Profiling versus fingerprinting analysis of sesquiterpene hydrocarbons for the geographical authentication of extra virgin olive oils. (1st March 2020)

• Record Type: Journal Article
• Authors: Quintanilla-Casas, Beatriz; Bertin, Sofia; Leik, Kerstin; Bustamante, Julen; Guardiola, Francesc; Valli, Enrico; Bendini, Alessandra; Toschi, Tullia Gallina; Tres, Alba; Vichi, Stefania
• Is Part Of: Food chemistry. Volume 307(2020)
• Publication Date: 2020-03-01
• Subjects: Fingerprinting -- Geographical origin -- Virgin olive oil -- Sesquiterpene -- Food authentication; Food -- Analysis -- Periodicals; Food -- Composition -- Periodicals; 664
• DOI: 10.1016/j.foodchem.2019.125556
• Languages: English
• ISSNs: 0308-8146

Abstract

Highlights: Geographical authentication models developed with virgin olive oil sesquiterpene (SH) data. The suitability of SH as virgin olive oil geographical markers was confirmed. Better classification by SH fingerprinting (100%) than by profiling (46–100%). SH fingerprinting set a promising scenario for downscaling the model to smaller regions. The efficiency of the model by geographical origin was independent from the cultivar. Abstract: The verification of the geographical origin of extra virgin (EVOO) and virgin olive oil (VOO) is crucial to protect consumers from misleading information. Despite the large number of studies performed, specific markers are still not available. The present study aims to evaluate sesquiterpene hydrocarbons (SHs) as markers of EVOO geographical origin and to compare the discrimination efficiency of targeted profiling and fingerprinting approaches. A prospective study was carried out on 82 EVOOs from seven countries, analyzed by Headspace Solid Phase Microextraction-Gas Chromatography-Mass Spectrometry (HS-SPME-GC-MS). Classification models were developed by Partial Least Square-Discriminant Analysis (PLS-DA) and internally validated (leave 10%-out cross-validation). The percentage of correct classification was higher for the fingerprinting (100%) than for the profiling approach (45.5–100%). These results confirm the suitability of SHs as EVOO geographical markers and establish the fingerprinting as the most efficient approach for the treatment of SH analytical data with this purpose up to date.