RESUMO
An enzyme-linked immunosorbent assay (ELISA) was developed to screen for the presence of botulinal toxin types A, B, and E in inoculated food studies. A commercially available trivalent antitoxin (Connaught Laboratories, Ontario) was used as a capture antibody and biotinylated for use as a secondary antibody. An avidin-alkaline phosphatase conjugate coupled with an enzymebased amplification system resulted in a high degree of sensitivity. Detection levels of purified neurotoxins in gelatin phosphate buffer were 9 LD50 for type A and <1 intraperitoneal mouse LD50 for types B and E, respectively. Toxin produced by two-type F strains (proteolytic and nonproteolytic) was detected in a liquid laboratory medium. In a comparative study of over 490 samples of ground turkey meat inoculated with C. botulinum types E and nonproteolytic B, the ELISA gave no false negatives and 91 false positives. False positives were thought to be due to the presence of inactivated toxin or toxin levels insufficient to cause mouse death. Statistical analysis of these data showed an ELISA sensitivity of 100%, specificity of 70.6%, and an efficiency of 81.4% when compared to the mouse bioassay for detection of botulinal toxins types B and E. Coffee intermediates inoculated with proteolytic Clostridium botulinum types A and B caused nonspecific death in mice but were negative for presence of toxin by ELISA.
