DI/LC-MS/MS-based metabolomics identifies early predictive serum biomarkers for ketosis in dairy cows

Abstract Text:

Subclinical ketosis is a prevalent metabolic disease in transition dairy cows that affects 30-40% of the cows during early lactation. Cows with ketosis have lower milk yield and reproductive performance, greater risk of other periparturient diseases, and higher culling rate. The objectives of this study were to retrospectively evaluate alterations of metabolites in the serum of dairy cows with ketosis before, during, and after the diagnosis of disease and identify monitoring and diagnostic serum metabolite biomarkers for ketosis. One hundred transition dairy cows were sampled and 20 healthy cows (CON) and 6 cows with ketosis were sampled during -8, -4, disease diagnosis, +4 and +8 wks relative to parturition. One hundred and twenty-eight serum metabolites were quantitatively profiled in CON and ketosis cows using a targeted metabolomics approach based on DI/LC-MS/MS at all time points. Univariate and multivariate data analyses were conducted at each time point to examine alterations of serum metabolites throughout the progress of ketosis. Significant changes were detected in the concentrations of several molecular species of amino acids, glycerophospholipids, sphingolipids, acylcarnitines, biogenic amines, and hexose in the serum of cows with ketosis during the entire experimental period. Multivariate analysis (i.e., PCA and PLS-DA) also showed clear distinctions between the two groups on the basis of the measured 128 serum metabolites at five time points. Furthermore, several metabolic pathways including Lys degradation, biotin metabolism, Try metabolism, urea cycle, Arg-Pro metabolism, protein biosynthesis, Met metabolism, phospholipid biosynthesis, Val-Leu-Ile degradation, betaine metabolism, Asp metabolism, His metabolism, and beta-Ala metabolism were perturbed in cows with ketosis during the onset and progression of disease. These new findings give insights into further understanding of the pathobiology of ketosis in dairy cows. Biomarker analysis showed that AUCs for ROC curves were 0.996 (95% CI, 0.969-1) at -8 wks, 0.995 (95% CI, 0.938-1) at -4 wks, 0.99 (95% CI, 0.882-1) at disease wk, 1 (95% CI: 1-1) at +4 wks and 0.985 (95% CI: 0.806-1) at +8 wks, respectively, which suggest that serum biomarkers identified have pretty accurate predictive, diagnostic, and prognostic abilities for ketosis in transition dairy cows.

Keywords: Amino acid, biomarkers, dairy cows, ketosis, lipid profiles.