氧化应激相关基因预测急性ST段抬高型心肌梗死患者心源性死亡

Oxidative stress-related genes predict cardiogenic death in patients with acute ST elevation myocardial infarction

目的对接受直接经皮冠状动脉介入治疗的急性ST段抬高型心肌梗死(STEMI)患者构建基于差异表达的氧化应激相关基因(DEOSRGs)的生存预测模型。方法从基因表达综合(GEO)数据库检索并筛选符合选择、排除标准的外周血mRNA表达谱数据集。使用关键词“急性心肌梗死”和“急性冠状动脉综合征”,并限定物种为“人”。使用GEO2R工具鉴定DEOSRGs;使用最小绝对收缩和选择算子惩罚的Cox比例风险回归建立最优模型;使用风险因子分布图、KaplanMeier生存分析和受试者工作特性曲线评估其临床预测价值。结果选择样本量较小的数据集GSE60993和GSE61144数据集作为训练集,具有完整预后信息的GSE49925数据集用于模型构建和预测价值的验证。通过与氧化应激相关基因集取交集,筛选出10个DEOSRGs,包括9个上调和1个下调的基因。上调基因分别为基质金属蛋白酶9基因(MMP9)、精氨酸酶1基因(ARG1)、FBJ鼠骨肉瘤病毒癌基因同源物基因(FOS)、血管非炎性分子1基因(VNN1)、Toll样受体4基因(TLR4)、白细胞介素18受体辅助蛋白基因(IL18RAP)、中性粒细胞胞质因子1基因(NCF1);花生四烯酸5-脂氧合酶基因(ALOX5)、髓系细胞白血病1基因(MCL1)。下调基因为解偶联蛋白2基因(UCP2)。MMP9和UCP2是STEMI患者早期阶段最高上调和下调的DEOSRGs。构建的预后模型(风险得分)为(年龄×0.31)+(Gensini评分×0.02)+(血清肌酐×1.1)+(MMP9表达水平×2.52)+[ARG1表达水平×(-0.22)]+[IL18RAP表达水平×(-0.51)]-44.3。风险因子分布图、Kaplan-Meier生存曲线分析和受试者工作特性曲线分析显示该模型可有效预测STEMI患者发病2年内的心源性死亡事件。曲线下面积(AUC)分别为0.846、0.816。结论实验研究对STEMI患者构建了基于DEOSRGs的预后预测模型,可对发病2年内的心源性死亡风险进行有效分层。

Objective To construct survival prediction model based on differentially expressed oxidative stress-related genes(DEOSRGs)for patients with acute ST-segment elevation myocardial infarction(STEMI)perform direct percutaneous coronary intervention.Methods Peripheral blood mRNA expression profile datasets met the criteria were retrieved and screened from Gene Expression Omnibus(GEO)database.The keywords“acute myocardial infarction”and“acute coronary syndrome”were used,and the species was defined as“human”.The DEOSRGs was identified with GEO2R tool;the optimal model was established by Cox proportional hazards regression with minimum absolute shrinkage and selection operator penalty;risk factor distribution,Kaplan-Meier survival analysis and receiver operating characteristic curve were used to evaluate its clinical predictive value.Results The smaller sample size data sets GSE60993 and GSE61144 were selected as training set,and GSE49925 data set with complete prognostic information was used for model construction and predictive value verification.By intersecting with oxidative stress-related gene set,10 DEOSRGs were screened,which included 9 up-regulated and 1 down-regulated genes.The up-regulated genes were matrix metallopeptidase 9 gene(MMP9),arginase 1 gene(ARG1),FBJ murine osteosarcoma viral oncogene homolog gene(FOS),vanin 1 gene(VNN1),Toll-like receptor 4 gene(TLR4),interleukin 18 receptor accessory protein gene(IL18RAP),neutrophil cytosolic factor 1 gene(NCF1),Arachidonate 5-lipoxygenase gene(ALOX5)and myeloid cell leukemia 1 gene(MCL1).The down-regulated gene was uncoupling protein 2 gene(UCP2).MMP9 and UCP2 were the highest upregulated and down-regulated DEOSRGs in the early stage of STEMI patients.The constructed prognostic model(risk score)was(age×0.31)+(Gensini score×0.02)+(serum creatinine×1.1)+(MMP9 expression level×2.52)+[ARG1 expression level×(-0.22)]+[IL18RAP expression level×(-0.51)]-44.3.The risk factor scattergram,Kaplan-Meier survival curve and receiver operating characteristic curve analysis results showed that the model was effective in predicting cardiac death events in STEMI patients within 2 years of onset.The area under the curve(AUC)were 0.846 and 0.816,respectively.Conclusion It is demonstrated that prognostic prediction model based on DEOSRGs was constructed for STEMI patients,which could effectively stratify the risk of cardiac death within 2 years of onset.