Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, September 2002, p. 2948-2955, Vol. 46, No. 9
0066-4804/02/$04.00+0 DOI: 10.1128/AAC.46.9.2948-2955.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.
Institute of Microbial Technology, Chandigarh,1 National Institute of Cholera and Enteric Diseases, Calcutta, India,2 International Centre for Diarrhoeal Disease Research, Dhaka 1000, Bangladesh3
Received 11 December 2001/ Returned for modification 10 April 2002/ Accepted 31 May 2002
|
|
|---|
|
|
|---|
) (31) and some other virulence genes (27). Based on these findings, it was concluded that some strains of V. cholerae belonging to different serotypes can cause diarrhea clinically indistinguishable from that associated with cholera (5) by a mechanism that could be distinct from that employed by the toxigenic V. cholerae O1 and O139 strains. The nomenclature "enteropathogenic V. cholerae" (EPVC) was proposed to include these serotypes (27). The incidence of EPVC had shown an upward trend from 1997 that continued into 1998. In the months of July and August 1998, the EPVC strains constituted one-third of the V. cholerae strains isolated from hospitalized patients (12). Recently a comparative study of clinical and environmental isolates of non-O1, non-O139 V. cholerae strains belonging to matching serogroups from our laboratories revealed that, despite sharing the same serogroups, the environmental and clinical isolates were genetically heterogeneous and also that the resistance to multiple antibiotics was more common among the clinical isolates (5). Multiple-antibiotic-resistant isolates of non-O1, non-O139 V. choelrae strains were identified in children with diarrhea in Bangkok, Thailand (7).
Reports of drug-resistant V. cholerae strains are appearing with increasing frequency (20). Emergence of resistance to multiple drugs is a serious clinical problem in the treatment and containment of the disease, as reflected by the increase in the fatality rate from 1% to 5.3% after the emergence of drug-resistant strains in Guinea-Bissau during the 1996-1997 epidemic of cholera (10). The molecular mechanisms responsible for the emergence of multiple-antibiotic-resistant V. cholerae are not very well worked out. Acquisition of antibiotic resistance genes across genera and species is mediated through horizontal and lateral gene transfer. Plasmids, conjugative transposons, and integrons are all vehicles for the acquisition of resistance genes (13, 14, 24, 29, 32). Integrons are genetic elements capable of integrating and mobilizing individual gene cassettes by a site-specific recombination mechanism involving a DNA integrase intI and an att1 site recognized by the integrase (13). Among the different integron families, class I integrons have been found to be most prevalent among clinical isolates. The 5' conserved segment (5'-CS) of the class I integrons contains the integrase gene (intI1) and the recombination site attI1 (6). The 3'-CS of class I integrons normally carries the antiseptic resistance gene, qacE
1,and the sulfonamide resistance gene, sul1. It has been observed, however, that class I integrons do not always contain the entire 3'-CS (6, 22, 25, 28).
Class I integrons are widely present in multidrug-resistant enteropathogens (15-17) and have been identified in V. cholerae O1 strains isolated in Vietnam (8) and Albania and Italy (11), as well as in non-O1, non-O139 strains in Thailand (9). A 150-kb conjugative plasmid was found to contain class I integron in epidemic strains of V. cholerae O1 isolated in Guinea-Bissau (10). So far, there has been no report on the presence of class I integrons in strains of V. cholerae isolated in India.
In the present study, 94 strains of V. cholerae (17 of which were environmental isolates) belonging to serogroups other than O1 and O139 isolated in Calcutta, India, during the period 1997 to 1998, were analyzed for the prevalence of class I integrons and the nature of the antibiotic resistance genes carried by them. They were also investigated for the presence of plasmids and the conjugative transposon SXT element (32), which could also contribute to the multidrug resistance observed in these strains.
|
|
|---|
Bacteriology and serogrouping. V. cholerae isolates from stool and environmental samples were isolated on thiosulfate citrate-bile salts sucrose agar (TCBS agar) as described previously (7). Identification of V. cholerae was then performed (3, 5,12, 27).
Antimicrobial susceptibility. All of the V. cholerae isolates included in this study were examined for resistance to ampicillin (10 µg), chloramphenicol (30 µg), ciprofloxacin (5 µg), furazolidone (50 µg), gentamicin (10 µg), neomycin (30 µg), nalidixic acid (30 µg), streptomycin (10 µg), sulfamethizole-sulfadiazine (100 µg), tetracycline (30 µg), and trimethoprim (25 µg) by using commercial discs (Hi Media, Bombay, India) as described previously (5) according to the interpretation criteria recommended by the World Health Organization (33).
Bacterial genomic DNA extraction. Genomic DNA was extracted from the V. cholerae isolates following the method of Murray and Thompson (21) with minor modifications. In brief, V. cholerae organisms from 1.5 ml of an 18-h Luria broth (LB) culture were collected and resuspended in 567 µl of TE buffer (10 mM Tris HCl, 1 mM EDTA [pH 8.0]). This was followed by the addition of 30 µl of 10% sodium dodecyl sulfate and 3 µl of proteinase K (20 µg/ml) and incubation at 37°C. After 1 h, 100 µl of 5 M NaCl was added, followed by the addition of 80 µl of 10% CTAB (cetyltrimethylammonium bromide) in 0.7 M NaCl. The mixture was incubated at 65°C for 10 min and then extracted, first with chloroform and then with phenol chloroform. DNA from the aqueous phase was precipitated with an equal volume of isopropanol. The pellet after washing with 70% ethanol and drying was dissolved in TE buffer and treated with RNase for 30 min at 37°C.
Preparation of DNA probes and Southern and colony hybridizations.
Ribotyping and Southern hybridization analysis of the various genes (toxR, toxT, RS1, hlyA, and hlyU) were performed with probes and methods essentially as described previously (27). The SXT probe was a 0.8-kb PCR product representing a part of the SXT integrase (int) gene sequence. It was amplified from V. cholerae strain MO10 by using the primers SXT-F and SXT-B (Table 1) according to the manufacturer's instructions (Promega, Madison, Wis.). The annealing temperature used was 50°C. The probe was radiolabeled and used in colony hybridization following standard protocols (26). The 0.8-kb PCR amplicon obtained with the primers qacE
1-F and sul1-B was used as the class I integron-specific probe for Southern hybridization of the pulsed-field gel electrophoresis (PFGE) gel of 14 V. cholerae strains to find out whether the class I integron was localized on the chromosome or on plasmids.
|
View this table: [in a new window] |
TABLE 1. PCR primers used in this study
|
Bacterial transformation and plating on selective media. Transformation of Escherichia coli JM109 with the plasmid preparations from V. cholerae was performed by electroporation. E. coli cells were prepared for electroporation according to the manufacturer's recommendations (Bio-Rad Laboratories, Richmond, Calif.). Approximately 20 to 50 ng of the plasmid preparations was used. Electroporated cells, after recovery, were screened for antibiotic resistance on LB plates containing different antibiotics. Transformants appearing on each selecting antibiotic plate were restreaked on the appropriate drug plates for confirmation.
PCR.
Primers qacE
I-F and sul1-B (Table 1) were used in the PCR to identify class I integron-bearing strains of V. cholerae. In order to amplify the antibiotic resistance gene cassettes present in the class I integron-bearing strains, primers in-F (5'-CS) and in-B (3'-CS) (Table 1) were used. PCR amplifications were performed according to the method of Aarestrup et al. (1) with a few modifications. Fifty nanograms of genomic DNA was used as a template in the PCR, and Taq DNA polymerase (Promega) was used for amplification according to the manufacturer's instructions. For PCR with the qacE
1-F and sul1-B primers, the annealing temperature was kept at 58°C, and an extension time of 1 min was used. Strains yielding a 0.8-kb PCR product with the class I primers were used further for PCR with the integron primers in-F and in-B. These primers amplify the region between the 5'-CS and the 3'-CS, yielding products of various sizes, depending upon the length of the inserted gene cassettes. For PCR with this second set of primers, an annealing temperature of 55°C was used, and the extension in each cycle was carried out for 2.5 min. A 0.2-kb ladder (Promega) was used as the molecular size marker during electrophoresis of PCR products. All amplicons obtained with the integron primers in-F and in-B were sequenced to identify the antibiotic resistance gene cassettes.
DNA sequencing. Amplified products, after PCR amplification, were separated by electrophoresis. The bands of interest were cut out, and DNA was eluted with a QIA Quick gel extraction kit (Qiagen, Hilden, Germany). Nucleotide sequences were determined by the ABI Prism 310 genetic analyzer by using the Big Dye Terminator kit (Applied Biosystems). The identities of the sequences determined were established by comparing the sequences obtained with the gene sequences in databases by using the BLAST software (2).
PFGE. Bacterial cells were grown in LB at 37°C, to an optical density at 600 nm of 0.9. Cells from 1.0-ml culture were harvested by centrifugation. The pellet was resuspended in suspension buffer (10 mM Tris-HCl [pH 7.2], 20 mM NaCl, 50 mM EDTA [pH 8.0]) and equilibrated for 10 min at 50°C. An equal volume of 2.0% agarose (Bio-Rad Low Melt Preparative grade), prewarmed to 50°C, was mixed and cast in a mold. Blocks were treated with 1.0 ml of lysozyme solution (1 mg of lysozyme per ml in 10 mM Tris-HCl [pH 7.2], 50 mM NaCl, 0.2% Na-deoxycholate, 0.5% Na-laurylsarcosinate) for 12 h at 37°C. After washing with sterile distilled water twice, blocks were incubated in 1.0 ml of proteinase K solution (1 mg of proteinase K per ml in 100 mM EDTA [pH 8.0], 0.2% Na-deoxycholate, 1% Na-laurylsarcosinate) for 12 h at 50°C. Blocks were washed four times for 1 h each at room temperature with 1.5 ml of 1x wash buffer (20 mM Tris-HCl [pH 8.0], 50 mM EDTA [pH 8.0]) and then stored in 0.1x wash buffer at 4°C until use. Before the gel was run, blocks were equilibrated in 0.5x TBE (45 mM Tris, 45 mM boric acid, 1 mM EDTA [pH 8.3]) for 1 h. A 1.0% agarose gel (Bio-Rad PFG certified grade) in 0.5x TBE was run in CHEF MAPPER (Bio-Rad) at a voltage gradient of 6 V/cm, temperature of 14°C, included angle of 120, separation range of 10 to 600 kb, initial switch time of 0.47s, and final switch time of 44.17 s, for 20 h 18 min, along with the Lambda Ladder PFG marker (New England Biolabs). The gel was stained with 1 µg of ethidium bromide per ml in 0.5x TBE for 30 min and photographed.
Nucleotide sequence accession numbers. The nucleotide sequences of the aac(6')-Ib, dfrA1, aadA1, In0, dfr12, and aadA2 (with 5'-CS and 3'-CS), dfrA1, and an unknown gene, dfr5, as well as ereA2 and open reading frame III (ORFIII) and partial 3'-CS have been assigned GenBank accession no. AY103455, AY103456, AY103457, AY103458, AY103459, AY103460, AF455254, AF512546, and AF512547, respectively.
|
|
|---|
|
View this table: [in a new window] |
TABLE 2. Multidrug resistance combinations in isolates of non-O1, non-O139 V. cholerae
|
|
View this table: [in a new window] |
TABLE 3. Phenotypic and genotypic characterization of non-O1, non-O139 V. cholerae strains containing the class I integron
|
1-F and sul1-B. This product was then sequenced by using the same set of primers to confirm the existence of the qacE
1 and sul1 genes, found at the 3'-CS of the class I integrons in the 0.8-kb amplicon (22, 24, 27, 29, 32). All 22 isolates were of clinical origin. When the primers in-F and in-B were used to amplify the class I integron gene cassettes from these strains, at least five distinct products could be identified (see Fig. 2 and Table 3). Thirteen of 22 strains produced a 1.25-kb amplicon that was found to contain, by sequence analysis, the dfrA1 gene for trimethoprim resistance and orfC (24). Two strains had a 0.73-kb gene cassette, which upon sequence analysis showed the presence of the dfrA15 gene for trimethoprim resistance. Both the 1.25-kb amplicon and the 0.73-kb amplicon showed complete identity to the integron cassettes described by Dalsgaard et al. (9). The third amplicon obtained in three clinical isolates, was a 1-kb product that contains the aadA1 gene for aminoglycoside resistance (30). Sequence analysis of a 0.85-kb product obtained from two strains, PG170 and P107b, revealed the presence of the aac(6')-Ib gene, which codes for aminoglycoside 6'-N-acetyltransferase. It showed complete identity with the aac(6')-Ib gene of class I integrons described in K. pneumonia plasmid pY2 (accession no. AF227505). A 2-kb amplicon was obtained from two strains, namely PG153/1 and PL61, used in this study. Sequence analysis of the 2-kb amplicon from PG153/1 revealed that it had two ORFs: one containing dfrA5 for trimethoprim resistance (28) and the other containing ereA2 for erythromycin resistance (accession no. AF099140). The ORF for dfrA5 was present immediately downstream of the integrase I and upstream of the ereA2 gene, at the 3' end of the class I integron. Sequencing of the 2-kb amplicon from PL61 revealed the presence of the dfrA12 and aadA2 genes. A similar amplicon has been reported in the class I integron of Serratia marcescens (accession no. AF284063). Strain PL91 (Table 3) had a
0.2-kb amplicon, which upon sequencing was found to be a class I integron element without any gene cassette in it. A similar integron element called In0 in a Pseudomonas aeruginosa plasmid was described earlier (4). The nucelotide sequence of the 3-kb amplicon obtained from strain PG149b, obtained with the primer in-B (Table 3), showed identity to an unidentified ORF, ORFIII, present in a 5-kb class I integron cassette of the Acinetobacter sp. genome (accession no. AF369871). This ORF has also been reported in P. aeruginosa (accession no. AY029772) (18) and S. marcescens (accession no. AY030343). However, the 5' end of the nucleotide sequence obtained with the in-F primer showed no identity to the 5'-CS of class I integrons. Another interesting strain was PL107/4 (Table 3). Although this strain had the 3'-CS of the class I integron, as confirmed by nucleic acid sequencing of the 0.8-kb PCR product by using the primers qacE
1-F and sul1-B, no amplicon could be generated with in-F and in-B primers under the same conditions; a long PCR protocol may be necessary. It may be mentioned here that recently a hybrid integron has been described in Acinetobacter baumannii, which has the 3'-CS of class I integrons and the 5' intI2 of class II integrons (23). It would be of interest to see if the integrons present in PL107/4 and PG149b are also hybrid integrons.
![]() View larger version (19K): [in a new window] |
FIG. 2. Schematic representation of different types of gene cassettes identified in class I integrons in non-01, non-0139 strains of V. cholerae. The gene cassettes are represented by solid lines, and their 59-base elements are represented by solid circles. A hatched box represents the att1 recombination site. The bottom panel shows the locations of the 5'- and 3'-CS of class 1 integrons and those of the primers in-F, in-B, qacE 1-F, and sul1-B.
|
![]() View larger version (80K): [in a new window] |
FIG. 1. (A) Separation of large plasmids from chromosomal DNA by PFGE. Lanes1 to 14, strains PG262b, PG149b, PL78/6, PG9, PG92, PG153/1, PG170, PL1, PL61, PL96, PL105b, PL107b, AS634, and PL91, respectively. Lane M, lambda ladder molecular size marker. The positions of 50- to 450-kb bands are shown. (B) Autoradiogram of the gel described above hybridized with class I integron-specific probe. Arrows indicate the positions of the bands, which, although faint, were clearly visible on the autoradiogram.
|
|
View this table: [in a new window] |
TABLE 4. Antibiotic resistance markers present on plasmids obtained from V. cholerae strains carrying class I integrons on plasmids
|
Similarly, a plasmid preparation from PL105b conferred upon the recipient E. coli strain resistance to ampicillin, gentamicin, neomycin, streptomycin, sulfonamide, and tetracycline, but not to trimethoprim, despite the presence of the dfrA1 gene on the class I integron in this strain. Since PL105b was shown to have class I integrons on both the chromosome and the plasmid (described below), one trivial explanation for this result could be that the dfrA1 gene on the plasmid is nonfunctional or that the integron cassette of this plasmid could not be amplified under the PCR conditions used. However, no such explanation can be proffered for strain PL78/6, which is trimethoprim resistant and in which dfrA1 resides only on the plasmid. Plasmids of PG149b conferred upon E. coli JM109 resistance to ampicillin, streptomycin, sulfonamide, and trimethoprim. However, as discussed in the last section, the nature of the antibiotic resistance genes contained in the class I integron of this strain could not be identified. In repeated attempts, E. coli JM109 could not be transformed by plasmids of strain PL91, in which the class I integron resides only on the 169-kb plasmid (described below).
PFGE. On the basis of the results described in the previous sections, 14 plasmid-bearing strains positive for class I integrons were subjected to PFGE to separate the large plasmids from the chromosomes (Fig. 1A). Upon Southern hybridization with the 3' conserved region of class I integron as the probe, a band corresponding to chromosomal DNA lit up in all strains, except PG262(b), PG149(b), PL78/6, PL107(b), and PG170, in which a positive signal could be seen only with plasmid bands with sizes of 370, 355, 240, and 370 kb, respectively, in the first four strains and with two bands with sizes of 315 and 50 kb from PG170. This suggested that in these strains, class I integrons reside on the large plasmids (Fig. 1B). Interestingly, in three strains, namely PL61, PL105(b), and PL91, class 1 integrons were found on both the chromosomes and the large plasmids (Fig. 1B, lanes 9, 11, and 14).
Colony hybridization with SXT int probe. Out of the 94 strains analyzed for the possible presence of SXTa 62-kb conjugative transposon that is known to confer resistance to streptomycin, sulfonamide, and trimethoprim (32)12 strains hybridized with the SXT int-specific probe in colony hybridization experiments, suggesting the presence of this element in these strains. The strains that were positive for SXT int were PG262b, PG149b, PG153, PG171b, PG95, PG58, PL105b, PL96, PL80, PL78/6, PL61, and PL30.
|
|
|---|
The dfrA1 and dfrA15 gene cassettes, coding for trimethoprim resistance, found among several strains, have also been detected among the strains isolated in Thailand (9). Similarly, the aadA1 gene for aminoglycoside resistance (spectinomycin and streptomycin) found in three strains in our collection was detected by Falbo et al. (11) among the outbreak strains in Albania and Italy in 1994. Several gene cassettes were detected for the first time in V. cholerae: aac(6')-Ib coding for amikacin resistance in PG170 and PL107b, ereA2 coding for erythromycin resistance in PG153/1, and dfrA5 and dfrA12 coding for trimethoprim resistance in PG153/1 and PL61, respectively. It is worth mentioning here that ereA2 has so far been reported only in Providencia stuartii (accession no. AF099140). A 3-kb amplicon obtained from one strain, PG149(b), which showed identity to ORFIII (accession no. AF369871), and had the 59-base elements and a small part of 3'-CS of class I integrons, but did not have the 5'-CS of the class I integron, has also been identified by us in a non-O1, non-O139 strain of V. cholerae isolated in June 2001 (our unpublished observation), suggesting that the existence of such strains may not be very rare.
All V. cholerae strains positive for class I integrons were resistant to multiple antibiotics. However, as our data showed, the presence of class I integrons could account for the resistance to only a few drugs. This leads to the obvious conclusion that there exist other determinants of antibiotic resistance in these strains. In a few cases in which it was examined, it was found that some of these "determinants" reside on plasmids (Table 4). Since 43 out of the 94 strains examined in this study carried plasmids, it is possible that many of these plasmids carry drug resistance genes, irrespective of whether they carry class I integrons. Furthermore, the presence of class I integrons on the plasmids of around 8% of the strains examined (8 of 94) indicated that large plasmids carrying integrons may not be very rare among the non-O1, non-O139 strains of V. cholerae.
Colony hybridization of the 94 strains with the SXT integrase (int)-specific probe identified 12 strains that were positive for the presence of SXT int gene. The integrase gene of SXT element was used as an SXT-specific probe, because it does not have homology with the nucleotide sequence of other similar genes in the database. Thus, the strains that were identified as having the SXT int gene could actually harbor the 62-kb SXT element. All of these strains, except PL96, exhibited a drug resistance pattern characteristic of the SXT element. PL96 did not show resistance to streptomycin, suggesting the presence of a defective gene on the SXT. Seven of the strains that appeared to harbor the SXT element (PG262b, PG149b, PL78/6, PG95, PL61, PL96, and PL105b) also carried the class I integron. Almost all of these had the dfrA gene for trimethoprim resistance. Thus, in these strains, trimethoprim resistance was conferred by both the SXT element and the class I integron. The strain PL61 had the aadA2 gene for streptomycin in addition to the dfrA1 and dfrA12 genes for trimethoprim. The presence of multiple determinants for trimethoprim resistance is alarming, because this drug is used for the treatment of cholera in children and pregnant women. The presence of an ereA2 gene cassette is also of concern, because erythromycin also is used in the treatment of cholera in children.
Most of the strains included in the study were heterogeneous with respect to their serogroups and ribotypes, as well as a few genotypic markers examined: toxR, toxT, RS1, hlyA, and hlyU. V. cholerae strains PG224, PL134, and PL141 belonged to the same ribotype and displayed an identical genotype with respect to the markers mentioned above. Both PL134 and PL141 belonged to serogroup O2; the serotype of PG224 was not known. All three strains had class I integrons with a 1.25-kb cassette containing a dfrA1 gene for trimethoprim resistance and orfC (24). Although the antibiotic resistance patterns of these strains were not identical, it appeared from observations described above that these three isolates could have a common origin. In contrast, the strains PG9 and PG92, although they belonged to the same serogroup, displayed identical ribotypes, and had identical class I integron cassettes, differed from each other not only with respect to their antibiograms, but also with respect to their genotypes. V. cholerae strains PL78/6 and PG149b constituted an interesting pair in the sense that although they both had identical ribotypes and harbored the SXT element, they differed in all other attributes examined, including their class I integron cassette structure. Our findings thus showed that class I integrons, SXT elements, and plasmids bearing drug resistance markers were distributed fairly widely in the non-O1, non-O139 strains of V. cholerae, isolated from patients admitted into the Infectious Diseases Hospital, Calcutta, India. It also revealed for the first time the presence of gene cassettes, aac(6')-Ib, dfrA5, dfrA12, and ereA2, and also possibly a hybrid integron in V. cholerae. Since there are more and more reports of cholera-like diarrhea being caused by non-O1 strains (27), and since these strains appear to survive better than O1 strains in a wide range of foods (25), it is important to monitor the distribution of integrons and SXTs in emerging strains of V. cholerae.
M. Thungapathra gratefully acknowledges the Council of Scientific & Industrial Research, Government of India, for a Senior Research Associateship under the Scientists' Pool Scheme.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2010 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»