Abstract: 【Objective】 This trial was designed to screen small molecule metabolites with differences in quantity in the serum of cows with milk fever (MF), and to clarify the changes of metabolites in cows suffering from MF at the molecular level and from the angle of material, energy metabolism. 【Method】 Thirty-two multiple holstein cows with similar age, parity, body condition and milk yield at 0 day after parturition were selected, and then assigned into two groups according to their serum calcium concentration and the clinical symptoms. Among them, 24 cows were assigned to the healthy control group (Group1, serum calcium concentration> 2.5 mmol·L^-1, without any other symptoms) and 8 MF group (Group2, serum calcium concentrations <1.4mmol·L^-1, with obvious clinical symptoms of MF). Blood  at 10mL was collected from the jugular vein from each of the 32 cows before feeding and miking in the morning, placed in a centrifugal tube, centrifuged at 4℃, 1 500×g for 20min, and then the serum obtained by centrifugation was separately loaded in a 1.5 mL EP tube at -80℃ until the test. After thawing, 400 μL serums was taken from every EP tube, added 200 μL buffered salt solution, mixed well, and then 550 μL supernatant was extracted into 5 mm NMR tube after centrifugation to collect signals using the 500 MHz NMR spectrometer. Then MestReNova and Topspin softwares were applied on the Fourier transform of the signal, the correction of zero, the baseline and phase, removed water and urea peaks, and the integration of one-dimensional map for segmentation, and mapping information into a TXT file format to facilitate subsequent data analysis, and then all compounds were identified using the Chenomx software. Finally, the SIMCA-P software was used for the multivariate statistical analysis, including principal component analysis (PCA) and orthogonal signal correction-partial least squares-discriminant analysis (OSC-PLS-DA), combined with the P values obtained from one-way ANOVA of NMR data performed in SPSS software and Loading plot, eventually small molecule metabolites with differences in performance were filtered out. 【Result】 In this study, serum metabolic profiles and loading plot from MF group and control group were successfully obtained. PCA result showed that all samples were within 95% confidence interval, without excluding, and the contribution rate of the primary component was low, in which PC1 = 26.2%, PC2 = 16.7%, the different variables between the two groups can not be selected. OSC-PLS-DA result showed that the five orthogonal signal corrections were conducted to remove those components which displayed no significant grouping and maximize the differences between groups. Compared with the healthy control group, 9 differentially expressed serum metabolites were screened, which included 4 upregulated compounds, such as β- hydroxybutyrate, acetone, pyruvate and lysine, 5 downregulated compounds, such as glucose, alanine, glycerol, creatine phosphate and gamma-Aminobutyrate. They were mostly sugars and amino acids, formed an energy conversion network diagram between each, involved in various energy metabolism pathways in the body. Of which, the more specific metabolites, such as creatine phosphate, directly provided energy for the body, decrease of which may be associated with the performance of muscle weakness and paralysis on the cow. In addition, the gamma-Aminobutyrate(GABA) of cows with MF was decreased, and had clinical manifestations of depression, even coma, which was coincided with human performance of depression, but the relationship between decrease of GABA spirit of the cow and MF needs to be further confirmed by investigators. 【Conclusion】 ^1ders.H NMR technique can be applied to screen small molecule metabolites in cow serum with differences in quantity and the difference was significant. In this study, metabolites in cows with MF showed some pathological features of negative energy balance and fat mobilization, suggesting that MF is associated with energy metabolism disor

Key words: dairy cows, milk fever, ¹H Nuclear magnetic resonance, metabolic profiling, energy metabolism disorders