Ribotyping of Chryseobacterium meningosepticum: its use as an epidemiological tool and its correlation with serovars.

Chryseobacterium meningosepticum (basonym, Flavobacterium meningosepticum King 1959) is associated with neonatal meningitis and is isolated from normal and immunocompromised adults. AAF-labelled Escherichia coli 16 + 23S rRNA was used as a probe for ribotype analysis of 92 clinical isolates from tracheal exsudate, blood culture, cerebrospinal fluid, urine and pus. The 92 isolates belonged to the 15 described serovars of C. meningosepticum, and included 21 strains isolated during an outbreak in an intensive care unit, all belonging to serovar G. Three restriction endonucleases, EcoRI, HindIII and PstI, were selected for use in ribotyping after preliminary experiments. Epidemiologically unrelated isolates were discriminated by ribotyping and could be classified into 48 ribotypes according to the hybridization banding patterns obtained after restriction with the three enzymes. Strains which were not discriminated by combined ribotype analysis belonged to the same serovar, and were of identical geographic origin. In one case, analysis with an additional enzyme, PvuII, was necessary for separating strains from two different serovars. However, three strains from different serovars (two isolated from the same place and one elsewhere within eight years) showed the same combined ribotype. Analysis of the rRNA gene patterns revealed 6 different patterns for clinical isolates of the outbreak, suggesting unrelated sources of infection. In three patients, isolation of C. meningosepticum with different combined ribotypes suggested superinfection. Ribotyping enabled differentiation between isolates belonging to the same serovar as well as between isolates of different serovars and provided a useful molecular epidemiological tool for the study of C. meningosepticum. Combined ribotype analysis with several restriction endonucleases increased the discriminating power of the method. However, there was only a partial correlation between serovars and the extent of DNA relatedness.