATTCGAAGCGAGTAACGTAGCAC-3') and SSEDY-R (5'-ATCCTCGAGTTACCTCGTTAATGCCCG-3'), and ligated into the vector pGAD424. Constructs were co-transformed into PJ69-4A, and analysis of the transformants by growth of yeast on minimal medium to monitor both HIS3 and ADE1 reporter gene activation was performed as described previously (Hartland et al., 2000). To measure the induction of the lacZ reporter gene, {beta}-galactosidase assays of yeast strains grown in liquid culture were performed as described by Francis et al. (2000). The values shown in Results and Discussion for the {beta}-galactosidase assays represent the mean±SD (Miller units) corresponding to three measurements for each of three independent cultures assayed. The host strain PJ69-4A was co-transformed with a pGAD424 empty vector and the pGBT vector expressing the SseA fusion protein, and the resulting transformants were used as a negative control for both assays. The yeast two-hybrid host strain and the fusion vectors were kindly provided by G. Frankel (Centre for Molecular Microbiology and Infection, London, UK).

Bacterial infection of HeLa cells.
HeLa (clone HtTA1) cells were kindly provided by S. Méresse (Centre d'Immunologie de Marseille-Luminy, Marseille, France). Cells were grown in Dulbecco's modified Eagle medium (DMEM) supplemented with 10 % fetal calf serum (FCS) and 2 mM glutamine at 37 °C in 5 % CO2. HeLa cells were seeded onto glass coverslips (12 mm diameter) in 24-well plates at a density of 5x105 cells per well, 24 h before infection. Bacteria were grown in LB broth for 16 h at 37 °C with shaking, diluted 1 : 33 in fresh LB broth and grown in the same conditions for 3·5 h. The cultures were diluted in Earle's buffered salt solution, pH 7·4, and added to the cells at an m.o.i. of approximately 100 : 1. The infection was allowed to proceed for 15 min at 37 °C in 5 % CO2. The monolayers were washed once with DMEM containing FCS and 100 µg gentamicin ml-1 and incubated in this medium for 1 h, after which the gentamicin concentration was decreased to 16 µg ml-1. Cell monolayers were fixed at 8 h after invasion, in 3·7 % paraformaldehyde in PBS, pH 7·4, for 15 min at room temperature and then washed three times in PBS.

Antibodies and reagents.
The rabbit polyclonal anti-SseA (this work), anti-SseB (Beuzón et al., 1999), anti-SseC and anti-SseD (Nikolaus et al., 2001) antibodies were used at a dilution of 1 : 10 000. The mouse mAb anti-LAMP-1 H4A3 developed by J. T. August and J. E. K. Hildreth was obtained from the Developmental Studies Hybridoma Bank, developed under the auspices of the NICHD and maintained by the University of Iowa (Department of Biological Sciences, Iowa, IA, USA), and was used at a dilution of 1 : 2000. Anti-Salmonella goat polyclonal antibody CSA-1 was purchased from Kirkegaard and Perry Laboratories (Gaithensburg, MD) and was used at a dilution of 1 : 400. The mouse mAb anti-HA 16B12 was purchased from Covance (Berkeley Antibody Company) and was used at a dilution of 1 : 1000. Texas red sulfonyl chloride-conjugated donkey anti-mouse and FITC-conjugated donkey anti-goat antibodies were purchased from Jackson Immunoresearch Laboratories and were used at a dilution of 1 : 400. Horseradish peroxidase-conjugated anti-rabbit antibody was purchased from Amersham Life Sciences and was used at a dilution of 1 : 20 000.

Virulence tests.
Female BALB/c mice (20–25 g) were inoculated intraperitoneally with a 0·2 ml volume of physiological saline containing 105 bacteria. Bacteria were grown overnight at 37 °C in LB medium with aeration (150 r.p.m.), diluted into fresh medium (1 : 100) and grown until an OD550 value of between 0·35 and 0·6 was reached. Cultures were then diluted in physiological saline to a concentration of 2·5x105 bacteria per strain per ml, and thoroughly mixed before the infection (input). The number of colony-forming units (c.f.u.) of each strain in the input was enumerated by plating a dilution series of the inoculum and using the appropriate antibiotic to discriminate between the strains. Mice were killed 48 h after inoculation; their spleens were then removed, placed in sterile water and homogenized by mechanical disruption. Bacteria were pelleted by centrifugation at 13 000 g and resuspended in sterile water. Bacterial c.f.u. were enumerated by plating a dilution series onto LB agar and LB agar with the corresponding antibiotic. The competitive index (CI) was calculated as the ratio between the mutant and wild-type strain within the output (bacteria recovered from the host after infection), divided by their ratio within the input (Freter et al., 1981; Taylor et al., 1987).


   RESULTS AND DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
NOTE ADDED IN PROOF
REFERENCES
 
SseA is not secreted in SPI-2 in vitro inducing conditions
In an effort to analyse SseA expression and localization, a recombinant hexahistidine-tagged SseA protein was expressed in E. coli, purified and used to raise an anti-SseA antibody. The resulting antibody, when affinity-purified, was capable of detecting SseA expressed from the chromosome in the wild-type Salmonella strain 12023. The presence or absence of SseA was then examined by immunoblot analysis both in the wild-type and sseA mutant strains, in different culture fractions, after growth in MgM/MES at low pH, conditions that induce SPI-2-dependent secretion (Beuzón et al., 1999; Hansen-Wester et al., 2002; Klein & Jones, 2001). Each culture was fractionated into three samples corresponding to (a) the bacterial pellet, containing all non-secreted proteins, (b) the bacterial surface protein fraction (hexadecane fraction), where several SPI-2-secreted proteins are found, including the translocon components SseB, SseC and SseD (Beuzón et al., 1999; Hansen-Wester et al., 2002; Nikolaus et al., 2001), and (c) culture supernatant, where translocated effectors are also found (Hansen-Wester et al., 2002).

After growth at low pH to induce SPI-2-mediated secretion, SseA was detected in the bacterial pellets of the wild-type strain and the sseA mutant strain carrying a plasmid expressing SseA constitutively. However, it was not detected in the bacterial surface protein fraction or in the culture supernatant of either strain (Fig. 1a). In the same conditions, SseB was found on the bacterial surface of the wild-type strain, as expected (Fig. 1b). These results indicate that SseA is not a SPI-2 secreted protein.



View larger version (38K):
[in this window]
[in a new window]
 
Fig. 1. (a, b) Subcellular localization of SseA in SPI-2-inducing conditions. S. typhimurium wild-type (12023), sseA mutant ({Delta}sseA) or sseA mutant carrying the complementing plasmid ({Delta}sseA psseA) were cultured in MgM/MES at pH 5·0, and pellet (P), bacterial cell surface (H) and culture supernatant (S) fractions were analysed by immunoblotting with the anti-SseA (a) or anti-SseB (b) antibodies. (c) S. typhimurium bacteria carrying a mutation in sseA are completely deficient for the formation of Sifs, but remain enclosed within a vacuole. HeLa cells were infected with wild-type (12023), sseA mutant ({Delta}sseA) strains or the sseA mutant strain carrying the complementing plasmid ({Delta}sseA psseA). Cells were fixed 8 h after bacterial invasion and examined by confocal fluorescence microscopy. Sifs (marked with asterisks) and vacuolar membranes (marked with arrows) were labelled with an anti-LAMP-1 antibody (red), while bacteria expressed green fluorescent protein constitutively (green). Bar, 5 µm.

 
SseA is required for Sif formation and PipB translocation
To determine whether SseA is required for translocation of SPI-2 effectors, we analysed by confocal immunofluorescence microscopy the intracellular behaviour of the sseA mutant strain after infection of HeLa epithelial cells. As detected by LAMP-1 labelling, the sseA mutant strain remained within vacuoles, but was completely deficient for the formation of Sifs. The ability to produce Sifs was restored in an sseA mutant strain harbouring a plasmid expressing SseA constitutively (Fig. 1c). These results suggested that the sseA mutant might be defective for SPI-2-mediated translocation of effectors.

We therefore examined the requirement for SseA in the translocation of an HA-tagged version of the SPI-2 effector protein PipB (Knodler et al., 2002). This protein has been shown to be translocated specifically through the SPI-2 TTSS and to be localized in the vacuolar membrane and in Sifs (Knodler et al., 2002). HeLa cells were infected with wild-type, ssaV mutant (a SPI-2 null mutant; Beuzón et al., 1999) or sseA mutant strains, each carrying a plasmid expressing the HA-tagged version of PipB, and analysed by confocal immunofluorescence microscopy using an anti-HA antibody. Translocated PipB was detected in Sifs and on the membrane of the Salmonella-containing vacuole when expressed in the wild-type strain, but not when expressed in either the ssaV or sseA mutant strains (Fig. 2). These results indicate that SseA is required for efficient SPI-2-dependent translocation of effectors.



View larger version (40K):
[in this window]
[in a new window]
 
Fig. 2. SseA is required for efficient SPI-2-dependent translocation of PipB. HeLa cells were infected with wild-type (12023), sseA mutant ({Delta}sseA) or SPI-2 null mutant ({Delta}ssaV) strains, all carrying plasmid pACB C-2HA expressing an HA-tagged version of PipB. Cells were fixed 8 h after bacterial invasion and examined by confocal fluorescence microscopy. HA-tagged PipB was detected with an anti-HA antibody (red), while bacteria were detected with an anti-Salmonella antibody (green). HA-tagged PipB was detected in Sifs and on the vacuolar membrane only when expressed in wild-type bacteria. Bar, 5 µm.

 
Effect of the sseA mutation on intracellular levels of SseB, SseC and SseD
SseA has no sequence similarity with other TTSS components, but is relatively small (12·5 kDa) and has a C-terminal region predicted to form a coiled-coil structure (Hensel et al., 1998). These are all characteristics of TTSS chaperones (Page & Parsot, 2002). For these reasons, we investigated a possible role of SseA as a chaperone for the translocon components SseB, SseC and SseD, which are encoded immediately downstream of sseA within the same operon (Cirillo et al., 1998).

The intracellular levels of SseB, SseC and SseD were examined by immunoblot analysis using specific polyclonal antibodies, in the wild-type and the sseA mutant strains. As expected, the three proteins were readily detected in the wild-type strain (Fig. 3a, b, c). By contrast, SseB and SseD were not detected in the sseA mutant strain, while the levels of SseC remained unaffected (Fig. 3a, b, c). The presence of intracellular SseB and SseD was partially restored in an sseA mutant strain carrying a plasmid constitutively expressing SseA (psseA), and both proteins were found at levels similar to those of the wild-type when the sseA mutant strain carried a plasmid expressing SseB and SseD simultaneously (psseBD) (Fig. 3a, b).



View larger version (41K):
[in this window]
[in a new window]
 
Fig. 3. SseB and SseD are not detected intracellularly in an S. typhimurium sseA mutant strain after growth in SPI-2-inducing conditions. The wild-type strain (12023), sseA mutant strain ({Delta}sseA), sseA mutant strain carrying either the psseA complementing plasmid ({Delta}sseA psseA) or the psseBD complementing plasmid ({Delta}sseA psseBD), and sseB and sseD mutant strains carrying the psseBD complementing plasmid ({Delta}sseB psseBD and {Delta}sseD psseBD, respectively) were cultured in MgM/MES, pH 5·0, and the pellet (P) and hexadecane (H) fractions were analysed by immunoblotting. (a) Immunoblots probed with anti-SseB antibody. (b) Immunoblots probed with anti-SseD antibody. (c) Immunoblots probed with anti-SseC antibody.

 
However, SseB and SseD were not detected by immunoblot analysis of the corresponding bacterial surface protein fractions (hexadecane fraction) in the sseA mutant strain carrying either psseA or psseBD (Fig. 3a, b). In contrast, SseB and SseD were readily detected on the hexadecane fraction of the sseB and sseD single mutant strains carrying psseBD (Fig. 3a, b).

The effect of a mutation in sseA on the levels of SseB and SseD suggested that SseA might act as a chaperone for these two translocon components. The lack of effect on the intracellular levels of SseC rules out a polar effect of the sseA mutation since all these genes form part of the same transcriptional unit (Cirillo et al., 1998). The lack of effect of the sseA mutation on SseC was not unexpected, as the sscA gene, located immediately upstream of sseC, is predicted to encode a chaperone similar to that (SycD) which interacts with the SseC homologue (YopB) in Yersinia spp. (Neyt & Cornelis, 1999).

SseA interacts with SseB in a column binding assay
To determine if an interaction occurs between SseA and SseB, we examined the ability of His6SseA immobilized on an Ni2+–NTA column to bind SseB. When a hexadecane-extracted preparation of secreted proteins (see Methods) was applied to a column containing immobilized His6SseA, SseB was retained on the column and could only be eluted together with His6SseA by using a high concentration (250 mM) of imidazole (Fig. 4a). In contrast, SseB was not detected in the flow-through or wash fractions from the same column (Fig. 4a). In a control experiment, SseB was not retained on an Ni2+–NTA column that was not pre-bound with His6SseA (Fig. 4b), but instead flowed through the column with the wash buffer containing a lower concentration (20 mM) of imidazole.



View larger version (48K):
[in this window]
[in a new window]
 
Fig. 4. Immobilized SseA binds to SseB. Bacterial surface-associated proteins were obtained from S. typhimurium wild-type strain (12023) cultures grown in minimal medium at low pH. Proteins were subjected to a column binding assay with immobilized His6SseA. Flow-through (FT), washes and eluted fractions were analysed by immunoblotting with anti-SseA or anti-SseB antibodies. (a) Both His6SseA and SseB were detected in the eluate fractions. Wash (lanes 1, 2, 3): successive fractions obtained by washing the column with buffer supplemented with 20 mM imidazole. Eluate (lanes 1, 2, 3, 4): successive fractions obtained by washing the column with buffer supplemented with 250 mM imidazole. (b) SseB was not retained in an empty Ni2+–NTA column. Wash (lanes 1, 2, 3): as mentioned above. Eluate (lanes 1 and 2): fractions obtained by washing the column with buffer supplemented with 250 mM imidazole.

 
We were unable to use the same method to determine if an interaction occurs between His6SseA and SseD, because the anti-SseD antibody had a relatively high level of cross-reactivity to the large amounts of column-bound His6SseA which detach from the column with the high-imidazole concentration eluted fractions (Fig. 4a, and data not shown).

SseA and SseD interact in the yeast two-hybrid assay
As an alternative approach to establish if an interaction occurs between SseA and SseD, the corresponding genes were ligated into yeast two-hybrid vectors to generate chimeric proteins with the GAL4 binding and activation domains, respectively. Plasmids expressing these fusions were co-transformed into a Saccharomyces cerevisiae reporter strain. The ability of the SseA and SseD fusion proteins to interact with each other when expressed within the same yeast strain was confirmed by their ability to grow on minimal medium, resulting from ADE1 and HIS3 reporter gene activation, compared to the complete growth defect of the control strain (data not shown). Protein–protein interactions were also shown by measuring the induction of an independent reporter gene (lacZ), by performing {beta}-galactosidase assays of the transformed yeast strains after growth in liquid culture. The reporter strain co-expressing the SseA and SseD fusion proteins consistently had significant {beta}-galactosidase activity (6·27±0·37 Miller units), which was not detected in the control strain (0·48±0·07 Miller units). The direct interaction of SseA and SseD in the yeast two-hybrid assay supports the hypothesis that SseA acts as a chaperone for SseD.

Expression of SseB and SseD from a plasmid does not restore Sif formation or virulence of an sseA mutant
The effect of the sseA mutation on the levels of SseB and SseD could explain the virulence defect of the sseA mutant as well as its intracellular phenotypes, since SseB and SseD are necessary for translocation of SPI-2 TTSS effectors (Nikolaus et al., 2001). We have shown that the absence of detectable SseB and SseD in the sseA mutant strain can be fully complemented by simultaneous expression of these two proteins from a plasmid (psseBD). The absence of detectable SseB and SseD in the sseA mutant strain was partially complemented by expression of SseA from a plasmid (psseA) (Fig. 3). No extracellular SseB or SseD was detected in the sseA mutant strain carrying either psseA or psseBD (Fig. 3a, b), although it is possible that some protein is secreted under these conditions but is below the level of detection by the antibodies. Alternatively, some secretion might occur in infected cells. We therefore analysed this strain for the ability to induce Sifs in cultured HeLa cells and for virulence in the mouse model of infection.

HeLa cells were infected for 8 h with wild-type, sseA, sseB, sseD, sseA psseA, sseA psseBD, sseB psseBD or sseD psseBD strains, then fixed and labelled with an anti-LAMP1 antibody, and examined for Sif formation. Although the expression of SseA from a plasmid was capable of restoring wild-type levels of Sif formation in an sseA mutant strain, expression of SseB and SseD from a plasmid was only capable of partial complementation of the sseA mutant (Fig. 5a). However, expression of SseB and SseD from the plasmid fully complemented the defect in Sif formation of the sseB and sseD single mutant strains (Fig. 5a).



View larger version (20K):
[in this window]
[in a new window]
 
Fig. 5. Sif formation and virulence defects of S. typhimurium strains. (a) Sif formation was examined in HeLa cells infected with the wild-type (12023), sseA mutant ({Delta}sseA), sseB mutant ({Delta}sseB) or sseD mutant ({Delta}sseD) or these strains carrying the plasmids shown. Cells were fixed 8 h after bacterial invasion and examined by confocal fluorescence microscopy. Sifs were labelled with an anti-LAMP-1 antibody, while bacteria were detected with an anti-Salmonella antibody. The number of infected cells containing Sifs was determined by confocal microscopy examination (n=100 infected cells). Each infection was carried out in triplicate and the standard deviations from the means are shown. (b) Graphical representation of the CI analysis of the wild-type strain (12023) in mixed infections with strains carrying mutations in sseA, sseB or sseD (open bars), and with the mutant strains carrying plasmids expressing either SseA or SseB and SseD simultaneously (solid bars).

 
To analyse the ability of the psseA and psseBD plasmids to complement the virulence defect of the sseA mutant strain, the sseA mutant strain carrying either psseA or psseBD, or none, was used to inoculate mice intraperitoneally in a mixed infection with the wild-type strain. Mice were killed 48 h after inoculation, and their spleens were homogenized and plated to determine bacterial load of each strain (on the basis of antibiotic resistances), and to calculate the CI. In agreement with previous reports (Hensel et al., 1998), the sseA mutant was highly attenuated (CI=0·01) (Fig. 5b). However, whereas the sseA psseA strain gave a CI close to 1·0, indicating that it is as virulent as the wild-type strain, the virulence of sseA psseBD was only slightly higher than that of the sseA mutant strain (CI=0·023). As with Sif formation, psseBD was able to restore the virulence of the sseB and sseD single mutant strains back to wild-type levels (Fig. 5b).

The complete recovery of virulence and Sif formation in the sseA psseA strain is intriguing, since SseB and SseD were not detected extracellularly on immunoblots of proteins extracted from this strain grown in SPI-2-inducing conditions (Fig. 3). Presumably, the extracellular levels of SseB and SseD required to restore SPI-2 TTSS function in infected HeLa cells are below the threshold of detection by immunoblot with the anti-SseB and anti-SseD antibodies.

RT-PCR experiments showed that mutation of sseA did not affect sseB transcript levels (data not shown). Our results indicate that SseA is required for the stability of SseB and SseD, since these proteins were not detected on immunoblots from the sseA mutant strain grown in SPI-2-inducing conditions (Fig. 3). However, this is not the only role of SseA in the SPI-2 TTSS, since expression of SseB and SseD simultaneously from a plasmid does not restore Sif formation or virulence in an sseA mutant strain (Fig. 5). One possible explanation is that SseA may act as a chaperone for yet another SPI-2 protein. Although we cannot rule out this possibility, it seems unlikely that this is the case since the phenotypes of strains lacking other proteins encoded in the sseAG operon are different from those observed for the sseA psseBD strain. Strains defective in SseF and SseG are only mildly attenuated in virulence and partially affected in Sif formation (Guy et al., 2000; Hansen-Wester et al., 2002; Hensel et al., 1998), whereas a strain defective in SseE is not attenuated in virulence and has no other reported phenotype (Hensel et al., 1998). SscA is predicted to be a chaperone for SseC on the basis of sequence similarity (Cirillo et al., 1998; Hensel et al., 1998; Neyt & Cornelis, 1999) and SscB is also predicted to be a chaperone (Cirillo et al., 1998; Hensel et al., 1998). Furthermore, type III chaperones of the translocators studied to date bind one or two substrates but not more (Page & Parsot, 2002).

However, in view of the lack of detectable extracellular SseB and SseD displayed in vitro by the sseA mutant strain carrying psseBD, a direct role of SseA in secretion of SseB and SseD is possible. Alternatively, SseA could be required for the correct folding of SseB and SseD, which would be essential for the proper assembly of the translocon, and consequently for SPI-2-mediated translocation of effectors into the host cell. Based on its similarity to EspA, SseB is predicted to be a major component of the SPI-2 translocon, and to bind to itself as well as to the other components of the translocon (Delahay et al., 1999; Hartland et al., 2000; Hensel et al., 1998). Supporting this idea, SseB (and also SseD) has a domain with the potential to form a coiled-coil structure, which could be involved in these interactions (Hensel et al., 1998). It is conceivable that chaperone binding is necessary to prevent incorrect folding of SseB and SseD or their premature association with themselves or other translocon components, events which could also account for the deficient secretion of both proteins observed in the sseA mutant strain carrying psseBD. However, the experiments presented in this study do not discriminate between a direct role of SseA in secretion of SseB and SseD, versus a role in facilitating their correct folding which may be required for efficient secretion and function. Additional research into the interaction of SseA with SseB and SseD, as well as the interaction of SseB with itself and other proteins, will further our understanding of the role of SseA as a SPI-2 chaperone.


   NOTE ADDED IN PROOF
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
NOTE ADDED IN PROOF
REFERENCES
 
After acceptance of this manuscript, SseA was shown to be a chaperone for SseB (Zurawski & Stein, 2003).


   ACKNOWLEDGEMENTS
 
This work was supported by a grant from the Medical Research Council (UK) to David Holden. We would like to dedicate this paper to the memory of Geoff Banks, who made the initial observation of the effect of SseA on the intracellular levels of SseB. We are very grateful to Leigh Knodler for kindly providing us with pACB C-2HA and to Gadi Frankel for the yeast two-hybrid system host strain and vectors. We also thank James Kaper for advice with the column binding assays, Miranda Batchelor for advice with the yeast two-hybrid assay and Kate Unsworth for critical reading of the manuscript.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
NOTE ADDED IN PROOF
REFERENCES
 
Beuzón, C. R., Banks, G., Deiwick, J., Hensel, M. & Holden, D. W. (1999). pH-dependent secretion of SseB, a product of the SPI-2 type III secretion system of Salmonella typhimurium. Mol Microbiol 33, 806–816.[CrossRef][Medline]

Beuzón, C. R., Meresse, S., Unsworth, K. E., Ruiz-Albert, J., Garvis, S., Waterman, S. R., Ryder, T. A., Boucrot, E. & Holden, D. W. (2000). Salmonella maintains the integrity of its intracellular vacuole through the action of SifA. EMBO J 19, 3235–3249.[Abstract/Free Full Text]

Brumell, J. H., Goosney, D. L. & Finlay, B. B. (2002). SifA, a type III secreted effector of Salmonella typhimurium, directs Salmonella-induced filament (Sif) formation along microtubules. Traffic 3, 407–415.[CrossRef][Medline]

Chang, A. C. & Cohen, S. N. (1978). Construction and characterization of amplifiable multicopy DNA cloning vehicles derived from the P15A cryptic miniplasmid. J Bacteriol 134, 1141–1156.[Medline]

Cirillo, D. M., Valdivia, R. H., Monack, D. M. & Falkow, S. (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.[CrossRef][Medline]

Deiwick, J., Nikolaus, T., Erdogan, S. & Hensel, M. (1999). Environmental regulation of Salmonella pathogenicity island 2 gene expression. Mol Microbiol 31, 1759–1773.[CrossRef][Medline]

Delahay, R. M., Knutton, S., Shaw, R. K., Hartland, E. L., Pallen, M. J. & Frankel, G. (1999). The coiled-coil domain of EspA is essential for the assembly of the type III secretion translocon on the surface of enteropathogenic Escherichia coli. J Biol Chem 274, 35969–35974.[Abstract/Free Full Text]

Elliott, S. J., Hutcheson, S. W., Dubois, M. S., Mellies, J. L., Wainwright, L. A., Batchelor, M., Frankel, G., Knutton, S. & Kaper, J. B. (1999). Identification of CesT, a chaperone for the type III secretion of Tir in enteropathogenic Escherichia coli. Mol Microbiol 33, 1176–1189.[CrossRef][Medline]

Francis, M. S., Aili, M., Wiklund, M. L. & Wolf-Watz, H. (2000). A study of the YopD–lcrH interaction from Yersinia pseudotuberculosis reveals a role for hydrophobic residues within the amphipathic domain of YopD. Mol Microbiol 38, 85–102.[CrossRef][Medline]

Freter, R., Allweiss, B., O'Brien, P. C., Halstead, S. A. & Macsai, M. S. (1981). Role of chemotaxis in the association of motile bacteria with intestinal mucosa: in vitro studies. Infect Immun 34, 241–249.[Medline]

Galan, J. E. (2001). Salmonella interactions with host cells: type III secretion at work. Annu Rev Cell Dev Biol 17, 53–86.[CrossRef][Medline]

Garcia-del Portillo, F., Zwick, M. B., Leung, K. Y. & Finlay, B. B. (1993). Intracellular replication of Salmonella within epithelial cells is associated with filamentous structures containing lysosomal membrane glycoproteins. Infect Agents Dis 2, 227–231.[Medline]

Guy, R. L., Gonias, L. A. & Stein, M. A. (2000). Aggregation of host endosomes by Salmonella requires SPI2 translocation of SseFG and involves SpvR and the fmsaroE intragenic region. Mol Microbiol 37, 1417–1435.[CrossRef][Medline]

Hansen-Wester, I., Stecher, B. & Hensel, M. (2002). Type III secretion of Salmonella enterica serovar Typhimurium translocated effectors and SseFG. Infect Immun 70, 1403–1409.[Abstract/Free Full Text]

Hartland, E. L., Daniell, S. J., Delahay, R. M., Neves, B. C., Wallis, T., Shaw, R. K., Hale, C., Knutton, S. & Frankel, G. (2000). The type III protein translocation system of enteropathogenic Escherichia coli involves EspA–EspB protein interactions. Mol Microbiol 35, 1483–1492.[CrossRef][Medline]

Hensel, M., Shea, J. E., Gleeson, C., Jones, M. D., Dalton, E. & Holden, D. W. (1995). Simultaneous identification of bacterial virulence genes by negative selection. Science 269, 400–403.[Medline]

Hensel, M., Shea, J. E., Waterman, S. R. & 7 other authors (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.[CrossRef][Medline]

Hueck, C. J. (1998). Type III protein secretion systems in bacterial pathogens of animals and plants. Microbiol Mol Biol Rev 62, 379–433.[Abstract/Free Full Text]

James, P., Halladay, J. & Craig, E. A. (1996). Genomic libraries and a host strain designed for highly efficient two-hybrid selection in yeast. Genetics 144, 1425–1436.[Abstract/Free Full Text]

Klein, J. R. & Jones, B. D. (2001). Salmonella pathogenicity island 2-encoded proteins SseC and SseD are essential for virulence and are substrates of the type III secretion system. Infect Immun 69, 737–743.[Abstract/Free Full Text]

Knodler, L. A., Celli, J., Hardt, W. D., Vallance, B. A., Yip, C. & Finlay, B. B. (2002). Salmonella effectors within a single pathogenicity island are differentially expressed and translocated by separate type III secretion systems. Mol Microbiol 43, 1089–1103.[CrossRef][Medline]

Kuhle, V. & Hensel, M. (2002). SseF and SseG are translocated effectors of the type III secretion system of Salmonella pathogenicity island 2 that modulate aggregation of endosomal compartments. Cell Microbiol 4, 813–824.[CrossRef][Medline]

Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685.[Medline]

Michiels, T., Wattiau, P., Brasseur, R., Ruysschaert, J. M. & Cornelis, G. (1990). Secretion of Yop proteins by Yersiniae. Infect Immun 58, 2840–2849.[Medline]

Neyt, C. & Cornelis, G. R. (1999). Role of SycD, the chaperone of the Yersinia Yop translocators YopB and YopD. Mol Microbiol 31, 143–156.[CrossRef][Medline]

Nikolaus, T., Deiwick, J., Rappl, C., Freeman, J. A., Schroder, W., Miller, S. I. & Hensel, M. (2001). SseBCD proteins are secreted by the type III secretion system of Salmonella pathogenicity island 2 and function as a translocon. J Bacteriol 183, 6036–6045.[Abstract/Free Full Text]

Ochman, H., Soncini, F. C., Solomon, F. & Groisman, E. A. (1996). Identification of a pathogenicity island required for Salmonella survival in host cells. Proc Natl Acad Sci U S A 93, 7800–7804.[Abstract/Free Full Text]

Page, A. L. & Parsot, C. (2002). Chaperones of the type III secretion pathway: jacks of all trades. Mol Microbiol 46, 1–11.[CrossRef][Medline]

Sambrook, J. & Russell, D. W. (2001). Molecular Cloning: a Laboratory Manual, 3rd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Shea, J. E., Hensel, M., Gleeson, C. & Holden, D. W. (1996). Identification of a virulence locus encoding a second type III secretion system in Salmonella typhimurium. Proc Natl Acad Sci U S A 93, 2593–2597.[Abstract/Free Full Text]

Stein, M. A., Leung, K. Y., Zwick, M., Garcia-del Portillo, F. & Finlay, B. B. (1996). Identification of a Salmonella virulence gene required for formation of filamentous structures containing lysosomal membrane glycoproteins within epithelial cells. Mol Microbiol 20, 151–164.[Medline]

Taylor, R. K., Miller, V. L., Furlong, D. B. & Mekalanos, J. J. (1987). Use of phoA gene fusions to identify a pilus colonization factor coordinately regulated with cholera toxin. Proc Natl Acad Sci U S A 84, 2833–2837.[Abstract]

Valdivia, R. H. & Falkow, S. (1996). Bacterial genetics by flow cytometry: rapid isolation of Salmonella typhimurium acid-inducible promoters by differential fluorescence induction. Mol Microbiol 22, 367–378.[Medline]

Wallis, T. S. & Galyov, E. E. (2000). Molecular basis of Salmonella-induced enteritis. Mol Microbiol 36, 997–1005.[CrossRef][Medline]

Yu, X. J., Ruiz-Albert, J., Unsworth, K. E., Garvis, S., Liu, M. & Holden, D. W. (2002). SpiC is required for secretion of Salmonella pathogenicity island 2 type III secretion system proteins. Cell Microbiol 4, 531–540.[CrossRef][Medline]

Zurawski, D. V. & Stein, M. A. (2003). SseA acts as the chaperone for the SseB component of the Salmonella pathogenicity island 2 translocon. Mol Microbiol 47, 1341–1351.[CrossRef][Medline]

Received 16 December 2002; revised 27 January 2003; accepted 29 January 2003.