文献类型：期刊文献

中文题名：基于基因表达数据库的前列腺癌细胞自噬预后分析及C-Met调节机制研究

英文题名：Prospective analysis of autophagy in prostate cancer cells based on gene expression databases and investigation of the C-Met regulatory mechanism

作者：张茹[1,2];谢永强[1,3];赵强[3];柴克强[3];刘昱林[3]

第一作者：张茹

机构：[1]甘肃中医药大学第一临床医学院,兰州730000;[2]甘肃中医药大学第三附属医院皮肤科,甘肃白银730900;[3]甘肃中医药大学第三附属医院泌尿外科,甘肃白银730900

第一机构：甘肃中医药大学临床医学院

年份：2025

卷号：41

期号：10

起止页码：750

中文期刊名：免疫学杂志

外文期刊名：Immunological Journal

基金：白银市科技计划项目(2023-2-16Y)。

语种：中文

中文关键词：前列腺癌;基因表达数据库;自噬;间质表皮转化因子;预后

外文关键词：prostate cancer;gene expression database;autophagy;stromal epidermal transition factor;prognosis

摘要：目的基于基因表达数据库(GEO)探究前列腺癌(PCa)中线粒体自噬相关基因(MRGs)的预后评估价值,揭示其与间质表皮转化因子(C-Met)的调控关系。方法从GEO的GSE153892数据集中获取3个PCa样本的单细胞RNA测序(scRNA-seq)数据,并从GeneCards数据库及既往文献中收集MRGs。使用Seurat软件包对scRNA-seq数据进行处理和分析,包括质量控制、基因表达筛选、细胞类型注释、差异表达基因(DEGs)识别及与MRGs的交集分析。从癌症基因组图谱数据库-PRAD队列中下载PCa及对照样本的转录组数据,并进行差异表达分析和拷贝数变异分析。采用非负矩阵分解算法对PCa样本进行聚类分析,以鉴定不同的PCa亚型。构建基于交集基因的预后风险模型,并通过Kaplan-Meier生存曲线分析和时间依赖性受试者工作特征(ROC)曲线分析模型的预测能力。进行独立预后分析,构建基于风险评分和临床特征的列线图模型,并评估其预测患者生存率的能力。利用单样本基因集富集分析(ssGSEA)算法和TIDE数据库对PCa样本的免疫浸润和肿瘤免疫逃逸可能性进行评估。采用Pearson相关分析交集基因与C-Met表达的关系。结果scRNAseq数据分析鉴定出B细胞、上皮细胞、单核细胞、自然杀伤细胞和T细胞等5种细胞类型,并发现了在不同细胞类型中高表达的交集基因。通过差异表达分析,筛选出与PCa患者预后显著相关的基因,并构建预后风险模型。通过LASSO分析保留ADH5、CAT等6个基因,构建诊断模型并分组,内部测试集两组生存时间差异明显(P<0.05)。ROC曲线评估显示,模型对1、3和5年生存率有较好预测能力,外部测试集验证交集基因表达有统计学差异(P<0.05)。独立预后分析确定T分期和风险评分为独立预后因素,构建列线图模型,校准曲线和ROC曲线分析显示,该模型预测能力优于单纯风险模型。ssGSEA分析显示两组免疫细胞浸润丰度及免疫功能评分差异,多数免疫细胞、免疫功能、风险评分与建模基因相关,高、低风险组TIDE评分及多种免疫检查点存在明显差异(P<0.05)。BCAT2、DCXR、OGT和FUS与C-Met表达呈正相关,ADH5、CAT与C-Met表达呈负相关(P<0.05)。结论基于交集基因的预后风险模型能够有效地预测PCa患者的预后,且风险评分和T分期是PCa的独立预后因素。交集基因与C-Met表达的相关性分析为PCa的靶向治疗提供了新的思路。
Objective To investigate the prognostic value of mitochondrial autophagy-related genes(MRGs)in prostate cancer(PCa),and to reveal their regulatory relationship with interstitial epidermal transforming factor(C-Met)based on the Gene Expression Database(GEO).Methods Single-cell RNA sequencing(scRNA-seq)data of three PCa samples were obtained from the GSE153892 dataset of GEO,and MRGs were collected from the Genecards database and previous literature.The scRNA-seq data were processed and analyzed using the Seurat software package,including quality control,gene expression screening,cell type annotation,differentially expressed genes(DEGs)identification,and intersection analysis with MRGs.The transcriptome data of PCa and control samples were downloaded from the Cancer Genome Atlas Database(TCGA)-PRAD cohort,and differential expression analysis and copy number variation analysis were conducted.The non-negative matrix factorization algorithm is adopted to conduct cluster analysis on PCa samples to identify different PCa subtypes.A prognostic risk model based on intersection genes was constructed,and the predictive ability of the model was analyzed through Kaplan-Meier survival curve analysis and time-dependent receiver operating characteristic(ROC)curve analysis.Conduct independent prognostic analysis,construct a nomogram model based on risk scores and clinical characteristics,and evaluate its ability to predict patient survival rates.The possibility of immune infiltration and tumor immune escape in PCa samples was evaluated by using the single-sample Gene Set Enrichment analysis(ssGSEA)algorithm and the TIDE database.The relationship between intersection genes and C-Met expression was analyzed using Pearson correlation analysis.Results scRNA-seq data analysis identified five cell types including B lymphocytes,epithelial cells,monocytes,natural killer cells and T lymphocytes,and discovered the intersection genes that were highly expressed in different cell types.Through differential expression analysis,genes significantly related to the prognosis of PCa patients were screened out,and a prognostic risk model was constructed.Six genes such as ADH5 and CAT were retained through LASSO analysis.A diagnostic model was constructed and grouped.There was a significant difference in survival time between the two groups in the internal test set(P<0.05).ROC curve evaluation showed that the model had a good predictive ability for 1-,3-,and 5-year survival rates.The external test set verified that there was a statistically significant difference in the expression of intersection genes(P<0.05).Independent prognostic analysis identified T stage and risk score as independent prognostic factors.A nomogram model was constructed.Calibration curve and ROC curve analyses showed that the predictive ability of this model was superior to that of the simple risk model.ssGSEA analysis revealed differences in the abundance of immune cell inflammation and immune function scores between the two groups.Most immune cells,immune function,and risk scores were related to the modeling genes.There were significant differences in TIDE scores and multiple immune checkpoints between the high-risk and low-risk groups(P<0.05).BCAT2,DCXR,OGT and FUS were positively correlated with the expression of C-Met,while ADH5 and CAT were negatively correlated with the expression of C-Met(P<0.05).Conclusion The prognostic risk model based on intersection genes can effectively predict the prognosis of patients with PCa,and the risk score and T stage are independent prognostic factors for PCa.The correlation analysis of intersection genes and C-Met expression provides a new idea for the targeted therapy of PCa.