Strain characterisation, antibiotic resistance and Meca Gene analysis of Staphylococci
S. aureus, is undoubtedly the most pathogenic of the Staphylococcus species, having the ability to produce invasive and toxigenic infections. Historically, the less virulent coagulase-negative staphylococci (ENS) were regarded as clinically insignificant contaminants but they have become increasingly implicated as opportunistic nosocomial pathogens. The increasing frequency of methicillin and multiple-antibiotic resistance in staphylococci over the last four decades has seriously compromised therapeutic options. The study was designed to (a) identify and type staphylococcal species, (b) undertake standardized antimicrobial susceptibility testing, and (c) determine the prevalence of methicillin resistance in staphylococcal isolates. Presumptive staphylococcal strains were isolated from the diagnostic microbiology laboratories of Pelonomi (147 strains) and Universitas (144 strains) hospitals. Subsequently, these strains were identified using conventional biochemical methods. Species-specific PCR identification assays were performed on selected ENS strains. Antimicrobial susceptibilities were determined for 13 clinically available antibiotics on 144 staphylococcal isolates and on selected strains for 5 developmental agents. RAPD and plasmid profiles were generated to assess possible epidemiological strain relatedness. For the detection of methicillin resistance in staphylococci the following methods were used: (a) oxacillin MICs detecting phenotypic methicillin resistance levels (b) a multiplex-PCR detecting the mecA gene, and (c) a slide agglutination test (MASTALEX-MRSA) detecting PBP2' production. The inclusion of bile-aesculin agar plates and a bacitracin susceptibility test into the diagnostic laboratory protocol for the identification of staphylococci would reduce misidentification of non-staphylococcal isolates by 12.8%. Colony morphology in combination with the coagulase test could be instrumental in the improved differentiation of S. aureus from CNS. Although expensive, when a rapid and fairly comprehensive identification of CNS species is required, the STAPH ID 32 API system was found to be satisfactory. Due to the apparent inaccuracy of the PCR identification assay based on API, its use in the clinical microbiology laboratory would be argued against; although if standardized and expanded it could be considered for future incorporation in routine practice. The presence of unique RAPD profiles for each specific Staphylococcus species suggests RAPD profiling could offer a molecular identification technique for the majority of commonly isolated CNS in the clinical microbiology laboratory. Good typeability was observed for Primer I and III in CNS strains, however, for S. aureus strains, poor typeability and discrimination was observed. It has been found by other researchers that longer oligonucleotide primers (>10 bp in length) are more efficient for S. aureus strain typing, but to the contrary in the present study primers ERIC 1 and 2 were totally unsatisfactory. Combined susceptibility data and plasmid profile analysis revealed strain relatedness in S. haemolyticus isolates but RAPD Primers I and III indicated otherwise. All staphylococcal strains isolated were vancomycin-susceptible. Of the staphylococci isolated in the Universitas hospital, 34.3% were oxacillin-resistant. Similarly, 30.1% staphylococci isolated in Pelinomi hospital were oxacillinresistant. Resistance to ciprofloxacin, erythromycin, gentamicin and clindamycin was found in 49% of oxacillin-resistant staphylococci. In comparison to the other quinolones tested, moxifloxacin showed superior activity against oxacillin-resistant CNS. The glycylcyclines, LY333328 and Q/D may well be considered excellent alternatives to vancomycin for the treatment of MRSA. Of the developmental agents investigated, linezolid showed consistent in vitro activity against all staphylococci. The inadequacy of a single diagnostic method for the detection of methicillin resistance in staphylococci was evident when comparing (a) susceptibility data, (b) multiplex-PCR for mecA gene detection, and (c) PBP2' detection. None of these methods was seen to correlate with each other at the 100%-level. The detection of PBP2' was rapid although, in comparison to mecA gene detection and antimicrobial susceptibility tests, inaccurate for the identification of methicillin resistance in staphylococci. DNA sequencing of a fragment of the mecA gene in selected staphylococcal strains revealed minimal sequence variation. This was an indication that variable levels of methicillin resistance in staphylococci can be attributed to different mechanisms of methicillin resistance or variations in the expression of the mecA gene, rather than mutations within the gene itself. The low sequence variation observed in the mecA gene is primarily responsible for initial assumptions of a clonal origin for methicillinresistance in staphylococci. As of yet, pharmaceutical companies have failed to produce an analogous antimicrobial agent to β-Iactam agents that would be able to specifically target PBP2'. The development of such an agent would be instrumental in the reduction of glycopeptide selection pressure. It is imperative that correct identification, strain typing, susceptibility testing and methicillin resistance detection is performed to direct therapy and epidemiologically monitor methicillin-resistant strain types.