Human MRSA Isolates with Novel Genetic Homolog, Germany

To the Editor: Methicillin-resistant Staphylococcus aureus (MRSA) represents a major cause of hospital-, community- and livestock-acquired infections that are increasingly difficult to manage (1–3). Detection and identification of MRSA by culture and nucleic acid–based methods is challenged by heterogeneous penicillin-binding protein 2a (PBP2a) expression and variability of the staphylococcal cassette chromosome (SCCmec) elements. Recently, a new SCCmec element (XI) carried in bovine and human isolates was described (4,5). This SCCmec element contains a novel mecA homolog, designated mecALGA251, that is not detectable by usual mecA-specific PCR approaches and PBP2a agglutination tests. Garcia-Alvarez et al. reported this novel mecA homolog exhibited 70% identity at DNA level to the mecA gene, and suggested these strains were transmitted from livestock to humans (4).

30%-50% of immunosuppressed patients with coccidioidomycosis (3). Disseminated coccidioidomycosis typically involves the skin, meninges, or bone (3); however, intraocular involvement has also been described (1). A review of the literature shows 25 reported cases of intraocular coccidioidomycosis. When present, intraocular involvement is associated with serious consequences, frequently leading to eye enucleation; 1 case series described eventual enucleation in 50% of reported patients who did not die from disseminated coccidioidal infection (2).
For the patient in our report, in the setting of reported trauma and negative metastatic work-up results, it is unclear whether ocular disease resulted independently as an exogenous infection or from endogenous lymphatic and/ or hematogenous spread from the patient's lung. Diagnosis of coccidioidal endophthalmitis can be diffi cult, often relying on serum or nonocular tissue evaluation (4). Intraocular coccidioidal involvement usually occurs with widespread infection (5). Thus, even with apparent isolated ocular fi ndings, evaluation for disseminated disease is warranted, including a careful history and physical examination, CT chest scan, bone scan, intracranial imaging, and lumbar puncture. Evaluation for immunosuppression, including HIV status, is warranted.
The optimal systemic antifungal therapy for intraocular coccidioidal infection is unclear, although fl uconazole is the drug of choice for extrapulmonary coccidioidomycosis, including meningitis (3). Fluconazole has good ocular penetration; however, voriconazole also achieves excellent intraocular levels (6) at lower 90% minimum inhibitory concentration levels (7). Furthermore, Gabrielian and Hariprasad (8)  The patient in our report received systemic voriconazole for 4 weeks plus repeated intravitreal voriconazole injections on follow-up. It is possible that this initial therapy had an effect on his positive outcome and the avoidance of eye enucleation. The optimal length of therapy is unclear; however, this patient will receive prolonged treatment (>1 year) with high-dose fl uconazole, followed by a slow taper guided by serologic testing and regular ophthalmologic examination. Future research should evaluate which antifungal therapy is superior and the appropriate duration of treatment. identity at DNA level to the mecA gene, and suggested these strains were transmitted from livestock to humans (4).

Michael L Cheng, Matthew Leibowitz, and Edward Ha
To search for isolates possessing the novel mecA LAG251 , we screened S. aureus databases for those entries describing oxacillin/cefoxitin-resistant phenotypes that were negative for mecA by PCR (6) or harbored S. aureus protein A gene (spa) types known to be associated with the occurrence of mecA LGA251 (4,5). The databases of the University Hospital Münster contain S. aureus spa typing results of S. aureus isolates obtained from hospital admission screenings and specimens from patients treated at University Hospital Münster. Moreover, they include isolates derived from human and animal subjects, respectively, of 2 cross-border projects between the Netherlands and Germany: MRSA-net EUREGIO Twente/Münsterland and SafeGuard MRSA vet-net (2,7).
The presence of mecA LGA251 was verifi ed by using a specifi c PCR that applied newly designed primers: mecAL1 (5′-AGC TGG CCA TCC CTT TAT TT-3′) and mecAL2 (5′-CTG GCA TAT GGA GAA GAA GAA A-3′), derived from the sequence of S. aureus LGA251 provided by M. Holden (Wellcome Trust Sanger Institute, Hinxton, UK; accession no. FR821779). The sensitivities and specifi cities of primers were checked by applying S. aureus and other staphylococcal isolates of different clonal backgrounds (8,9). Positive PCR products were sequenced to confi rm identifi cation of mecA LGA251 ; the isolates were then characterized by typing the SCCmec region with specifi c primers for mecR1, mecI, blaZ, ccrA, and ccrB related to type XI SCCmec as described by García-Álvarez et al. (4). Identifi ed isolates were tested for PBP2a by using a latex agglutination assay (Oxoid Deutschland GmbH, Wesel, Germany). We used Etest (bioMérieux SA, Marcy-l'Étoile, France) for antibacterial agent susceptibility testing of β-lactams and other antibacterial agents.
We report on 16 (clinically derived, n = 14; ovine origin, n = 2) oxacillin/ cefoxitin-resistant S. aureus isolates possessing the recently described mecA LGA251 isolate, but lacking the classical mecA gene currently defi ning classic MRSA (Table). The isolates belong to spa types t843, t978, t1535, t1773, and t7189. Concurring with the fi ndings in the United Kingdom and Denmark, we found t843 to be the most prevalent spa type. Results of the PBP2a latex agglutination assay were negative for all isolates except for 1 (no. 14), which was indeterminable. García-Álvarez et al. described negative results for all tested isolates (4); Shore et al. reported inconsistent results with this test (5).
Until mecA LGA251 is included as a diagnostic target in molecular MRSA detection tests, oxacillin/cefoxitinresistant isolates determined to be methicillin-susceptible by traditional, culture-based susceptibility testing methods should not be disregarded, even if mecA and/or PBP2a tests fail to detect their targets. Susceptibility patterns of mecA LGA251 -positive S. aureus isolates revealed low MICs of oxacillin compared with those for MRSA of the classical mecA type. We presume this indicates an altered affi nity of β-lactam antibacterial agents to the putative mecA LGA251 gene product or a divergent expression of the gene. The choice and the dosage of antibacterial agents applicable for S. aureus infections should be reconsidered in light of this novel mecA homolog in molecular screening and identifi cation tests. Studies are warranted to investigate the prevalence of this novel MRSA entity in and outside of hospitals in the human population and in livestock, its clinical effects, and its response to antibacterial agent therapy.