应用核磁共振技术对类风湿关节炎血清代谢物图谱研究

摘要

目的 应用基于核磁共振(NMR)的代谢组学方法,研究类风湿关节炎(RA)患者血清特征的代谢物表型.方法 收集65例血清标本,其中RA患者30例、健康对照35例.用600 mHz NMR采集血清代谢图谱,结合模式识别法分析RA内源性代谢物的差异及构建诊断模型.结果 RA患者糖类、脂类、氨基酸及能量代谢途径均受影响.主成分分析法(PCA)模型可以区分疾病组与健康对照组.偏最小二乘法判别分析(PLS-DA)模型具有较好的可解释性(R2X =64.2％,R2Y=7.16％)及可预测性(Q2 =0.652).而正交偏最小二乘法判别分析(OPLS-DA)模型的敏感性及特异性达到96.7％和91.4％.结论 基于NMR的血清代谢组学技术有望成为RA诊断的新方法并在临床应用中蕴涵巨大潜力.%Objective Using nuclear magnetic resonance ( NMR) spectroscopy that depended on metabolomics approach to investigate the characteristic metabolites phenotype in serum from patients with rheumatoid arthritis. Methods 65 serum samples were collected including 30 RA patients and 35 healthy controls. Serum metabolites spectroscopy were obtained with 600 mHz NMR, the difference of serum metabolic profiles of RA patients were identified and consequently the diagnosis models were established combined with pattern recognizable methods. Results Diversities occurred in the metabolism of carbohydrate, lipids, amino acids and energy in RA patients according to metabolic profiles. Principal component analysis ( PCA) model is capable of distinguishing RA from healthy control. Partial least-squares discriminant analysis (PLS-DA) model has a good interpretability (R2X = 64. 2% , R2Y = 7.16%) and predictability ( Q2 =0.652). And the sensitivity and specificity of Orthogonal PLS-DA (OPLS-DA) model is up to 96. 7% and 91.4% separately. Conclusion NMR involved in metabolomics approach has a great potential to be a new diagnosis tool for RA.