INTRODUCTION

Top Abstract Introduction Materials and Methods Results and Discussion References

Cholera is a severe dehydrating diarrhea caused by toxigenic strains of the gram-negative bacterium Vibrio cholerae. The profuse watery diarrhea is mainly due to an enterotoxin, cholera toxin (CT), produced by V. cholerae (4, 21). The ctxAB operon, which encodes the A and B subunits of CT, resides in the genome of CTX, a lysogenic filamentous bacteriophage (27). In the natural habitat, CTX may infect nontoxigenic V. cholerae strains, leading to the origination of novel toxigenic strains (5, 6). Following entry into the recipient cells, the phage genome either integrates into the chromosome at a specific attachment site (attRS), forming stable lysogens, or exists as a plasmid referred to as the replicative form (RF) of the phage genome (27). Attenuated live vaccine strains are supposed to be protected from lysogenic conversion by CTX if the attRS sequence is deleted, thus impairing the chromosomal integration of the CTX genome. Recently potential vaccine strains have been developed which lack the entire CTX element, including attRS sequences (1, 14, 25, 28). A previous study has shown that V. cholerae cells carrying the RF of the CTX genome can produce CT under in vitro laboratory conditions (15). However, the diarrheagenic potential of such strains in vivo was not examined.

V. cholerae strains belonging to the O1 or O139 serogroup are normally associated with epidemic cholera, whereas other serogroups of V. cholerae have been mostly associated with sporadic cases of diarrhea (11). Recent studies have recorded incidences of diarrhea outbreaks associated with non-O1 non-O139 V. cholerae, but these strains have been found to be nontoxigenic, and the pathogenic determinants for the diarrhea have not been identified (23). It is not clear, however, whether transient acquisition of the CTX genome by these strains in vivo might have led to the diarrhea. In the present study CTX was introduced into attRS-negative non-O1 non-O139 V. cholerae of clinical or environmental origin as well as into attenuated (core attRS) derivatives of clinical V. cholerae O1 and O139 strains. The resulting strains, which carried the RF of the CTX genome, were tested for production of a diarrheal response in adult rabbits and for stability of the phage genome. This study was designed to examine whether potential vaccine strains carrying a deletion of attRS or naturally occurring attRS-negative V. cholerae strains can become transient diarrheal pathogens by harboring the RF of the CTX genome.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results and Discussion References

Bacterial strains, phages, and plasmids. The V. cholerae non-O1 non-O139 strains included in this study were either isolated from surface water samples collected in Dhaka, Bangladesh, or obtained from patients with diarrhea who attended the treatment center of the International Centre for Diarrhoeal Disease Research, Bangladesh (ICDDR,B), located in Dhaka. A total of 37 non-O1 non-O139 strains were first tested for possession of different virulence genes and the attRS sequence, and strains which were negative for attRS were further analyzed in this study. The O1 and O139 strains were attenuated derivatives of toxigenic clinical strains described previously (1, 17, 20, 25). Strains were stored either in lyophilized form or in sealed deep nutrient agar at room temperature until used for the present study. Before use, the identities of the cultures were verified by biochemical and serological methods (29) and by using specific DNA probes as described below. The genetically marked phage MSF8.2 used in this study was a derivative of an El Tor CTX which carried a functional ctxAB operon as well as a kanamycin resistance (Km^r) determinant. The strategy for the construction of pMSF8.2 is described below. Relevant characteristics of bacterial strains, phages, and plasmids used in this study are summarized in Table 1.

Recombinant DNA procedures. For in vitro DNA manipulations, pUC18, a chromogenic substrate (X-Gal [5-bromo-4-chloro-3-indolyl--D-galactopyranoside]), and DNA restriction and DNA-modifying enzymes were purchased from Bethesda Research Laboratories (Gaithersburg, Md.) and used in accordance with the manufacturer's suggestions. The strategy for the construction of pMSF8.2, a genetically marked derivative of the RF of the CTX genome carrying a functional ctxAB operon, is shown in Fig. 1. CTX-Km isolated from strain SM44 (10, 27) was used to infect the classical biotype strain O395. The RF of the phage genome, pCTX-Km isolated from strain O395, carried a Km^r determinant in place of ctxAB genes. The entire ctxAB operon was obtained from another recombinant plasmid, pRT41 (26), and reinserted into pCTX-Km using a number of cloning steps to construct pMSF8.2. Briefly, pCTX-Km was digested with BamHI, and a 1.3-kb fragment encoding Km^r and the remaining 6.3-kb fragment of the phage genome were isolated. The 1.9-kb BamHI-EcoRI insert carrying the entire ctxAB operon, including the wild-type promoter, was isolated from pRT41. The DNA fragments carrying the ctxAB operon and Km^r were sequentially ligated to BamHI-cleaved dephosphorylated pUC18. The ligated DNA was isolated, and protruding ends were filled in using the Klenow fragment of Escherichia coli DNA polymerase I. The resulting blunt ends were then ligated, and the ligation mixture was used to electroporate E. coli DH5. Colonies which were resistant to both kanamycin and ampicillin were screened for the presence of a pUC18 derivative with a 3.2-kb BamHI insert carrying the ctxAB genes and the gene encoding Km^r, and this plasmid was designated pMSF8.1. The 3.2-kb BamHI fragment of pMSF8.1 was isolated and ligated with the 6.3-kb BamHI fragment of pCTX-Km. The ligated DNA was used to electroporate V. cholerae strain O395, and colonies were selected for resistance to kanamycin. The final plasmid construct, designated pMSF8.2, thus consisted of a functional ctxAB operon, a Km^r cassette, and the RF DNA of CTX and was able to support the morphogenesis of infectious phage particles.

View larger version (36K): [in this window] [in a new window]   FIG. 1.   Strategy for construction of pMSF8.2, a derivative of pCTX-Km carrying a functional ctxAB operon. Restriction sites: B, BamHI; X, XbaI; E, EcoRI. See the text for details.

Probes and PCR assays. The gene probe used in this study to detect the CTX genome was a 0.5-kb EcoRI fragment of pCVD27 carrying part of the ctxA gene (12). All strains were also tested for the presence of genes encoding the toxin-coregulated pilus (TCP) (which is the receptor for CTX), the virulence regulatory gene toxR, and the CTX attachment sequence attRS. The presence of the TCP pathogenicity island was determined by PCR assays specific for the tcpA, tcpI, and acfB genes as described previously (8, 13). The toxR gene probe was a 2.4-kb BamHI fragment of pVM7 (19), and the 18-bp attRS sequence was identified using a synthetic oligonucleotide corresponding to the attRS sequence (20).

Assay for CT production. Production of CT by V. cholerae strains harboring the RF of CTX as well as the CTX lysogens was determined by the G_M1 ganglioside-dependent enzyme-linked immunosorbent assay (ELISA) and the rabbit ileal loop assay as described previously (2, 8, 22). For each round of CT assay, 5 ml of AKI medium (1.5% Bacto Peptone, 0.4% yeast extract, 0.5% NaCl, 0.3% NaHCO₃ [pH 7.4]) was inoculated with approximately 10³ bacterial cells. For strains carrying the Km^r-labeled phage genome, kanamycin (50 µg/ml) was added to the culture medium to retain the phage genome. All cultures were grown for 16 h at 30°C with shaking. The culture was centrifuged at 4,000 × g for 5 min, and the supernatant was collected and filtered through 0.22-µm-pore-size Millipore filters. Aliquots of the undiluted supernatant, 10-fold and 100-fold dilutions of the supernatant, and dilutions of purified CT (Sigma) were used for the toxin assay. Quantification of CT production was done using a standard curve prepared for each batch of assay mixture. The amount of CT produced by each strain was the mean value from five different assays with the same strain and culture conditions. Toxigenic El Tor strain P-27459 and nontoxigenic El Tor strains SA-317 and SM44 were included as positive and negative control strains in each round of the assay.

Ileal loop assay. Culture filtrates prepared for the ELISA were also tested in ileal loops of adult New Zealand White rabbits. A maximum of six ileal loops of approximately 10 cm in length were made in each rabbit (rabbits had previously been fasted for 48 h), and 1 ml of the filtrate was inoculated into each loop as described previously (2). After 18 h, rabbits were sacrificed and the loops were examined for fluid accumulation. The results were expressed as milliliters of fluid accumulated per centimeter of loop.

Assay for diarrhea in rabbits. Diarrheal responses to V. cholerae strains were assayed in adult rabbits by using the removable intestinal tie-adult rabbit diarrhea model (24). Adult New Zealand White rabbits weighing 1.5 to 2.7 kg were used to prepare the model. Rabbits were starved for the previous 24 h, and surgery was done under a local anesthetic. The cecum of each animal was ligated to prevent it from retaining fluid secreted by the small intestine, and a temporary removable tie of the small bowl was introduced at the time of challenge. Strains were grown in Casamino Acids-yeast extract broth as described previously (24), and cells were precipitated by centrifugation and resuspended in 10 mM phosphate-buffered saline (pH 7.4) at a concentration of approximately 10⁹ cells per ml. One millliliter of the suspension was injected into the lumen of the anterior jejunum. The removable tie in the intestine was removed after 2 h of inoculation. Each strain was inoculated in at least five different rabbits. Rabbits were observed for overt diarrhea and for death, and stools or rectal swabs were cultured on gelatin agar plates and a duplicate plate containing kanamycin (50 µg/ml) whenever appropriate to monitor shedding of the challenge organisms. Observations were made at 6-h intervals during the 7 days following inoculation; the numbers of rabbits developing moderate to severe diarrhea were arbitrarily scored, and the numbers of deaths were recorded. Rabbits that died with or without diarrhea were subjected to postmortem examinations to check for the presence of fluid in the intestine.

Stability of the CTX genome. To determine the stability of the CTX genome in V. cholerae cells in vivo, the ratio of Km^r colonies to the total number of colonies recovered from stools or rectal swabs of rabbits challenged with each strain was calculated and expressed as the percentage of cells retaining the phage genome. To test the stability of the CTX genome under in vitro conditions, representative colonies of the infected recipient were grown in aliquots of Luria broth either containing kanamycin (50 µg/ml) or without kanamycin. Serial dilutions of the cultures were plated on Luria agar plates containing kanamycin and on a duplicate set of Luria agar plates without the antibiotic to determine the proportion of cells retaining the phage genome.

Statistical analysis. Statistical comparison of CT production between two groups of strains was carried out by the Mann-Whitney test. For comparison between the responses of different proportions of rabbits to different challenge strains, the 2 statistic or Fisher's exact test was used. Differences were considered to be significant when the P value was 0.05. Data analyses were done by using statistical software (Sigmastat for Windows, version 2.03; Jandel Scientific, San Rafael, Calif.).

	RESULTS AND DISCUSSION

Top Abstract Introduction Materials and Methods Results and Discussion References

We evaluated the diarrheagenic potential of strains carrying the RF of the CTX genome to investigate whether strains carrying a deletion of attRS or naturally occurring attRS-negative V. cholerae strains can become transient diarrheal pathogens by temporarily harboring the CTX genome. In this study, V. cholerae strains carrying the RF of the CTX genome were constructed by infecting attRS-negative V. cholerae strains with a genetically marked phage, MSF8.2. Infected cells were first selected by their expression of the Km^r phenotype. While all O1 and O139 strains which carried genes for the CTX receptor TCP were infected by the phage, 4 of 29 TCP-negative, attRS-negative non-O1 strains were also infected. Previous studies have also reported that a small proportion of TCP-negative strains were susceptible to the phage (6). Molecular analysis of the infected strains showed the presence of RF DNA of the phage in all attRS-negative strains, although in control attRS-positive strains the phage genome integrated into the chromosome (data not shown). All O1 and O139 strains as well as the four infected non-O1 non-O139 strains (Tables 2 and 3) were further analyzed for production of CT and for causing diarrhea in the adult rabbit model. All of these strains carried the toxR gene, which is required for expression of major virulence factors in V. cholerae.

View this table: [in this window] [in a new window]   TABLE 2.   Production of CT by native V. cholerae strains and their derivatives carrying the RF DNA of a genetically marked CTX

View this table: [in this window] [in a new window]   TABLE 3.   Diarrheal response of adult rabbits to toxigenic V. cholerae strains (CTX lysogens) and their derivatives carrying the RF of the phage

Expression of virulence. Production of CT by the infected cells was initially studied in vitro by G_M1-based ELISA, using an antibody against the B subunit of CT (Table 2). These assays showed that strains carrying the RF of CTX produced significantly more CT than the lysogens (P < 0.01). We suspect that the higher levels of CT produced by strains carrying the RF were due to possession of multiple copies of the phage genome and hence multiple copies of the ctxAB operon. In CTX lysogens, the expression of CT is regulated by the transcriptional activator ToxR, whereas in strains carrying the RF of the CTX genome, expression of CT is also known to occur independently of ToxR (15). Thus, the ToxR-independent pathway of expression might also have contributed to the high levels of in vitro expression of CT by strains carrying the RF of the phage genome. To further ascertain whether the toxin was biologically active, we used the ligated ileal loop assay in rabbits and observed fluid accumulation. All culture supernatants which were positive for CT in the ELISA also caused fluid accumulation in the ileal loops of rabbits. Culture supernatants of strains carrying the RF of CTX caused somewhat more fluid accumulation in the rabbit ileal loops than the lysogens did (Table 2), although the difference was not statistically significant. Considering possible variation in response to CT among individual rabbits, the number of observations was possibly less than optimum for observation of a statistically significant difference between the two groups. Nevertheless, these results confirmed a previous observation (15) that strains carrying the RF of CTX produce biologically active CT in their culture supernatants.

View this table: [in this window] [in a new window]   TABLE 4.   Loss of the CTX genome from toxigenic V. cholerae strains (CTX lysogens) and their derivatives carrying the RF of the CTX genome in the gastrointestinal tracts of rabbits

Diarrheal response to non-O1 non-O139 strains. In the present study, although the native non-O1-non-O139 strains did not cause diarrhea in the rabbit model (Table 3), 14 of 22 rabbits (63.6%) inoculated with V. cholerae non-O1 non-O139 strains carrying pMSF8.2 developed mild to moderate diarrhea. While the O1 and O139 strains carried genes for the colonization factor TCP, the non-O1 non-O139 strains were TCP negative. Apparent colonization of rabbit intestines by these TCP-negative strains suggests that these strains probably produce some other, unknown colonization factors. Further studies are under way in our laboratory to characterize possible new colonization factors produced by the TCP-negative clinical strains. The diarrhea caused by non-O1 non-O139 strains carrying the RF of the CTX derivative was of short duration (12 to 36 h), and the strains when excreted were mostly CT negative due to in vivo loss of the RF DNA. This scenario resembles clinical cases of diarrhea due to non-O1 vibrios, when strains cultured from the stool are usually CT negative (23). Besides the possibility of the diarrhea being induced by additional, unknown virulence factors, it also seems possible that in the natural habitat some non-O1 strains become transiently toxigenic by acquisition of CTX. Such strains are not normally detected until they infect a host and produce diarrhea. The CTX genome is probably lost in the intestinal tract following infection and production of mild diarrhea by such strains. However, it is not clear what determines the stability of the unintegrated CTX genome in these cells under environmental conditions prior to infecting a mammalian host. An alternative explanation may be that such strains acquire the phage genome while inside the intestine of the host and transiently produce CT.

Selection of CTX lysogens in the intestine. Previous studies have suggested that in the gastrointestinal environment expression of CT confers a survival advantage to V. cholerae and that hypertoxigenic strains are selected in vivo by a need to upregulate the expression of CT. In the present study, strains carrying the RF of CTX rapidly lost the phage genome in vivo, whereas strains carrying the integrated form of the phage retained the phage genome. We also examined the stability of the RF in V. cholerae cells under in vitro conditions. Although normally the Km^r-labeled phage genome was retained by the recipient cells when grown in the presence of kanamycin, a proportion of the cells lost the phage genome in the absence of kanamycin. However, the loss of the phage genome in vitro was more gradual than that in vivo. After 24 h of culture in the absence of kanamycin, 15 to 32% of cells lost the phage genome in vitro, whereas the frequency of loss in the rabbit intestine was between 75 and 90% for all strains tested after the first 24 h. This was unexpected, particularly since in the intestine, where TCP is adequately expressed, the CTX is expected to be maintained in the cells due to continuous reinfection (15). Thus, the gastrointestinal tract seems to select CTX lysogens rather than strains carrying the RF, although the latter strains produced higher levels of CT when assayed in vitro. This may explain why all naturally occurring toxigenic strains of V. cholerae carry the CTX genome in the lysogenic form and particularly in serogroups which are capable of causing human disease. CTX is different from the well-characterized filamentous bacteriophages derived from E. coli in that this phage has evolved to possess genes for a site-specific integration system. Integration of the CTX genome into the host chromosome seems to allow it to withstand the intestinal selection, whereas the ability of the prophage to enter the replicative state maximizes horizontal propagation.

Implications for vaccine development. The development of live oral vaccines against cholera invariably involves the deletion of the ctxA gene encoding the enzymatic subunit of CT or of the entire CTX element, thus rendering the strain nontoxigenic. The discovery of CTX has shown that the possibility of reversion of attenuated strains by reacquisition of the CTX genome is real. The receptor for CTX for entry into a recipient V. cholerae cell is the TCP, which is also the major colonization factor. Since to elicit an adequate immune response live vaccines must colonize the intestine, TCP is retained in vaccine strains. A recent approach to protect attenuated live vaccine strains from lysogenic conversion by CTX involves deleting the attRS sequence, thus impairing the chromosomal integration of the CTX genome. Thus, although the attenuated strains remain susceptible to infection by CTX, in strains with the attRS sequence deleted the phage genome is expected to exist as a replicative plasmid. In the present study, we found that such strains with the RF of CTX lose the phage genome when introduced into the intestines of rabbits. The diarrheal response, if any, was mild, and the excreted strains were mostly nontoxigenic. These results confirmed that attenuated vaccine strains with attRS deletions are considerably protected from generating stable toxigenic revertants.