Detection in Italy of two clinicalEnterococcus faeciumisolates carrying both the oxazolidinone and phenicol resistance geneoptrAand a silent multiresistance genecfr: Table 1.

Sir, In a century in which the issue of emerging antibiotic resistance is being dominated by severe concerns chiefly regarding Gram-negative organisms, the multiresistance gene cfr is probably the greatest emerging problem in Gram-positive pathogens, particularly staphylococci and enterococci. The concern over this problem is motivated not only by the fact that the resistance involves linezolid—widely used in serious infections caused by MDR Gram-positive organisms, often as a last-resort drug—but also, and critically, by the fact that the frequent location of cfr on conjugative plasmids makes the resistance transferable. Now, the report in China of a second plasmid-borne transferable gene, optrA, conferring efflux-mediated oxazolidinone (including second-generation tedizolid) and phenicol resistance in enterococcal isolates1,2 adds further to the concern. As soon as the first report, with the sequence of an Enterococcus faecalis plasmid (pE349) carrying optrA (accession no. KP399637),1 became available as an Advance Access article in the Journal of Antimicrobial Chemotherapy, we decided to test for the optrA gene in 81 Enterococcus isolates from blood samples, which make up the first batch of an enterococcal collection we had recently started for a study including cfr screening. Identification at the species level was performed using VITEK 2 (bioMérieux, Marcy-l'Étoile, France). The cfr and optrA genes were sought by PCR using primer pairs internal to either gene: respectively, the known pair cfr-fw and cfr-rv, yielding a 746 bp amplicon,3 and the specially designed pair optrA-fw and optrA-rv, which yielded a 422 bp amplicon (Table S1, available as Supplementary data at JAC Online). Two Enterococcus faecium isolates were positive for cfr and two were positive for optrA. Much to our surprise, there were in fact only two positive isolates (E20818 and E35048), since each carried both optrA and cfr. The antibiotic MICs and other features for the two isolates are reported in Table 1. Both isolates had a relatively low linezolid MIC, 4 mg/L, a value that is regarded as ‘susceptible’ according to EUCAST4 and ‘intermediate’ according to CLSI.5 Both had a tedizolid MIC of 2 mg/L; breakpoints for resistance have recently been established by EUCAST4 for staphylococci and β-haemolytic streptococci (>0.5 mg/L) and for viridans group streptococci (>0.25 mg/L). The two isolates were also examined for mutations in 23S ribosomal RNA (not detected) and for the phenicol exporter genes fexA and fexB (not detected). Demographic data, antibiotic MICs and susceptibility profiles for the two E. faecium clinical isolates under study AMP, ampicillin; SAM, ampicillin/sulbactam; CXM, cefuroxime; IPM, imipenem; GEN (HLAR), gentamicin (high-level aminoglycoside resistance); STR (HLAR), streptomycin (high-level aminoglycoside resistance); ERY, erythromycin; CLI, clindamycin; Q/D, quinupristin/dalfopristin; LZD, linezolid; TZD, tedizolid; CHL, chloramphenicol; FFC, florfenicol; VAN, vancomycin; TEC, teicoplanin; TGC, tigecycline; SXT, trimethoprim/sulfamethoxazole; CVC, central venous catheter. aGentamicin MIC: ≤128 mg/L denotes no high-level resistance.4 bStreptomycin MIC: >512 mg/L denotes high-level resistance.4 Demographic data, antibiotic MICs and susceptibility profiles for the two E. faecium clinical isolates under study AMP, ampicillin; SAM, ampicillin/sulbactam; CXM, cefuroxime; IPM, imipenem; GEN (HLAR), gentamicin (high-level aminoglycoside resistance); STR (HLAR), streptomycin (high-level aminoglycoside resistance); ERY, erythromycin; CLI, clindamycin; Q/D, quinupristin/dalfopristin; LZD, linezolid; TZD, tedizolid; CHL, chloramphenicol; FFC, florfenicol; VAN, vancomycin; TEC, teicoplanin; TGC, tigecycline; SXT, trimethoprim/sulfamethoxazole; CVC, central venous catheter. aGentamicin MIC: ≤128 mg/L denotes no high-level resistance.4 bStreptomycin MIC: >512 mg/L denotes high-level resistance.4 The two isolates exhibited closely related SmaI-PFGE profiles; one (E35048) was investigated for molecular traits. Sequencing demonstrated that optrA and cfr displayed high-level DNA identities (98% and 99%, respectively) to the respective reference sequences (accession numbers KP3996371 and AJ579366). Three amino acid changes were detected in the protein sequence of cfr and 21 (4 of which were already reported in Chinese isolates)2 in the protein sequence of optrA compared with the respective reference sequences. The results of long PCR assays seeking a possible linkage between optrA and cfr were negative. The genetic contexts of both genes proved capable of undergoing excision in circular form, and were completely sequenced. The sequence of the minicircle containing optrA (3350 bp), deposited under accession no. KT892063, included a transposase gene downstream of optrA. This transposase gene exhibited 70% DNA identity and 65% amino acid identity to a chromosomal transposase from Clostridium sticklandii (accession no. FP565809). The minicircle (3405 bp) containing the cfr gene and one intact IS, ISEnfa5, was almost identical to a cfr genetic context described in Staphylococcus lentus (accession no. KF049005).7 Considering the low MICs of linezolid, florfenicol and chloramphenicol (Table 1), in spite of the prezsence of two resistance genes acting by different mechanisms (cfr perturbing the ribosome function8 and optrA providing for active efflux1), RT–PCR experiments were performed to check the actual transcription of the two genes (Supplementary Data). We found that optrA was transcribed, whereas cfr was not. Although the exact mechanism of non-transcription is still being investigated, preliminary data indicate a 52 bp deletion in the regulatory region upstream of cfr. Interestingly, a cfr gene failing to mediate resistance to oxazolidinones and phenicols has been described in a porcine E. faecalis isolate in China.9 Our collection of Enterococcus blood isolates is still in progress, and the overall results of the survey will be assessed and described later. The present, partial data are reported in advance to alert the scientific community to at least three important new facts: (i) that the optrA gene just described in enterococci in China is also found in enterococci isolated in Italy; (ii) that its genetic context may be unstable and yield a minicircle, a cause for concern in view of possible resistance spread; and (iii) that the optrA gene can be found in enterococcal strains that also carry cfr. Although in this first report cfr was silent, meaning that only optrA accounted for the oxazolidinone MICs, we cannot exclude the possibility that strains carrying both functional genes are circulating. This study was supported by internal funding. None to declare.