This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Johnson, T. L.
Right arrow Articles by Sandkvist, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Johnson, T. L.
Right arrow Articles by Sandkvist, M.

 Previous Article  |  Next Article 

Journal of Bacteriology, December 2007, p. 9082-9089, Vol. 189, No. 24
0021-9193/07/$08.00+0     doi:10.1128/JB.01256-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Mapping Critical Interactive Sites within the Periplasmic Domain of the Vibrio cholerae Type II Secretion Protein EpsM{triangledown}

Tanya L. Johnson, Maria E. Scott,{dagger} and Maria Sandkvist*

Department of Microbiology and Immunology, University of Michigan Medical School, 1150 West Medical Center Drive, Ann Arbor, Michigan 48109

Received 3 August 2007/ Accepted 26 September 2007


arrow
ABSTRACT
 
The type II secretion (T2S) system is present in many gram-negative species, both pathogenic and nonpathogenic, where it supports the delivery of a variety of toxins, proteases, and lipases into the extracellular environment. In Vibrio cholerae, the T2S apparatus is composed of 12 Eps proteins that assemble into a multiprotein complex that spans the entire cell envelope. Two of these proteins, EpsM and EpsL, are key components of the secretion machinery present in the inner membrane. In addition to likely forming homodimers, EpsL and EpsM have been shown to form a stable complex in the inner membrane and to protect each other from proteolytic degradation. To identify and map the specific regions of EpsM involved in protein-protein interactions with both another molecule of EpsM and EpsL, we tested the interactions of deletion constructs of EpsM with full-length EpsM and EpsL by functional characterization and copurification as well as coimmunoprecipitation. Analysis of the truncated EpsM mutants revealed that the region of EpsM from amino acids 100 to 135 is necessary for EpsM to form homo-oligomers, while residues 84 to 99 appear to be critical for a stable interaction with EpsL.


arrow
INTRODUCTION
 
Vibrio cholerae is a pathogen that causes cholera, an acute intestinal infection in humans, when ingested. Cholera toxin is the major virulence factor of V. cholerae and is responsible for the explosive watery diarrheal symptoms characteristic of cholera (16). The virulence of V. cholerae is dependent on the transport of cholera toxin across the outer membrane of the bacterial cell. Transport of the toxin out of the cell and into the extracellular environment occurs by way of the type II secretion (T2S) system, a pathway that is present in many gram-negative pathogens in addition to V. cholerae and that supports the secretion of key virulence factors (7, 9, 26). The secretion apparatus produced by V. cholerae is encoded by the extracellular protein secretion (eps) operon, which gives rise to 12 gene products, EpsC to EpsN, and by pilD (vcpD) (10, 17, 31). Based on studies of the Eps proteins and their homologous counterparts in Pseudomonas aeruginosa, Klebsiella oxytoca, and Erwinia chrysanthemi, these components are suggested to assemble into a multiprotein complex that spans the entire cell envelope (8, 15, 20, 21, 23, 24). Although the function and protein-protein interactions of some members of the T2S complex have been described, the configuration of this apparatus and the fundamental mechanisms involved in its assembly are still unknown. Numerous interactions between T2S proteins still need to be uncovered to better understand how this large, highly organized, multiprotein complex is constructed in the cell envelope.

Two of the components that make up the T2S system, EpsL and EpsM, are inner membrane proteins that have been shown to form a stable protein complex that can be immunoprecipitated with either anti-EpsL or anti-EpsM antibodies (29). EpsL and EpsM, as well as their respective homologs in P. aeruginosa, K. oxytoca, and E. chrysanthemi, are also capable of protecting each other from proteolytic degradation (18, 21, 23-25, 29). It is likely that EpsL and EpsM directly interact, as no other Eps protein is needed for their assembly when they are expressed in Escherichia coli (29). The EpsL homologs are made up of a large N-terminal cytoplasmic domain, a single transmembrane helix, and a smaller periplasmic domain (4, 14). The cytoplasmic domain of EpsL binds to and localizes another protein present in the secretion complex, the ATPase EpsE, to the cytoplasmic membrane, a finding that has been confirmed with homologous T2S components from other species (3, 21, 22, 27, 30). The X-ray crystal structure of a complex formed between subdomains of EpsL and EpsE confirmed results obtained with EpsL chimeras showing that two regions, both located in the cytoplasmic domain of EpsL, are involved in interaction with EpsE (1, 30). In addition, we have recently shown that a region of the cytoplasmic domain of EpsL adjacent to the predicted transmembrane helix appears to "fine tune" the interaction of EpsE with the membrane and to stimulate its ATPase activity (5).

The EpsM homologs are inner membrane proteins with a short segment in the cytoplasm, one transmembrane helix, and a periplasmic C terminus (4). By taking advantage of the species specificity of the T2S pathway, chimeric proteins composed of EpsL and its homologue, ExeL, from Aeromonas hydrophila were used to map the EpsM binding domain in EpsL (30). Although this approach may not have identified all contacts between these proteins, an EpsM species-specific binding site was mapped to a region of EpsL between residues 216 and 296, which also contains the membrane-spanning helix of EpsL (30). The corresponding EpsL binding site on EpsM, however, is not known.

Besides its ability to stabilize EpsL and protect it from proteolytic degradation (29), little has been uncovered about the role that EpsM plays in the T2S apparatus. All T2S systems contain an EpsM homolog (26) that is essential for function, however. The X-ray crystal structure of a large portion of the periplasmic domain of EpsM has been determined and represents a novel version of the ferredoxin fold, but the extensive variety of functions that ferredoxin-like proteins perform lends few clues to the function of EpsM (2). In solution, this periplasmic EpsM construct, which consists of residues 65 to 165, forms dimers (2). This suggests that neither the transmembrane domain nor the short N terminus of EpsM is needed for this process. It remains to be confirmed, however, that the full-length protein present in the membrane forms the same dimer interface as that observed in the X-ray structure. The possibility that larger oligomers of native EpsM are formed in vivo also needs to be considered.

Because EpsL and EpsM are links between the energy-providing EpsE protein in the cytoplasm, the rest of the T2S machinery in the cytoplasmic membrane, and the secretion pore in the outer membrane, a more detailed knowledge of how these proteins interact is crucial for understanding how the energy generated by ATP hydrolysis is used in the assembly and/or function of the T2S machinery. Using C-terminal deletion constructs of EpsM, we mapped the region within EpsM that contains residues involved in its dimerization to the C-terminal region between amino acids 100 and 135, thus confirming the X-ray crystal structure model. Additionally, we took advantage of the ability of EpsL and EpsM to protect each other from proteolytic degradation and tested the interactions between EpsL and both N-terminal and C-terminal deletion constructs of EpsM by coimmunoprecipitation and functional analyses to identify and map the region of EpsM interacting with EpsL.


arrow
MATERIALS AND METHODS
 
Bacterial strains and growth conditions. Table 1 describes relevant features of the strains and plasmids used in this study. V. cholerae and E. coli strains were grown either to stationary phase at 37°C in Luria broth (LB) supplemented with thymine for V. cholerae or to mid-logarithmic phase at 37°C in M9 growth medium (32) supplemented with 4% Casamino Acids and 0.4% glucose. Where indicated, overexpression of green fluorescent protein (GFP) constructs was obtained by induction with isopropyl-β-D-thiogalactopyranoside (IPTG). In coexpression studies, 0.5% glycerol replaced glucose and optimal expression of genes carried on the pBAD33 vector was obtained with 1 mM arabinose. Final antibiotic concentrations were 100 µg/ml of ampicillin, 100 µg/ml of carbenicillin, and 30 µg/ml of chloramphenicol, as needed.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Strains and plasmids used for this study

Plasmid construction. Table 2 lists the PCR primers used in this study. Plasmids pGFP, pGFP-EpsM, and pBADEpsM were constructed as previously described (32). For plasmid GFP-EpsM{Delta}82, we used primers Gfp03 and EpsM29 and plasmid pGFP-EpsM in a PCR to amplify a fragment carrying gfp-epsM(1-83), flanked by EcoRI and PstI restriction sites. After digestion of pMMB66 with EcoRI and PstI, gfp-epsM(1-83) was inserted. Plasmids GFP-EpsM{Delta}66, GFP-EpsM{Delta}30, and GFP-EpsM{Delta}14 were prepared in a similar manner, with changes in the PCR primers used but with the same template (pGFP-EpsM). Specifically, PCR amplification used primer Gfp03 in conjunction with primers EpsM04, EpsM05, and EpsM06 to create GFP-EpsM{Delta}66, GFP-EpsM{Delta}30, and GFP-EpsM{Delta}14, respectively. To construct {Delta}83-EpsM, pGFP-EpsM was used as a template with primers EpsM30 and EpsM31 to amplify an epsM(82-165) fragment flanked by SacI and HindIII restriction sites. After digestion of pMMB68 with SacI and HindIII, epsM(82-165) was inserted, thus replacing the region of the etxB gene that encodes the mature portion of the E. coli heat-labile enterotoxin and fusing the epsM(82-165) portion in frame with the signal peptide sequence.


View this table:
[in this window]
[in a new window]

  		TABLE 2. PCR primers used to construct hybrids and truncated proteins

For plasmid pBADEpsL, we used primer pair EpsL01 and EpsLSAL and the template pMS58 in a PCR to generate epsL with flanking SacI and SalI restriction sites. The resulting epsL fragment was ligated into the SacI and SalI sites of pBAD33.

Construction of pBADEpsMHis6 occurred in three steps. First, plasmid pMMB587 was digested with EcoRI and then overhangs were blunt ended with T4 polymerase. Next, the pMMB587 fragment was cut with HindIII to liberate a 577-bp segment that contained the C-terminal hexahistidine-tagged EpsM coding sequence (epsMHis6). Lastly, the epsMHis6 fragment was cloned into the compatible SmaI and HindIII sites of pBAD33.

Complementation of epsM mutants. GFP-EpsM fusions were expressed in V. cholerae epsM mutant strains PU3 and PU5, with and without induction by IPTG, and were tested for protease secretion using a modified fluorescence-based assay (5, 31). Briefly, supernatants from overnight cultures grown in LB were assayed in 5 mM HEPES, pH 7.5, and 0.05 mM N-tert-butoxy-carbonyl-Gln-Ala-Arg-7-amido-4-methyl-coumarin (Sigma-Aldrich, St. Louis, MO) for 10 min at 37°C, using excitation and emission wavelengths of 385 nm and 440 nm, respectively. One fluorescence unit was defined as the amount of fluorescence produced by 50 picograms of 7-amido-4-methyl-coumarin.

Triton X-100 cell extraction protocol. E. coli strains expressing various GFP fusions from pMMB66 were grown overnight at 37°C in M9 growth medium. When GFP fusions were coexpressed with proteins encoded on pBAD33, cells were grown with glycerol instead of glucose. Overnight cultures were diluted 1:50 in fresh medium and grown to an optical density at 600 nm of 0.4, at which time the expression of proteins was induced. After 90 min of induction, cells were harvested by centrifugation at 2,500 x g for 10 min. Cell pellets were suspended in 50 µl of 200 mM Tris (pH 8.0). Spheroplasts were prepared by sequential addition of 100 µl of 200 mM Tris (pH 8.0) with 1 M sucrose, 10 µl of 10 mM EDTA, 10 µl of 10-mg/ml lysozyme, and 300 µl of water to the cell suspension. Cysteine and serine protease inhibitor cocktail and the protease inhibitor phenylmethylsulfonyl fluoride (Roche Diagnostics, Indianapolis, IN) were added, followed by incubation on ice for 10 min. Next, DNase at a final concentration of 10 µg/ml and 500 µl of buffer containing 2% Trition X-100 and 10 mM MgCl2 in 50 mM Tris (pH 8.0) were added to the spheroplast suspension. The suspension was incubated on ice for 30 min and then centrifuged at 16,000 x g for 10 min at 4°C. The supernatant fraction collected was then subjected to ultracentrifugation at 160,000 x g for 1 h at 4°C. Following ultracentrifugation, the supernatant contained Triton X-100-soluble proteins, including periplasmic, cytoplasmic, and cytoplasmic membrane proteins.

Coimmunoprecipitation and metal-affinity chromatography. For coimmunoprecipitation experiments, equal volumes of anti-EpsL antiserum and a suspension of protein G-Sepharose beads (GE Biosciences, Buckinghamshire, United Kingdom) in 50 mM Tris-buffered saline (pH 8.0) were mixed overnight at 4°C. Following incubation, beads were washed three times with Tris-buffered saline to remove unbound antibody. To evaluate interactions between GFP-EpsM fusions and native EpsL, 200 µl of Triton X-100 cell extract was incubated with 10 µl of anti-EpsL protein G-Sepharose beads. To detect EpsM-EpsMHis6 interactions, Triton X-100 cell extracts were instead incubated with 10 µl cobalt-immobilized metal-affinity chromatography resin (IMAC beads; BD Biosciences, San Jose, CA) that recognizes the histidine tag of EpsMHis6. Coimmunoprecipitation and metal-affinity chromatography reactions occurred in binding buffer (50 mM Tris [pH 8.0]-1% Triton X-100) for 2 h with rocking at 4°C. After incubation, samples were centrifuged at 3,000 x g for 1 min, the supernatant was removed, and beads were washed three times with binding buffer and once with 50 mM Tris (pH 8.0). Sample buffer (20 µl) containing sodium dodecyl sulfate (SDS)-dithiothreitol was added to the beads and boiled for 10 min prior to centrifugation. Samples were subjected to SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotted with biotinylated anti-GFP (Rockland Inc., Gilbertsville, PA) or biotinylated anti-EpsM immunoglobulin G (29) and horseradish peroxidase-conjugated streptavidin (Bio-Rad Laboratories, Hercules, CA). Western blots were developed with a chemiluminescent substrate (Pierce, Rockford, IL). Densitometric analysis of images was performed with the ImageQuant program (Molecular Dynamics, Sunnyvale, CA).

Statistical analysis. Student's t test was applied for all statistical analyses, and differences were considered significant if the P value was <0.05.


arrow
RESULTS
 
Two mutants of V. cholerae, PU3 and PU5, with transposon-induced disruptions of the epsM gene, were previously found to be defective in secretion of an extracellular protease, chitinase, and the cholera toxin B subunit homolog EtxB (19, 31). The transposon was inserted at gene positions 247 and 339 in PU3 and PU5, respectively, and may have resulted in the production of truncated forms of EpsM lacking either 82 or 52 C-terminal residues. Extracellular secretion was restored to different levels in these mutants when attempts were made to complement the secretion defects with plasmid-carried epsM (19). The differential complementation may have been a result of negative dominance of the longer EpsM variant in PU5. Given that EpsM interacts with both another molecule of EpsM and EpsL (2, 29), it is possible that the truncated form of EpsM in PU5 interferes with one or both of these interactions of the plasmid-encoded native EpsM protein. To test this hypothesis and to map binding sites in EpsM, we subjected EpsM to C-terminal deletion mutagenesis and tested the EpsM truncations for the ability to complement the secretion defect in the epsM mutant PU3 and to inhibit secretion in the wild-type V. cholerae strain TRH7000. In addition, we determined the ability of the truncated forms of EpsM to interact with full-length EpsM and EpsL. Since important epitopes for anti-EpsM antibodies may be lost in the shorter EpsM deletions, we used an easily detectable GFP fusion, GFP-EpsM, in our mapping studies.

Plasmid-encoded GFP-EpsM could complement the secretion defect in PU3, and the level of extracellular protease (19, 31) detected in the culture supernatant was the same as that of the wild-type strain, indicating that GFP-EpsM is fully functional and interacts with the rest of the Eps apparatus (Fig. 1). Consistent with the earlier findings mentioned above, the level of protease secreted from PU5 was not restored to wild-type levels when efforts were made to complement the secretion defect with GFP-EpsM (Fig. 1). Mutant PU5 could, however, be complemented fully when the level of GFP-EpsM was increased by the addition of IPTG to the cultures (Fig. 1). This suggests that the truncated form of EpsM produced as a consequence of the transposon insertion in PU5 may still be capable of dimerization and/or EpsL interaction and thereby competes with plasmid-encoded GFP-EpsM for important binding sites within the T2S apparatus.


Figure 1
View larger version (17K):
[in this window]
[in a new window]

  		FIG. 1. Complementation of epsM mutant strains PU3 and PU5 with GFP-EpsM. Wild-type V. cholerae TRH7000 and its isogenic epsM mutant strains PU3 and PU5, expressing either GFP alone or GFP-EpsM, were grown in LB and carbenicillin at 37°C overnight. To increase the expression of GFP-EpsM in PU5, the growth medium was supplemented with IPTG at a final concentration of 10 µM. Culture supernatants were separated from cells by centrifugation and tested for the presence of extracellular protease, using the proteolysis substrate N-tert-butoxy-carbonyl-Gln-Ala-Arg-7-amido-4-methyl-coumarin. The rate of hydrolysis is presented as the mean for three independent experiments ± standard error of the mean (SEM). There was a statistically significant difference between the protease activities of the wild type and PU5 expressing GFP-EpsM without IPTG (P < 0.004), while there was no difference between the wild type and PU5 expressing GFP-EpsM following induction with IPTG (P > 0.05). FU, fluorescence units; OD600, optical density at 600 nm.

No truncated constructs of GFP-EpsM could restore secretion in the epsM mutant strain PU3. In order to identify the minimal region of EpsM necessary to interact with EpsL and to homo-oligomerize, we first constructed C-terminal truncations of GFP-EpsM and compared their properties to those of the full-length GFP-EpsM fusion in V. cholerae. The diagram in Fig. 2A shows a schematic representation of the full-length and truncated forms of GFP-EpsM.


Figure 2
View larger version (20K):
[in this window]
[in a new window]

  		FIG. 2. C-terminal deletions prevent GFP-EpsM chimeras from supporting protease secretion. (A) Schematic representation of native EpsM and the deletion fragments within the cytoplasmic membrane (CM). Residues 24 to 41 constitute the transmembrane helix (hatched box). GFP (not shown) was fused to the N-terminal residue of EpsM. (B) Wild-type V. cholerae TRH7000 and epsM mutant strain PU3 expressing either GFP alone, full-length GFP-EpsM, or the GFP-EpsM truncated constructs were grown in LB and carbenicillin at 37°C overnight. Culture supernatants were separated from cells by centrifugation and analyzed for the presence of extracellular protease as described in the legend to Fig. 1. The means ± SEM for three independent experiments are presented. There was a statistically significant difference between the protease activities of PU3 expressing full-length GFP-EpsM and PU3 producing truncated GFP-EpsM variants (P < 0.0005). FU, fluorescence units; OD600, optical density at 600 nm. (C) epsM mutant strain PU3 containing vector only (pMMB) or vector expressing full-length GFP-EpsM or the GFP-EpsM truncated constructs was grown in LB supplemented with carbenicillin and IPTG at a final concentration of 10 µM at 37°C to mid-log phase. Cell lysates were immunoblotted with anti-GFP antibodies to compare relative amounts of full-length and truncated GFP-EpsM proteins expressed. Positions of molecular weight markers are shown.

Expression in the epsM mutant PU3 allowed for functional characterization of the truncated GFP-EpsM fusions. While full-length GFP-EpsM could complement the secretion defect in PU3, none of the truncated forms could restore secretion of the protease (Fig. 2B), despite all being produced in similar amounts (Fig. 2C). The results suggest that removal of C-terminal residues from EpsM interferes with its ability to support secretion.

Inhibition of secretion from wild-type V. cholerae cells. The truncated GFP-EpsM fusions were then tested for the ability to inhibit protease secretion from wild-type V. cholerae TRH7000 cells. While expression of full-length GFP-EpsM following the addition of 10 µM IPTG did not affect the level of protease secretion in strain TRH7000, GFP-EpsM{Delta}14, GFP-EpsM{Delta}30, and GFP-EpsM{Delta}66 expression resulted in less protease being detected in the supernatants, suggesting that the truncated forms of GFP-EpsM competed with native EpsM for functional dimerization or interaction with EpsL and/or the rest of the Eps apparatus (Fig. 3). In contrast, removal of the C-terminal 82 residues in GFP-EpsM{Delta}82 resulted in a truncation mutant that had lost the ability to inhibit secretion from wild-type cells, possibly due to its failure to interact with EpsM or EpsL (Fig. 3). These findings are consistent with the suggestion that transposon insertion in PU5 results in a 52-amino-acid, C-terminally deleted form of EpsM that has a dominant-negative effect on secretion, while the production of an 82-residue deleted EpsM variant as a consequence of transposon insertion in PU3 does not interfere with protease secretion.


Figure 3
View larger version (21K):
[in this window]
[in a new window]

  		FIG. 3. Dominant-negative effects of truncated GFP-EpsM chimeras in wild-type V. cholerae. Wild-type V. cholerae TRH7000 expressing GFP alone, GFP-EpsM, or the GFP-EpsM deletion constructs was cultured in LB containing carbenicillin and supplemented with IPTG at a final concentration of 10 µM at 37°C overnight. Cells were pelleted by centrifugation, and the resulting supernatants were tested for the presence of extracellular protease as described in the legend to Fig. 1. The results presented are the means and corresponding SEMs for three independent experiments. The differences in inhibition of secretion between full-length GFP-EpsM and GFP-EpsM{Delta}14, GFP-EpsM {Delta}30, and GFP-EpsM {Delta}66 were statistically significant (P < 0.03), while the difference between full-length GFP-EpsM and GFP-EpsM{Delta}82 was not significant (P > 0.05). FU, fluorescence units; OD600, optical density at 600 nm.

A region of EpsM between amino acids 100 and 135 is important for homo-oligomer formation. The ability of the periplasmic domain of EpsM, consisting of residues 65 to 165, to form dimers, according to gel filtration analysis of recombinant material from E. coli, indicated that native EpsM is capable of forming dimers or larger oligomers (2). To determine whether deletions at the C terminus of EpsM affect its ability to oligomerize and to map the site of EpsM-EpsM interaction, we expressed the GFP-EpsM variants in the presence of EpsMHis6 in E. coli. We reasoned that histidine-tagged EpsM would bind to cobalt IMAC beads and, consequently, that any GFP-EpsM proteins capable of interacting with EpsMHis6 would be copurified. Samples eluted from IMAC beads were loaded onto SDS-PAGE gels, followed by immunoblotting with anti-GFP antibody to detect GFP-EpsM proteins capable of binding to histidine-tagged EpsM. Full-length GFP-EpsM was recovered when it was coexpressed with EpsMHis6, as shown in Fig. 4 (lane 9). In contrast, GFP-EpsM expressed alone did not react with IMAC beads and therefore was not detected (lane 8). GFP-EpsM{Delta}14 and GFP-EpsM{Delta}30 also bound to EpsMHis6, as they were purified on the IMAC beads (Fig. 4). Binding of GFP-EpsM{Delta}66 and GFP-EpsM{Delta}82 to the IMAC resin was not detectable; therefore, these deletions likely cannot form stable oligomers with EpsMHis6. These results suggest that the C-terminal 30 residues are not necessary for EpsM dimerization, while removal of the last 66 residues prevents EpsM from stably interacting with another EpsM molecule, thus confirming the findings from the X-ray crystallography study of the periplasmic domain of EpsM (2). Although GFP-EpsM{Delta}66 was unable to stably interact with EpsMHis6, it inhibited secretion when expressed in wild-type V. cholerae, suggesting that this deletion construct is still capable of interacting with another binding partner in the T2S system, likely EpsL.


Figure 4
View larger version (40K):
[in this window]
[in a new window]

  		FIG. 4. Mapping of EpsM-EpsM interaction site. Full-length and truncated GFP-EpsM constructs were copurified with EpsMHis6 from E. coli Triton X-100 extracts by metal-affinity chromatography and subjected to SDS-PAGE and immunoblotting with biotinylated anti-GFP antibodies. Full-length GFP-EpsM was induced at a final concentration of 20 µM IPTG, while the truncations were induced with a higher concentration of IPTG (50 µM). Lanes 1 to 6, GFP-EpsM hybrids detected in Triton X-100 extracts; lanes 8 to 13, GFP-EpsM proteins that bound to hexahistidine-tagged EpsM. GFP-EpsM alone did not react with the metal-affinity resin and served as the negative control for the procedure (compare lanes 1 and 8). The GFP antibodies did not cross-react with the hexahistidine-tagged EpsM proteins present in all lanes, except for lanes 1 and 8. Positions of molecular weight markers are shown.

EpsL interacts with and stabilizes all of the truncated GFP-EpsM fusions, except for GFP-EpsM{Delta}82. The above data suggested that the inhibition of protease secretion by some of the truncated GFP-EpsM mutants might result from competition with native EpsM for EpsL binding, since GFP-EpsM{Delta}66 was not able to oligomerize with EpsMHis6 but was able to inhibit secretion in wild-type V. cholerae. To test this hypothesis, we determined the ability of the various truncations to interact with EpsL. This was first accomplished by determining whether EpsL was capable of stabilizing and protecting the various deletion mutants from proteolysis in E. coli. When total E. coli cell extracts containing the GFP-EpsM fusions were subjected to SDS-PAGE and examined by Western blot analysis, we noted that all of the truncated fusions were present at a lower level than that of full-length GFP-EpsM (Fig. 5, even-numbered lanes). To determine if the reduced levels of the truncated GFP-EpsM fusions were due to proteolysis and whether they could be rescued by EpsL, we also coexpressed the GFP-EpsM variants with EpsL. The results shown in Fig. 5 (odd-numbered lanes) indicate that EpsL stabilized all of the GFP-EpsM proteins and protected them from proteolysis, with the exception of GFP-EpsM{Delta}82. This stabilization with EpsL was particularly noticeable with GFP-EpsM{Delta}30 and GFP-EpsM{Delta}66, whose levels were barely detected in the absence of EpsL (compare lane 5 to lane 6 and lane 7 to lane 8). These results confirmed that the truncated GFP-EpsM variants are susceptible to proteolysis when expressed in E. coli and showed that coexpression with native EpsL resulted in increased levels of the GFP-EpsM fusions, except for GFP-EpsM{Delta}82. This suggests that all deletion mutants but GFP-EpsM{Delta}82 are capable of interaction with EpsL and that the extended deletion in GFP-EpsM{Delta}82 may have removed residues important for interaction with EpsL.


Figure 5
View larger version (39K):
[in this window]
[in a new window]

  		FIG. 5. Stabilization of truncated GFP-EpsM chimeras by EpsL coexpressed in E. coli. E. coli cell lysates were immunoblotted with anti-GFP antibodies to compare relative amounts of GFP-EpsM protein expressed in the presence (+) or absence (–) of EpsL. Lanes 1 and 2, GFP-EpsM; lanes 3 and 4, GFP-EpsM{Delta}14; lanes 5 and 6, GFP-EpsM{Delta}30; lanes 7 and 8, GFP-EpsM{Delta}66; lanes 9 and 10, GFP-EpsM{Delta}82. Odd-numbered lanes contain GFP-EpsM fusions expressed in the presence of native EpsL. Positions of molecular weight markers are shown.

Coimmunoprecipitation of EpsL and GFP-EpsM truncated proteins. To further test the ability of the truncated forms of GFP-EpsM to interact with EpsL, we subjected them to coimmunoprecipitation. Previously, we have shown that EpsL and EpsM interact with each other to form a stable complex that can be extracted with the nonionic detergent Triton X-100 from the cytoplasmic membranes of both V. cholerae and E. coli and coimmunoprecipitated with either anti-EpsL or anti-EpsM antibodies (29). We coexpressed the truncated GFP-EpsM proteins with EpsL in E. coli and subjected the cells to Triton X-100 extraction followed by immunoprecipitation with anti-EpsL antibodies. The precipitated material was then analyzed by SDS-PAGE and immunoblotting with biotinylated anti-GFP antibody. The Triton X-100 starting material and samples recovered from the immunoprecipitation with anti-EpsL antibodies are shown in Fig. 6. Anti-EpsL antibodies do not recognize EpsM, and therefore GFP-EpsM did not precipitate when extracts did not contain EpsL (Fig. 6A, lane 7, and B, lane 4). In contrast, when EpsL was present, GFP-EpsM coimmunoprecipitated with EpsL, as did all the GFP-EpsM variants, except for GFP-EpsM{Delta}82 (Fig. 6B, lane 6). Densitometric analysis of the immunoblot indicated that truncated EpsM hybrids missing 14, 30, and 66 amino acids, as well as full-length GFP-EpsM, were immunoprecipitated with EpsL. Taken together, the combined data presented in Fig. 5 and 6 indicate that removal of up to 66 residues from the C terminus of EpsM does not interfere with the ability of EpsL to bind to and stabilize the GFP-EpsM fusions, while removal of 82 residues most likely eliminates an EpsL binding site.


Figure 6
View larger version (22K):
[in this window]
[in a new window]

  		FIG. 6. Mapping of EpsL binding site in EpsM. (A) Triton X-100 cell extracts were prepared from E. coli cells that expressed full-length or C-terminally deleted GFP-EpsM (induced with 10 µM IPTG) in combination with native EpsL. Immunoprecipitation of cell extracts with anti-EpsL antibodies coupled to protein G-Sepharose was followed by immunoblot analysis with biotinylated anti-GFP antibodies. Lanes 1 to 5, GFP-EpsM hybrids present in Triton X-100 cell extracts; lanes 7 to 11, GFP-EpsM fusions recovered following immunoprecipitation (co-IP) with anti-EpsL. Lanes 1 and 7 contain samples from cells expressing only GFP-EpsM, which were the negative control for the procedure. Biotinylated antibody to GFP detected EpsM hybrids but not native EpsL (lanes 6 and 12). Positions of molecular weight markers are shown. (B) Since GFP-EpsM{Delta}82 is expressed at a lower level due to its inability to be stabilized by EpsL, this construct required induction with a higher concentration of IPTG (50 µM). Lanes 1 to 3, full-length EpsM and the GFP-EpsM{Delta}82 hybrid present in Triton X-100 cell extracts; lanes 4 to 6, GFP-EpsM fusions recovered following immunoprecipitation (co-IP) with anti-EpsL. Lanes 1 and 4 contain samples from cells expressing only GFP-EpsM, which were the negative control for the procedure.

{Delta}83-EpsM is capable of interacting with EpsL. Py and colleagues used yeast two-hybrid studies to determine which portion of GspM, the Erwinia chrysanthemi EpsM homologue, interacts with GspL, the EpsL homologue (23). By testing of three large GspM fragments, it was concluded that a fragment that contained GspM residues 84 to 162 (GspM contains 162 amino acids) is capable of interaction with the periplasmic domain of GspL, indicating that residues 1 to 83 are dispensable for GspL interaction (23). Since our truncation studies revealed that residues 100 to 165 of EpsM are not necessary for interaction with EpsL, we speculated that the EpsL binding site is localized to a region that includes residues 84 to 99. To test this possibility and to confirm the results obtained with the homologous Gsp proteins expressed in the yeast two-hybrid system, we constructed an N-terminal 83-amino-acid truncation of EpsM and examined its ability to interact with EpsL. Because it does not contain a transmembrane domain, {Delta}83-EpsM was fused to the signal peptide of the E. coli heat-labile enterotoxin EtxB (28) in order to facilitate translocation of {Delta}83-EpsM across the cytoplasmic membrane. The diagram in Fig. 7A shows schematic representations of full-length EpsM and {Delta}83-EpsM.


Figure 7
View larger version (23K):
[in this window]
[in a new window]

  		FIG. 7. N-terminal deletion mutant of EpsM interacts with EpsL. (A) Schematic representation of native EpsM and {Delta}83-EpsM. Residues 24 to 41 constitute the transmembrane helix (hatched box). (B) Triton X-100 cell extracts were prepared from E. coli cells that expressed full-length EpsM or {Delta}83-EpsM in combination with native EpsL. Immunoprecipitation (co-IP) of cell extracts with anti-EpsL antibodies coupled to protein G-Sepharose was followed by immunoblot analysis with biotinylated anti-EpsM antibodies. Lanes 1 and 2 and lanes 5 and 6 show full-length EpsM and {Delta}83-EpsM, respectively, present in Triton X-100 cell extracts. Samples shown in lanes 3, 4, 7, and 8 represent proteins recovered following immunoprecipitation with anti-EpsL. Lanes with cells not expressing EpsL served as the negative controls for the procedure. Positions of molecular weight markers are shown. The arrowhead indicates the position of {Delta}83-EpsM.

We coexpressed either native EpsM or {Delta}83-EpsM with EpsL in E. coli and subjected the cells to Triton X-100 extraction followed by immunoprecipitation with anti-EpsL antibodies as described above. The precipitated material was then analyzed by SDS-PAGE and immunoblotting with biotinylated anti-EpsM antibody (29). The Triton X-100 starting material and samples recovered from the immunoprecipitation with anti-EpsL antibodies are shown in Fig. 7B. {Delta}83-EpsM coimmunoprecipitated with EpsL (lane 8), although not as efficiently as full-length EpsM (lane 4). This finding confirms the results from the yeast two-hybrid system and suggests that the EpsL binding site involves a region of EpsM that includes residues 84 to 99.


arrow
DISCUSSION
 
In order to better understand the mechanism of secretion via the T2S pathway, in this study we have analyzed important protein-protein interactions between individual members of the secretion apparatus. EpsM and EpsL are key components of the secretion machinery present in the inner membrane. By establishing a link between the ATPase, EpsE, in the cytoplasm and the rest of the T2S apparatus in the inner and outer membranes of the cell, they likely play a major role in transducing energy to power the assembly or function of the secretion machinery. The main objective of this study was to confirm which residues are involved in EpsM-EpsM homo-oligomerization and to determine the minimal region of V. cholerae EpsM required for interaction with EpsL.

Results from gel permeation chromatography of the soluble periplasmic construct of EpsM (residues 65 to 165) used for X-ray crystallography suggested that EpsM's native state is a dimer or larger oligomer. In the crystal structure model of the dimer, subunit-subunit interactions occur primarily through contacts between residues 123 to 133 and 147 of the EpsM monomer (2). Removal of these residues should result in a form of EpsM that does not dimerize, consistent with our experimental finding that deletion of residues 100 to 165 to create GFP-EpsM{Delta}66 prevented interaction with full-length EpsM (Fig. 4). When we constructed a truncated GFP variant of EpsM missing the final 30 residues of the protein (deleting residues 136 to 165), most of this dimerization domain remained intact, explaining why this deletion construct was still capable of oligomerization. Of the residues pinpointed in the X-ray structure as being involved in dimerization, only Arg147 is missing in GFP-EpsM{Delta}30; what remains is a form of the protein still retaining the majority of residues necessary for dimerization to occur. Removal of 14 amino acids from the C terminus of EpsM results in a truncated protein that retains all the residues necessary for dimerization, as GFP-EpsM{Delta}14's dimerization ability was similar to that of full-length GFP-EpsM. These results with full-length EpsM and the truncated variants confirm that EpsM in the membrane likely forms the same dimer interface as that deduced from the X-ray structure of the soluble periplasmic domain (2).

The sequence homology between EpsM and its family members is concentrated in the predicted transmembrane helix and two blocks of conserved residues present in the periplasmic region of the polypeptide chain (2). One of these conserved blocks includes the region that we have shown to be necessary for dimerization. According to the X-ray structure of periplasmic EpsM, residues 122 to 135 within this region constitute an {alpha}-helix. Interestingly, although this region of EpsM shows a particularly high degree of sequence conservation with other members of the M protein family, the most highly conserved residues are mostly buried within the EpsM monomer and therefore are not involved in interaction with another monomer. The residues that in fact form the dimer interface are conserved only among the closest homologs of EpsM (2).

The GFP-EpsM fragment missing the last 66 amino acid residues bound EpsL as well as full-length EpsM. Therefore, the last 66 C-terminal residues of EpsM are not required for stable interaction with EpsL. GFP-EpsM{Delta}82, however, was no longer protected from proteolytic degradation by EpsL, nor was it coimmunoprecipitated by anti-EpsL antibodies. Reciprocally, when the {Delta}83-EpsM construct consisting of the last 82 residues of EpsM was subjected to coimmunoprecipitation with anti-EpsL antibodies, it was found to interact with EpsL. {Delta}83-EpsM did not appear to interact with EpsL as well as full-length EpsM, however, suggesting that residues 1 to 82, while not absolutely necessary for interaction with EpsL, may play some role in the correct folding or positioning of the EpsL binding site. The results of these experiments are consistent with our suggestion that removal of 82 residues from the C terminus of EpsM results in a protein which is no longer able to interact with EpsL, and taken together with the data for the other C-terminally truncated variants, they suggest that residues 83 to 99 in EpsM are critical for a stable interaction with EpsL. These results are in agreement with the finding that all truncated GFP-EpsM chimeras but GFP-EpsM{Delta}82 exert a dominant-negative effect on secretion in wild-type V. cholerae. The results are also in agreement with yeast two-hybrid studies, which showed that residues 84 to 162 of GspM, the E. chrysanthemi EpsM homologue, interact with GspL, the EpsL homologue (23).

The X-ray structure of the periplasmic domain of EpsM begins with residue 86 of the full-length protein (2). In the structural model, residues 88 to 99 form an {alpha}-helix on the opposite side of the EpsM dimer interface (highlighted in yellow in Fig. 8). We propose that this {alpha}-helix interacts directly with EpsL. Similar to the region important for dimerization of EpsM, residues 88 to 99 are located in a region of the protein that displays high sequence homology to other members of the M protein family (2). Whether the most conserved residues are the ones directly involved in binding of EpsL remains to be determined.


Figure 8
View larger version (23K):
[in this window]
[in a new window]

  		FIG. 8. Structure of EpsM dimer indicating the putative EpsL binding site. The ribbon structure (Protein Data Bank accession no. 1UV7) of the periplasmic EpsM dimer is shown, with individual monomers colored in blue and green while the putative EpsL binding site is highlighted in yellow (3D-Mol Viewer).

Using peptide linker insertional mutagenesis of XcpZ, the EpsM homologue of Pseudomonas aeruginosa, Robert and colleagues identified four mutant proteins whose sensitivity to proteolysis could not be rescued by the EpsL homolog XcpY. Therefore, insertions at positions 65, 127, 128, and 141 were suggested to prevent contact with XcpY (25). The crystal structure of the periplasmic domain of EpsM and our findings that GFP-EpsM{Delta}30 is capable of interaction with full-length EpsM, while GFP-EpsM{Delta}66 is not, suggest that the insertions at positions 127, 128, and 141 of XcpZ may interfere with its ability to dimerize. Since the structure of the upstream region of the periplasmic domain of EpsM that corresponds to position 65 in XcpY is not known, the effect of the insertion of a peptide at position 65 is not as easily explained; however, the insertion may sterically hinder the interaction between XcpY and the XcpY-binding {alpha}-helix of XcpZ.

In this study, we have mapped critical sites of interaction between EpsM and both EpsL and another molecule of EpsM to the C-terminal residues 84 to 99 and 100 to 135, respectively. It is interesting, however, that while the last 14 residues of EpsM were not necessary for either interaction with EpsL or dimerization of EpsM, the GFP-EpsM{Delta}14 construct could not complement the epsM mutant and inhibited secretion when expressed in wild-type cells. The ability of GFP-EpsM{Delta}14 to form stable homodimers and interact with EpsL yet be unable to support secretion suggests that EpsM may harbor an additional interactive site that is critical for its function. Since it is localized to the very C terminus of EpsM, it may be the site of protein-protein interactions with another Eps component. Intriguingly, when the crystal structure of the periplasmic domain of EpsM was solved, the C-terminal portion was observed to form a cleft that may constitute a binding site for a ligand (2). The identity of the putative ligand and, furthermore, the function of this cleft remain to be determined.


arrow
ACKNOWLEDGMENTS
 
This work was supported by grant AI49294 from the National Institutes of Health (to M.S.). T.L.J. was supported in part by National Institutes of Health training grant AI007258.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Department of Microbiology and Immunology, University of Michigan Medical School, 1150 West Medical Center Drive, Ann Arbor, MI 48109. Phone: (734) 764-3552. Fax: (734) 764-3562. E-mail: mariasan{at}umich.edu Back

{triangledown} Published ahead of print on 5 October 2007. Back

{dagger} Present address: Western Michigan University, Kalamazoo, MI. Back


arrow
REFERENCES
 
    1
  1. Abendroth, J., P. Murphy, M. Sandkvist, M. Bagdasarian, and W. G. J. Hol. 2005. The X-ray structure of the type II secretion system complex formed by the N-terminal domain of EpsE and the cytoplasmic domain of EpsL of Vibrio cholerae. J. Mol. Biol. 348:845-855.[CrossRef][Medline]
    2. 2
  2. Abendroth, J., A. E. Rice, K. McLuskey, M. Bagdasarian, and W. G. J. Hol. 2004. The crystal structure of the periplasmic domain of the type II secretion system protein EpsM from Vibrio cholerae: the simplest version of the ferredoxin fold. J. Mol. Biol. 338:585-596.[CrossRef][Medline]
    3. 3
  3. Ball, G., V. Chapon-Herve, S. Bleves, G. Michel, and M. Bally. 1999. Assembly of XcpR in the cytoplasmic membrane is required for extracellular protein secretion in Pseudomonas aeruginosa. J. Bacteriol. 181:382-388.[Abstract/Free Full Text]
    4. 4
  4. Bleves, S., A. Lazdunski, and A. Filloux. 1996. Membrane topology of three Xcp proteins involved in exoprotein transport by Pseudomonas aeruginosa. J. Bacteriol. 178:4297-4300.[Abstract/Free Full Text]
    5. 5
  5. Camberg, J. L., T. L. Johnson, M. Patrick, J. Abendroth, W. G. J. Hol, and M. Sandkvist. 2007. Synergistic stimulation of EpsE ATP hydrolysis by EpsL and acidic phospholipids. EMBO J. 26:19-27.[CrossRef][Medline]
    6. 6
  6. Casadaban, M. J., and S. N. Cohen. 1980. Analysis of gene control signals by DNA fusion and cloning in Escherichia coli. J. Mol. Biol. 138:179-207.[CrossRef][Medline]
    7. 7
  7. Cianciotto, N. P. 2005. Type II secretion: a protein secretion system for all seasons. Trends Microbiol. 13:581-588.[CrossRef][Medline]
    8. 8
  8. Douet, V., L. Loiseau, F. Barras, and B. Py. 2004. Systematic analysis, by the yeast two-hybrid, of protein interaction between components of the type II secretory machinery of Erwinia chrysanthemi. Res. Microbiol. 155:71-75.[Medline]
    9. 9
  9. Filloux, A. 2004. The underlying mechanisms of type II protein secretion. Biochim. Biophys. Acta 1694:163-179.[Medline]
    10. 10
  10. Fullner, K. J., and J. J. Mekalanos. 1999. Genetic characterization of a new type IV-A pilus gene cluster found in both classical and El Tor biotypes of Vibrio cholerae. Infect. Immun. 67:1393-1404.[Abstract/Free Full Text]
    11. 11
  11. Furste, J. P., W. Pansegrau, R. Frank, H. Blocker, P. Scholz, M. Bagdasarian, and E. Lanka. 1986. Molecular cloning of the plasmid RP4 primase region in a multi-host-range tacP expression vector. Gene 48:119-131.[CrossRef][Medline]
    12. 12
  12. Guzman, L. M., D. Belin, M. J. Carson, and J. Beckwith. 1995. Tight regulation, modulation, and high-level expression by vectors containing the arabinose p-BAD promoter. J. Bacteriol. 177:4121-4130.[Abstract/Free Full Text]
    13. 13
  13. Hirst, T. R., J. Sanchez, J. B. Kaper, S. J. Hardy, and J. Holmgren. 1984. Mechanism of toxin secretion by Vibrio cholerae investigated in strains harboring plasmids that encode heat-labile enterotoxins of Escherichia coli. Proc. Natl. Acad. Sci. USA 81:7752-7756.[Abstract/Free Full Text]
    14. 14
  14. Housby, J. N., J. D. Thomas, S. D. Wharam, P. J. Reeves, and G. P. Salmond. 1998. Conditional mutations in OutE and OutL block exoenzyme secretion across the Erwinia carotovora outer membrane. FEMS Microbiol. Lett. 165:91-102.[CrossRef][Medline]
    15. 15
  15. Johnson, T. L., J. Abendroth, W. G. J. Hol, and M. Sandkvist. 2006. Type II secretion: from structure to function. FEMS Microbiol. Lett. 255:175-186.[CrossRef][Medline]
    16. 16
  16. Kaper, J. B., J. G. Morris, Jr., and M. M. Levine. 1995. Cholera. Clin. Microbiol. Rev. 8:48-86.[Abstract]
    17. 17
  17. Marsh, J. W., and R. K. Taylor. 1998. Identification of the Vibrio cholerae type 4 prepilin peptidase required for cholera toxin secretion and pilus formation. Mol. Microbiol. 29:1481-1492.[CrossRef][Medline]
    18. 18
  18. Michel, G., S. Bleves, G. Ball, A. Lazdunski, and A. Filloux. 1998. Mutual stabilization of the XcpZ and XcpY components of the secretory apparatus in Pseudomonas aeruginosa. Microbiology 144:3379-3386.[Abstract/Free Full Text]
    19. 19
  19. Overbye, L. J., M. Sandkvist, and M. Bagdasarian. 1993. Genes required for extracellular secretion of enterotoxin are clustered in Vibrio cholerae. Gene 132:101-106.[CrossRef][Medline]
    20. 20
  20. Possot, O. M., M. Gerard-Vincent, and A. P. Pugsley. 1999. Membrane association and multimerization of secreton component PulC. J. Bacteriol. 181:4004-4011.[Abstract/Free Full Text]
    21. 21
  21. Possot, O. M., G. Vignon, N. Bomchil, F. Ebel, and A. P. Pugsley. 2000. Multiple interactions between pullulanase secreton components involved in stabilization and cytoplasmic membrane association of PulE. J. Bacteriol. 182:2142-2152.[Abstract/Free Full Text]
    22. 22
  22. Py, B., L. Loiseau, and F. Barras. 1999. Assembly of the type II secretion machinery of Erwinia chrysanthemi: direct interaction and associated conformational change between OutE, the putative ATP-binding component and the membrane protein OutL. J. Mol. Biol. 289:659-670.[CrossRef][Medline]
    23. 23
  23. Py, B., L. Loiseau, and F. Barras. 2001. An inner membrane platform in the type II secretion machinery of gram-negative bacteria. EMBO Rep. 2:244-248.[CrossRef][Medline]
    24. 24
  24. Robert, V., A. Filloux, and G. P. Michel. 2005. Subcomplexes from the Xcp secretion system of Pseudomonas aeruginosa. FEMS Microbiol. Lett. 252:43-50.[CrossRef][Medline]
    25. 25
  25. Robert, V., F. Hayes, A. Lazdunski, and G. P. Michel. 2002. Identification of XcpZ domains required for assembly of the secreton of Pseudomonas aeruginosa. J. Bacteriol. 184:1779-1782.[Abstract/Free Full Text]
    26. 26
  26. Sandkvist, M. 2001. Type II secretion and pathogenesis. Infect. Immun. 69:3523-3535.[Free Full Text]
    27. 27
  27. Sandkvist, M., M. Bagdasarian, S. P. Howard, and V. J. Dirita. 1995. Interaction between the autokinase EpsE and EpsL in the cytoplasmic membrane is required for extracellular secretion in Vibrio cholerae. EMBO J. 14:1664-1673.[Medline]
    28. 28
  28. Sandkvist, M., T. R. Hirst, and M. Bagdasarian. 1987. Alterations at the carboxyl terminus change assembly and secretion properties of the B subunit of Escherichia coli heat-labile enterotoxin. J. Bacteriol. 169:4570-4576.[Abstract/Free Full Text]
    29. 29
  29. Sandkvist, M., L. P. Hough, M. M. Bagdasarian, and M. Bagdasarian. 1999. Direct interaction of the EpsL and EpsM proteins of the general secretion apparatus in Vibrio cholerae. J. Bacteriol. 181:3129-3135.[Abstract/Free Full Text]
    30. 30
  30. Sandkvist, M., J. M. Keith, M. Bagdasarian, and S. P. Howard. 2000. Two regions of EpsL involved in species-specific protein-protein interactions with EpsE and EpsM of the general secretion pathway in Vibrio cholerae. J. Bacteriol. 182:742-748.[Abstract/Free Full Text]
    31. 31
  31. Sandkvist, M., L. O. Michel, L. P. Hough, V. M. Morales, M. Bagdasarian, M. Koomey, V. J. Dirita, and M. Bagdasarian. 1997. General secretion pathway (eps) genes required for toxin secretion and outer membrane biogenesis in Vibrio cholerae. J. Bacteriol. 179:6994-7003.[Abstract/Free Full Text]
    32. 32
  32. Scott, M. E., Z. Y. Dossani, and M. Sandkvist. 2001. Directed polar secretion of protease from single cells of Vibrio cholerae via the type II secretion pathway. Proc. Natl. Acad. Sci. USA 98:13978-13983.[Abstract/Free Full Text]

Journal of Bacteriology, December 2007, p. 9082-9089, Vol. 189, No. 24
0021-9193/07/$08.00+0     doi:10.1128/JB.01256-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Lybarger, S. R., Johnson, T. L., Gray, M. D., Sikora, A. E., Sandkvist, M. (2009). Docking and Assembly of the Type II Secretion Complex of Vibrio cholerae. J. Bacteriol. 191: 3149-3161 [Abstract] [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Johnson, T. L.
Right arrow Articles by Sandkvist, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Johnson, T. L.
Right arrow Articles by Sandkvist, M.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»