Objective: To describe antimicrobial resistance and molecular epidemiology of methicillin‐resistant Staphylococcus aureus (MRSA) isolated in community settings in Australia. Design and setting: Survey of S. aureus isolates collected prospectively Australia‐wide between July 2004 and February 2005; results were compared with those of similar surveys conducted in 2000 and 2002. Main outcome measures: Up to 100 consecutive, unique clinical isolates of S. aureus from outpatient settings were collected at each of 22 teaching hospital and five private laboratories from cities in all Australian states and territories. They were characterised by antimicrobial susceptibilities (by agar dilution methods), coagulase gene typing, pulsed‐field gel electrophoresis, multilocus sequence typing, SCCmec typing and polymerase chain reaction tests for Panton–Valentine leukocidin (PVL) gene. Results: 2652 S. aureus isolates were collected, of which 395 (14.9%) were MRSA. The number of community‐associated MRSA (CA‐MRSA) isolates rose from 4.7% (118/2498) of S. aureus isolates in 2000 to 7.3% (194/2652) in 2004 (P = 0.001). Of the three major CA‐MRSA strains, WA‐1 constituted 45/257 (18%) of MRSA in 2000 and 64/395 (16%) in 2004 (P = 0.89), while the Queensland (QLD) strain increased from 13/257 (5%) to 58/395 (15%) (P = 0.0004), and the south‐west Pacific (SWP) strain decreased from 33/257 (13%) to 26/395 (7%) (P = 0.01). PVL genes were detected in 90/195 (46%) of CA‐MRSA strains, including 5/64 (8%) of WA‐1, 56/58 (97%) of QLD, and 25/26 (96%) of SWP strains. Among health care‐associated MRSA strains, all AUS‐2 and AUS‐3 isolates were multidrug‐resistant, and UK EMRSA‐15 isolates were resistant to ciprofloxacin and erythromycin (50%) or to ciprofloxacin alone (44%). Almost all (98%) of CA‐MRSA strains were non‐multiresistant. Conclusions: Community‐onset MRSA continues to spread throughout Australia. The hypervirulence determinant PVL is often found in two of the most common CA‐MRSA strains. The rapid changes in prevalence emphasise the importance of ongoing surveillance.

Clinical application of antimicrobial peptides (AMPs), as with conventional antibiotics, may be compromised by the development of bacterial resistance. This study investigated AMP resistance in methicillin resistant Staphylococcus aureus, including aspects related to the resilience of the resistant bacteria toward the peptides, the stability of resistance when selection pressures are removed, and whether resistance can be overcome by using the peptides with other membrane-permeabilising agents. Genotypically variant strains of S. aureus became equally resistant to the antibacterial peptides melittin and bac8c when grown in sub-lethal concentrations. Subculture of a melittin-resistant strain without melittin for 8 days lowered the minimal lethal concentration of the peptide from 170 μg ml-1 to 30 μg ml-1. Growth for 24 h in 12 μg ml-1 melittin restored the MLC to 100 μg ml-1. Flow cytometry analysis of cationic fluorophore binding to melittin-naïve and melittin-resistant bacteria revealed that resistance coincided with decreased binding of cationic molecules, suggesting a reduction in nett negative charge on the membrane. Melittin was haemolytic at low concentrations but the truncated analog of melittin, mel12-26, was confirmed to lack haemolytic activity. Although a previous report found that mel12-26 retained full bactericidal activity, we found it to lack significant activity when added to culture medium. However, electroporation in the presence of 50 μg ml-1 of mel12-26, killed 99.3% of the bacteria. Similarly, using a low concentration of the non-ionic detergent Triton X-100 to permeabilize bacteria to mel12-26 markedly increased its bactericidal activity. The observation that bactericidal activity of the non-membranolytic peptide mel12-26 was enhanced when the bacterial membrane was permeablized by detergents or electroporation, suggests that its principal mechanism in reducing bacterial survival may be through interaction with intracellular organelles or processes. Additionally, our results showed that the haemolytic peptide bac8c, had increased antibacterial activity at non-haemolytic concentrations when used with membrane-permeabilizing surfactants.

Multiple studies have shown that the antibacterial dressing Acticoat can inhibit growth of bacteria but is unable to completely clear a wound of infection, which could leave patients vulnerable to sepsis. Agar inoculated with four different Staphylococcus aureus strains and overlain with Acticoat showed growth inhibition beneath and within a 1 mm perimeter of the dressing after 24 h. When lifted from inoculated agar and briefly blotted onto fresh agar plates, Acticoat transferred viable bacteria. Scanning electron microscopy of the surface of Acticoat that overlaid meticillin-resistant S. aureus for 24, 48 and 72 h showed dense clusters of apparently undamaged bacteria distributed across the mesh. The number of bacteria growing on inoculated pig skin, underneath and on the surface of Acticoat, was lower than on controls for the first 8 h, but after 24 h the number of bacteria on the skin was 2.3-fold greater than the untreated controls. In contrast, after 24 h the number of bacteria surviving on the surface of the Acticoat was 11.9 % of controls. Acticoat moistened with 10 % glycerol plus antimicrobial peptides (AMPs) mel12-26 or bac8c (50 mg ml 21 ) reduced the numbers of bacteria on the dressing and on the skin underneath to below 10 % and 0.01 % of the controls, respectively. When lysozyme (1 mg ml 21) was added to Acticoat wetted with glycerol and the AMP bac8c, the dressing was able to prevent the survival of bacteria on densely inoculated pig skin and on the surface of Acticoat for up to 24 h. In effect, biocompatible solvents and AMPs significantly enhance the bactericidal efficacy of Acticoat.