Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. 2003 Feb 4; 100(3): 1304–1309.
Published online 2003 Jan 21. doi:  10.1073/pnas.0337468100
PMCID: PMC298768
Inaugural Article
Evolution, Microbiology

Emergence and evolution of Vibrio cholerae O139


The emergence of Vibrio cholerae O139 Bengal during 1992–1993 was associated with large epidemics of cholera in India and Bangladesh and, initially, with a total displacement of the existing V. cholerae O1 strains. However, the O1 strains reemerged in 1994 and initiated a series of disappearance and reemergence of either of the two serogroups that was associated with temporal genetic and phenotypic changes sustained by the strains. Since the initial emergence of the O139 vibrios, new variants of the pathogen derived from multiple progenitors have been isolated and characterized. The clinical and epidemiological characteristics of these strains have been studied. Rapid genetic reassortment in O139 strains appears to be a response to the changing epidemiology of V. cholerae O1 and also a strategy for persistence in competition with strains of the O1 serogroup. The emergence of V. cholerae O139 has provided a unique opportunity to witness genetic changes in V. cholerae that may be associated with displacement of an existing serogroup by a newly emerging one and, thus, provide new insights into the epidemiology of cholera. The genetic changes and natural selection involving both environmental and host factors are likely to influence profoundly the genetics, epidemiology, and evolution of toxigenic V. cholerae, not only in the Ganges Delta region of India and Bangladesh, but also in other areas of endemic and epidemic cholera.

Vibrio cholerae is the causative agent of cholera, an acute dehydrating diarrhea that occurs in epidemic and pandemic forms (1, 2). Seven distinct pandemics of cholera have occurred since the onset of the first pandemic in 1817 (3). Except for the seventh pandemic which originated in Indonesia, six of the pandemics arose from the Indian subcontinent, usually from the Ganges Delta region, and reached to other continents (2). The species V. cholerae is classified on the basis of its somatic antigens (O-antigens) into serogroups, and there are at least 206 known serogroups of V. cholerae (4). Until the emergence of V. cholerae O139 in late 1992, the serogroup O1 was supposed to include all strains responsible for epidemic and endemic cholera. The emergence of V. cholerae O139 attracted worldwide attention, particularly because this was the first non-O1 serogroup associated with widespread epidemics of cholera (5, 6). Extensive outbreaks have occurred in various regions of Bangladesh and India, and cases caused by V. cholerae O139 have been reported in Pakistan, Nepal, China, Thailand, Kazakhstan, Afghanistan, and Malaysia (59). Imported cases have been reported in the United Kingdom and the United States (9, 10). Epidemics of cholera caused by this new serogroup continue to occur, apparently representing the beginning of an eighth cholera pandemic (10). Recent trends in India (11) and Bangladesh (S.M.F., M. A. Salam, A. Faruque, G.B.N., and D.A.S., unpublished data) show an escalating association of the O139 serogroup with outbreaks of cholera.

Since the initial emergence of V. cholerae O139, new variants of the pathogen with altered genetic and phenotypic characteristics have appeared frequently. These include strains with new ribotypes, CTX genotypes, and altered antimicrobial resistance (1214). Attempts have been made to characterize the new variants as well as the original O139 isolates to determine the origin of the O139 serogroup. Clinical and epidemiological characteristics of these strains have also been studied. Thus, the emergence of V. cholerae O139 has provided a unique opportunity to witness epidemiological and genetic changes associated with strains initiating and sustaining a new cholera pandemic. The purpose of this review is to summarize available information on the epidemiology, genetics, and evolution of V. cholerae O139. Special emphasis has been made to compile scientific data obtained from studies on various aspects of V. cholerae O139 to provide insight into the possible origin of V. cholerae O139, as well as the significance of emerging clonal diversity within the O139 serogroup of V. cholerae.

The Emergence of V. Cholerae O139

In late 1992, epidemics of severe acute watery diarrhea, clinically resembling cholera and mainly affecting adults, was reported in Madras, a southern port city of India, and in Southern Bangladesh (5, 6). The epidemics later spread to other parts of both countries and to some of the neighboring countries of the region (7, 9, 12, 13). The bacterium responsible for the epidemics resembled V. cholerae 01 in cultural and biochemical characteristics, but did not agglutinate with V. cholerae 01 antisera (5, 6). Primers specific for the cholera toxin (CT) genes of V. cholerae 01 amplified sequences corresponding to CT in these strains in PCR (5), and all strains tested also were positive for CT production by standard bioassays for CT. However, this bacterium did not belong to any of the 138 O serogroups for V. cholerae described until then; the conclusion was that it belonged to a new serogroup (15). The new epidemic strain of V. cholerae was later serogrouped as O139 and given the synonym “Bengal” in recognition of the first appearance of this serogroup in regions in the vicinity of the Bay of Bengal. Before the emergence of V. cholerae O139, non-01 serogroups of V. cholerae were not known to be associated with such large outbreaks of diarrhea. Moreover, they were known to produce CT at a very low frequency (16), unlike the O139 serogroup, given that almost all tested isolates of this serogroup produced the toxin. Since then, V. cholerae O139 has persisted as a second etiologic agent of cholera. There are now two serogroups, O1 and O139, that have been associated with cholera epidemics, and the simple distinction between O1 and non-O1 V. cholerae regarding epidemic potential has, therefore, become obsolete.

Genesis of V. Cholerae O139

Because V. cholerae O139 was recorded as the only non-O1 V. cholerae capable of causing epidemic outbreaks, immediately after the emergence of the O139 serogroup comparative analyses of V. cholerae O1 and O139 strains were carried out to investigate the origin of this new serogroup (17, 18). The early studies indicated that O139 strains were closely related to O1 El Tor biotype strains, which is responsible for the seventh pandemic of cholera, and the initial O139 strain may have emerged from serotype-specific genetic changes in an ancestral El Tor strain. More detailed molecular epidemiological analyses, such as zymovar analysis, ribotyping, and pulsed-field gel electrophoresis showed that V. cholerae O139 Bengal strains are closely related to O1 El Tor strains (19, 20). Furthermore, V. cholerae O139 strains had all of the virulence factors normally found in O1 El Tor strains, and both V. cholerae O1 and O139 Bengal cause cholera of comparable clinical severity (21, 22). However, in contrast to O1 strains, O139 strains are encapsulated (23), and the O139 serogroup antigen includes an O-antigen capsule and bacterial lipopolysaccharide (LPS; refs. 23 and 24). The LPS of serogroup O139 does not contain any long O-antigen side chains, whereas O1 strains have a core substituted with an average of 17 repeat units of 4-NH2-4,6-dideoxymannose, each substituted with 3-deoxy-l-glycero-tetronic acid (25). The O139 LPS appears to be an efficiently substituted core polysaccharide, even though it possesses only a few additional sugar moieties (23). Interestingly, these changes have rendered the O139 vibrios immunologically distinct from the O1 El Tor strains.

Analysis of the genetic regions associated with O-antigen biosynthesis in O1 and O139 strains (2628) suggested that the conversion of the ancestral El Tor strain involved insertion of a large foreign genomic region encoding the O139-specific genes and simultaneous deletion of most of the O1-antigen-specific gene cluster. Later, more detailed analysis and sequencing of the wbf genes responsible for the biosynthesis of O-antigen and genes downstream of the wbf gene cluster of V. cholerae O139 have been reported (2932). These studies have now characterized the major genetic differences accounting for the phenotypically distinct surface polysaccharide of O1 El Tor and O139 Bengal. In brief, the genes responsible for the synthesis of O-antigen are present in a cluster designated as the wb* region. A large portion of DNA corresponding to the wbe region of O1 strains was found to be missing in O139 strains, and O139 strains were found to have acquired a unique DNA region (2729, 32). It was also shown that the serogroup O139 resulted from a precise 22-kb deletion of the wbe (rfb) region of O1, with replacement by a 35-kb wbf region (wbfA through wbfX) encoding the O139 O-antigen (30). The schematic diagram of the wbf gene cluster and its flanking genetic regions in V. cholerae O139 are shown in Fig. Fig.1.1.

Figure 1
Schematic diagram of the O-antigen-specific gene cluster of V. cholerae O139 showing the insertion of the O139 DNA into a probable ancestral strain. The entire region shown here is 41,221 bp in length, of which the O139-specific wbf region is 35,807 bp ...

The complete DNA sequence of the O1 wbe region was previously determined by Stroeher et al. (33). The sequenced wb* regions of V. cholerae O1 and O139 and the V. anguillarum serogroups O1 and O2 all have the gmhD gene at the left junction (29, 34). The sequenced right junction of the O1 wbe cluster has a 30-bp overlap with that of the O139 wbf right junction (35). The intervening regions are divergent in the two serogroups except for the presence of an insertion sequence (IS) element, IS1358.

Two hypotheses have been proposed to explain the emergence of V. cholerae O139. The first proposes that a transposition event, mediated by the IS element IS1358, resulted in the replacement of the O1 wbe genes with the O139 wbf genes (26, 31, 35, 36). The second hypothesis involves a homologous recombination event resulting in the replacement of the entire O1 wbe region by the O139 wbf region (30, 34, 35). The genes downstream of the right junction of the insertion were sequenced and analyzed by Sozhamannan et al. (32). Sequence of the genes downstream of wbfX and analysis of genes flanking the O-antigen region (gmhD and rjg) in other serogroups showed that all serogroups analyzed to date have an organization of the wb* region similar to that of O1 and O139 strains. These results supported the latter hypothesis and raised the possibility that pathogenic strains belonging to non-O1/non-O139 serogroups can also emerge by homologous recombination via the junction genes. The preferential linkage of IS1358 to the wb* region in most V. cholerae strains analyzed further supported the idea that IS1358 was acquired by homologous recombination, rather than by transposition, because a random transposition event is expected to deliver the IS element to any part of the chromosome (32).

Source of the wbf Cluster

Although the possible mechanisms of genetic changes leading to the emergence of the O139 serogroup have been revealed to some extent, the donor for the O139 specific DNA in this gene transfer event remains unidentified. Several studies have suggested possible nonpathogenic serogroups of V. cholerae as donor for the wbf cluster and involvement of generalized transducing phages or conjugative plasmids as vectors for the large DNA region in the recombination event (32, 35). A comparative chemical analysis of the LPS of V. cholerae O139 and O22 revealed that the sugar composition of the O22 LPS was quite similar to that of O139 LPS. Each contained d-glucose, l-glycero-d-manno-heptose, colitose (3,6-dideoxy-l-galactose), d-fructose, d-glucosamine, d-quinovosamine, and d-galacturonic acid.

The O-antigenic relationship between the O22 and O139 strains was also analyzed by passive hemolysis (PH) and PH inhibition tests with the respective LPS being used as antigens. The LPS was used to sensitize sheep red blood cells (SRBC), and as inhibitors in a PH system that consisted of LPS-sensitized SRBC, guinea-pig complement and anti-O139 or anti-O22 antiserum, both unabsorbed and absorbed with the heterologous antigen. Results indicated that the O-antigen of O139 is closely related to that of O22 in an a,b–a,c type of relationship, where “a” is common antigenic factor, “b” is an O139-specific antigenic factor, and “c” is an O22-specific antigenic factor (37).

The gene cluster responsible for O-antigen biosynthesis of the O139 serogroup of V. cholerae has also been found to be closely related to that of O22 serogroup, when DNA fragments derived from O139 O-antigen biosynthesis gene region were used to probe reference strains of V. cholerae representing serogroups O1-O193 (38). PCR amplification of a long defined region of the genome was used to determine if a simple deletion or substitution was involved to account for the difference between O139 and O22 specific O-antigen genes. A product of ≈15 kb was amplified when O139 DNA was used as the template, whereas a product of ≈12.5 kb was amplified when O22 DNA was used as the template, indicating that substitution but not deletion could account for the difference in the region between O22 and O139 serogroups. To compare precisely between the genes responsible for O-antigen biosynthesis of O139 and O22, the region responsible for O-antigen biosynthesis of O22 serogroup was cloned and analyzed. In concurrence with the results of the hybridization test, most of the regions were well conserved in O22 and O139 serogroups (38), suggesting that the gene clusters responsible for O139 O-antigen biosynthesis have striking similarity to those of O22. These results indicated that a strain of the O22 serogroup might have been a donor of the O139-specific genes in the gene transfer event that led to the origination of the O139 serogroup from a progenitor El Tor strain of V. cholerae O1. However, further studies are required to confirm this assumption.

Multiple Progenitors of V. Cholerae O139

Several studies have suggested that V. cholerae O139 that emerged as the predominant epidemic strain 1992 was derived from a seventh-pandemic El Tor clone by horizontal gene transfer. This was suggested not only because the important virulence factors, specifically the CT and toxin coregulated pili of V. cholerae O139 are indistinguishable from a typical V. cholerae O1 biotype El Tor strain, but the identity of ribotypes between an O139 and an El Tor strain has also been documented (17, 39). However, molecular analysis of representative V. cholerae O139 strains isolated during the last decade has revealed the existence of genetically diverse strains belonging to the O139 serogroup, including nontoxigenic variants. These nontoxigenic O139 strains included N[1-(thienyl)cyclohexyl]piperidine (TCP)-positive variants that were susceptible to toxigenic conversion by CTX phage (39, 40) as well as TCP-negative strains. The later strains were more widely distant from the El Tor strains but agglutinated with O139-specific antiserum and reacted with O139-specific DNA probes.

PCR-based analysis using specific primers corresponding to six defined regions of the wbe gene cluster and flanking sequences of V. cholerae O1 showed that a large region of chromosomal DNA representing O1 specific wbe gene cluster was absent in all O139 strains. Regions corresponding to gmhD gene of V. cholerae O1, located upstream of wbeA gene, and those corresponding to the insertion sequence IS1358, located upstream of wbeU were present in the O139 strains. The gmhD which is found upstream of manC in V. cholerae O1 and upstream of wbf in V. cholerae O139 was assumed to have served as a site of homologous recombination in the gene-transfer event that led to the origination of V. cholerae O139 (34). Although the gmhD locus, amplified from all toxigenic O139 and El Tor strains, and CTXφ-susceptible nontoxigenic O139 strains were similar in size, the amplification products of some CT-negative O139 strains were smaller in size and were similar to those produced by non-O1 non-O139 strains isolated from the environment (39). Like most non-O1 non-O139 strains, these CT-negative O139 strains that were isolated in Argentina, Bangladesh, and India lacked tcpA as well as the tcpI and acfB genes, and presumably the entire TCP pathogenicity island. This finding provided further evidence that these CT-negative O139 strains possibly originated from non-O1 progenitors, and that the gmhD genes in these strains were remnants of that of the progenitor strains. These findings provided evidence that the gene-transfer event that caused the origination of the O139 serogroup was not a unique event, but similar serotype conversions of different progenitor strains may have been occurring continually, resulting in the spread of the O139-antigen among different lineages of V. cholerae. Hence, the O139-antigen is present in different lineages, and this serogroup appears to comprise epidemic and nonepidemic strains derived separately from different progenitors.

Evolution of V. Cholerae O139

The emergence of the O139 serogroup has prompted a renewed effort to investigate the evolution and epidemiology of toxigenic V. cholerae by using available molecular technology and, thus, conceptualize the molecular basis for the establishment of a newly emerged bacterial pathogen as an endemic strain.

Since the initial emergence of V. cholerae O139 in late 1992, almost continuous epidemiological monitoring of V. cholerae O139 has been carried out by the International Center for Diarrhoeal Disease, Bangladesh (ICDDR,B), and the National Institute of Cholera and Enteric Diseases in Calcutta, India. Both these laboratories used recently developed molecular approaches to analyze representative strains that provided insights into the evolution of the O139 serogroup. These studies have documented temporal changes in genetic and phenotypic properties and the emergence of new clones within the O139 serogroup. The CTXφ prophage (40) which encodes CT is carried by all toxigenic clones, but important changes, including amplification or rearrangement of the CTX prophage as well as acquisition of new CTX prophage, has been documented (1214). Other prominent changes include restriction fragment-length polymorphisms in conserved rRNA genes (ribotype), as well as rapid changes in antibiotic-resistance phenotypes.

The O139 strains isolated during 1992–93 have two copies of the El Tor type CTX prophage (CTXETφ) connected by an RS1 element (Fig. (Fig.2),2), whereas the O139 strains collected in Calcutta in 1996 have three copies of the CTX prophage arranged in tandem (41, 42). However, the O139 strains which reemerged in 1996 have two types of CTX prophages (Fig. (Fig.2);2); whereas the first of the three copies is an El Tor-type CTX prophage, the second and third copies of the CTX prophage are a new type of CTX prophage referred to as the Calcutta type (CTXcalφ) CTX prophage (43). The CTXcalφ differs from the CTXETφ primarily in the rstR gene, the gene that codes for the repressor protein of CTXφ. Most O139 strains isolated in Calcutta in 1998 have a single copy of the CTX prophage, whereas those isolated from other parts of India either have the arrangement of the O139 strains isolated in 1996 or have two tandemly arranged copies of the CTX prophage.

Figure 2
Schematic diagram of the arrangement of V. cholerae O139 CTX prophage. Restriction endonuclease sites shown are H, HindIII; BI, BglI; BII, BglII; and P, PstI.

Although most of the early studies considered the O139 isolates from the first epidemic period (1992–1994) to belong to a unique clone derived from an O1 El Tor strain, subsequent studies suggested the existence of greater genetic variability within the O139 population (19, 44, 45). Molecular analyses have identified two different ribotypes (19, 46) and four pulsed-field gel electrophoresis patterns (19) in O139 isolates of the initial period. These results are also in agreement with the diversity, determined by multilocus enzyme electrophoresis, that was observed in the O139 population (47). Moreover, a comparative molecular analysis of gene fragments of six housekeeping loci, distributed around the two chromosomes in V. cholerae O139 Bengal strains isolated in India during 1992 and 1993 showed a higher genetic diversity than that reported previously in other molecular studies. The O139 strains clustered in several lineages of the dendrogram generated from the matrix of allelic mismatches between the different genotypes, a finding which does not support the hypothesis that the O139 serogroup is a unique clone.

Several studies also indicate continuous genetic changes in V. cholerae O139, leading to the emergence of yet new clones of toxigenic O139 vibrios. This process is likely to involve defined genetic reassortments and a natural selection possibly involving unidentified ecological factors and immunity of the host population (12, 13, 39, 46, 4850). Alternatively, the genetic reassortments observed here may also be random genetic changes, and strains which gained advantage as a result of these changes may have infected humans more efficiently, got enriched inside the gastrointestinal tract, and eventually became detectable as new strains. Change in the antimicrobial resistance pattern is also likely to influence the emergence and prevalence of particular clones, and the emergence of a new clone of V. cholerae associated with a change in antibiogram has been documented (14, 41). In many bacteria, changes in antibiogram are mediated by plasmids. However, in V. cholerae, plasmids are relatively rare, and there are only a few documented instances of plasmid-mediated drug resistance in V. cholerae. More often, antibiotic resistance in V. cholerae has been shown to be encoded by horizontally transferable conjugative transposons which reside in the chromosome of the host bacterium (5153). It is also possible that a change in the antibiogram can occur in V. cholerae because of the presence of integrons (54, 55), which are gene-expression elements that can acquire structural genes, including those encoding antimicrobial resistance from foreign organisms and convert these into functional operons. The observed diversity in the genomes of V. cholerae O139 may have substantially contributed to mobile genetic elements containing particular antibiotic resistance genes. Factors which determine the emergence and domination of particular clones of toxigenic V. cholerae and displacement of existing clones remain to be elucidated further.

Molecular Epidemiology

Epidemiological data on the emergence and prevalence of V. cholerae O139 and its coexistence with the O1 El Tor strains are available primarily from Bangladesh and India (Fig. (Fig.3)3) through systematic surveillance studies. In the Ganges Delta region of India and Bangladesh, epidemics of cholera occur with a regular seasonality, but temporal variation in the prevalence of the two epidemic serogroups O1 and O139 have been noticed (4850). The emergence of V. cholerae O139 initially caused a complete displacement of the El Tor biotype strains in both these countries. However, V. cholerae O139 was again displaced in 1994 by a new genetic variant of the O1 strain, and this variant strain dominated until 1996 in India. In August, 1996, a new variant of the O139 strain emerged, and cholera caused by the new O139 genetic variant dominated for a year, until September, 1997 in Calcutta. Similarly in neighboring Bangladesh, during 1994 and till the middle of 1995, in most northern and central areas of the country, the O139 vibrios were replaced by a new clone of V. cholerae O1 of the El Tor biotype, whereas in the southern coastal regions, the O139 vibrios continued to exist (12, 13). During late 1995 and through 1996, cases of cholera caused by both V. cholerae O1 and O139 were again detected in various regions of Bangladesh. However, since 1996, choleras in Bangladesh were caused mostly by V. cholerae O1 of the El Tor biotype, whereas only a few cases were caused by strains of the O139 serogroup. This changing epidemiology of cholera in Bangladesh shifted further recently, and a large outbreak of cholera caused predominantly by V. cholerae O139 occurred in the capital city of Dhaka and adjoining areas during the first half of 2002 (S.M.F., M. A. Salam, A. Faruque, G.B.N., and D.A.S., unpublished data).

Figure 3
Isolation of V. cholerae O1 and O139 from patients admitted to the Infectious Diseases Hospital in Calcutta, India (Upper) and from those admitted to the ICDDR,B Hospital in Dhaka, Bangladesh (Lower) between 1992 and 2000.

Analyses of strains collected during the past 9 years since the first detection of the O139 serogroup have demonstrated the emergence of new clones of V. cholerae O139 and their association with different epidemic outbreaks. Analysis of restriction fragment-length polymorphisms in conserved rRNA genes (ribotypes) as well as CTX genotypes have been used to track clonal relationships among outbreak strains. At least seven different ribotypes have been reported among O139 strains isolated until 2000, and since then, two new ribotypes have been detected (Fig. (Fig.4).4). Within the short span of emergence of O139, the genomic instability of O139 vibrios, at least considering that of the rrn operons as a marker, appears to be more than that of V. cholerae O1 El Tor strains. The observed mutations or rearrangement of rrn operons, allow ribotype diversity that can be exploited to monitor strains with epidemic potential. Yearly isolation of different ribotypes of V. cholerae O139 in Bangladesh and India (Table (Table1)1) clearly shows a striking difference in the proportion of strains belonging to different ribotypes, as well as the distribution of the ribotypes between these two neighboring countries. Since 1993, a decline in the proportion of strains belonging to ribotype B-I and an increase in the proportion of strains belonging to ribotype B-II was noticed both in India and in Bangladesh, and subsequently, in some districts of Bangladesh, strains belonging to ribotype B-III increased considerably and caused an outbreak in 1997 (13). This finding shows the dynamic nature of the epidemiology of V. cholerae O139 and the different clones involved in the initiation of epidemics.

Figure 4
BglI ribotype patterns of V. cholerae O139 isolated between 1992 and 2000. Ribotype designations are shown on top of each lane.
Table 1
Yearly isolation of different ribotypes of V. cholerae O139 in Bangladesh and India between 1992 and 2002

Restriction fragment-length polymorphisms in ctxA genes and its flanking sequences (CTX genotypes) have also been used to define different clones. A total of 11 distinct CTX genotypes were identified among six ribotypes until the year 2000 (46). Although the structural genes for CT are identical in different strains, the observed diversity of CTX genotypes may result from duplication of the CTX prophage. In 1998, the prevalence of two new clones of V. cholerae O139 were observed at two cholera endemic areas, including Calcutta and Alleppey, in southern India (56). Whereas most of the O139 strains from Calcutta have only the El Tor-type CTX prophage, the strains from Alleppey were found to contain the unique arrangement of the CTX prophage of the O139 isolates of 1996. Thus, two different clones of O139 were prevalent at two regions in India, demonstrating contrasting distribution of different O139 clones in different geographic locations.

Preliminary analysis of the newly emerged O139 strains in 2002 in Bangladesh showed that these strains carry a Calcutta-type CTXCal prophage (S.M.F., M. A. Salam, A. Faruque, G.B.N., and D.A.S., unpublished data). It is not clear how the CTX genotype diversity can contribute to an increased incidence of cholera. It may be noted that O139 strains prevailing in Calcutta during 1996 carried the CTXCal prophage (42, 43), and the incidence of O139 cholera in Calcutta was higher than in Dhaka during that period (57).

Antimicrobial Resistance

Analysis of O139 strains isolated during the last 9 years revealed interesting patterns of antibiotic resistance to various common antibiotics. Although the strains remained largely susceptible to ciprofloxacin, tetracycline, and gentamicin, resistance to ampicillin and susceptibility to cotrimoxazole (sulfamethoxazole, trimethoprim), chloramphenicol, and streptomycin varied during this period. The O139 serogroup of V. cholerae that emerged during 1992 and 1993 was sensitive to tetracycline and resistant to trimethoprim-sulfamethoxazole (SXT) and streptomycin. Waldor et al. (51) reported the presence of a 62-kb self-transmissible transposon-like element (SXT element) encoding resistance to sulfamethoxazole, trimethoprim, and streptomycin in V. cholerae O139 strains isolated from this epidemic. The SXT element could be conjugally transferred from V. cholerae O139 to V. cholerae O1 and E. coli strains, where it integrated into recipient chromosomes in a site-specific recA-independent manner (51). Strains isolated from an O139 outbreak in Bangladesh in 1997 were found to be mostly sensitive to SXT and streptomycin (13). In keeping with the observation in Bangladesh, comparison of the antibiotic resistance patterns between the O139 strains isolated during 1992 and 1993 and those isolated in 1996 and 1997 in India also showed that the later strains were susceptible to SXT (14). Analysis of genetic changes associated with the observed SXT sensitivity showed that sensitivity to SXT and streptomycin was associated with a deletion of an ≈3.6-kb region of the SXT element in strains that were sensitive to SXT and streptomycin (13). Since 1997, the O139 strains isolated in India also showed an increased trend of resistance to ampicillin and neomycin and susceptibility to chloramphenicol and streptomycin (41).

This pattern of rapid shift in antimicrobial resistance is consistent with previous reports indicating substantial mobility of genetic elements, which confers resistance to antimicrobials, a phenomenon which has also been observed in V. cholerae O1 strains (2, 58). A multiple antibiotic-resistance plasmid belonging to incompatibility group C has also been associated with drug resistance of V. cholerae O139 (59).

Influence of Genetic Variation on the Epidemiology of Cholera

Although substantial information is available on the epidemiology of V. cholerae and the emergence of new epidemic clones, it is not clear what drives the frequent emergence of new clones often associated with epidemics and the replacement of existing clones. Analyzing genetic variation in isolates of V. cholerae O1 and O139 from successive outbreaks of cholera and the determination of whether these genetic variations contribute to the emergence of new clones of V. cholerae can be an important step in understanding the evolution of new pathogenic strains. An important area that needs to be addressed is whether preexisting immunity against one clone of either O1 or O139 can provide protection completely against another emerging clone. From an epidemiological viewpoint, it certainly appears that genetic rearrangement fosters some advantage to the emerging clone. It is also crucial to understand whether the observed genetic reassortment in the O1 and O139 strains is accompanied by other undiscovered discreet changes in the organism that enables the organism to escape the immune pressure of the host population against previously existing clones of toxigenic V. cholerae. There is no cross protection between V. cholerae O1 and O139 in the animal models. The predominantly adult population infected by O139 cholera also provides evidence that O1 strains do not protect efficiently against O139 strains. Considering the extent of longevity of a clone and the frequency of appearance and reappearance of different clones in a cholera-endemic area, as well as the occurrence of epidemics of cholera with seasonal regularity, it appears that the clonal turnover has greater implications. We hypothesize that the continual emergence of new toxigenic strains and their selective enrichment during cholera outbreaks constitute an essential component of the natural ecosystem for the evolution of epidemic V. cholerae strains to ensure its continued existence. The molecular epidemiological studies indicating continuously developing genetic diversity among clones of V. cholerae O1 and O139 are likely to complicate further the development of an effective cholera vaccine.

Concluding Remarks

Despite efforts to control cholera, the disease continues to occur as a major public health problem in many developing countries. Numerous studies over more than a century have made advances in our understanding of the disease and ways of treating patients, but the mechanism of emergence of new epidemic strains, and the ecology supporting the regular epidemics, remain mysterious and challenging to investigators in the field. V. cholerae provides a natural system to study the coevolution of bacteria and the virulence-associated genetic elements, and the mutual benefits imparted to each other in terms of attaining greater evolutionary fitness. The emergence of V. cholerae O139 has provided an opportunity to study the coevolution of two different serogroups of epidemic V. cholerae strains, apparently driven by competition for survival and thereby attaining enhanced fitness. A continuous surveillance for genetic changes in V. cholerae O139, as well as V. cholerae O1, can provide more insights into the evolution of this serogroup and those factors leading to its establishment as the eighth pandemic strain of cholera.


We thank Afjal Hossain for secretarial assistance. Research work in International Center for Diarrhoeal Disease, Bangladesh, is supported by countries and agencies that share its concern for the health problems of developing countries. Current donors providing unrestricted support include the aid agencies of the governments of Australia, Bangladesh, Belgium, Canada, Japan, Kingdom of Saudi Arabia, The Netherlands, Sweden, Sri Lanka, Switzerland, and the United States of America.


CTcholera toxin
LPSbacterial lipopolysaccharide
ISinsertion sequence


This contribution is part of the special series of Inaugural Articles by members of the National Academy of Sciences elected on April 30, 2002.


1. Kaper J B, Morris J G, Levine M M. Clin Microbiol Rev. 1995;8:48–86. [PMC free article] [PubMed]
2. Faruque S M, Albert M J, Mekalanos J J. Microbiol Mol Biol Rev. 1998;62:1301–1314. [PMC free article] [PubMed]
3. Pollitzer R. In: Cholera. Pollitzer R, editor. Geneva: W.H.O.; 1959. pp. 11–50.
4. Yamai S, Okitsu T, Shimada T, Katsube Y. J Jpn Assoc Infect Dis. 1997;71:1037–1045. [PubMed]
5. Cholera Working Group, International Center for Diarrhoeal Disease Research, Bang-ladesh. Lancet. 1993;342:387–390. [PubMed]
6. Ramamurthy T, Garg S, Sharma R, Bhattacharya S K, Nair G B, Shimada T, Takeda T, Karasawa T, Kurazano H, Pal A, Takeda Y. Lancet. 1993;341:703–704. [PubMed]
7. Chongsa-nguan M, Chaicumpa W, Moolasart P, Kandhasingha P, Shimada T, Kurazono H, Takeda Y. Lancet. 1993;342:430–431. [PubMed]
8. Centers for Disease Control. Morbid Mortal Wkly Rep. 1993;42:501–503. [PubMed]
9. Nair G B, Ramamurthy T, Bhattacharya S K, Mukhopadhyay A K, Garg S, Bhattacharya M K, Takeda T, Shimada T, Takeda Y, Deb B C. J Infect Dis. 1994;169:1029–1034. [PubMed]
10. Swerdlow D L, Ries A A. Lancet. 1993;342:382–383. [PubMed]
11. Sinha S, Chakraborty R, De K, Khan A, Datta S, Ramamurthy T, Bhattacharya S K, Takeda Y, Nair G B. J Clin Microbiol. 2002;40:2635–2637. [PMC free article] [PubMed]
12. Faruque S M, Ahmed K M, Siddique A K, Zaman K, Alim A R M A, Albert M J. J Clin Microbiol. 1997;35:2299–2306. [PMC free article] [PubMed]
13. Faruque S M, Siddique A K, Saha M N, Asadulghani, Rahman M M, Zaman K, Albert M J, Sack D A, Sack R B. J Clin Microbiol. 1999;37:1313–1318. [PMC free article] [PubMed]
14. Mitra R, Basu A, Dutta D, Nair G B, Takeda Y. Lancet. 1996;348:1181. [PubMed]
15. Shimada T, Nair G B, Deb B C, Albert M J, Sack R B, Takeda Y. Lancet. 1993;341:1347.
16. Janda J M, Powers C, Bryant R G, Abbott S L. Clin Microbiol Rev. 1988;1:245–267. [PMC free article] [PubMed]
17. Faruque S M, Alim A R M A, Roy S K, Khan F, Nair G B, Sack R B, Albert M J. J Clin Microbiol. 1994;32:1050–1053. [PMC free article] [PubMed]
18. Calia K E, Murtagh M, Ferraro M J, Calderwood S B. Infect Immun. 1994;62:1504–1506. [PMC free article] [PubMed]
19. Popovic T, Fields P I, Olsvik O, Wells J G, Evins G M, Cameron D N, Farmer J J, III, Bopp C A, Wachsmuth K, Sack R B, et al. J Infect Dis. 1995;171:122–127. [PubMed]
20. Berche P, Poyart C, Abachin E, Lelievre H, Vandepitte J, Dodin A, Fournier J-M. J Infect Dis. 1994;170:701–704. [PubMed]
21. Bhattacharya S K, Bhattacharya M K, Nair G B, Dutta D, Deb A, Ramamurthy T, Garg S, Saha P K, Dutta P, Moitra A, et al. J Infect Dis. 1993;27:11–15. [PubMed]
22. Morris J G, Jr, Losonsky G E, Johnson J A, Tacket C O, Nataro J P, Panigrahi P, Levine M M. J Infect Dis. 1995;171:903–908. [PubMed]
23. Johnson J A, Salles C A, Panigrahi P, Albert M J, Wright A C, Johnson R J, Morris J G., Jr Infect Immun. 1994;62:2108–2110. [PMC free article] [PubMed]
24. Waldor M K, Colwell R, Mekalanos J J. Proc Natl Acad Sci USA. 1994;91:11388–11392. [PMC free article] [PubMed]
25. Weintraub A, Widmalm G, Jansson P-E, Jansson M, Hultenby K, Albert M J. Microb Pathog. 1994;16:235–241. [PubMed]
26. Stroeher U H, Jedani K E, Dredge B K, Morona R, Brown M H, Karageorgos L E, Albert M J, Manning P A. Proc Natl Acad Sci USA. 1995;92:10374–10378. [PMC free article] [PubMed]
27. Waldor M K, Mekalanos J J. Lancet. 1994;343:1366. [PubMed]
28. Bik E M, Bunschoten A E, Gouw R D, Mooi F R. EMBO J. 1995;14:209–216. [PMC free article] [PubMed]
29. Bik E M, Bunschoten A E, Willems R J L, Chang A C Y, Mooi F R. Mol Microbiol. 1996;20:799–811. [PubMed]
30. Comstock L E, Johnson J A, Michalski J M, Morris J G, Jr, Kaper J B. Mol Microbiol. 1996;19:815–826. [PubMed]
31. Stroeher U H, Parasivam G, Dredge B K, Manning P A. J Bacteriol. 1997;179:2740–2747. [PMC free article] [PubMed]
32. Sozhamannan S, Deng Y K, Li M, Sulakvelidze A, Kaper J B, Johnson J A, Nair G B, Morris J G., Jr Infect Immun. 1999;67:5033–5040. [PMC free article] [PubMed]
33. Stroeher U H, Karageorgos L E, Morona R, Manning P A. Proc Natl Acad Sci USA. 1992;89:2566–2570. [PMC free article] [PubMed]
34. Stroeher U H, Jedani K E, Manning P A. Gene. 1998;223:269–282. [PubMed]
35. Mooi F R, Bik E M. Trends Microbiol. 1997;4:161–165. [PubMed]
36. Stroeher U H, Manning P A. Trends Microbiol. 1997;5:178–180. [PubMed]
37. Isshiki Y, Kondo S, Iguchi T, Sano Y, Shimada T, Hisatsune K. Microbiology. 1996;142:1499–1504. [PubMed]
38. Yamasaki S, Shimizu T, Hoshino K, Ho S T, Shimada T, Nair G B, Takeda Y. Gene. 1999;237:321–332. [PubMed]
39. Faruque S M, Saha M N, Asadulghani, Sack D A, Sack R B, Takeda Y, Nair G B. J Infect Dis. 2000;182:1161–1168. [PubMed]
40. Waldor M K, Mekalanos J J. Science. 1996;272:1910–1914. [PubMed]
41. Mukhopadhyay A K, Basu A, Garg P, Bag P K, Ghosh A, Bhattacharya S K, Takeda Y, Nair G B. J Clin Microbiol. 1998;36:2149–2152. [PMC free article] [PubMed]
42. Sharma C, Maiti S, Mukhopadhyay A K, Basu A, Basu I, Nair G B, Mukhopadhyaya R, Das B, Kar S, Ghosh R K, Ghosh A. J Clin Microbiol. 1997;35:3348–3350. [PMC free article] [PubMed]
43. Kimsey H H, Nair G B, Ghosh A, Waldor M K. Lancet. 1998;353:457–458. [PubMed]
44. Farfán M, Miñana-Galbis D, Fusté M C, Lorén J G. J Bacteriol. 2002;184:1304–1313. , [PMC free article] [PubMed]
45. Beltrán P, Delgado G, Navarro A, Trujillo F, Selander R K, Cravioto A. J Clin Microbiol. 1999;37:581–590. [PMC free article] [PubMed]
46. Faruque S M, Saha M N, Asadulghani, Bag P K, Bhadra R K, Bhattacharya S K, Sack R B, Takeda Y, Nair G B. FEMS Microbiol Lett. 2000;184:279–284. [PubMed]
47. Farfán M, Miñana D, Fusté M C, Lorén J G. Microbiology. 2000;146:2613–2626. [PubMed]
48. Basu A, Garg P, Datta S, Chakraborty S, Bhattacharya T, Khan A, Ramamurthy T, Bhattacharya S K, Yamasaki S, Takeda Y, Nair G B. Emerg Infect Dis. 2000;6:139–147. [PMC free article] [PubMed]
49. Faruque S M, Ahmed K M, Alim A R M A, Qadri F, Siddique A K, Albert M J. J Clin Microbiol. 1997;35:624–630. [PMC free article] [PubMed]
50. Faruque S M, Roy S K, Alim A R M A, Siddique A K, Albert M J. J Clin Microbiol. 1995;33:2833–2838. [PMC free article] [PubMed]
51. Waldor M K, Tschape H, Mekalanos J J. J Bacteriol. 1996;178:4157–4165. [PMC free article] [PubMed]
52. Glass R I, Huq I, Alim A R M A, Yunus M. J Infect Dis. 1980;142:939–942. [PubMed]
53. Glass R I, Huq M I, Lee J V, Threlfall E J, Khan M R, Alim A R M A, Rowe B, Gross R J. J Infect Dis. 1983;147:204–209. [PubMed]
54. Baker A, Clark C A, Manning P A. J Bacteriol. 1994;176:5450–5458. [PMC free article] [PubMed]
55. Mazel D, Dychinco B, Webb V A, Davies J. Science. 1998;280:605–608. [PubMed]
56. Garg P, Nandy R K, Chaudhury P, Chowdhury N R, De K, Ramamurthy T, Yamasaki S, Bhattacharya S K, Takeda Y, Nair G B. J Clin Microbiol. 2000;38:4249–4253. [PMC free article] [PubMed]
57. Basu A, Mukhopadhyay A K, Sharma C, Jyot J, Gupta N, Ghosh A, Bhattacharya S K, Takeda Y, Faruque A S G, Albert M J, Nair G B. Microb Pathog. 1998;24:175–183. [PubMed]
58. Mukhopadhyay A K, Garg S, Mitra R, Basu A, Dutta D, Bhattacharya S K, Shimada T, Takeda T, Takeda Y, Nair G B. J Clin Microbiol. 1996;34:2537–2543. [PMC free article] [PubMed]
59. Yamasaki S, Nair G B, Bhattacharya S K, Yamamoto S, Kurazono H, Takeda Y. Microbiol Immunol. 1997;41:1–6. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Cited in Books
    Cited in Books
    NCBI Bookshelf books that cite the current articles.
  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles
  • Taxonomy
    Taxonomy records associated with the current articles through taxonomic information on related molecular database records (Nucleotide, Protein, Gene, SNP, Structure).
  • Taxonomy Tree
    Taxonomy Tree

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...