Estimating size and diversity of nitrifying communities in deodorizing filters using PCR and immunofluorescence

Thirty isolates of Listeria monocytogenes and 18 of L. innocua obtained from different short-ripened cheeses manufactured in Asturias (northern Spain), were compared with each other and with reference strains using serotype, phage type and pulsed-field restriction endonuclease digestion profiles analysis of the total DNA. Restriction enzymes Apa I and Sma I defined five clusters in L. monocytogenes ( m1 to m5 ) and two main clusters in L. innocua ( i1 and i2 ). Cluster i2 was further arranged into three subclusters ( i2a , i2b and i2c ) based on the different Eco 52I ( Xma III) and Crf 42I ( Sac II) patterns of its isolates. Clusters of L. innocua were clearly different whereas those of L. monocytogenes were more closely related to each other. In this latter species, serotype 4b isolates ( m4 and m5 ) constituted a more homogeneous group than serogroup 1 isolates ( m1 , m2 and m3 ). Cluster m3 contained two strains of serotype 1/2a whereas m1 and m2 harboured strains of both serotypes, 1/2a and 1/2b. Therefore, the combined use of restriction patterns and serotype may be useful to differentiate L. monocytogenes strains showing identical restriction profiles but differing in serotype. The cheese source of Listeria strains proved that isolates from cluster m1 were repeatedly detected as a contaminant in the same type of cheese. Comparison of L. monocytogenes Apa I profiles showed a genetic proximity of m4 and m5 to the recognized pathogenic strains ATCC 13932 and NCTC 11994, responsible for meningitis cases in other countries. Finally, bacteriophage typing data indicated that m4 , the sole phage typable group, had a phage type resembling that of strains causing the Auckland (New Zealand) outbreak of listeriosis in 1969. These data suggest a wide distribution of closely related types which might cause, under several circumstances, sporadic cases of listeriosis.