Determination of capsaicinoids and carotenoids for the characterization and geographical origin authentication of paprika by UHPLC–APCI–HRMS

The production area mislabeling of a food product is considered a fraudulent practice worldwide. In this work, a method that uses ultra-high-performance liquid chromatography coupled to high-resolution mass spectrometry using atmospheric pressure chemical ionization (UHPLC–APCI–HRMS) was used for th...

Descripción completa

Detalles Bibliográficos
Autores: Arrizabalaga Larrañaga, Ane, Campmajó, Guillem, Saurina, Javier, Núñez, Oscar, Santos, Francisco Javier, Moyano, Encarnación
Tipo de recurso: artículo
Fecha de publicación:2020
País:España
Institución:Universidad del País Vasco
Repositorio:Addi. Archivo Digital para la Docencia y la Investigación
OAI Identifier:oai:dnet:addi________::718ea2e84f38565b4c6ff859dba67112
Acceso en línea:http://hdl.handle.net/10810/78780
Access Level:acceso abierto
Descripción
Sumario:The production area mislabeling of a food product is considered a fraudulent practice worldwide. In this work, a method that uses ultra-high-performance liquid chromatography coupled to high-resolution mass spectrometry using atmospheric pressure chemical ionization (UHPLC–APCI–HRMS) was used for the geographical origin authentication of paprika based on the determination of capsaicinoids and carotenoids. Satisfactory instrumental method performance was obtained, providing good linearity (R2 > 0.998), run-to-run and day-to-day precisions (%RSD < 15 and 10%, respectively), and trueness (relative errors < 10%), while method limits of quantification were between 0.21 and 51 mg·kg–1. Capsaicinoids and carotenoids were determined in 136 paprika samples, from different origins (La Vera, Murcia, Hungary, and the Czech Republic) and types (hot, sweet, and bittersweet). The composition of capsaicinoids and carotenoids was used as chemical descriptors to achieve paprika authentication through a classification decision tree built by partial least squares regression−discriminant analysis (PLS-DA) models and reaching a rate of 80.9%.