Development of an alternative approach for detecting botulinum neurotoxin type A in honey: Analysis of non-toxic peptides with a reference labelled protein via liquid chromatography-tandem mass spectrometry

Hiroshi Koike,Maki Kanda,Hairoshi Hayashi,Yoko Matsushima,Souichi Yoshikawa,Yumi Ohba,Momoka Hayashi,Chieko Nagano,Kotaro Sekimura,Kenji Otsuka,Junichi Kamiie,Takeo Sasamoto,Tsuneo Hashimoto

DOI: https://doi.org/10.1080/19440049.2020.1766121

2020-06-09

Abstract:In this study, we developed a reference labelled protein containing the partial amino acid sequence of botulinum neurotoxin type A (BoNTA). We also applied it as an internal standard to detect specific and non-toxic peptides originated from BoNTA in honey with the use of liquid chromatography–tandem mass spectrometry (LC–MS/MS). Original proteins in the honey sample were collected through a two-step process that included solubilisation and trichloroacetic acid (TCA) precipitation. Solubilisation by adding water enabled processing of proteins in honey. TCA precipitation collected proteins without specific binding. The combination of protein alkylation and an appropriate enzyme-to-protein ratio ensured feasibility of tryptic digestion. A desalting process eliminated a large amount of salts and other tryptic peptides in the honey sample. The use of the reference labelled protein enabled compensation for tryptic digestion efficiency and electrospray ionisation efficiency based on LC-MS/MS measurement. After the peptide selection and protein BlastP analysis, five unique peptides were chosen. The non-toxic peptides originating from BoNTA were reliably detected using LC–MS/MS based on a multiple-reaction monitoring mode. Detection of several peptides ensured screening of BoNTA in honey samples. Based on the responses, the proteotypic peptide LYGIAINPNR was selected as the quantitative peptide. Due to maintaining the relative ion ratios, the selective transition completely identified the non-toxic peptides. The intensity of the transitions established a detection limit of BoNTA estimated to be 9.4 ng mL⁻¹. Although extraction efficiency was not evaluated using the BoNTA standard, the results suggested this method may be used for quantification of BoNTA in honey. The method was applied to 19 honey samples purchased in Tokyo; none of them was found to contain the target toxin. Overall, the method is expected to accelerate BoNTA monitoring for food safety.