Clinical Isolates of Vibrio cholerae O1 El Tor Ogawa of 2009 from Kolkata, India: Preponderance of SXT Element and Presence of Haitian ctxB Variant

Background Increase in the number of multidrug resistant pathogens and the accompanied rise in case fatality rates has hampered the treatment of many infectious diseases including cholera. Unraveling the mechanisms responsible for multidrug resistance in the clinical isolates of Vibrio cholerae would help in understanding evolution of these pathogenic bacteria and their epidemic potential. This study was carried out to identify genetic factors responsible for multiple drug resistance in clinical isolates of Vibrio cholerae O1, serotype Ogawa, biotype El Tor isolated from the patients admitted to the Infectious Diseases Hospital, Kolkata, India, in 2009. Methodology/Principal Findings One hundred and nineteen clinical isolates of V. cholerae were analysed for their antibiotic resistance phenotypes. Antibiogram analysis revealed that majority of the isolates showed resistance to co-trimoxazole, nalidixic acid, polymixin B and streptomycin. In PCR, SXT integrase was detected in 117 isolates and its sequence showed 99% identity notably to ICEVchInd5 from Sevagram, India, ICEVchBan5 from Bangladesh and VC1786ICE sequence from Haiti outbreak among others. Antibiotic resistance traits corresponding to SXT element were transferred from the parent Vibrio isolate to the recipient E. coli XL-1 Blue cells during conjugation. Double-mismatch-amplification mutation assay (DMAMA) revealed the presence of Haitian type ctxB allele of genotype 7 in 55 isolates and the classical ctxB allele of genotype 1 in 59 isolates. Analysis of topoisomerase sequences revealed the presence of mutation Ser83 → Ile in gyrA and Ser85→ Leu in parC. This clearly showed the circulation of SXT-containing V. cholerae as causative agent for cholera in Kolkata. Conclusions There was predominance of SXT element in these clinical isolates from Kolkata region which also accounted for their antibiotic resistance phenotype typical of this element. DMAMA PCR showed them to be a mixture of isolates with different ctxB alleles like classical, El Tor and Haitian variants.


Introduction
Vibrio cholerae is a Gram-negative pathogen that causes cholera, an acute dehydrating diarrhoea which is globally important as it occurs in endemic, epidemic and pandemic forms [1,2]. V. cholerae has been classified on the basis of its somatic O-antigen and more than 200 serogroups have been identified. Out of these, only O1 and O139 are epidemic [1,2]. The emerging multiple drug resistance in all the bacterial pathogens including V. cholerae is complicating the treatment of diseases and therefore is a major public health concern. Chromosome-borne and/or mobile genetic element-borne genes contribute to the drug resistance phenotype of a bacterium. The acquisition and dissemination of antibiotic resistance genes is mediated by mobile genetic elements like plasmids, integrons and transposons [3]. One such transposon is SXT element, an integrative conjugative element (ICE) that integrates and replicates with the host chromosome, can excise itself and be transferred between bacteria by conjugation [4]. ICEs are known to transfer a diverse array of functions including antibiotic resistance genes [4]. SXT element was first reported in 1993 from India; strain O139, MO10 which encoded resistance to sulfamethoxazole, trimethoprim, chloramphenicol and streptomycin [5]. SXT MO10 -related elements are now present in most V. cholerae O139 and O1 clinical isolates [4][5][6][7]. For evolutionary reasons, V. cholerae strains have been continuously changing from classical to El Tor, from O1 to O139, from Ogawa to Inaba, SXT M010 /R391 hybrids and from plain ctxB to ctxB hybrids [6][7][8]. India and Bangladesh have been the haven for evolutionary optimisation of this pathogen and SXT-related ICEs have been characterized from these regions [7,9]. The ongoing Haiti outbreak has also been predicted to originate from Southeast Asian region [7,[10][11][12][13][14][15][16] though the controversies still remain regarding the precise geographical source and the etiological agent [15,16].
In earlier studies from this laboratory, various genetic factors like efflux pumps, plasmids, integrons, qnr genes and mutations in topoisomerases were evaluated for their role in conferring antibiotic resistance [17][18][19][20]. In the present study, V. cholerae O1 Ogawa isolated from the patients of Infectious Diseases Hospital (IDH) of Kolkata, India, in 2009, were examined for genetic factors governing their antibiotic resistance profiles. Results revealed the prevalence of SXT element and the absenceof integrons in these isolates. Antibiotic resistance traits and their transferability by conjugation also corroborated the presence of this mobile genetic element. Interestingly, Double-Mismatch-Amplification Mutation Assay (DMAMA) showed the presence of classical, El Tor as well as Haitian ctxB variants in these isolates. Mutations in topoisomerase genes gyrA and parC governed the quinolone resistance phenotype in these isolates.

Bacterial Strains, Genomic and Plasmid DNA Isolation
One hundred and nineteen isolates of V. cholerae O1 Ogawa were obtained from patients with acute cholera admitted to the Infectious Diseases Hospital (IDH), Kolkata, India, in 2009 and these patient samples were anonymized. The participants provided their written consent for participating in the study and in case of children, written consent was obtained from their parents. The consent procedure was approved by the Institutional Ethical Clearance Committee of National Institute of Cholera and Enteric Diseases (NICED), Kolkata, from where the samples were obtained for this study. The study was also approved by the Institutional Biosafety Committee (IBSC) of Indian Institute of Advanced Research, Gandhinagar, and the Review Committee on Genetic Manipulation (RCGM) governed by guidelines laid down by Department of Biotechnology, Govt. of India.
V. cholerae strains MO10, O1 El Tor N16961, O1 classical Inaba strain 569B were used as controls in various experiments. Escherichia coli XL-1 Blue cells were used as recipient in conjugation experiments. Genomic and plasmid DNA isolations were done as described previously [21].

Polymerase Chain Reactions
Primer pairs L2/L3, qacED1/Sul1B, In-F/In-B were used for detection and characterization of class 1 integrons using the same conditions as described before [17]. Amplification of SXT integrase was carried out using the forward primer 59-ATGGCGTTATGAGTTAGCTC-39 and the reverse primer 59-GCGAAGATCATGCATAGAC-39. For amplification of SXT integrase, conditions used for PCR were similar as described earlier [17] except that annealing was carried out at 57uC for 30 s and extension was carried out at 72uC for 1 min. DMAMA-PCR was carried out to screen for the type of ctxB gene present using the primers ctxB-F3, ctxB-F4, Fw-con, Rv-El Tor and Rv-Cla as described recently [23]. PCR amplifications for four topoisomerase genes (gyrA, gyrB, parC, parE) were carried out as described earlier [18]. PCR reactions were performed using a PTC-225 DNA Engine Tetrad TM Cycler (MJ Research Inc., MA, USA) and Pfu (Fermentas International Inc., Ontario, Canada) or Taq DNA polymerases (Bangalore Genei, Bangalore, India).

Bacterial Conjugation
Two representative SXT-positive isolates (IDH01572 and IDH01738) were tested for their ability to transfer SXT-borne resistance traits to the recipient strain in conjugation experiments according to the published protocols [18]. SXT-negative strain (IDH02095) was included as control in this experiment using the same recipient. The strains IDH01738 (streptomycin resistant), IDH01572 (streptomycin resistant) and IDH02095 (ampicilin and nalidixic acid resistant) were used as donors while E. coli XL1-Blue (tetracycline, nalidixic acid resistant and streptomycin sensitive) was used as recipient. Briefly, the recipient and donor strains were mixed in a ratio of 2:1 on a sterile 0.45 mm nylon membrane (Nytran N, Whatman, PA, USA) and incubated overnight for mating on LB agar at 37uC. The transconjugants were selected on LB agar plates containing appropriate antibiotics. For conjugation between IDH01738, IDH01572 and XL1-Blue, streptomycin (20 mg/mL) and tetracycline (25 mg/mL) were used. For conjugation between IDH02095 and XL1-Blue, selection of ampicillin (25 mg/mL) and tetracycline (25 mg/mL) was used. The transconjugants were confirmed for their authenticity by determination of their antibiotic susceptibility profiles and by PCR for SXT integrase using the donor and recipient strains as controls in both these assays.

DNA Sequence Analysis
DNA segments amplified from SXT integrase and topoisomerase genes were sequenced and subsequently, the sequences were assembled for each amplicon. The assembled sequences were analysed by nucleotide BLAST search at the National Center for Biotechnology Information (NCBI) website (http://www.ncbi.nlm. nih.gov). Alignment of the topoisomerase sequences used DNAS-TAR. All new data corresponding to the DNA sequences of SXT integrase and topoisomerase genes were deposited in GenBank and the accession numbers have been mentioned in the appropriate places in results section.

Presence of SXT Element and Class 1 Integron
One hundred and seventeen of the 119 isolates produced a1.0 kb amplicon with primers specific for SXT integrase (Figure 2). Sequence analysis of this amplicon was done using BLAST and sequence for the isolate IDH02596 was deposited into GenBank (JQ013431). Results revealed that the SXT sequence had 99% identity to SXT integrase from many other isolates including ICEVchInd5 from Sevagram, India (GQ463142), ICEVchBan5 from Bangladesh (GQ463140) and VC1786ICE sequence from Haiti outbreak (JN648379). In all the 117 isolates, presence of SXT element correlated with resistance to cotrimoxazole. All the isolates were negative for class 1 integron in PCR assay.

Transfer of Resistance Traits by Conjugation
To confirm the transferability of SXT element, conjugation experiments were carried out with two representative SXTpositive isolates (IDH01572 and IDH01738) as donors using E. coli XL1-Blue cells as recipients. IDH02095, a SXT-negative isolate, was taken as negative control in this experiment. Transconjugants were obtained with SXT-positive isolates whereas SXT-negative isolates could not yield any transconjugants. PCR and antibiogram analysis of the transconjugants indicated the transfer of SXT element and resistance traits harboured by them (STR, TRI, SUL and COT), from the donor to the recipient cell (Table 1; Figure 2).

Presence of Haitian Variant of ctxB
DMAMA-PCR was carried out to discriminate the classical, Haitian and El Tor ctxB alleles present in these V. cholerae isolates as described in a recent report [23]. This assay distinguishes the three ctxB alleles based on the mutations specific to each type. Haitian variant carries a mutation at 58 th nucleotide corresponding to 20 th amino acid (His20 in classical and El Tor R Asn in Haitian allele) which forms the basis of primer ctxB-F3 for Haitian ctxB and primer ctxB-F4 for classical ctxB. The reverse primer Rv-Cla would anneal to both Haitian as well as classical ctxB [23]. El Tor allele would not show amplification in DMAMA-PCR as neither of the forward primers (ctxB-F3 or ctxB-F4) nor the reverse primer (Rv-Cla) would anneal to this ctxB variant. PCR results revealed that this population of 119 clinical isolates was a mixture of Haitian (genotype 7) and non-Haitian (genotype 1) classical ctxB gene. The Haitian allele was present in 46.2% of the isolates (55 out of 119) that yielded a 191-bp fragment in a PCR with ctxB-F3 and Rv-Cla primers. Rest of the isolates showed either classical ctxB allele (59 out of 119) that yielded 191-bp fragment with the primer pair ctxB-F4 and Rv-Cla or El Tor ctxB allele (5 out of 119) that did not yield any amplicon in the two PCR assays mentioned above.

Mutations in Topoisomerases
Out of 119 strains, few representative strains were selected for amplification and sequencing of the Quinolone-Resistance-Determining Regions (QRDRs) from the four topoisomerase genes for GyrA, GyrB, ParC and ParE. Sequences of these genes for the isolate IDH02431were deposited in GenBank. (JX081540-JX081543). Results revealed that these isolates carried the mutations encoding Ser83R Ileu in gyrA and Ser85R Leu in parC genes. No mutations were detected in gyrB and parE genes. The nucleotide BLAST analysis of the sequences from all four topoisomerase genes from Kolkata isolates showed 99% identity with many sequences including the ones from the strains 2010EL-

Discussion
With the dismal scenario of an alarming increase in the drug resistance of various infectious pathogens, continuous surveillance becomes imperative to understand the pathogens and their changing drug resistance profiles for an effective treatment [3]. The present study was undertaken to determine the drug resistance profiles of one hundred and nineteen clinical isolates of V. cholerae O1 El Tor Ogawa from Kolkata, in 2009, and unravel some of the mechanisms that could be responsible for their observed drug resistance phenotypes. Presence of SXT element in majority of the isolates and their resistance to drugs characteristic of SXT element clearly showed the circulation of this genetic factor in the clinical isolates of V. cholerae. Sequences of topoisomerases from the representative isolates in this Indian population indicated the presence of a mutation in GyrA (Ser83R Ileu) and another mutation in ParC (Ser85R Leu). Though the mutation Ser83R Ileu in GyrA accounted for the observed nalidixic acid resistance in 100% of the isolates, the effect of the other mutation in ParC could not be explained as the isolates were not resistant to fluoroquinolones like ciprofloxacin and norfloxacin for which this mutation is known to contribute [24,25]. Resistance profile of these isolates was very similar to the one described earlier for V. cholerae strains circulating in India from the year 2004-2007 and also in Haiti [6,26]. A recent report has described DMAMA-PCR as an effective tool to study the emergence and dissemination of Haitian ctxB allele in India [23]. Our study using the DMAMA-PCR also established that the 2009 V. cholerae isolates had 46.2% of the Haitian ctxB allele corresponding to genotype 7, a new variant of ctxB that has been reported from Orissa, India, in 2007 and later in Kolkata from 2006-2011 [6,23].
Though there are few reports describing different number of V. cholerae clinical isolates from India in recent years [6,7,23,25,[27][28][29][30], our report adds another dimension to these finding from India and abroad to once again show that a large number of the Indian strains as recent as 2009 had similar characteristics of the strain that caused Haiti outbreak. It is established now that UN peacekeepers from Nepal were the actual carriers of the pathogen in Haiti but such isolates have also been circulating in the Southeast Asian regions like India and Bangladesh since last few years [10][11][12][13][14][15][16]29,30]. An earlier report had established the genetic relatedness of the ICEVchHAI 1, an ICE element from the strain that caused Haiti outbreak, with that of ICEVchInd5, an ICE from an Indian isolate from Sevagram, India, in 1994 [25]. BLAST search of the sequence of SXT integrase in our study showed 99% identity with ICEVchHAI, ICEVchInd5, ICEVflInd1 (from V. fluvialis) and ICEVchBan5; all belonging to the same group 2 of SXT/R391 ICEs. Only 95% identity with ICEVchInd4 (from O139) with origin in Kolkata (group 1of ICEs) and no relation/ identity was observed with the third group of ICEs containing the Matlab variant/ICEVchBan9 [7,9,30]. Persistence of this ICEV-chInd5 element in India till 2005 has also been documented [7]. In this study, the continued presence of ICEVchInd5 till 2009 and its preponderance in V. cholerae O1 El Tor Ogawa clinical isolates is shown from Kolkata. It would be pertinent to state here that in this laboratory, earlier work with V. fluvialis clinical isolate BD146 had indicated horizontal transfer of a plasmid between V. fluvialis and V. cholerae O1 [17]. Results described in this paper again indicate the possibility of transfer of an SXT element between these two Vibrio species. Though partially proved, further work would be required to prove the sibling relationships of these ICEs and complete characterization of the genetic content of these ICE elements probed in the present study [7]. These studies aimed at understanding the molecular nature of antibiotic resistance do not really help clinical manangement of the diseases but they are an important insight into the evolution and dissemination of the deadly pathogens that traverse the varied geographical and climatic conditions to give rise to an outbreak or a pandemic.