Metabolomics profiling of four biofluids from dairy cow fed different forages using gas chromatography–time of flight/mass spectrometry

Abstract Text: The quality of forage can significantly impact milk production and quality. This study aims to investigate the effect of type of forage on metabolites in biofluids of dairy cow and their potential role in lactation. Sixteen multiparous Holstein dairy cows were blocked based on days in milk (164 ± 27.5 d, mean ±SD) and milk yield (29.4 ± 2.16 kg, mean ± SD), and were randomly assigned into one of two treatments. Isonitrogenous diets with a ratio of forage-to-concentrate at 45:55 contained similar concentrate mixtures and 15% corn silage, with two different forage sources (DM basis): 23% alfalfa hay and 7% Chinese wild rye hay (AH); and 30% corn stover (CS). After a 65-d feeding, four biofluids (rumen fluid, milk, serum, and urine) were collected from all cows to characterize the metabolites using a novel metabolomics method based on gas chromatography–time of flight/mass spectrometry). A total of 165, 195, 218, and 156 metabolites were identified in rumen fluid, milk, serum, and urine, respectively. Among them, 29 metabolites were mutual in all four biofluids. The chromatograms among four biofluids were obviously different, with a clear discrimination between diets CS and AH. Principal component analysis yielded the separated clusters of metabolite profiles between diets AH and CS. Partial least squares discriminant analysis also showed segregation in the metabolites from four biofluids between two diets. There were 56, 8, 31, 31 metabolites in four biofluids, respectively, that differed significantly between diets AH and CS (VIP>1 & P<0.05). These metabolites were mainly involved in amino acid metabolism and carbohydrate metabolism. Among 56 dissimilar metabolites in the rumen, 90% was lower for diet CS than for AH, while 74% of 31 dissimilar urinary metabolic wastes was higher in CS. The cumulative explanation rate (R²Y) of orthogonal to partial least squares discriminant analysis was 0.899, 0.967, 0.899, and 0.982, respectively, suggesting that biofluid metabolomics combining metabolic profiles with multivariate analysis can be used to investigate the cow complex metabolic alteration in response to different type of forages with successful metabolites models.

Keywords: biofluids, metabolomics, dairy cow