Identification by DNA hybridization of enterotoxigenic Escherichia coli in a longitudinal study of villages in Thailand.

Radioactively labeled enterotoxin genes were used to study the epidemiology of enterotoxigenic Escherichia coli infections in two Thai villages. When E. coli that were isolated from 674 specimens were fixed on nitrocellulose paper and examined for hybridization with E. coli enterotoxin gene probes in Bangkok, the technique had a sensitivity of 94% (31 of 33) and a specificity of 100% (641 of 641), when compared with tests of E. coli for enterotoxin production in the Y-1-adrenal cell and suckling-mouse assays. However, when the same specimens were fixed directly onto nitrocellulose paper at a field laboratory and transported to the reference laboratory for assay with the gene probes, 27 specimens that contained enterotoxigenic E. coli did not hybridize with the E. coli gene probes. Enterotoxigenic E. coli that hybridized with the LT, ST-H, and ST-P probes were identified in 10% (17 of 177) of villagers with diarrhea, 7% (8 of 108) of contacts of individuals with diarrhea caused by enterotoxigenic E. coli, and 3% (32 of 1,199) of persons not associated with cases of diarrhea caused by enterotoxigenic E. coli. Enterotoxigenic E. coli that hybridized with the ST-II probe was not a cause of diarrhea. Alternative methods of retaining DNA on filters under field conditions are needed before this technique can be used for direct examination of specimens with enterotoxin gene probes.