Abstract
PCR for the detection of botulinum neurotoxin gene types A to E was used in the investigation of a case of equine botulism. Samples from a foal diagnosed with toxicoinfectious botulism in 1985 were reanalyzed by PCR and the mouse bioassay in conjunction with an environmental survey. Neurotoxin B was detected by mouse bioassay in culture enrichments of serum, spleen, feces, and intestinal contents. PCR results compared well with mouse bioassay results, detecting type B neurotoxin genes in these samples and also in a liver sample. Other neurotoxin types were not detected by either test. Clostridium botulinum type B was shown to be prevalent in soils collected from the area in which the foal was raised. Four methods were used to test for the presence of botulinum neurotoxin-producing organisms in 66 soil samples taken within a 5-km radius: PCR and agarose gel electrophoresis (types A to E), PCR and an enzyme-linked assay (type B), hybridization of crude alkaline cell lysates with a type B-specific probe, and the mouse bioassay (all types). Fewer soil samples were positive for C. botulinum type B by the mouse bioassay (15%) than by any of the DNA-based detection systems. Hybridization of a type B-specific probe to DNA dot blots (26% of the samples were positive) and PCR-enzyme-linked assay (77% of the samples were positive) were used for the rapid analysis of large numbers of samples, with sensitivity limits of 3 x 10(6) and 3,000 cells, respectively. Conventional detection of PCR products by gel electrophoresis was the most sensitive method (300-cell limit), and in the present environmental survey, neurotoxin B genes only were detected in 94% of the samples.