基于LC-Q/TOF-MS代谢组学方法筛选奶牛妊娠识别阶段乳样生物标志物的研究

Selection of Milk Biomarkers of Pregnancy Recognition in Dairy Cows Based on LC-Q/TOF-MS Metabolomics

本研究旨在筛选奶牛妊娠识别阶段乳汁生物标志物,为早期妊娠诊断提供新的思路和方法。选择胎次和体况相近的荷斯坦健康高产奶牛10头,同期发情后,采集人工授精当天和第17天乳样,采用高效液相色谱-四极杆-飞行时间质谱(LC-Q/TOF-MS)代谢组学技术,结合主成分分析(PCA)和正交偏最小二乘法-判别分析(OPLSDA)对两组乳样代谢轮廓进行模式识别;根据OPLS-DA模型中的VIP值(VIP>1)和Kruskal-Wallis检验的P值(P<0.05)筛选两组乳样差异代谢物;ROC曲线进一步考察这些差异代谢物对两组乳样分类的识别能力。结果表明:1)人工授精当天与第17天乳样代谢轮廓存在明显差异;2)10个代谢物(甲羟戊酸内酯、色胺、L-酪氨酸、磷酸二氢新蝶呤、乙酰半胱氨酸、N-乙酰基-L-酪氨酸、L-脯氨酸、四氢1-苯甲基-1,2,3,4-四氢异喹啉、鸟苷、十七烷基肉碱)被初步确定为两组乳样的差异代谢物;3)这10个差异代谢物中仅有6个代谢物(L-酪氨酸、N-乙酰基-L-酪氨酸、L-脯氨酸、四氢1-苯甲基-1,2,3,4-四氢异喹啉、鸟苷和十七烷基肉碱)对两组乳样具有显著的区分能力,最终被视为妊娠识别阶段潜在乳样生物标志物。综上表明,牛奶中L-酪氨酸、N-乙酰基-L-酪氨酸、L-脯氨酸、四氢1-苯甲基-1,2,3,4-四氢异喹啉、鸟苷和十七烷基肉碱有望成为妊娠识别期妊娠诊断指标,也可为深入研究奶牛早期妊娠识别机理提供新的视角。

This study aimed to select milk biomarkers of pregnancy recognition in dairy cows and provide a new insight and method for early pregnancy diagnosis.Ten healthy Holstein high yield dairy cows with similar body condition and parity were selected.All cows were oestrus synchronized and their milk samples were collected at day 0 and day 17 of artificial insemination.Metabolic profiles of the 2 groups of milk samples were recognized with LC-Q/TOF-MS metabolomics combined with principal component analysis（PCA）and orthogonal partial least squares-discriminant analysis（OPLS-DA）;Differentiated metabolites for the 2 groups of milk samples were selected according to VIP values（VIP1）from OPLS-DA model and P values（P〈0.05）from Kruskal-Wallis test;Their discriminating ability to classification of the 2 groups of milk samples were further observed by ROC curve.The results showed that：1）Milk metabolic profiles at day 0 were obviously different from those at day 17 after artificial insemination.2）Ten metabolites（mevalonolactone,tryptamine,L-tyrosine,dihydroneopterin phosphate,acetylcysteine,N-Acetyl-L-tyrosine,L-proline,1 BnTIQ,guanosine,heptadecanoyl carnitine）were first confirmed as differentiated metabolites between the 2 groups of milk samples.3）Only 6 of the 10 differentiated metabolites including 1 BnTIQ,N-Acetyl-L-tyrosine,L-tyrosine,guanosine,L-proline and heptadecanoyl carnitine had the obviously separating capacity for the 2 groups of milk samples and were finally viewed as potential milk biomarkers of pregnancy recognition.In conclusion,the 6 metabolites could be better indicators for dairy cows pregnancy diagnosis at pregnancy recognition stage,and provide a new insight into early pregnancy recognition mechanism.