New approach for the rational selection of markers to identify botulinum toxins.

The application of mass spectrometry (MS) to detect unique peptide markers has been widely employed as a means of identifying bacterial proteins. Botulinum neurotoxins (BoNTs) are bacterial proteins that cause the life-threatening disease botulism. BoNTs are divided into several antigenically distinct serotypes and several dozen subtypes. The toxins' molecular heterogeneity makes their detection highly challenging. In this study, we describe a new LC-MS/MS-based platform for the direct identification of proteins derived from various species and subspecies in a single assay, as exemplified by BoNTs. The platform employs a rational down-selection process through several steps based on a combination of bioinformatics, tryptic digestion, and LC-MS, each leads to the final panel of markers. This approach has been demonstrated for all 8 subtypes of botulinum serotype A (BoNT/A). Ab-independent and Ab-dependent assays were developed based on the identification of 4 rationally selected markers or a combination of some of them, which enables full selectivity coverage. The Ab-independent assay, which is highly simple and rapid, has a sample-to-result turnaround time of approximately 40 min and enables the identification of 500 MsLD50/mL (5 ng/mL) BoNT/A in complex environmental matrices. The Ab-dependent assay, which is based on toxin's specific enrichment, has a turnaround time of 100 min, but enables improved sensitivity (50 MsLD50/mL, 0.5 ng/mL). Both assays were verified and validated using various environmental samples. This approach can easily be expanded to other botulinum serotypes and exhibits the potential for even further extension as a highly multiplexed assay for protein-based toxins, viruses, and organisms.