In vitro assays for estimating the activity of therapeutic preparations of botulinum toxin

The potency of novel botulinum neurotoxin-derived therapeutic products, now in routine medical use, is determined exclusively by in vivo methods. The currently accepted assay, which is dependent on lethality as an endpoint, has been identified as a priority for replacement by ECVAM. Immunoassays have been shown not to be able to differentiate between active and inactive neurotoxin, which limits their applicability for monitoring type A botulinum toxin (BoNT/A) activity in therapeutic formulations, where it is essential to measure biological activity. The discovery that SNAP-25, a synaptosomal membrane-associated protein, is the intracellular target for BoNT/A has provided the basis for the development of mechanism-based, in vitro assay methods. Using recombinant and chemical methods, we have prepared fragments of SNAP-25 containing the BoNT/A cleavage site at bond Q(197)-R-198. The proteolytic cleavage of recombinant (SNAP-25(134-206)) and synthetic (SNAP-25(137-206)) peptides was monitored either by capillary zone electrophoresis or by immunochemical methods using targeted antibodies. The specificity of the reaction was confirmed by molecular weight analysis of the two products and the inhibitory profile of known inhibitors of zinc-dependent endopeptidases. Both assays were sufficiently sensitive for measuring BoNT/A activity in therapeutic formulations, with a limit of detection of <0.5 mouse LD50/ml. Using immunodetection, we obtained an excel; lent correlation with the in vivo bioassay for different clinical formulations (r = 0.93, slope = 1.03 +/- 0.02, n = 32). It is possible to estimate activity of BoNT/A by using nonanimal models. Although the alternative approach described looks promising, further validation is required, as the activity measured reflects only the intracellular mode of action of the neurotoxin.