基于非靶标代谢组学方法的原因不明复发性流产患者血浆代谢标志物的研究

Study of plasma metabolic markers in unexplained recurrent spontaneous abortion based on non-target metabolomics approach

目的应用非靶标代谢组学技术筛选原因不明复发性流产(URSA)患者的血浆代谢标志物。方法收集2022年9月至2023年5月在甘肃省妇幼保健院就诊的URSA且妊娠早期出现先兆流产症状的孕妇23例(URSA组), 并选取同期于本院进行产前检查的健康妊娠早期孕妇22例(对照组), 采集两组孕妇的血浆, 采用超高效液相色谱-质谱(UPLC-MS)联用技术对血浆代谢物进行代谢组学分析, 应用差异表达倍数分析、主成分分析和偏最小二乘法判别分析方法筛选差异代谢物, 并应用受试者工作特征(ROC)曲线评价差异代谢物的诊断效能, 应用通路富集分析方法筛选与URSA发生相关的通路。结果 URSA组与对照组孕妇的年龄、体重指数、孕周均无显著差异(P均>0.05)。使用UPLC-MS联用技术进行代谢组学分析显示, 从血浆中共检测到526种代谢物, 根据筛选条件发现其中33种是与URSA相关的差异代谢物。通过ROC曲线分析确定了曲线下面积(AUC)较大的6种差异代谢物, 包括磷脂酰乙醇胺(AUC=0.972, 95%CI为0.920~1.000)、檀萜烯水合物(AUC=0.902, 95%CI为0.786~0.982)、L-亮氨酸(AUC=0.884, 95%CI为0.772~0.960)、西松烯(AUC=0.881, 95%CI为0.758~0.956)、咖啡因(AUC=0.875, 95%CI为0.756~0.962)、4-羟基苯甲酸丙酯(AUC=0.864, 95%CI为0.732~0.946)。6种差异代谢物联合诊断URSA的AUC为0.983(95%CI为0.929~1.000)。对差异代谢物进行通路富集分析显示, URSA的发生与多条代谢通路相关, 包括咖啡因代谢、甘油磷脂代谢、不饱和脂肪酸的生物合成等。结论妊娠早期正常妊娠与URSA孕妇的血浆代谢谱有差异, 6种潜在的差异代谢物包括磷脂酰乙醇胺、檀萜烯水合物、L-亮氨酸、西松烯、咖啡因和4-羟基苯甲酸丙酯及其代谢通路可能参与URSA的发生过程。

Objective:To screen plasma metabolic markers in patients with unexplained recurrent spontaneous abortion(URSA)by non-target metabolomics approach.Methods:From September 2022 to May 2023,the plasma of 23 URSA pregnant women with threatened abortion who visited the outpatient clinic of Gansu Provincial Maternity and Child-care Hospital in the first trimester(URSA group)was collected,and the plasma of 22 healthy pregnant women in the first trimester who underwent prenatal examination during the same period(normal control group)was collected.Plasma metabolomics was analyzed by ultra performance liquid chromatography(UPLC)coupled with mass spectrometry(MS),fold change analysis,principal component analysis and partial least square discriminant analysis were applied to screen for differential metabolites,and the metabolites and their pathways associated with URSA were screened using receiver operating characteristic(ROC)curve and pathway enrichment analysis.Results:There were no significant differences in age,body mass index and gestational weeks between URSA and normal control group(all P<0.05).Metabolomics analysis using UPLC-MS showed that a total of 526 metabolites were detected from plasma,of which 33 were found to be differential metabolites associated with URSA based on the screening standards.Six potential metabolites with large area under the curve(AUC)were identified by ROC curve analysis,including phosphatidylethanolamine(AUC=0.972,95%CI:0.920-1.000),santene hydrate(AUC=0.902,95%CI:0.786-0.982),L-leucine(AUC=0.884,95%CI:0.772-0.960),cembrene(AUC=0.881,95%CI:0.758-0.956),caffeine(AUC=0.875,95%CI:0.756-0.962),and 4-hydroxybenzoic acid propyl ester(AUC=0.864,95%CI:0.732-0.946).The AUC for the combined diagnosis of URSA by the six metabolites was 0.983(95%CI:0.929-1.000).Pathway enrichment analysis of the differential metabolites showed that the pathogenesis of URSA was associated with a variety of metabolic pathways including caffeine metabolism,glycerophospholipid metabolism,and unsaturated fatty acid biosynthesis.Conclusion:The plasma metabolic profiles of pregnant women with normal pregnancies versus URSA differ in early pregnancy,and six potential metabolites such as phosphatidylethanolamine,santene hydrate,L-leucine,cembrene,caffeine,4-hydroxybenzoic acid propyl ester,and their metabolic pathways may be involved in the pathogenesis of URSA.