COMPARISON OF 3 TECHNIQUES FOR MOLECULAR DIFFERENTIATION OF FOODBORNEAND CLINICAL LISTERIA STRAINS - SINGLE-STRAND CONFORMATION POLYMORPHISMS, RANDOM AMPLIFIED POLYMORPHIC DNA POLYMORPHISMS AND PULSED-FIELD GEL-ELECTROPHORESIS

The type of molecular subtyping technique used in tracing and epidemio logy depends largely on discriminatory ability, particularly when the focus is on problematically species like Listeria monocytogenes where threat to human health is assumed to be caused by clones of enhanced p athogenicity. Traditional subtyping techniques like serotyping fail to reveal these clones. Molecular techniques employing amplification (Ra ndom amplification of polymorphic DNA, Single strand conformation poly morphisms-PCR technique) or restriction of genomic DNA (Pulsed field g el electrophoresis) excel in tracing for contaminants or clinical isol ates. The goals of this study were: (i) to evaluate three subtyping te chniques - Random amplification of polymorphic DNA (RAPD), PCR-Single strand conformation polymorphisms-technique (PCR-SSCP) and Pulsed fiel d gel electrophoresis (PFGE) in a set of 66 Listeria isolates originat ing from both foodstuffs and clinical specimens, (ii) to calculate the numerical index of discriminatory ability (D-value) and (iii) to appo int the reproducibility of each method by distributing randomly seven genetically identical strain couples into the set of isolates. Pulsed field gel electrophoresis using restriction endonucleases Sma I, Asc I , Not I and Apa I revealed numbers of PFGE-subtypes to be 29, 30, 30 a nd 34, respectively. The index of discriminatory ability (D-value) was calculated 0.963 using Sma I, 0.97 using Asc I, 0.97 using Not I and 0.9711 using Apa I. RAPD analysis with primer sequence HLWL 74 reveale d 23 subtypes, therefore D-value was calculated 0.93. Conversely, SSCP typing showed only 4 subtypes amplifying intergenetic region V6 (D-va lue: 0.27) and no discrimination amplifying intergenetic region V3. Re producibility of the three methods used was shown to be 100 %: all thr ee methods successfully identified each of the seven randomly distribu ted strain-couples as identical.