Random amplification of polymorphic DNA versus pulsed field gel electrophoresis of SmaI DNA macrorestriction fragments for typing strains of vancomycin-resistant enterococci.

Genetic typing of vancomycin-resistant enterococci (VRE) can be performed using a variety of methods, but comparative analyses of the quality of these methods are still relatively scarce. We here compare random amplification of polymorphic DNA (RAPD) analysis with pulsed field gel electrophoresis (PFGE) of DNA macrorestriction fragments as examples of two of the recent and well-accepted molecular typing methods. For the latter method, empirical guidelines for the interpretation of the DNA fingerprints have been proposed in the international literature. Based on our experimental analyses, we define similar criteria for RAPD fingerprinting. A collection of 100 strains of VRE, comprising Enterococcus faecium, Enterococcus faecalis, Enterococcus avium, Enterococcus gallinarum and Enterococcus casseliflavus, was assembled. Fifty isolates were Dutch, another 50 were isolated in the UK. Strains were selected on the basis of previously determined putative identity, close relatedness or uniqueness. The strains were analysed using well-standardised RAPD and PFGE protocols. Resulting fingerprints were interpreted with computerised methods involving band positioning and we show that typing of VRE by PFGE and RAPD generates highly congruent DNA fingerprint clustering. When the proposed international criteria for interpretation of PFGE fingerprints were applied in our case, 86% PFGE homology as discriminating value between close relatedness and uniqueness, a 75% homology cut-off for the comparison of the RAPD-generated DNA fingerprints revealed essentially identical strain clusters. As a spin-off it is revealed that strains from the different species can be efficiently discriminated, that strains from the UK and The Netherlands form separate clusters and that strains from veterinary origin can be identified separately as well.