Comparison of pulsed-field gel electrophoresis and whole-genome-sequencing-based typing confirms the accuracy of pulsed-field gel electrophoresis for the investigation of local Pseudomonas aeruginosa outbreaks

Abstract

To determine whether pulsed-field gel electrophoresis (PFGE) accurately recognizes isolates belonging to clusters defined by techniques based on whole-genome sequencing (WGS) using Pseudomonas aeruginosa as a model. We selected 65 isolates of ST395 P.aeruginosa isolated in seven European hospitals between 1998 and 2012. Isolates were typed by PFGE and sequenced by WGS. A core genome multi-locus sequence typing (cgMLST) analysis based on 3831 genes was performed with a homemade pipeline. PFGE identified eight pulsotypes and cgMLST differentiated nine clusters and nine singletons. Five cgMLST clusters and pulsotypes (31/65 isolates) coincided perfectly. Isolates without evident epidemiological links grouped by PFGE were separated by cgMLST (16/65 isolates) differentiating cities, suggesting that PFGE should be kept for the investigation of local outbreaks. Importantly, hypermutator isolates still shared the pulsotype with their parents (16/65 isolates), whereas they were not recognized by cgMLST. This shows that PFGE was less affected than WGS-based typing by the accelerated genetic drift that occurs in epidemic P.aeruginosa. although WGS-based typing has logically become the new reference standard, we show here that the PFGE can be used with confidence for the investigation of local outbreaks caused by P.aeruginosa.