肝细胞癌m6A相关免疫基因预后模型构建及DDX1在肝细胞癌中的作用

Construction of a prognostic model based on m6A-related immune genes and the role of DDX1 in hepatocellular carcinoma

目的筛选肝细胞癌中N6-甲基腺苷(m6A)相关免疫基因构建预后模型,并探讨其关键基因DDX1在肝癌细胞中的作用。方法本研究选取癌症基因组图谱(TCGA)数据库中368例肝癌样本数据作为训练数据集,基因表达综合数据库(GEO)GSE 14520数据集中242例肝癌样本数据作为验证数据集,并基于表达相关性进行m6A相关的免疫基因筛选。通过LASSO Cox回归模型筛选与预后相关的m6A免疫基因,并构建风险评分(RS)模型。根据RS中位数值,分别将训练集和验证集中所有样本分为高与低风险组,Kaplan-Meier法评估高低风险组的生存预后差异。使用实时荧光定量PCR(RT-qPCR)检测肝癌细胞系(HuH-7、SK-HEP-1、PLC/PRF/5和SNU-449)以及人正常肝细胞系THLE-2中筛选的关键基因表达水平;挑选表达变化较显著的基因,运用功能加强和缺失实验对其功能进行细胞验证。one-way ANOVA法比较多组间统计学差异,其中两两比较采用Tukey法。结果基于共表达分析共获得130个m6A相关免疫基因,最终筛选了5个基因(DDX1、HDAC1、HDAC2、NRAS和PLK1)用于RS模型构建。构建的模型显示,与低风险组比较,高风险组的生存率较低,在TCGA(P<0.001)和GSE 14520数据集(P=0.003)中,差异有统计学意义。与其他基因相比,DDX1在肝癌细胞中高表达。CCK-8实验结果显示,过表达DDX1可促进PLC/PRF/5和SNU-449细胞增殖,差异有统计学意义,P<0.05。Transwell结果显示,与阴性对照组相比,DDX1过表达可促进PLC/PRF/5和SNU-449细胞的迁移(105.00±12.53 vs 173.67±12.22;100.00±10.15 vs 147.33±12.10)和侵袭(83.00±10.81 vs 131.00±13.23;85.33±10.01 vs 135.00±12.49)。DDX1过表达可升高细胞丙二醛(1.01±0.10 vs 1.73±0.13)和活性氧(17.54±1.30 vs 25.97±2.24)水平,但降低超氧化物歧化酶(1.03±0.12 vs 0.52±0.12)的水平。DDX1敲除得到了相反的结果。结论本研究基于5个m6A相关免疫基因构建的预后模型可以有效预测肝癌患者的预后以及对免疫治疗的反应。其中DDX1可促进肝癌进展。

Objective To screen m6A-related immune genes in liver cancer to construct a prognostic model,and to investigate the role of the key gene DDX1 in hepatocellular carcinoma cells.Methods This study used the data from the cancer genome atlas(TCGA)database containing 368 cancer samples as a training dataset,and the data from the gene expression comprehensive database(GEO)GSE14520(accession number)containing 242 cancer samples as a validation dataset.Based on the expression correlation,m6A related immune gene screening was conducted,and a risk score(RS)model for evaluating the prognosis of model genes was established.According to RS median,samples in training dataset and validation dataset were divided into high and low risk groups.Kaplan-Meier method was used to compare the difference of survival rate between the two groups.qRT-PCR was used to detect the expression level of key genes in cancer cell lines(HuH-7,SK-HEP-1,PLC/PRF/5 and SNU-449)and normal liver cells THLE-2.We performed gain-function and loss-function experiments to explore the role the candidate gene with most prominently changed expression.one-way ANOVA was used to compare multi-group differences,followed by Tukey method.Results Based on co expression analysis,a total of 130m6Arelated immune genes were obtained,of which five(DDX1,HDAC1,HDAC2,NRASand PLK1)were ultimately selected for RS model construction.The survival rate of patients in the high-risk group was worse than that in the low-risk group based on TCGA(P<0.001)and GSE14520(P=0.003)datasets.CCK-8assay showed that overexpression of DDX1promoted cell proliferation in PLC/PRF/5and SNU-449cells.Compared with negative control group,DDX1overexpression enhanced migration(105.00±12.53 vs 173.67±12.22;100.00±10.15 vs 147.33±12.10)and invasion(83.00±10.81 vs 131.00±13.23;85.33±10.01 vs 135.00±12.49)of PLC/PRF/5and SNU-449cells.DDX1overexpression increased malondialdehyde(1.01±0.10 vs 1.73±0.13)and ROS(17.57±1.30 vs 25.97±2.24)levels,but decreased SOD level(1.03±0.12 vs 0.52±0.12).Conclusion The prognostic model based on five m6Arelated immune genes can effectively predict the prognosis in patients with hepatocellular carcinoma cancer.DDX1can promote the progress of hepatocellular carcinoma.