Abstract: 【Objective】 Fatty liver is an important metabolic disorder of dairy cows in transition period, and 50% of postpartum cows are affected. Productivity redution, poor reproductive performance, weak immune function, failure of liver function and premature death are related to fatty liver in postpartum cows. Surface-enhanced laser desorption/ionization (SELDI) time-of-fight (TOF) mass spectrometry (MS) is a sensitive proteomics technology. The aim of this study was to determine the differences between the plasma proteomic profiles of health cows and those with fatty liver. 【Method】  Forty Holstein cows at 7-28 d postpartum with 1-2 parties were selected from an intensive dairy farm in Heilongjiang province. Ten milliliters of blood was taken from the jugular vein of the 40 cows, the blood was placed in an anticoagulant tube and were mixed with the anticoagulant evenly. Blood samples were centrifuged at 3000r/min for 5 min, the serum were stored at −80°C. According to the concentration of serum glycerin trilaurate (TG>0.20mmol·L^-1) and beta-hydroxybutyric acid (BHBA>0.97mmol·L^-1), 20 cows were assigned to the fatty liver group (T) and 20 to the control group (C). Plasma proteins mass spectra were tested bysurface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF-MS). The peak figures of the two groups were analyzed using the Ciphergen ProteinChip Software. The peak data of the two groups were analyzed using wilcoxon rank sum test, p values were calculated to confirm the difference of the two groups. Different peaks were selected if P<0.01. m/z values of different peaks were compared with theoretical m/z values using Swissport database. The analogous proteins were identified as a prediction result. 【Result】The 39 significantly different peptide peaks which were identified in the T group were compared with the theoretical m/z values in Swissport database. A total of 26 predictable differences peaks were obtained, 11 kinds of proteins were predicted. Results showed that the levels of these 11 proteins were reduced in T group compared with C group. The 11 kinds of protein networks figure were analyzed by bioinformatics analysis (Network，GO，Pathway) and cattle genetic network search, networks analysis results were got. The two differential protein amyloid precursor protein (APP) and fibrinogen alpha chain (FGA) were found in database, relative Network analysis results were obtained. The BinGO plug-ins in Cytoscape software were used to analyze the 11 kinds of protein and GO analysis results were got. There were 9 kinds of different proteins were found by the software search, including FGA, serum amyloid A protein (SAA), plasma protease c1 inhibitor (SERPING1 or C1INH), apolipoprotein c-Ⅲ (Apo-CⅢ), hepcidin (HAMP), osteopontin (OPN or SPP1), transthyretin (TTR), Cystatin-c (CysC, CST), neurosecretory protein (VGA). There were 6 kinds of different proteins were searched in the KEGG pathway database, including FGA, SERPING1, Apo-CⅢ, APP, CysC and SPP1. They may related to the pathogenesis of fatty liver of dairy cows. 【Conclusion】The differential expression protein in serum between health cows and fatty liver cows were separated effectively by SELDI-TOF-MS technology, which may play an important role in the process of metabolism of liver and development of fatty liver. Therefore, this study has important theoretical values to reveal new mechanism of fatty liver and its effect on bio-function of dairy cows. Further studies need to be conducted to reveal the impact of different proteins on the pathogenesis of fatty liver in dairy cows.

Key words: dairy cows, fatty liver, SELDI-TOF-MS, plasma proteomics, differential expression protein