Comparison of amplified ELISA and mouse bioassay procedures for determination of botulinal toxins A, B, E, and F.

The amplified enzyme-linked immunosorbent assay (amp-ELISA) was compared to the mouse bioassay for determination of botulinal neurotoxin types A, B, E, and F. Twelve different toxin-producing type A, 13 proteolytic type B, 9 nonproteolytic type B, 16 type E, 8 proteolytic type F, 5 nonproteolytic type F, and 6 nontoxigenic clostridial strains were tested. The cultures were inoculated into cooked meat medium (CMM) and tryptone-peptone-glucose-yeast extract (TPGY) medium, incubated for 5 days, and then examined for biological toxicity in mice and amp-ELISA endpoints. The amp-ELISA was less sensitive in detecting toxins produced by nonproteolytic than proteolytic strains of type B and F organisms. All of the toxin-producing strains tested were positive by the AOAC method and the amp-ELISA in either undiluted TPGY or CMM culture fluids regardless of mouse toxicity level, source, or strain. Cross-reactivity was observed between some but not all of the botulinal strains tested. None of the nontoxigenic strains were positive by the amp-ELISA. Purified botulinal toxins were also assayed using these 2 methods. The sensitivity of the amp-ELISA using purified neurotoxins was about 0.1 ng/mL for types A, B, and E and about 1.0 ng/mL for type F.