ORIGINAL RESEARCH article

Front. Microbiol., 06 September 2019

Sec. Antimicrobials, Resistance and Chemotherapy

Volume 10 - 2019 | https://doi.org/10.3389/fmicb.2019.02072

Altered Integrative and Conjugative Elements (ICEs) in Recent Vibrio cholerae O1 Isolated From Cholera Cases, Kolkata, India

  • 1. Division of Bacteriology, National Institute of Cholera and Enteric Diseases, Kolkata, India

  • 2. Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama University, Okayama, Japan

  • 3. Collaborative Research Center of Okayama University for Infectious Diseases in India, National Institute of Cholera and Enteric Diseases, Kolkata, India

  • 4. Center for Human Microbial Ecology, Translational Health Science and Technology Institute, Faridabad, India

Abstract

The self-transferring integrative and conjugative elements (ICEs) are large genomic segments carrying several bacterial adaptive functions including antimicrobial resistance (AMR). SXT/R391 family is one of the ICEs extensively studied in cholera-causing pathogen Vibrio cholerae. The genetic characteristics of ICE-SXT/R391 in V. cholerae are dynamic and region-specific. These ICEs in V. cholerae are strongly correlated with resistance to several antibiotics such as tetracycline, streptomycin and trimethoprim-sulfamethoxazole. We screened V. cholerae O1 strains isolated from cholera patients in Kolkata, India from 2008 to 2015 for antibiotic susceptibility and the presence of ICEs, and subsequently sequenced their conserved genes. Resistance to tetracycline, streptomycin and trimethoprim-sulfamethoxazole was detected in strains isolated during 2008–2010 and 2014–2015. The genes encoding resistance to tetracycline (tetA), trimethoprim-sulfamethoxazole (dfrA1 and sul2), streptomycin (strAB), and chloramphenicol (floR) were detected in the ICEs of these strains. There was a decrease in overall drug resistance in V. cholerae associated with the ICEs in 2011. DNA sequence analysis also showed that AMR in these strains was conferred mainly by two types of ICEs, i.e., ICETET (comprising tetA, strAB, sul2, and dfrA1) and ICEGEN (floR, strAB, sul2, and dfrA1). Based on the genetic structure, Kolkata strains of V. cholerae O1 had distinct genetic traits different from the ICEs reported in other cholera endemic regions. Transfer of AMR was confirmed by conjugation with sodium azide resistant Escherichia coli J53. In addition to the acquired resistance to streptomycin and trimethoprim-sulfamethoxazole, the conjugally transferred (CT) E. coli J53 with ICE showed higher resistance to chloramphenicol and tetracycline than the donor V. cholerae. Pulsed-field gel electrophoresis (PFGE) based clonal analysis revealed that the V. cholerae strains could be grouped based on their ICEs and AMR patterns. Our findings demonstrate the epidemiological importance of ICEs and their role in the emergence of multidrug resistance (MDR) in El Tor vibrios.

Introduction

The Gram-negative pathogen Vibiro cholerae O1 has caused seven pandemics in the history of cholera and tends to cause several epidemics in developing countries (). This pathogen has more than 200 serogroups, but only the serogroups O1 and O139 are associated with epidemic cholera (). The ongoing seventh pandemic is linked with the El Tor biotype of serogroup O1 that has spread in the cholera endemic regions of the world (). The emergence and spread of antimicrobial resistant (AMR) V. cholerae, especially those resistant to nalidixic acid, tetracycline, and trimethoprim-sulfamethoxazole, has been reported since the 1980s (). Resistance to these antimicrobials has been strongly associated with the presence of integrative and conjugative elements (ICEs) of the SXT/R391 family and its discovery has greatly changed the understanding of AMR in V. cholerae.

SXT/R391 ICEs have been characterized/classified based on the conserved core genes, and their integration into the 5′-end of the prfC gene that encodes peptide chain release factor 3 (). More than 1000 ICEs have been updated in the ICEberg database1. Mobility of SXT/R391 ICEs occurs between bacteria by conjugation, resulting in the transfer of several functions including AMR, resistance to heavy metals, regulation of motility and biofilm formation (; ). Five insertion hotspots (H1 to H5) and four variable regions (VRI to VRIV) are also carried by the ICEs (). The structure of ICEs changes periodically contributing to the differences in AMR profiles of V. cholerae. More than 50 ICEs have been grouped within the SXT/R391 family, of which 30 are reported in clinical and environmental V. cholerae strains (). Between 1992 and 2001, 15 ICEs were identified in India and Bangladesh, of which six (SXTMO10, ICEVchInd4, ICEVchBan5, ICEVchBan10, ICEVchBan9, and ICEVchInd5) were completely sequenced and annotated ().

Tetracycline has been the drug of choice in treating cholera cases for a long time (). A sudden upsurge in the tetracycline resistance (TetR), from 1% in 2004 to 76% in 2007, was reported among V. cholerae in Kolkata and it decreased to about 50% in 2009 (). Similar trends have been observed previously in large cholera epidemics in Tanzania and Madagascar due to extensive prophylactic use of tetracycline (; ). Only a few studies have been carried out to understand the mechanisms of AMR due to ICEs in India (; ; ). In this study, we screened the AMR patterns of V. cholerae O1 Ogawa strains isolated from cholera patients in Kolkata, India from 2008 to 2015 and examined the type of ICEs present by analyzing their backbone genes. Our study revealed the differences between the sequence types of ICEs and recent changes in AMR patterns of V. cholerae.

Materials and Methods

Clinical Specimens and Bacterial Strains

Stool specimens were collected from the Infectious Diseases Hospital (IDH) and B. C. Roy Children Hospital (BCH), Kolkata, before the patients were treated with antibiotics. Clinical symptoms of diarrheal patients included loose/watery stools with or without dehydration, abdominal cramps, vomiting and fever. Dysentery patients had frequent passage of stool with blood/mucus and mild to severe abdominal pain. For the isolation of V. cholerae, all the stool specimens/rectal swabs were enriched in alkaline peptone water (pH 8.0) (Difco, Sparks, MD, United States) for 6 h, followed by inoculation and overnight incubation in thiosulphate citrate bile-salts sucrose agar (TCBS, Eiken, Tokyo, Japan) plates. Sucrose-positive strains were confirmed serologically using commercially available V. cholerae O1 poly and monovalent antisera (Denka-Seiken, Tokyo, Japan). To obtain the AMR pattern from 2008 to 2015, 546 out of 1591 strains were randomly selected covering each month of the study period. Sodium azide resistant (AzR) Escherichia coli J53 () was used for the conjugation experiments. All the strains were preserved in Luria Bertani (LB) broth (Difco) containing 15% glycerol at −80°C. E. coli ATCC 25922 () was used as a control strain in antimicrobial susceptibility testing.

Antibiotic Susceptibility Testing

Susceptibilities of V. cholerae strains to ampicillin (AMP, 10 μg), ceftriaxone (CRO, 30 μg), chloramphenicol (CHL, 30 μg), nalidixic acid (NA, 30 μg), ciprofloxacin (CIP, 5 μg), ofloxacin (OFX, 5 μg), norfloxacin (NOR, 10 μg), imipenem (IPM, 10 μg), streptomycin (STR, 10 μg), azithromycin (AZM, 15 μg), tetracycline (TET, 30 μg), trimethoprim-sulfamethoxazole (SXT, 1.25 and 23.75 μg) and gentamicin (GEN, 10 μg), were determined by Kirby-Bauer disk diffusion technique using commercial disks (BD, Sparks, MD, United States) as per the Clinical and Laboratory Standards Institute guidelines (, ).

Detection of Antibiotic Resistance Encoding Genes

Total nucleic acid of V. cholerae strains was extracted using a QIAamp DNA mini kit (Qiagen, Hilden, Germany) following the manufacturer’s instructions. The integrase gene (intSXT) present in ICE was amplified by PCR using previously described primer pair int1-F and int1-B (). Beside intSXT, PCR was also performed to detect the presence of resistance encoding genes for chloramphenicol (floR and cat), streptomycin (strA and strB), and sulfonamide (sul1 and sul2) (). Primer pairs VCtetA.F-(5′- ACGGTATCCTGCTGGCACTGTATG-3′) and VCtetA.R-(5′- CATCCATATCCAGCCATCCCAACT-3′) and VctetR.F-(5′-GA AGTGGGAATGGAAGGGCTGAC-3′) and VctetR.R-(5′-AG CCTCTGTGCCATCATCTTG-3′) were designed to detect the TetR encoding gene (tetA), and the repressor protein (tetR) for a regulatory portion of resistance cassettes, respectively. Representative amplicons were purified using a PCR product purification kit (Qiagen) and sequenced using the ABI Big Dye terminator cycle sequencing ready reaction kit, version 3.1 (Applied Biosystems, Foster City, CA, United States) in an automated DNA sequencer (ABI 3730, Applied Biosystems). The sequences were assembled and analyzed using DNASTAR software (DNASTAR Inc., Madison, WI, United States).

Conjugation

To test the mobility of the ICEs, conjugation assay was carried out using a representative ICE-positive V. cholerae O1 strain as donor with E. coli J53 (AzR, ). In brief, overnight cultures of the bacteria were mixed at 1:2 donor-to-recipient ratios in 1 ml of LB broth and allowed to grow overnight at 37°C. The donor and recipient suspensions were diluted serially in phosphate buffer saline (PBS) and plated on TCBS and MacConkey agar plates, respectively, to confirm the purity and count the number of colonies. To detect the conjugally transferred E. coli J53 (CT-E. coli J53), MacConkey agar supplemented with streptomycin (100 μg/ml) and sodium azide (AZD, 100 μg/ml) was used. Transconjugants were confirmed as ICE-positive by PCR analysis, followed by PCR amplicon sequencing. To confirm the resistance phenotype, antibiotic susceptibility patterns of the donor, recipient and transconjugants were determined after their growth on Mueller-Hinton (MH, Difco) agar by disk diffusion method. An increase in resistance of transconjugants was quantified by determining the MICs of CHL, STR, TET, and SXT using E-test strips (AB bioMérieux, Solna, Sweden).

Pulsed-Field Gel Electrophoresis (PFGE)

Clonal analysis of representative V. cholerae O1 strains isolated between 2008 and 2015 was made following the PulseNet protocol (). V. cholerae O1 strains were used after digesting the DNA with NotI [New England Biolabs (NEB), Ipswich, MA, United States]. XbaI (NEB) digested Salmonella Braendruff H-9812 was used as a DNA size marker. The PFGE run conditions were generated by the auto-algorithm mode of the CHEF Mapper system (Bio-Rad, Hercules, CA, United States). PFGE profiles were analyzed by the BioNumerics version 4.0 software (Applied Maths, Sint-Martens-Latem, Belgium) using the Dice coefficient and unweighted pair group method using arithmetic averages (UPGMA).

Whole Genome Sequence Analysis

The whole genome sequences submitted from our previous study () were used in the analysis. The open reading frames (ORFs) from the contigs were generated by contig integrator for sequence assembly (CISA) using Glimmer-MG program2. Nucleotide sequences and amino acid sequences were obtained from these ORFs and translated in the appropriate frame. The predicted ORFs were annotated using CANoPI (Contig Annotator Pipeline) that also includes BlastX search for each ORF sequence against the “nr” database of NCBI3. From the whole genome sequence data of representative strains (TetR IDH 1986 and TetS IDH 4268), we have used part of the ICE region in the analysis. The contigs were aligned, assembled and compared with SEQMAN, assembly module of DNASTAR’s LASERGENE with published sequences like ICEVchInd5 (GQ463142), ICEVchBan5 (GQ463140), MO10 (AY055428), etc. For confirmation, PCR was performed targeting important short regions of the ICEs (rumAB, traI, traC, setR, traA-traC, and traG) with previously described primers (). Published ICE sequences were used for homology search. ORF search and gene prediction were performed for the complete ICE region with EditSEQ, Lasergene software (DNASTAR), and pairwise alignment was analyzed by blastN and blastP homology search using the NCBI database.

Nucleotide Sequence Submission

The AMR encoding gene cassettes and their flanking sequences of representative ICE of TetR and TetSV. cholerae O1 have been submitted in GenBank (Accession numbers MK165649 and MK165650, respectively).

Ethics and Biosafety Statements

The Ethics and Biosafety Committees of National Institute of Cholera and Enteric Diseases, Kolkata approved this study (A:1/2015-IEC). Each participant/parent in the case of children gave written informed consent. All the experiments were performed following Biosafety Level-2 standards.

Results

Prevalence of Cholera

During 8 years of surveillance from 2008 to 2015, the isolation rate of V. cholerae O1 Ogawa was about 11% (1591 of 14237 tested samples) (Figure 1). The incidence of this pathogen in BCH samples was very low (∼2%) but was found to be much higher (∼18%) in IDH samples. As shown in Figure 1, the mean incidence of cholera in IDH/BCH fluctuated between 4.9% (2014) and 27.2% (2009). Except for children ≤5 years, V. cholerae O1 remained one of the important bacterial pathogens. The incidence of V. cholerae O1 varies in certain extent from year to year (Figure 1).

FIGURE 1

Antimicrobial Resistance

All the V. cholerae O1 strains isolated were consistently resistant to NA. TetR gradually decreased from 58% in 2008 to 48% in 2009, followed by a further drop in 2010 (9%). Thereafter, all the strains isolated between 2011 and 2013 were found to be susceptible to TET (Table 1). Remarkably, TetR trait increased again in 2015 (56%). There was a marked change in AMP resistance each year with highest in 2010 (94%) and lowest in 2012 (21%) (Table 1). About three fourth of the strains were resistant to AMP in 2009 and 2011 (>76%). Thereafter, most of the V. cholerae isolated from 2013 to 2015 were found to be susceptible to AMP.

TABLE 1

Year (n)% of resistance
TETCHLSTRSXTAMPNA
2008 (76)5833929253100
2009 (120)4845969976100
2010 (53)991989894100
2011 (52)025233177100
2012 (48)058676921100
2013 (87)09198990100
2014 (44)2869193098
2015 (66)563894920100

Resistance of V. cholerae O1 Ogawa against different antibiotics.

Except three, rest of the strains were intermediate (i) to CHL.

Throughout the study period, only three V. cholerae strains were found to be fully resistant to CHL and the rest of the floR containing strains showed intermediate resistance [CHL(i)] to this antibiotic. Interestingly, resistance to TET was found to be inversely proportional to CHL(i), i.e., strains showing TetR had intermediate resistance to CHL. The CHL(i) trait increased in 2010 (91%) when TetR was very low (9%) but dropped to 38% with the re-emergence of TetR in 2015 (56%). Resistance to STR and SXT were detected in most of the V. cholerae O1 strains. Resistance to these antimicrobials was >90% from 2008 to 2010 and 2013 to 2015. Interestingly, there was a sudden decrease in STR and SXT resistance (23 and 31%, respectively, in 2011) followed by an increase in 2012 (67 and 69%, respectively) (Table 1).

This study shows the changing profile of MDR in V. cholerae from Kolkata; MDR profiles NA-STR-SXT-TET-AMP and NA-STR-SXT-TET were predominant during 2008, 2009 and 2015 (Table 2), while from 2009 to 2010 and 2012 to 2014 the MDR profiles NA-STR-SXT-CHL(i), and NA-STR-SXT-CHL(i)-AMP were found in more than 50% of the V. cholerae O1 strains.

TABLE 2

Resistance profile/Year2008 (n = 76)2009 (n = 120)2010 (n = 53)2011 (n = 52)2012 (n = 48)2013 (n = 87)2014 (n = 44)2015 (n = 66)
NA-STR-SXT-TET-AMP27.65835.8483.890.020.053
NA-STR-SXT-TET30.312.55.72.352.9
NA-STR-SXT-CHL(i)-AMP18.43440.05188.78919.6246.4720.01000.0980.046
NA-STR-SXT-CHL(i)15.810.84.366.010097.745.7
NA-AMP6.680.811.9256.57614.9280.01
NA1.30.00.019.612.81.0

Percentage of resistance pattern in V. cholerae O1 strains during 8 years in Kolkata.

(i), intermediate resistance for CHL. Numbers in bold represents cumulative percentage of resistance patterns.

ICE Comprising Antimicrobial Resistance Genes

While analyzing the sequences of the resistance gene clusters, two types of ICEs could be detected, i.e., ICETET (Acc No. MK165649; TetR IDH 1986) and ICEGEN (Acc No. MK165650; TetS IDH 4268). The superscript “GEN” stands for “general.” Although the ICEGEN was very similar to the ICEVchInd5 with 99% identity at 100% query coverage, the ICETET had only 99% identity at 70% query coverage. The structure of these two ICEs with ORFs is shown in Figure 2. The ICEGEN was found to be larger (96.7 kb) than ICETET (91.5 kb). SXT and STR resistant V. cholerae O1 strains were positive for intSXT. Detection of ICEs was >90% in 2008 and 2009, with highest in 2010 (98%), followed by an abrupt decrease in 2011 (23%). However, in 2012, 68% of the V. cholerae O1 strains harbored the ICEs. Interestingly, except for NA, the intSXT negative strains were susceptible to most of the antimicrobials tested in this study. In the 1st type, ICETET carried a TET efflux pump encoding gene (tetAR; tetA is a gene encoding TET efflux pump and tetR is a repressor protein regulating the tetA expression) and in the 2nd type, ICEGEN harbored CHL efflux pump encoding gene (floR). ICEGEN has high similarity (99%) with the ICEVchInd5, the most common ICE detected among seventh-pandemic El Tor vibrios (; ). This ICE also has very high similarity to the ICEVchHai1 from the Haitian V. cholerae lineage ().

FIGURE 2

The ICEGEN and ICETET had sul2, strBA in the AMR gene cluster conferring resistance to SXT and STR, respectively. Generally, in V. cholerae, the presence of tet alleles within the ICE gene clusters is uncommon. In the prototype SXTMO10, resistance gene cluster comprised dfr18, floR, strBA, sul2 encoding resistance to trimethoprim, CHL, STR, and sulfamethoxazole, respectively (Table 3). In ICEVchInd4, there was a major deletion of dfr18 gene in the cluster. In IDH1986 and IDH14268 strains, a class 4 integron carrying the trimethoprim resistance encoding dfrA1 was identified in H3 located within the s073-traF locus. Such arrangement exists in ICEVchInd5 backbone (Figure 2) and ICEVchInd1. But, tetA gene was absent in these ICEs.

TABLE 3

ICEHost strainCountry and year of isolationSize (bp)Resistance gene contentGenBank accession numberReferences
ICEVchMex1Vibrio cholerae non O1-O139Mexico 200182839GQ463143
ICETETVibrio cholerae O1 (IDH1986)India 200991463tetAR, strBA, sul2, dfrA1MK165649In this study
ICEGENVibrio cholerae O1 (IDH4268)India 201296718floR, strBA, sul2, dfrA1MK165650In this study
ICEVchInd4Vibrio cholerae O139India 199795491floR, strBA, sul2GQ463141
ICEVchInd5Vibrio cholerae O1India 199497847floR, strBA, sul2, dfrA1GQ463142
ICEVchBan5Vibrio cholerae O1Bangladesh 1998102131floR, strBA, sul2, dfrA1GQ463140
ICEPalBan1Providencia alcalifaciensBangladesh 199996586floR, strBA, sul2, dfrA1GQ463139
ICEVflInd1Vibrio fluvialisIndia 200291369dfr18, floR, strBA, sul2GQ463144
ICEVchMoz10/ICEVchB33Vibrio cholerae O1Mozambique 2004104495floR, strBA, sul2, tetA’ACHZ00000000
ICEPmiUsa1Proteus mirabilisUnited States 198679733AM942759
ICEVchBan9Vibrio cholerae O1Bangladesh 1994106124floR, strBA, sul2, dfrA1, tetA’CP001485
ICEVchBan8Vibrio cholerae non O1-O139Bangladesh 2001105790NZ_AAUU00000000
SXTMO10Vibrio cholerae O139India 200299452dfr18, floR, strBA, sul2AY055428
R391Providencia rettgeriSouth Africa 196788532kanR, merRTPCAAY090559
ICEPdaSpa1Photobacterium damselaeSpain 2003102985tetARAJ870986
ICESpuPO1Shewanella putrefaciensPacific Ocean 2000108623CP000503

Comparison of the ICE gene cluster with the other SXT/R391 ICE family members.

Detection of ICETET in V. cholerae O1 decreased from 2008 (58% TetR) to 2010 (9% TetR). All the V. cholerae O1 strains isolated during 2011–2013 lacked ICETET. In 2015, however, the tetAR was again detected in a higher number of strains (56% TetR). In contrast, ICEGEN was detected throughout the study period. AMR gene cassettes located within the rumB locus are also different. From 2011 to 2013, the tetAR locus in ICETET was replaced by floR gene of ICEGEN. This feature marked the difference of ICETET from ICEVchLao1, where floR and tetA were concurrently present.

Based on the presence of the AMR encoding genes harbored by these elements, the genetic background of ICETET appears to be very different from the other ICEs carrying the tet. The ICEPdaSpa1 was found to have only the TET resistance determinant located within rumBA operon (Table 3). Whereas, in the ICEVchLao1, resistance genes of CHL (floR), STR (strBA) and sulfamethoxazole (sul2) were present along with tetA. But, the ICEVchLao1 did not carry dfrA1 or dfr18 that confer resistance to trimethoprim in SXTET and SXTMO10, respectively. Within the resistance gene cluster of 2008–2010 strains of V. cholerae in Kolkata, a deletion of floR gene, which was present upstream of the tetA gene in ICEVchLao1 and ICEVchBan9 was detected.

Genetic Structure of the ICEs

Generally, the genetic organization of ICETET and ICEGEN was similar to that of the other members of this family. Many ORFs were commonly shared by these ICEs; most of them being in the conserved core genes (). Five conserved insertion hotspots are located between s043 (traJ) and traL (H1), traA and s054 (H2), s073 and traF (H3), traN and s063 (H4), and s025 and traID (H5) ().

Five ORFs were found in the H1 of ICETET that include tbp (integrase catalytic subunit), a hypothetical protein (HP), transposase, ISPsy4 transposition helper protein and DNA helicase family protein. These ORFs present in H1 are unique compared to other reported ICEs. Instead of mosA, mosT that encode toxin-antitoxin reported in the H2 of other ICEs, the ICEGEN and ICETET have 3 ORFs with ynd (transcriptional regulator with AbiEi antitoxin N-terminal domain), ync (nucleotidyl transferase AbiEii/AbiGii toxin family protein) and dsbC (disulfide isomerase DsbC). H3 of ICEGEN and ICETET contains 7 ORFs with bleR (glyoxalase/bleomycin resistance), araC (AraC family transcriptional regulator; helix-turn-helix domain protein), a hypothetical protein, XRE family transcriptional regulators, a putative membrane protein, dfrA1 (trimethoprim-resistance) and intI4 (site-specific recombinase IntI4). Of these, AraC, XRE, and DFRA1 were reported in ICEVchMoz10. H3 in ICEGEN and ICETET is varied from ICEVchInd4, SXTMO10, ICER391 ICEVchMex1, ICEVflInd1, ICEPmiUSA1, ICESpuPO1 (). H4 of ICETET was small with 2 ORFs, whereas the ICEGEN had 5 ORFs with two SMC (structural maintenance of chromosome) domain proteins, istB (ATP binding domain), istA (integrase catalytic subunit) and deoxyribonuclease I. The ORF content of H4 in these ICEs is different from the others. In ICEGEN and ICETET, the H5 has 10–11 gene combinations with the new ORFs of WYL domain protein, N-6 DNA methylase, restriction endonuclease subunit S, BstXI (restriction endonuclease protein), ATPases associated with diverse cellular activities (AAA) family protein, McrC (putative protein) in ICETET and WYL domain-containing protein with three conserved amino acids, BrxC (BREX system P-loop protein), PglX (BREX-1 system adenine-specific DNA-methyltransferase) and abortive phage resistance protein in ICEGEN. These changes in the hotspot regions may not have an obvious effect on the ICE, as they did not influence its transfer. VR-II has an insertion of single ORF, mutL similar to the ICE contigs circulating in India and Bangladesh. In the VRIII of ICETET, 12 ORFs [Tn3 (transposase), tnpA (transposase), tnpB (InsA transposase), truncated virD2, tetA, tetR, IS91 transposase, strB, strA, sul2, tnpA tn3 transposase, s021] were identified within the two rumB portions. In the case of ICEGEN, 14 ORFs [Tn3 (trnansposase), tnpA (transposase), tnpB (InsA transposase), virD2 (relaxase), floR, LysR family protein, truncated transpoase, strB, strA, sul2, tnpA tn3 transposase, truncated s021, putative transpoase, truncated mutL] have been detected.

The restriction-modification system is composed of genes encoding the functions of DNA modification, recombination, and repair (). ICEGEN and ICETET were found to have a type I restriction-modification system in the H5. In the ICE backbones, there were sequences in the ORFs located between s024 and traI in Kolkata strains (Figure 2). In ICEGEN carrying strains, after the traN locus, there was an insertion of istBA gene flanked by gene encoding SMC domain protein. This arrangement was not observed in V. cholerae strains with ICETET. Though these two types of ICEs had same traFHG locus, ORFs encoding transposases and ATPase were found incorporated between the traD and traE locus only in ICETET. In contrast, the ICEGEN possessed an intact transfer region (Figure 2). In ICEVchInd4, there was a major deletion of dfr18 gene in the cluster. In strains with ICEGEN or ICETET, a class 4 integron carrying the trimethoprim resistance encoding dfrA1 was identified in the H3 region located within the s073-traF locus. Similar gene configuration exists in the ICEVchInd1 and ICEVchInd5 backbones. In the 2008–2010 strains of V. cholerae in Kolkata, TetR in ICE was primarily due to tetA, whose presence was previously reported in ICEPdaSpa1 of Photobacterium damselae, ICEVchLao1 and ICEVchBan9 of V. cholerae O1 from Laos and Bangladesh, respectively (Table 3).

The tra loci appeared to be derived from a common ancestor and were mostly present in ICEs of V. cholerae strains. These loci are crucial for the transfer of ICEs and generating the conjugation machinery (). Similar to the other ICEs backbone, the tra genes are arranged in four clusters in IDH1986 and IDH4268 strains, spanning more than 25 kb. Cluster 1 contains the genes and sequences necessary for transfer initiation, the nickase (encoded by traI), and the coupling protein (encoded in the traD). The mating pair formation function is controlled by three gene clusters: (i) traLEKBVA, (ii) traC/trhF/traWUN, and (iii) traFHG (Figure 2).

Comparison of Conserved Genes in the ICEs

ICETET and ICEGEN shared the same exclusion group (EexR). This EexR system might have been transferred from R391 type ICEs (). The site-specific integration of the ICE is mediated through integrase enzyme encoded in the int. The int of ICETET and ICEGEN harboring V. cholerae O1 is identical to those present in the strains that have ICEPalBan1 of P. alcalifaciens, ICEVfInd1 of V. fluvialis and ICEVchBan5, ICEVchBan9 and ICEVchInd5 of V. cholerae (Figure 3). These ICEs are distinct from those reported in Proteus mirabilis, Providencia rettgeri, Shewanella putrefaciens, P. damselae as well as in other V. cholerae with ICEVchMex1, ICEVchInd4, and SXTMO10. SetR and SetC/D are the key regulators of ICEs, which are closely followed by the genes encoding for inner membrane proteins (Eex and TraG) of the donor and recipient cells. Eex and TraG facilitate entry-exclusion in the SXT/R391 family of ICEs. In the cluster tree, eex genes of the ICETET and ICEGEN showed high homology with ICE identified in ICEVchBan5, ICEVchBan9, ICEVchInd5, but was distantly related to other ICEs of V. cholerae and other species (Figure 4). setR in the ICETET and ICEGEN are identical with that in ICEVchInd4, ICEVchInd5, ICEVchBan5, ICEVchBan9, SXTMO10, ICEVfInd1, ICEPalBan1 but different from ICEVchMex1 and ICEs of other species (Figure 5).

FIGURE 3

) evolutionary relationships of taxa of int of V. cholerae O1 strains.

FIGURE 4

) evolutionary relationships of taxa of eex of V. cholerae O1 strains.

FIGURE 5

) evolutionary relationships of taxa of setR of V. cholerae O1 strains.

Transfer of ICEs

To test the transferability of the V. cholerae ICEs, we selected ICETET and ICEGEN carrying strains (IDH1986 and IDH1439, respectively). Both the types of ICEs could be transferred to E. coli J53 by conjugation. The transconjugants acquired additional resistance against SXT and STR (Table 4). Remarkably, CT-E. coli J53 from ICEGEN was highly resistant to CHL compared to the donor V. cholerae O1 strain, which showed reduced susceptibility to this antibiotic. Similarly, CT-E. coli J53 from ICETET expressed more resistance against TET than the donor Vibrio (Table 4). The frequency of transfer ranged from 3 × 10–5 to 5 × 10–6 transconjugants/recipient.

TABLE 4

StrainResistance profileMIC value (μg/ml)
SXTSTRTETCHL
IDH1986 (V. cholerae O1 Ogawa)NA-TET-SXT-STR>32192161
CT-E. coli J53/ICETET (Transconjugant)TET-SXT-STR-AZD>32 (>600 fold)48 (24 fold)24 (48 fold)3
E. coli J53 (Recipient)AZD0.04720.53
CT-E. coli J53/ICEGEN (Transconjugant)CHL-SXT-STR-AZD>32 (>600 fold)64 (32 fold)0.5>256 (>85 fold)
IDH1439 (V. cholerae O1 Ogawa)NA-SXT-STR-{CHL(i)}>321280.58

Increased resistance attributed by acquired ICE in transconjugants.

Increase in fold compared to the recipient.

PFGE Analysis

Pulsed-field gel electrophoresis was performed to identify the clonal relationship between ICETET and ICEGEN carrying V. choleare strains. It was found that the V. cholerae O1 strains displayed clonal clusters reflecting their MDR profile, which indirectly revealed the composition of AMR encoding genes in the ICEs (Figure 6). Cluster A represented Vibrio strains devoid of the ICEs. These strains were only resistant to NA. Strains with ICEGEN were present in cluster B. These strains are resistant to NA, SXT and exhibited intermediate susceptibility to CHL. Cluster C contained the ICETET harboring strains that showed resistance to NA, SXT, and TET (Figure 6).

FIGURE 6

Discussion

Cholera is endemic in the Indian subcontinent and it has spread to several other parts of the world (). In Kolkata, MDR V. cholerae is associated with sporadic cholera for many years (; ). V. cholerae O1 was susceptible to several antibiotics before 1980s, but developed resistance to SXT in the following years (). V. cholerae O1 El Tor biotype that re-emerged in 1994 may have acquired SXT resistance phenotype from the O139 serogroup (). Investigations conducted almost during the same period in several cholera endemic regions in India showed that the isolation rate of V. cholerae O1 was lesser than Kolkata, but the AMR pattern followed nearly the same trend, especially to tetracycline (; ; ; ; ; ; ; ; ; ; ).

From 2010 to 2012, V. cholerae strains with AMR profiles of NA-STR-SXT-TET-AMP and NA-STR-SXT-TET were completely replaced with NA-STR-SXT-CHL(i)-AMP and NA-STR-SXT-CHL(i) along with NA-AMP and NA. Strains with the AMR profile of NA-STR-SXT-TET appeared again in 2015 (53%). Though the number of V. cholerae strains with the NA-SXT-STR-CHL(i) profile was highest from 2013 to 2014 (98–100%), it has reached to 46% with the re-emergence of TetR in 2015. The appearance of TetR in V. cholerae O1 Ogawa in 2008 has been reported from northern parts of India (). TetR has been previously reported mostly in Inaba serotype (; ). Presence of tetA, floR, strBA, sul2, dfrA1 within the AMR gene cassettes has positive correlation with the phenotypic expression of drug resistance against TET, CHL, STR, and SXT (; ; ). It is interesting to note that although dfrA18 conferring resistance to trimethoprim was reported in MO10, later it was replaced by the dfrA1 allele in a class IV integron located in the H3 ().

In our study, floR and tetA genes were not found to coexist within the VRIII present in the rumB locus. Previous reports, however, had shown the presence of both floR and tetA in the V. cholerae ICEVchLao1 isolated from the Laos, ICEVchB33 from Beira, Mozambique (; ). Depending upon the presence of resistance cassettes in the ICEs, we found two types of ICEs in our study namely ICEGEN and ICETET. Though the ICE backbone of ICEGEN was similar to those of SXTMO10 and SXTET, it had 99% structural similarity to ICEVchInd5. Lineages of ICEVchInd5 of V. cholerae O1 strains causing epidemics in the Indian subcontinent might have spread to Africa ().

ICEGEN circulating in V. cholerae strains from Kolkata belonged to the group 1 ICE, which comprised ICEVchInd5 (India, 1994–2005), ICEVchBan5 (Bangladesh, 1998), ICEVchHai1 (Haiti, 2010), ICEVchNig1 (Nigeria, 2010), and ICEVchNep1 (Nepal, 1994) (). Type I restriction-modification system systems of ICEGEN and ICETET were also reported in the other ICEs families, such as ICEVchMex1 and ICESpuPO1 (; ). ICEs are constantly spreading in different geographical areas. ICEVchB33, which is different from other ICEs of SXT/R391 was first identified in V. cholerae O1 strains from India in 1994 and then Mozambique in 2004 (). Similar to V. cholerae O1 from India with ICEVchInd1, the other ICEs identified in Vietnam, Laos, and Mozambique (ICEVchVie1, ICEVchLao1, and ICEVchB33, respectively) lack the trimethoprim resistance encoding dfr18, but carried virD2 and floR, conferring resistance to CHL (). Majority of the V. cholerae O1 isolated in Kolkata from 1989 to 1990 had STXMO10/ICEVchInd4. This ICE was replaced by ICEVchInd5/ICEVchBan5 in the subsequent years (, ).

In this study, the ICETET detected in V. cholerae O1 strains had significant structural dissimilarities with ICEVchBan9 (Bangladesh, 1994), ICEVchMoz10 (Mozambique, 2004), ICEVchB33 (Beira, 2004), and ICEVchLao1 (; ; ). Nevertheless, structural variations, unstable core region, and the transfer region of both the ICEs found in our study were very much similar and shared a common ancestral backbone. In many ICEs, the core genes such as int, bet, exo, and setR are usually associated with phages, and genes such as tra are associated with plasmids (; ). Having the same exclusion group (eexR1), ICEGEN and ICETET were mutually exclusive and therefore did not co-exist in a strain. ICE sequences reconfirmed that there were two ICE types that kept emerging in different years. The key modifications between them indicated that they may have diverse origins or be derived from a common ancestor and could have later evolved independently.

We could transfer the ICEGEN and ICETET from V. cholerae O1 to E. coli J53 by conjugation. The frequency of transfer observed was high (10–5 to 10–6), indicating that the ICEs were promiscuous due to the presence of an active tra region (; ). Our study showed that only the resistances conferred by genes present in ICE were transferable and that the level of expression was different, being more in the transconjugants with respect to the donor vibrios. This could be due to “gene dosage” effect or absence of repressor in the new genetic environment of the recipient E. coli. Transconjugants showing higher drug resistance have been described in the previous reports as well (; ). The co-existence of ICEs with plasmids and class 1 integrons in clinical as well as environmental V. cholerae has been reported (; ). The involvement of plasmids carrying the ICEs was not tested in this study. We also observed that resistance to NA and AMP were not transferable, indicating that the resistance to these antimicrobials could be contributed by the chromosomal factors such as mutations and efflux pumps ().

As shown in the PFGE analysis, the clonal relatedness of V. cholerae strains isolated during different years corresponded with the MDR profiles. ICE integrase-negative strains isolated in 2008, 2011, and 2012 were found to cluster together (cluster A). V. cholerae O1 strains harboring either ICEGEN or ICETET were also grouped in different clusters (B and C, respectively). A similar observation was made with the outbreak strains of V. cholerae O1 in Kenya ().

In conclusion, our findings revealed the existence of two types of ICEs in V. cholerae O1 strains from Kolkata. The ICEGEN that contained conserved backbone genes was most commonly detected in V. cholerae O1 circulating around Kolkata. Features of the Kolkata V. cholerae O1 strains with ICE carrying the TetR encoding genes are unique and the sequence of the ICETET had several variations from other sequenced ICEs. Also the ICETET harboring V. cholerae O1 strains reappeared after 4 years of disappearance in Kolkata. Unique PFGE clusters of V. cholerae O1 harboring different ICEs are linked with the AMR patterns. The primer pair designed in this study may be useful in the detection of ICEs carrying the tet. The transmission potential of ICEs identified in this study was very high, as evidenced from the conjugation assay. Therefore, the impact of ICE regulation and interactions between bacteria prevailing in the same ecological niches should be explored in detail. Emergence of new types of ICEs may pose challenges in the existing cholera management strategies.

Statements

Author contributions

AG, TR, and KO conceived and designed the experiments. AS, DM, and GC performed the experiments. KO contributed reagents, materials, and analysis tools. TR and AM analyzed the data. AS and TR wrote the manuscript. All authors discussed the results, and reviewed and commented on the manuscript.

Funding

This work was supported in part by the Department of Biotechnology, New Delhi, India (Grant No. BT/MB/THSTI/HMC-SFC/2011), the Japan Initiative for Global Research Network on Infectious Diseases (J-GRID), the Ministry of Education, Culture, Sports, Science and Technology in Japan, the Japan Agency for Medical Research and Development (AMED; Grant No. JP18fm0108002), and the Indian Council of Medical Research. AG is J. C. Bose Chair Professor of the National Academy of Sciences, India.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The reviewer AB declared a past co-authorship with several of the authors, GC, AM, TR, and AG, to the handling Editor.

References

Summary

Keywords

cholera, V. cholerae O1, tetracycline, antimicrobial resistance, multidrug resistance, integrative conjugative element

Citation

Sarkar A, Morita D, Ghosh A, Chowdhury G, Mukhopadhyay AK, Okamoto K and Ramamurthy T (2019) Altered Integrative and Conjugative Elements (ICEs) in Recent Vibrio cholerae O1 Isolated From Cholera Cases, Kolkata, India. Front. Microbiol. 10:2072. doi: 10.3389/fmicb.2019.02072

Received

16 November 2018

Accepted

22 August 2019

Published

06 September 2019

Volume

10 - 2019

Edited by

Rustam Aminov, University of Aberdeen, United Kingdom

Reviewed by

Pramod Kumar, All India Institutes of Medical Sciences, New Delhi, India; Ashima Kushwaha Bhardwaj, Independent Researcher, Gurugram, India

Updates

Copyright

*Correspondence: Thandavarayan Ramamurthy, ;

These authors have contributed equally to this work

This article was submitted to Antimicrobials, Resistance and Chemotherapy, a section of the journal Frontiers in Microbiology

Disclaimer

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

Outline

Figures

Cite article

Copy to clipboard


Export citation file


Share article

Article metrics