Relevance of isotopic and molecular biomarkers for the authentication of milk according to production zone and type of feeding of the cow

The first objective of the present paper was to assess the potential of both isotopic ( (18)O/ (16)O in milk water) and molecular biomarkers (terpenes, fatty acids, carotenoids, and vitamins) and milk color to discriminate the production zone (lowland or upland areas) from which 49 tanker bulk milks were collected over one year from a total of 204 farms. The milk water (18)O enrichment was higher in lowland (<500 m altitude) than in upland (>700 m altitude), but the delta (18)O values failed to discriminate systematically the production zone at the scale of the year because of its high variability related to the sampling period. In contrast with vitamins A and E, carotenoids, and milk color measurements, terpenes and fatty acids were confirmed to be relevant tracers of the production zone. The milk compounds with the strongest discriminative potential were fatty acids, which were determined by high-resolution gas chromatography. The calculation of fatty acid ratios, which permits the limitation of using fatty acid relative quantity expressed in percentage of total fatty acids to be overcome, was shown to be particularly relevant in discriminating upland from lowland milk ratios. The selection of two pairs of ratios, namely, iso-C17:0/C18:3 n-3 and iso-C15:0/iso-C14:0, enabled the authentication of 100% of the highland versus lowland milks whatever the season. The second objective was to evaluate the relevance of fatty acid composition to discriminate milks according to the proportion of corn silage in the diets of dairy cows. The selection of two fatty acids ratios, namely, trans11 cis15-C18:2/trans11-C18:1 and cis9-C16:1/iso-C16:0, enabled the correct classification of 100% of the milk samples according to the proportion of corn silage in the basic fodder rations (<25% vs >30%). The relationship between the milk production zone and the type of forage fed to the cows is discussed.