Verification of causal relationships between Listeria monocytogenes isolates implicated in food-borne outbreaks of listeriosis by randomly amplified polymorphic DNA patterns.

Food and clinical isolates of Listeria monocytogenes recovered from four different outbreaks of listeriosis were analyzed by their PCR-based randomly amplified polymorphic DNA (RAPD) patterns to verify their causal relationships. The generation of DNA fingerprints by PCR-based RAPD analysis is a fast and sensitive method for the epidemiological tracking and identification of bacteria implicated in food poisoning outbreaks. The L. monocytogenes strains used in the study were obtained from the following four outbreaks: California, 1985, Mexican-style cheese; Canadian Maritime Provinces, 1981, coleslaw; Canada, 1989, brie cheese; and Canada, 1989, alfalfa tablets. RAPD profiles were generated by using random 10-mer primers for at least one food and one clinical isolate recovered from each outbreak. Identical profiles for 20 different primers were observed for each pair of food and clinical isolates from two of the four outbreaks. Isolates from the outbreak involving alfalfa tablets exhibited identical patterns for 19 primers; however, primer OPA-1 produced one additional 1.8-kb fragment, designated OPA-1-1.8, that was found in the food isolate but not in the corresponding clinical isolate. Hybridization analysis revealed that the absence of the OPA-1-1.8 polymorphic fragment in the clinical isolate was due to a deletion of at least 1.8 kb. Loss of the OPA-1-1.8 polymorphic fragment could not be induced by infective passage of the L. monocytogenes isolate from the alfalfa tablet through a mouse or by growth of this isolate under selective conditions. This suggests that the isolate recovered from the food was not identical to the isolate recovered from the patient. The ability to produce unique RAPD patterns allows for the discrimination between isolates even if they are of the same serotype and multilocus enzyme electrophoretic type.