In order to check details obtain clear and reproducible PFGE banding patterns using Cfr9I as restriction enzyme, the Harmony PFGE protocol had to be adjusted. This resulted in the following protocol: From each isolate, 100 μl bacterial suspension of an overnight Trypton Soy Broth (TSB) culture, was embedded in a plug mold

(Biorad) with 1.2% low-melting-point agarose (Seakem gold®, Biorad). Then, 500 μl lysostaphine (100 μg/ml, Sigma) was added and incubated for 6 h at 37°C. Subsequently, the plugs were incubated overnight at 55°C with 500 μl Proteinase K (50 μg/ml, Merck). The plugs were then washed, 6 to 10 times in a shaking incubator for 30 min. in 1 × Tris-EDTA buffer (Fluka, pH 7) at 50°C in order to remove cell debris. Finally, the plugs were equilibrated in 1 × Cfr9I buffer (Fermentas, Ontario, Canada) for 15 min. at room temperature prior to digestion and then submerged in QNZ concentration 200 μl of 1 × Cfr9I reaction buffer containing 40 U of Cfr9I restriction enzyme (Fermentas, Ontario, Canada). The reaction tubes were incubated overnight at 37°C in a shaking incubator. Further steps were carried out according to

the Harmony protocol [26]. Briefly, a 1% agarose gel was poured into a gel tray and positioned in a contour-clamped homogeneous electric field (CHEF) (Biorad) tank and submerged in 1,700 ml of 0.5 × Tris-Borate-EDTA (TBE). The total run time was 22 h at 14°C with an initial pulse time of 5 s, a final pulse time of 50 s and a voltage of 6 V/cm or 200 V. Gels were stained in PI3K inhibitor ethidium bromide (1 μg/ml, Invitrogen) and viewed

and photographed with UV transillumination. Digital images were analyzed using Bionumerics software, version 5.1. If a difference in PFGE pattern was observed, a new pulsed field type was assigned. The definition of a PFGE cluster was based on a similarity cutoff of 80% [28] (Dice coefficient, represented by UPGMA, 0.5% optimization and 1.0% tolerance). Different PFGE clusters were given in alphabetical order. Every band difference within PRKACG a PFGE cluster resulted in adding a numerical order to the pulsed field cluster. Results Optimization and validation of the Cfr9I PFGE method In the initial experiments the SmaI restriction enzyme was replaced by Cfr9I and exactly the same conditions were used as in the original PFGE protocol. This led to uninformative PFGE patterns consisting mainly of smears and faint bands obtained through partial digestion of the genomic DNA. A higher lysostaphine concentration (100 μg/ml), longer incubation steps for lysis (6 h), proteinase K and digestion overnight and hot washes at 50°C – instead of washes at room temperature – produced clear and reproducible banding profiles. After optimizing the PFGE method with Cfr9I, high quality banding patterns from all selected (n = 124) previously non-typeable ST398 MRSA isolates were obtained.