Discrimination of Atlantic salmon origins using untargeted chemical fingerprinting

• Published in: Food chemistry 2022-11, Vol.394, p.133538-133538, Article 133538
• Main Authors: Chang, Wen-Hsin, Soon Ling, Yee, Wang, Ko-Chih, Nan, Fan-Hua, Chen, Wen-Ling
• Format: Article
• Language: English
• Subjects: Atlantic salmon; chemical fingerprints; geographic origin; high-resolution mass spectrometry; untargeted analysis
• ISSN: 0308-8146; 1873-7072
• DOI: 10.1016/j.foodchem.2022.133538

[Display omitted] •Small-molecular features in Atlantic salmon were acquired using LC-HRMS.•Origin discrimination was accomplished using untargeted fingerprinting.•95 candidate markers of origins were selected by univariate analyses.•37 confirmed markers were related to feed formulations and environment adaptation.•None of the markers were residues or contaminants of potential food safety concern. Mislabelling the geographic origin of same-species aquaculture products is difficult to identify. This study applied untargeted small-molecule fingerprinting to discriminating between Atlantic salmon originating from Chile and Norway. The acquired liquid chromatography-high-resolution mass spectrometry data from Chilean (n = 32) and Norwegian (n = 29) salmon were chemometrically processed. The partial least squares discriminant analysis (PLS-DA) models successfully discriminated between Chilean and Norwegian salmon at both positive and negative ionisation modes (R2 > 0.96, Q2 > 0.81). Univariate analyses facilitated the selection of approximately 100 candidate markers with high statistical confidence (> 95%). Of these, 37 confirmed markers of Chilean and Norwegian salmon were primarily associated with feed formulations, including lipid derivatives and feed additives. None of the markers were residues or contaminants of potential food safety concern.