基于非靶向代谢组学技术的胆石症血浆代谢标志物的初步探讨

A Pilot Study Based on Non-targeted Metabolomics Revealed Perturbed Plasma Metabolites for Cholelithiasis

背景:胆石症是胆道系统最常见的疾病之一,有关胆石症代谢组学的研究目前尚无报道。目的:应用非靶向代谢组学技术筛选胆石症患者的血浆代谢标志物。方法:基于上海市男、女性健康队列开展胆石症血浆代谢谱预初实验。应用超高效液相色谱-四级杆飞行时间质谱(UPLC-QTOFMS)和气相色谱-飞行时间质谱(GC-TOFMS)技术检测43例基线期胆石症患者和43名年龄配对健康对照者的血浆代谢物,采用主成分分析(PCA)、正交偏最小二乘判别分析(OPLS-DA)以及t检验筛选胆石症患者的特异性代谢物。通过Logistic回归分析构建胆石症预测模型,以ROC曲线对模型进行初步评价。结果:UPLC-QTOFMS和GC-TOFMS共检测出血浆代谢物618个,根据OPLSDA VIP> 1. 0且t检验P <0. 05的筛选条件,筛选出差异代谢物30个,由其中的2-甲基-4-戊烯酸、磷脂酰乙醇胺PE(p-16∶0e/0∶0)、顺式-4,7,10,13,16,19-二十二碳六烯酸、2-氨基丁酸、尿酸、瓜氨酸、S-异戊烯基-L-半胱氨酸、羟环己脲和2,3-环氧甲基萘醌九个代谢物构建的胆石症预测模型ROC曲线下面积(AUC)为0. 97,敏感性和特异性分别为95. 4%和88. 4%。结论:胆石症患者的血浆代谢谱发生明显变化,由其中9个差异代谢物构建的预测模型能较好地区分胆石症患者与健康对照者。

Background: Cholelithiasis is one of the most commonly seen diseases of biliary tract. However,the metabolomics study of cholelithiasis has not yet been reported. Aims: To identify perturbed plasma metabolites by non-targeted metabolomics technique in patients with cholelithiasis. Methods: A case-control study nested in the Shanghai Men ’s Health Study and the Shanghai Women’s Health Study was conducted to preliminarily investigate the plasma metabolic profile of cholelithiasis. Plasma metabolites from 43 patients with cholelithiasis that diagnosed at baseline and 43 agematched healthy subjects were evaluated using ultra-high performance liquid chromatography quadruple time-of-flight mass spectrometry( UPLC-QTOFMS) and gas chromatography time-of-flight mass spectrometry( GC-TOFMS). Principal component analysis( PCA),orthogonal partial least squares-discriminant analysis( OPLS-DA) and t-test were carried out to compare the case-control differences. A prediction model for cholelithiasis was constructed using logistic regression analysis,and its diagnostic performance was estimated by ROC curve analysis. Results: A total of 618 plasma metabolites derived from the UPLC-QTOFMS and GC-TOFMS were included. OPLS-DA( VIP > 1. 0) and t-test( P < 0. 05) identified a panel of 30 metabolites as potential biomarkers. A prediction model for cholelithiasis including nine of these metabolites( 2-methyl-4-pentenoic acid,PE( p-16∶ 0e/0∶ 0),cis-4,7,10,13,16,19-docosahexaenoic acid,2-aminobutyric acid,uric acid,citrulline,S-prenyl-L-cysteine,hydroxyhexamide and 2,3-epoxymenaquinone) yielded an area under the ROC curve( AUC) of 0. 97,with sensitivity of 95. 4% and specificity of 88. 4%. Conclusions: The plasma metabolic profile changes significantly in cholelithiasis patients. The prediction model constructed by 9 perturbed metabolites might be used to distinguish cholelithiasis patients from healthy subjects.