Genetic Mapping of Secretion and Functional Determinants of the Vibrio cholerae TcpF Colonization Factor

Bacterial strains and growth conditions.All bacterial strains and plasmids used in this study are shown in Table 1. Strains were stored at −70°C in Luria-Bertani (LB) broth containing 30% (vol/vol) glycerol. V. cholerae was grown in LB broth with a starting pH of 6.5 at 30°C for 16 h (TCP-inducing conditions). When necessary, antibiotics were used in culture at the following concentrations: ampicillin (Ap), 100 μg/ml; gentamicin (Gm), 30 μg/ml; chloramphenicol (Cm), 15 μg/ml; streptomycin (Sm), 100 μg/ml generally or 1 mg/ml for selection for loss of integrated plasmids from V. cholerae during allelic exchange (43). 5-Bromo-4-chloro-3-indolyl-β-d-galactoside (X-Gal) was used in LB agar at 40 μg/ml for enumeration in the competitive index experiments. Arabinose was used at a final concentration of 0.01%. Primers for the construction of each plasmid or strain are listed in Table 2.

Construction of tcpF internal deletions. V. cholerae tcpF internal deletions were constructed as follows. A sequence from the V. cholerae database identified by a BLAST search as encoding TcpF provided the basis for the design of oligonucleotide primers F25A, F25B, F25C, F25D, F80A, F80B, F80C, F80D, F140A, F140B, F140C, F140D, F221A, F221B, F221C, F221D, F260A, F260B, F260C, F260D, F270A, F270B, F270C, F270/290D, F290A, F290B, and F290C (Table 2), which were used to amplify chromosomal sequences from O395 that flank the specific internal deletion sites within the tcpF gene. The upstream fragment of approximately 500 bp and the downstream 500-bp fragment were inserted into the allelic exchange vector pKAS32. The tcpF internal deletions were then introduced into the chromosome of V. cholerae strain O395 by allelic exchange (43), and the constructs were verified by PCR.

Linker scanning mutagenesis.Random mutagenesis of V. cholerae tcpF was achieved using the GPS Linker Scanning system (New England Biolabs) and has been described previously by Nye et al. (30). Briefly, TnsABC* transposase was used to insert a Tn7-based transprimer into pTK10 at random locations. The transposition reaction causes a staggered cut at the transprimer insertion site that results in a 5-bp duplication. The resultant plasmids were selected on Cm and screened by PCR for insertions within the tcpF gene. Subsequent PmeI digestion and religation were performed to remove the bulk of the transprimer, after which plasmids were designated with the suffix “P.” Religation of the mutagenized plasmids ultimately leaves a 15-bp insertion 5 bp from the duplication at the transprimer insertion site and 10 bp from the PmeI site. In four out of six frames, these 15 bp cause an insertion of 5 amino acids. In two out of six frames, an ocher mutation is created. Mutagenized plasmids were characterized by DNA sequencing of tcpF using primers FHISA and FHISB to determine the location of the 5-amino-acid insertion or nonsense codon. Plasmids encoding representative mutations were analyzed in V. cholerae O395 ΔtcpF. Pentapeptide insertions in the TcpF protein are named by the amino acid residue N-terminal to the insertion followed by the amino acid sequence of the insertion itself. Induction of the pBAD22-based TcpF derivatives was achieved using 0.01% arabinose with V. cholerae strains.

Cellular fractionation.Cells grown under TCP-expressing conditions were harvested by centrifugation, and culture supernatant fractions were removed and passed through a 0.4-μm-pore-size filter. Cellular fractionation was performed as previously described by Bose and Taylor (2). Briefly, pellets were washed, suspended in phosphate-buffered saline, and treated with polymyxin B sulfate (10 mg/ml) for 10 min on ice to release periplasmic contents. Spheroplasts were separated from the soluble periplasmic fraction by centrifugation at 10,000 × g for 10 min. In order to obtain total membranes, the spheroplast fraction was resuspended in 10 mM Tris-Cl (pH 8.0) and passed through a French pressure cell at 8,000 lb/in² to lyse the cells. After low-speed centrifugation at 10,000 × g at room temperature for 10 min to remove any unbroken cells, the lysate was centrifuged at 100,000 × g at 4°C for 30 min to pellet the membrane fraction. The inner and outer membranes were differentially solubilized as follows. The membrane pellet was resuspended in 10 mM Tris-Cl-100 mM NaCl containing 2.5% Sarkosyl, held at room temperature for 30 min, and centrifuged at 200,000 × g at 12°C for 1 h. The supernatant was retained as the inner membrane fraction, and the pellet containing the outer membrane was resuspended in 200 μl of 10 mM Tris-Cl (pH 8.0).

As previously described by Tripathi and Taylor (48), the purity of subcellular fractions was determined by immunoblotting and enzyme assays for known localization markers, as follows. EpsL and TcpC immunoblots were used as inner membrane and outer membrane localization markers, respectively (2, 38). As a cytoplasmic marker to test for cell lysis, glucose-6-phosphate dehydrogenase (G6PDH) activity was assayed (14). G6PDH substrate vials containing glucose-6-phosphate along with the phenazine methosulfate NADP, substrates, and 2,6-dichlorophenol indophenol (DCP-IP) (Sigma) were reconstituted with 0.3 M Tris-Cl (pH 8.0) just prior to use, as recommended by the manufacturer. One milliliter of a 1:2 dilution of the cytoplasmic fraction or 1 ml of a culture supernatant or outer membrane fraction was added to a 4.5-ml cuvette containing the reconstituted substrate, immediately followed by an overlay of 2 ml of mineral oil. The decrease in the optical density at 600 nm was measured at 1-min intervals for 5 min to monitor the reduction of blue DCP-IP to a colorless form (1). Specific activity was calculated as nanomoles of DCP-IP reduced per minute per milligram of total protein. The culture supernatant, periplasmic, and outer membrane fractions from each strain were tested for the presence or absence of these inner membrane and outer membrane control proteins and for G6PDH activity.

Protein electrophoresis and Western blot analysis.Proteins from various cellular fractions were diluted in 2× sodium dodecyl sulfate-polyacrylamide gel electrophoresis loading buffer, and fractions containing equal amounts of total protein were separated on a sodium dodecyl sulfate-12.5% polyacrylamide gel. Protein levels were estimated using the bicinchoninic acid protein assay kit (Pierce) with bovine serum albumin as a standard. For total-protein detection, gels were stained with 0.1% Coomassie brilliant blue in 40% methanol for 1 min and destained with 50% methanol. For immunodetection, proteins were electroblotted onto a nitrocellulose membrane at 4°C in transfer buffer (25 mM Tris, 192 mM glycine, 20% methanol [pH 8.3]) with a wet transfer apparatus (Bio-Rad Laboratories) (46). The membrane was blocked with 3% bovine serum albumin in 1× Tris-buffered saline with 0.1% Tween (TBST) for 2 h. Antisera containing polyclonal antibodies raised against TcpF-glutathione S-transferase were then applied at a 1:10,000 dilution in blocking buffer. A horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin G secondary antibody was added at a 1:10,000 dilution in TBST. Reactive protein bands were detected using an ECL kit (Amersham).

In vivo competition index analysis.For the in vivo competitive index determinations, strains to be tested were grown under TCP-inducing conditions and then mixed with equal cell numbers of the reference strain, SJK9. Five- to seven-day-old CD1 mice from mixed litters were orally inoculated with 50 μl of the mixture with a final concentration of 1% arabinose and were incubated at 30°C for 24 h. The bacteria were then recovered by homogenizing harvested intestines in 3 ml of 10% glycerol in water with a Tissue Terror homogenizer (Fisher Scientific, Los Angeles, CA). The homogenate was appropriately diluted and plated onto solid medium containing streptomycin and X-Gal. The competitive index was calculated by dividing the output ratio (ratio of the output of the test strain to that of the reference strain recovered from the intestine) by the input ratio (ratio of the input of the test strain to that of the reference strain in the original inoculum). The competitive index is the average for at least five mice for each strain tested, and the Student t test was used to determine which strains were significantly defective in colonization compared to the positive control.

Construction of tcpF mutations using linker scanning mutagenesis.Since TcpF presently has no known homologs or obvious conserved domains, targeted mutagenesis of regions of the protein likely to be important for secretion through the TCP apparatus or for in vivo colonization could not be undertaken. Instead, we employed a random mutagenesis approach to identify regions involved in the secretion and function of TcpF by using the New England Biolabs GPS LS Linker Scanning system (see Materials and Methods). Mutagenesis generated 33 unique linker insertions encoding 5 amino acids each and 18 unique ocher truncation mutations in tcpF. Twenty of the 33 insertions produced stable protein derivatives as determined by anti-TcpF immunoblotting (data not shown) and were studied further. None of the 18 truncated proteins were found to be stable (data not shown), suggesting that the C terminus may contribute to protein stability. The plasmids expressing derivatives of TcpF that were stable in V. cholerae RT4372 were sequenced, and the location of each insertion was determined. A graphical representation of the locations of these TcpF variants derived from linker scanning mutagenesis is given in Fig. 1A, and their respective 5-amino-acid insertions are listed in Fig. 1B. Internal deletions (discussed below) were also constructed throughout tcpF and are represented in Fig. 1.

• Open in new tab
• Download powerpoint

FIG. 1.

Summary of the locations of tcpF mutations. (A) Amino acid sequence of mature TcpF. Shaded letters represent the Sec-dependent signal sequence. The locations of the linker scanning mutations are indicated above the TcpF sequence. Internal deletions are underlined and labeled below the TcpF sequence. (B) Alteration of each TcpF derivative. Plasmids carrying each tcpF linker scan (LS) mutation are listed, and the insertions are given according to the nomenclature described in Materials and Methods. Chromosomal internal deletions of tcpF are indicated by numbers 1 through 7, and the amino acids (aa) deleted are shown.

Regions near the C terminus of TcpF are required for its function in colonization.To begin to define the determinants important for the secretion and function of TcpF, V. cholerae strains expressing stable TcpF derivatives (Fig. 1) were first analyzed by Western blotting for their abilities to secrete the variant TcpF proteins into the culture supernatant. Eleven of the 20 TcpF derivatives were detected in the culture supernatant (Fig. 2A). As seen previously, TcpF is represented by multiple bands (19), and it is not yet known if one or all these forms of the protein are important for its function in colonization. The lack of G6PDH activity in these culture supernatant fractions compared to the cytoplasmic fraction suggests that little or no cell lysis, which could contribute to the detection of these different forms, is occurring (data not shown). Each of the strains expressing TcpF derivatives that were detected in the culture supernatant were competed in the infant mouse colonization model against the reference strain expressing wild-type TcpF, SJK9 (Fig. 2B). The positive control was competition of SJK7 against SJK9, each expressing wild-type TcpF, which gave a competitive index of about 1, indicating that these strains colonize the infant mouse equally. The negative control was competition of SJK71 (empty vector) against SJK9 (expressing wild-type TcpF), which gave a competitive index of approximately 1 log unit less than 1. Student's t test showed that this difference was significant, indicating that SJK71 is defective in colonization, as predicted. As shown in Fig. 2, plasmids encoding TcpF derivatives with insertions near the N terminus (pSJK18, pSJK20, pSJK26, pSJK14) were not defective in colonization. However, plasmids encoding TcpF derivatives with insertions from the middle of the protein to near the C terminus (pSJK43, pSJK21, pSJK55, pSJK16, pSJK42, pSJK44, pSJK19) were significantly defective in colonization (Fig. 2B). The amount of protein produced by the strain containing plasmid pSJK21 was less than the amount of protein produced by the reference strain in vitro, which could account for its defect in colonization in vivo. However, the strain containing plasmid pSJK18 also produced less protein than the reference strain, and its function in colonization was not affected, suggesting that the amount of protein expressed by the strain containing plasmid pSJK21 would be sufficient for colonization if the protein were functional. Taken together, these results suggest that regions near the C terminus are required for the function of TcpF in the colonization and pathogenesis of V. cholerae.

• Open in new tab
• Download powerpoint

FIG. 2.

Analysis of TcpF derivatives in colonization. TcpF derivatives constructed by linker scanning mutagenesis that were detected in the culture supernatant were analyzed for their functions in colonization. (A) Stable TcpF derivatives in the culture supernatant were loaded onto the protein gel according to the location of the 5-amino-acid insertion in TcpF, from the N terminus to the C terminus, and were detected by Western blotting using anti-TcpF polyclonal antibodies. Plasmids encoding the TcpF derivatives are given above the Western blot, followed by the alteration of TcpF in brackets. wt, wild type. (B) The competitive index was determined for each TcpF derivative by competing each strain in the infant mouse model against the reference strain, SJK9, expressing wild-type TcpF. The competitive index of each TcpF derivative was graphed from the N terminus to the C terminus according to the location of the 5-amino-acid insertion. Each circle represents a mouse, and each horizontal line represents the average competitive index for the five or six mice tested for one strain. The positive (+) control is the competition between SJK9 and SJK7, both expressing wild-type TcpF. The negative (−) control is the competition between SJK9, expressing wild-type TcpF, and SJK71, harboring the empty vector. Student's t test was used to determine which of the strains were significantly defective in colonization compared to the positive control. TcpF derivatives with 5-amino-acid insertions that mapped toward the C terminus were found to be significantly defective in colonization, as indicated by asterisks (P < 0.01).

Multiple regions throughout TcpF influence extracellular secretion.In order to elucidate the molecular determinants required for proper secretion of TcpF by the TCP apparatus, subcellular localization of TcpF derivatives was performed. The subcellular fractionation protocol was able to produce fairly pure fractions (data not shown), as indicated by the presence or absence of the inner membrane protein marker EpsL, the outer membrane marker TcpC, or the activity of G6PDH (see Materials and Methods). V. cholerae strains expressing the TcpF linker scan derivatives that could not be detected in the culture supernatant were examined to determine if these TcpF derivatives could be detected in the periplasm or the outer membrane. All of the known components of the TCP type IV pilus biogenesis apparatus are required for extracellular secretion of TcpF, including the pilin monomer, TcpA (19). In mutants missing any of these components, TcpF accumulates in the periplasm (19). Overexpression of tcpF in V. cholerae (SJK7) led to normal amounts of TcpF in the culture supernatant and also some periplasmic accumulation of the protein, likely due to higher than normal levels of TcpF that occur when the gene is expressed from the plasmid (Fig. 3). Of the 20 mutants that produced stable protein derivatives from linker scanning mutagenesis, 9 mutants were totally devoid of full-length protein in the culture supernatant. However, these protein derivatives yielded full-length products that were detected in the periplasm (Fig. 3). Among these secretion-defective TcpF derivatives, six had insertions that clustered in two main regions. Plasmids pSJK29, pSJK30, pSJK47, and pSJK41 encode TcpF derivatives with 5-amino-acid insertions located near the N terminus, and plasmids pSJK63 and pSJK52 encode TcpF derivatives with insertions located near the C terminus (Fig. 1). The remaining three plasmids (pSJK37, pSJK31, and pSJK61) encode TcpF derivatives with insertions distributed at various positions throughout the protein. Strains containing plasmid pSJK63 or pSJK52, encoding a TcpF derivative that contains an insertion located near the C terminus, or pSJK47, encoding a TcpF derivative that contains an insertion located near the N terminus, produced TcpF derivatives that accumulated in the outer membrane fraction to a greater degree than wild-type TcpF (Fig. 3). To determine if these protein derivatives were held in a form associated with the outer membrane secretin, TcpC, these genes were expressed in strains lacking tcpC and tcpF and were blotted with anti-TcpF polyclonal antibodies. However, these TcpF derivatives still accumulated in the outer membrane, even in the absence of the outer membrane secretin (data not shown). To determine if the 5-amino-acid insertion might cause these protein derivatives to form β-barrels, leading to their accumulation in the outer membrane independently of interactions with other proteins, PSORT (http://www.psort.org/psortb/index.html ) was utilized to analyze the amino acid sequences of these TcpF derivatives. Based on this computer algorithm, these TcpF derivatives had scores similar to that of wild-type TcpF and were not predicted to localize to the outer membrane. This finding suggests that either TcpF is associating with other components of the TCP biogenesis apparatus localized to the outer membrane or there are other components not encoded within the tcp operon that aid in mediating proper extracellular secretion of TcpF. Perhaps these three protein derivatives encoded on plasmids pSJK47, pSJK63, and pSJK52 retain some ability to initiate extracellular export but are defective in its completion, leading to a membrane-bound intermediate, whereas the other six protein derivatives are completely defective in interaction with the export machinery.

• Open in new tab
• Download powerpoint

FIG. 3.

Western blot analysis of TcpF linker scan derivatives and the mutant with internal deletion 1 for extracellular secretion. Western blotting with anti-TcpF polyclonal antibodies was performed on the culture supernatant (CS), periplasmic (P), and outer membrane (OM) fractions. (A) TcpF derivative fractions were loaded according to the location of the 5-amino-acid insertion, from the N terminus to the C terminus. Full-length TcpF derivatives constructed by linker scanning mutagenesis were detected in either the periplasm or the outer membrane fraction, or both, but were not detected in the culture supernatant. (B) The mutant with internal deletion 1 was detected in the periplasm but not in the outer membrane or the culture supernatant.

The six protein derivatives encoded on plasmids pSJK29, pSJK30, pSJK47, pSJK41, pSJK37, and pSJK31, which have 5-amino-acid insertions located close to the N terminus, were found to be partially degraded into two products in the periplasm: degradation product A (approximately 18 kDa) and degradation product B (approximately 13 kDa) (Fig. 3). Degradation product B was also detected at low levels in the culture supernatant. When mutants lacking the secretin, TcpC, were transformed with the plasmids encoding these protein derivatives, degradation product B was still detected in the culture supernatant (data not shown), indicating that degradation product B is released from the bacterial cell independently of the TCP apparatus. Degradation of these protein derivatives must have occurred in the periplasm and not in the supernatant, since full-length TcpF requires TcpC for secretion (19). Wild-type TcpF (SJK7) is not degraded into products A and B (Fig. 3); thus, the 5-amino-acid insertion in TcpF derivatives encoded on plasmids pSJK29, pSJK30, pSJK47, pSJK41, pSJK37, and pSJK31 may have caused changes in the protein structure, which could account for their lack of secretion. It has been well documented that accumulation of proteins in the periplasm can have harmful effects on cell viability (16, 18, 42, 44).

HtrA (DegP), characterized in E. coli, is a ubiquitous housekeeping protein found throughout the prokaryotes and eukaryotes. One of its primary functions in gram-negative prokaryotes is to act as a protease under conditions where proteins accumulate in the periplasm (5; reviewed in references 23 and 31). To determine whether HtrA has a role in degrading the unstable TcpF derivatives, a strain lacking both htrA and tcpF (SJK94) was transformed with plasmids encoding these proteins. In this strain, degradation products were no longer detected in the periplasm, and full-length protein was restored to wild-type levels, indicating that HtrA degrades these protein derivatives (Fig. 4). However, even though these TcpF derivatives were stable in the htrA background, they still could not be detected in the culture supernatant (Fig. 4), suggesting that the regions identified by these linker insertions are important for efficient secretion of TcpF.

• Open in new tab
• Download powerpoint

FIG. 4.

Lack of HtrA restores the stability of protein derivatives that are defective in extracellular secretion. The TcpF protein derivatives that were degraded into products A and B, as shown in Fig. 3, were analyzed in the presence and absence of HtrA by Western blotting of the periplasmic (P) and culture supernatant (CS) fractions with anti-TcpF polyclonal antibodies. Although HtrA restored these protein derivatives to a stable full-length product in the periplasm, they still could not be detected in the culture supernatant.

Limited regions of primary sequences common among secretion-proficient TcpF proteins from environmental isolates provide a way to identify secretion determinants.To date, tcpF genes from six pathogenic strains of V. cholerae have been sequenced, and all of these strains have identical TcpF amino acid sequences. However, a study by Mukhopadhyay et al. (26) found four different alleles of tcpF in strains isolated from the environment in Kolkata, India, which ranged from 31% to 70% identity at the protein level to TcpF proteins from pathogenic strains. Comparison of all five sequences in a Clustal W alignment (47) reveals 15% similarity throughout mature TcpF when identical, conserved, and semiconserved residues are considered. For clarity, TcpF from pathogenic strains is referred to below as “pathogenic TcpF,” and TcpF from environmental strains is referred to below as “environmental TcpF.” Alignment of these environmental TcpF proteins with the pathogenic TcpF protein revealed multiple regions throughout the protein that were conserved among the strains, and the N-terminal region was more conserved than the C-terminal region (Fig. 5). Due to the regional similarities of these environmental TcpF proteins to the pathogenic TcpF protein, we decided to utilize these natural derivatives of TcpF to further define secretion and function determinants within the highly conserved pathogenic TcpF by expressing the environmental TcpF proteins in the pathogenic strain and testing them for secretion and function.

• Open in new tab
• Download powerpoint

FIG. 5.

Alignment of the pathogenic TcpF protein with divergent environmental TcpF proteins. Clustal W alignment reveals residues that are identical (*), conserved (:), and semiconserved (.) between the pathogenic TcpF protein and environmental TcpF proteins. Plasmids encoding linker scan protein derivatives that were found to influence secretion are annotated above the alignment with respect to the location of the 5-amino-acid insertion. The level of identity between each environmental TcpF protein and pathogenic TcpF, as reported by Mukhopadhyay et al. in 2001 (26), is given on the left. (Adapted from reference 26 with permission.)

The three environmental tcpF genes with the least identity at the protein level to pathogenic TcpF (SCE4, SCE256, and SCE263, with 66%, 33%, and 31% identity, respectively), were cloned into pBAD22 and expressed in the pathogenic strain lacking endogenous tcpF (RT4372). The presence of a restriction enzyme site in SCE200 (with 70% identity to the pathogenic strain) prevented further analysis of tcpF from this strain. Each of the environmental TcpF proteins examined was secreted from the pathogenic strain, as revealed by Coomassie brilliant blue staining of the culture supernatant and periplasmic fractions (Fig. 6A). None of the environmental TcpF proteins accumulated in the periplasm to any greater degree than pathogenic TcpF encoded on plasmid pTK10, indicating that the environmental TcpF proteins are all secreted with the same efficiency as pathogenic TcpF (Fig. 6A). Secretion of these environmental TcpF proteins is also specific to the TCP apparatus, since a strain lacking the secretin, TcpC, could not mediate the secretion of these proteins (Fig. 6B). Consistent with this, several of the TcpF linker scan derivatives encoded on pSJK29, pSJK30, pSJK47, pSJK41, pSJK61 and pSJK63, which were found to have an influence on secretion, disrupted some of these conserved regions or conserved residues (Fig. 5).

• Open in new tab
• Download powerpoint

FIG. 6.

Secretion of environmental TcpF proteins by the pathogenic strain. (A) Coomassie brilliant blue staining of the culture supernatant (CS) and periplasmic (P) fractions revealed that the pathogenic strain mediates the secretion of the environmental TcpF proteins. (B) In the absence of TcpC, the TCP apparatus secretin, these proteins are no longer detected in the culture supernatant, indicating that this secretion is dependent on the TCP apparatus.

Internal deletions were also constructed throughout tcpF based on the locations of the conserved regions and the linker scan data (Fig. 1). These deletions were recombined into the chromosome by allelic exchange. As seen previously with the linker scanning mutagenesis, deletions in the C terminus that were constructed in SJK216 and SJK218 (internal deletions 6 and 7) resulted in unstable proteins, suggesting that the C terminus is required for the structural stability of the protein (data not shown). Internal deletions 2 through 5, constructed throughout the gene, also resulted in unstable proteins (data not shown); however, internal deletion 1 generated a stable protein that was detected in the periplasm (Fig. 3). Internal deletion 1 removed a region on the N terminus rich in the amino acids tyrosine, serine, and threonine, all of which contain hydroxyl side chains. Alignment of the environmental TcpF proteins with the pathogenic TcpF protein revealed that this region rich in hydroxyl residues is the most highly conserved among all these strains (Fig. 5). Western blotting revealed that this protein derivative accumulated in the periplasm to a greater extent than the wild type and was not detected in the culture supernatant or the outer membrane (Fig. 3), indicating that a region within the N terminus is important for the secretion of the protein. Taken together, these data suggest that multiple regions throughout TcpF are important for recognition by the TCP apparatus and for the ultimate secretion of this protein.

Lack of colonization mediated by environmental TcpF proteins suggests that regions divergent between the pathogenic and environmental proteins are important for TcpF function.Since these environmental proteins are secreted from the pathogenic strain, we wanted to determine if they could mediate the colonization of the infant mouse. Positive and negative controls were performed as described in the legend to Fig. 2. Each of the strains expressing environmental TcpF proteins was competed against SJK9, expressing pathogenic TcpF, and tested for its ability to colonize five infant mice. As seen in Fig. 7, each strain producing an environmental TcpF protein was significantly defective in colonization compared to the positive control (SJK7) (P < 0.016). The amount of environmental protein in the culture supernatants of the strains carrying pSJK245 and pSJK247 was slightly smaller than the amount of protein in the culture supernatant of the reference strain (Fig. 6A), which could account for their defect in colonization. However, the strain carrying pSJK244, which encodes a TcpF derivative with the highest identity to the pathogenic strain (66%), secretes quantities of TcpF similar to those secreted by the reference strain (Fig. 6A) and is still defective in colonization. Although the environmental TcpF proteins contain regions important for secretion, they must be divergent in regions important for function. Alignment of the TcpF protein of environmental strain SCE4 (which shares the highest identity with pathogenic TcpF), encoded on pSJK244, with pathogenic TcpF, encoded on pTK10, revealed regions divergent from one another that are localized mainly from the middle of the protein to the C-terminal region (Fig. 8). The strains carrying plasmids pSJK21, pSJK16, pSJK42, pSJK55, and pSJK19, which are defective in colonization (Fig. 2), express TcpF derivatives with 5-amino-acid insertions that disrupt, or are located near, regions divergent between the environmental and pathogenic TcpF proteins (Fig. 8). The correlation of these data sets provides further evidence suggesting that at least some of these regions define portions of the TcpF functional determinants with respect to mediating the colonization and pathogenesis of V. cholerae.

• Open in new tab
• Download powerpoint

FIG. 7.

Environmental TcpF proteins cannot mediate colonization in the infant mouse cholera model. The competitive index was determined for each environmental TcpF variant by competing each strain against the reference strain, SJK9, expressing wild-type TcpF. Each circle represents a mouse, and each horizontal line marks the average competitive index of the five or six mice tested for one strain. The positive control is the competition between SJK9 and SJK7, both expressing wild-type TcpF. The negative control is the competition between SJK9, expressing wild type TcpF, and SJK71, harboring the empty vector. The competitive index was determined for each mouse, and Student's t test was used to determine which strains were significantly defective in colonization compared to the positive control. Environmental TcpF proteins with various levels of divergence from pathogenic TcpF were found to be significantly defective in colonization compared to the positive control, as indicated by asterisks (P < 0.016).

• Open in new tab
• Download powerpoint

FIG. 8.

Clustal W alignment of SCE4 environmental TcpF, which is 66% identical to pathogenic TcpF, reveals regions that are divergent between these proteins. TcpF derivatives that were found to be defective in colonization (Fig. 2) are represented by the designations of the plasmids encoding them, given above the alignment at the location of the 5-amino-acid insertion. Residues in the SCE environmental TcpF protein that are identical (*), conserved (:), or semiconserved (.) with respect to the pathogenic TcpF protein are indicated below the sequence.