SME-2-Producing Serratia marcescens Isolate from Switzerland

Abstract

Carbapenemases in Serratia marcescens have been rarely reported, being related either to metallo-β-lactamases IMP-1, IMP-6, and VIM-2 (10) or to SME-type Ambler class A β-lactamases (6). SME-1 had been identified first from the carbapenem-resistant S. marcescens strain S6, isolated in London in 1982 (5), and then in carbapenem-resistant S. marcescens strains isolated in 1999 in the United States (1). It hydrolyzes penicillins, aztreonam, cephalosporins, and carbapenems and is inhibited by clavulanic acid (4). The blaSME-1 gene was chromosome encoded in isolate S6 (5). The SME-2 and SME-3 variants have been identified (being just point mutant analogues) from S. marcescens isolates recovered from the United States and United Kingdom (8, 9). Our study was initiated by the isolation in May 2006 at the University Hospital of Lausanne, Switzerland, of a carbapenem-resistant S. marcescens isolate. This isolate was from an exudate obtained after a parotidectomy of a 62-year-old patient who received a prophylaxis containing amoxicillin-clavulanate. Three days after surgery, the patient suffered from a pulmonary deficiency and received an imipenem-containing treatment. Two days later, culture of the exudate gave S. marcescens strain AW, which was resistant to penicillins and imipenem. The MICs of imipenem, meropenem, and ertapenem were 32, 8, and 4 μg/ml, respectively. It was also resistant to cefoxitin and aztreonam, whereas it remained susceptible in vitro to expanded-spectrum cephalosporins. The results for double-disk synergy testing, performed as described previously (2), were slightly positive with the clavulanate-imipenem combinations, indicating that S. marcescens AW produced a carbapenemase that was inhibited by clavulanic acid. PCR amplification using various primers, including blaSME-specific primers and whole-cell DNA of S. marcescens AW, followed by sequencing performed as described previously (7), identified a gene encoding SME-2. No other similar isolate was recovered from the same hospital during the same period of time, and no history of travel abroad or previous hospitalization was identified. The clonal relationship between isolate AW and isolate S6 from London was evaluated by pulsed-field gel electrophoresis as described previously (8), showing that isolate AW was not clonally related to isolate S6. By analyzing the sequences of the rpoB genes as reported previously (3), we found that the sequences obtained from isolates AW and S6 and also from two randomly selected carbapenem-susceptible and blaSME-negative S. marcescens isolates were identical. This observation likely rules out a possible identity for SME producers as part of a given subspecies. Conjugation, electrotransformation, and plasmid analysis, performed as described previously (5, 7), failed to identify a plasmid-borne location for the blaSME-2 gene, suggesting a likely chromosomal location for this gene. Further PCR mapping showed the presence of the LysR-type regulatory gene smeR upstream of blaSME-2, as previously identified (4). This was in accordance with the observation of a slight antagonism observed between the cefoxitin- and imipenem-containing disks. This report underlines the possible identification of SME-type S. marcescens producers worldwide. Along with other clavulanic acid-inhibited carbapenemases (NMC-A, IMI, KPC), SME-type enzymes may confer clinically significant resistance to carbapenems. Interestingly, SME producers may confer a lower level of resistance to ertapenem than to the other carbapenems.