Abstract

Sir, Carbapenems are considered to be one of the few drugs that are useful for the treatment of infections caused by multidrugresistant isolates with high levels of AmpC or extended-spectrum b-lactamase (ESBL) production. While carbapenem resistance has become a serious problem for non-fermenting bacteria, it has remained uncommon in Enterobacteriaceae. However, identification of carbapenem-resistant Enterobacteriaceae is increasingly common. Especially, reports of nosocomial infection due to carbapenem-resistant Serratia spp. have become significantly more common, which has mainly been attributed to production of b-lactamases that are capable of hydrolysing carbapenems such as IMP-1, IMP-6, VIM-2, and SME-1, 2 and 3. The KPC class A carbapenemases are capable of hydrolysing carbapenems, penicillins, cephalosporins and aztreonam, and are inhibited by clavulanic acid and tazobactam. The first report of this b-lactamase, KPC-1, was from a carbapenemresistant Klebsiella pneumoniae. KPC-2 was then found in isolates of K. pneumoniae, Salmonella enterica, Klebsiella oxytoca and Enterobacter spp. Soon afterward, KPC-3 was found in K. pneumoniae and Enterobacter cloacae from New York. Recently, KPCs were found in France, South America and Israel. We describe three clinical isolates of carbapenem-resistant Serratia marcescens that produce KPC-2. This is the first report of detection of a plasmid-encoded carbapenem-hydrolysing enzyme KPC-2 in S. marcescens. We collected three clinical isolates of carbapenem-resistant S. marcescens from our hospital in February 2006. The three isolates exhibited either resistance or reduced susceptibility to carbapenems with imipenem MICs of 8, 64 and 8 mg/L. The isolates were also resistant to penicillins, cephalosporins and aztreonam, but were susceptible to quinolones and aminoglycosides (Table 1). Following mating of the three S. marcescens isolates with an Escherichia coli (EC600) by conjugation, all of the resulting transconjugants exhibited a phenotype of resistance to b-lactams similar to that of the S. marcescens isolate used. The imipenem and meropenem MICs of transconjugants were 1 mg/L. They were also resistant to penicillins and aztreonam, and were resistant or intermediately resistant to cephalosporins, but were susceptible to quinolones and aminoglycosides (Table 1). The three S. marcescens isolates and three transconjugants, one derived from a mating with each parent strain, had identical plasmid profiles and restriction patterns (data not shown). Following the elimination of plasmids by repeated SDS treatment, the S. marcescens isolates became susceptible to all antibiotics except ampicillin. Isoelectric focusing analysis of crude enzyme extracts from the isolates (data not shown) demonstrated the production of two b-lactamases with apparent pIs of 6.5 and 6.7. The E. coli transconjugants produced the pI 6.7 b-lactamase only. Neither pI 6.7 nor pI 6.5 b-lactamase was detected in the three plasmid-eliminated S. marcescens strains. Given the pI of 6.7, we suspected the production of a KPC b-lactamase. To confirm the presence of a blaKPC gene, we conducted PCR amplification using the primers KPC-F and KPC-R. The reaction gave a positive result and DNA sequencing identified the gene as blaKPC-2. The data show that the carbapenem-hydrolysing enzyme in the three test S. marcescens isolates is KPC-2. The enzyme is encoded on a conjugative plasmid and mainly contributed to resistance to carbapenems. We presume that the unidentified pI 6.5 b-lactamase was either the degradation product or the precursor of KPC-2. Unidentified carbapenemase activity bands around pI 6.3–6.5 have also been detected from K. pneumoniae, which were presumed to be the degradation product of KPCs. The resistance against ampicillin remaining when the blaKPC-containing plasmid had been cured is probably due to the low activity of chromosomal AmpC in S. marcescens, which was not detected in IEF but confirmed through enzyme assay. The three S. marcescens isolates had identical plasmid profiles and restriction patterns, indicating that the same plasmid spread among these isolates. Most blaKPC-encoding plasmids could be conjugated into E. coli, except the blaKPC-1-encoding plasmid described by Yigit et al. Many blaKPC-encoding plasmids also encode blaTEM-1 and various ESBLs. However, only a blaKPC gene was identified from the plasmid in S. marcescens in this study. KPC family enzymes have been detected in a variety of Enterobacteriaceae but mainly in North America. The emergence of a KPC-2 type carbapenem-hydrolysing enzyme in China is alarming. It indicates that the spread of blaKPC genes is no longer a regional problem.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call