A 2-year-old child was admitted to the department of pediatrics at the Jawaharlal Institute of Postgraduate Medical Education and Research (JIPMER), Puducherry, India, with a provisional diagnosis of cholera. Before the child received antimicrobial therapy, a fecal specimen was submitted to the Department of Microbiology at JIPMER. The child was treated successfully with intravenous fluid supplements and ciprofloxacin.

Meanwhile, a strain of V. cholerae O1 El Tor Ogawa was isolated from the fecal specimen. This strain was biochemically identified (3) and confirmed by agglutination with specific antiserum (BD Difco, Becton Dickinson, Sparks, MD, USA). Antimicrobial drug susceptibility testing was conducted by Kirby–Bauer method in accordance with the Clinical and Laboratory Standards Institute (4) against ampicillin (10 μg), ceftriaxone (30 μg), ciprofloxacin (5 μg), furoxone (300 μg), cotrimoxazole (25 μg), and tetracycline (30 μg). The MIC for ceftriaxone was determined by the agar dilution method and Etest for the El Tor Ogawa strain. For the agar dilution method, we doubled dilutions of ceftriaxone sodium (Himedia, Mumbai, India) from 0.5 μg/mL to 128 μg/mL, which included the recommended break points (5). American Type Culture Collection Escherichia coli 25922 was spread on each plate as growth control. The Etest was performed according to the manufacturers’ instructions (AB bioMérieux, Solna, Sweden). We performed the combination disk test using cefotaxime and ceftazidime, alone and in combination with clavulanic acid, to detect extended-spectrum β-lactamase (ESBL) in accordance with Clinical and Laboratory Standards Institute guidelines (4). The AmpC disk test was performed as described (6). The modified Hodge test and imipenem-EDTA disk synergy test for the phenotypic detection of carbapenemase were performed as described (4,7). We extracted DNA from the strain by using the boiling method. PCR was performed to detect ESBL genes, i.e., blaCTX (8), blaSHV (8), and blaTEM (9); a multiplex PCR was used to detect the AmpC group of genes, i.e., blaMOX, blaCIT, blaDHA, and blaACC (10). A multiplex PCR to detect carbapenemase genes, i.e., blaKPC, blaIMP, blaVIM, and blaNDM, was performed (11). Sequencing was performed to identify AmpC β-lactamase gene and the carbapenemase gene. Sequencing of these genes was conducted by Macrogen Inc. (Seoul, South Korea). We used the BLASTN program (www.ncbi.nlm.nih.gov/BLAST) for database searching.

The strain was resistant to ampicillin, ceftriaxone, cotrimoxazole, and furoxone and sensitive only to ciprofloxacin and tetracycline. The MIC for ceftriaxone was 16 μg/mL. The phenotypic tests for ESBL detection were negative. PCRs for detection of the ESBL genes were negative. Because the strain was negative for the phenotypic and the genotypic tests for ESBL production, in our attempt to determine the reason for the increased MIC, we further tested for the AmpC type of β-lactamase production. The strain, an AmpC β-lactamase producer, was positive for the blaDHA gene (405 bp) by multiplex PCR. On sequencing, the blaDHA gene detected had 99% identity with Klebsiella pneumoniae β-lactamase blaDHA-1 gene (GenBank accession no. AY635140.1). Because carbapenems are considered the treatment of choice for AmpC-producing organisms, we tested the strain for the production of carbapenemase and found it to be a carbapenemase producer by a positive modified Hodge test and by the EDTA disk synergy test. The multiplex PCR for detection of carbapenemase genes yielded an amplicon of 660 bp, which was confirmed by sequencing to be an blaNDM-1 gene and had 100% identity with E. coli strain HK-01 plasmid pNDM-HK (GenBank accession no. HQ451074.1).

To demonstrate the presence of a plasmid bearing the blaNDM-1 gene, we isolated a plasmid using the alkaline lysis method (12) from the V. cholerae strain. This plasmid was subjected to the multiplex PCR to detect the carbapenemase genes mentioned earlier (11), which yielded an amplicon of 660 bp matching that of the blaNDM-1 gene amplicon. We transferred this plasmid by a method described by Lee et al. (13) and used E. coli J53 as the recipient, which was isolated on meropenem (4 mg/L) containing Luria-Bertani agar and MacConkey agar, meropenem (4 mg/L), and sodium azide (100 mg/L) (14). We also subjected the V. cholerae plasmid to the multiplex PCR to detect the AmpC gene, as mentioned earlier, but none of the genes were amplified. We failed to transfer the AmpC gene (blaDHA-1) by using the plasmid transfer experiment. Such failures of transfer of blaDHA-1 have been mentioned by Yan et al. (15).

Of the many AmpC genes known are 2 types of blaDHA genes: blaDHA-1 and blaDHA-2; both are inducible plasmid-mediated genes. Because the strain reported here was already resistant to a third-generation cephalosporin, i.e., ceftriaxone, we did not perform the inducible AmpC test and proceeded to perform the AmpC disk test. Plasmid-mediated inducible β-lactamases are extremely rare and are most often constitutive (15). The AmpC disk test does not differentiate between chromosomal and plasmid-mediated AmpC β-lactamases. We do not have complete information about the chromosomal AmpC genes in V. cholerae.

Carbapenems are the treatment of choice for AmpC-producing strains (6) and are hydrolyzed by carbapenemases, which have many families, including the metallo-β-lactamases (7). The most recently described bla_NDM-1 (New Delhi metallo-β-lactamase) is plasmid-mediated metallo-β-lactamase and has been isolated from the environment and from hospitals (14).

We isolated a clinical strain of V. cholerae producing an AmpC β-lactamase and a carbapenemase. Our findings, although perhaps not an issue for treatment of cholera, have other implications. The critical role of all bacteriology laboratories needs to be emphasized for determining not only resistance patterns but also the mechanisms of resistance. Health care–associated networks need to be strengthened to ensure justified and appropriate use of antimicrobial agents that will result in safe drinking water and improved sanitation; these can have remarkable effect in reducing the spread of many communicable diseases, such as cholera, and can go a long way in controlling the growing menace of antimicrobial drug resistance. In light of the above findings, the antimicrobial drug profile of organisms, such as V. cholerae, needs to be under constant surveillance in the community.