Differences in the serum metabolome profile of dairy cows according to the BHB concentration revealed by proton nuclear magnetic resonance spectroscopy (¹H-NMR)

Objectives: During the transition period, a higher nutrient demand and a reduction in dry matter intake induce to a negative energy balance (NEB). The nonesterified fatty acids (NEFA) are then mobilized by adipose tissue and they may be oxidized into ketone bodies (β-hydroxybutyrate (BHB), acetoacetate and acetone). This condition makes dairy cows more susceptible to metabolic diseases such as ketosis. The BHB concentration is commonly used for the diagnosis in high-yielding dairy cows, with a cutoff of blood BHB value above 1.0-1.4 mmol/L for subclinical ketosis or hyperketonemia without clinical signs. These conditions are not easily identifiable and are frequently related to other diseases that cause economic loss. Metabolomics is a new analytical approach that aims to measure simultaneously the entire metabolite profile of a biologic sample. The aim of this study was to analyze the serum metabolome using ¹H-NMR in dairy cows with different levels of BHB. Materials and methods: Animal care and procedures were in accordance with the European Directive 2010/63/EU and the national law D.L. 2014/26. Furthermore, The Ethics Statement was approved by the Animal Care and Use Committee of the University of Padua (ID number 91/2019 - “BovineOmics” Projects). Forty-nine Holstein Friesian dairy cows between 15 and 30 days in milk were enrolled from a single high-yielding dairy farm. The same total mixed ratio (TMR) was used for all enrolled animals. A cross-sectional experimental design was used. Each animal received a clinical examination by veterinarians at the University of Padua and animals with clinical signs of disease were excluded from the study. The blood samples were collected into tubes containing clot activator to obtain serum for biochemical and metabolomic analysis. According to the serum BHB concentration, the animals were divided into three groups: Group 0 (G0; 12 healthy animals; BHB≤0.50 mmol/L); Group 1 (G1; 19 healthy animals; 0.51≤BHB<1.0 mmol/L); and Group 2 (G2; 18 hyperketonemic animals; BHB≥1.0 mmol/L). The statistical differences for biochemical parameters were performed by one-way ANOVA, whereas a t-test was used to evaluate differences in metabolite concentration. A post hoc pairwise comparison among metabolite concentrations was performed using Bonferroni correction. A robust principal component analysis (rPCA), and the metabolic pathways overrepresentation analysis (ORA) were generated to summarize the structure of the data and to highlight the metabolic pathways influenced by BHB concentration. A p-value<0.05 was accepted, whereas a 0.05≤p-value≤0.10 was considered as trend to significance. Results: Among biochemical parameters, only NEFA showed a significant difference between groups, with a progressive increment according to BHB concentration. A total of fifty-seven metabolites were identified in serum samples: 27 amino acids and derivates, 10 organic acids, 5 alcohols, 4 carbohydrates, 2 amine and derivates, 2 fatty acids, 2 ketone bodies, 1 sulfone, 1 vitamin, 1 imidazole, 1 nucleoside, and 1 guanidine. The extreme groups (G0-G2) showed a statistical difference for thirteen metabolites, specifically: glutamate, proline, serine, aspartate, isovalerate, and choline showed a significant reduction in G2, whereas 3-hydroxybutyrate, 3-hydroxyisobutyrate, acetate, succinate, 2,3-butanediol, methanol, and methylsuccinate showed a significant increase. In addition, 11 metabolites showed a trend toward significance: lysine, alanine, arginine, formate, pyruvate, and dimethylsulfone were reduced in G2, whereas isoleucine, valine, ethanol, trimethylamine-N-oxide (TMAO), and acetone were increased. The rPCA analysis revealed three different structure of metabolome, with G1 values located between G2 and G0. The ORA analysis identified three metabolic pathways possibly responsible for changes in metabolome profile: lipid metabolism, synthesis of phosphatidylserine, and glycosaminoglycan metabolism. Conclusions: Metabolomic analysis through ¹H-NMR is a useful tool to achieve knowledge about metabolic profiling related to serum β-hydroxybutyrate modifications during the transition period in dairy cows. The metabolic state of our hyperketonemic cows suggests the mobilization of body resources, increased anaerobic fermentation, alteration of lipid metabolism, a potential oxidative stress state, and a possible alteration of inflammatory and healing processes. This study demonstrates that the metabolomic approach can be considered a significant means to achieve knowledge about dairy cow diseases and their pathogenesis.