【作者】 彭靖；

【导师】 王锷；

【作者基本信息】 中南大学 ， 麻醉学， 2024， 硕士

【摘要】 目的:心力衰竭是一种以心脏泵血功能障碍为核心的临床综合征,其病因多样且病程复杂,对临床治疗构成了极大的挑战。本研究的目标是系统性收集并分析公共数据库中不同心力衰竭类型患者心脏组织的转录组学数据,以揭示其免疫表型、细胞浸润特征及分子生物标志物的异同。本研究旨在通过对不同种类心力衰竭间共同生物学路径的深入分析,开发一种基于这些路径的高灵敏度诊断模型,并探讨该模型在心力衰竭进展过程中的潜在作用,以及涉及的相关基因的具体生物学功能。最后,本研究将在心力衰竭动物模型进行模型有效性的实验验证。方法:在此研究中,我们使用了基因表达综合数据库(Gene Expression Omnibus,GEO)的数据集进行分析,旨在识别心力衰竭相关的特异性基因表达模式。训练数据集包括多种表现心衰的心肌病类型及健康对照的心脏组织中的基因表达信息,具体为:扩张型心肌病(Dilated Cardiomyopathy,DCM)179个,肥厚型心肌病(Hypertrophic Cardiomyopathy,HCM)111个,缺血性心肌病(Ischemic Cardiomyopathy,ICM)223个,病毒性心肌病(Viral Cardiomyopathy,VCM)12个,以及正常对照(NC)202个样本。测试集由166个DCM、28个HCM和166个NC样本组成。通过选取差异表达基因并利用基因本体(Gene Ontology,GO)和Reactome数据库对这些差异性基因进行富集分析,从而揭示心力衰竭中的共同生物学通路。为了精细化诊断指标,本研究采用了包括随机森林、支持向量机(Support Vector Machine-Recursive Feature Elimination,SVM-RFE)和最小绝对收缩与选择算子(Least Absolute Shrinkage and Selection Operator,LASSO)回归模型在内的多种机器学习技术,对上述生物学通路进行了深入分析,从而识别出关键的诊断性基因。通过应用x Cell软件处理上述数据集,本研究确立了心力衰竭样本中的细胞浸润模式,并对细胞分布的差异进行了分析。我们采用箱线图和热图对不同分组样本中的细胞分布模式进行了可视化,并通过单个样本基因集富集分析(Single-sample Gene Set Enrichment Analysis,ss GSEA)及相关性分析,进一步阐明了所鉴定生物标志物的在心衰进展中的作用机制。最后,通过实施横向主动脉缩窄(Transverse Aortic Constriction,TAC)手术在大鼠中建立了左心室心力衰竭模型。利用M型超声技术在左心室短轴切面测定了左心室射血分数和左心室舒张末期内径等参数,以评估心力衰竭的严重程度。随后,通过q PCR对前期鉴定的生物标志物进行了表达水平的检测,以验证诊断模型的可靠性。结果:在本项研究中,我们鉴定出142个在心力衰竭共同上调的差异表达基因(Differentially Expressed Genes,DEGs)与181个共同下调的差异表达基因,通过富集分析深入揭示了在各类心肌病心力衰竭中肥大细胞激活的关键作用。利用LASSO回归以及SVM-RFE方法,从肥大细胞激活相关的基因中选择出5个关键生物标志物(FCER1A、KIT、NDRG1、FGR和GATA2),这些标志物被证实在心力衰竭的训练及验证数据集中具有显著的诊断能力。特别是,FCER1A基因的表达差异在所有鉴定的基因中最为显著。在细胞浸润特性分析中,ICM和DCM心力衰竭的组织样本显示,巨噬细胞相对丰度减少,同时成纤维细胞、肥大细胞和CD8~+T细胞的相对丰度则增加。值得注意的是,基因FCER1A的表达与心脏组织中巨噬细胞的相对丰度呈现负相关,而与成纤维细胞相对丰度表现为正相关。结论:本研究利用生物信息学技术探讨肥大细胞在心力衰竭中的作用,并鉴定了FCER1A、KIT、NDRG1、FGR和GATA2作为诊断标志物。通过分析发现,FCER1A的表达量与巨噬细胞相对丰度负相关,与成纤维细胞相对丰度正相关。这些结果为理解心力衰竭中肥大细胞的作用提供了新依据,提示了未来研究的方向。图14幅,表7个,参考文献61篇更多还原

【Abstract】 Objective:Heart failure is a clinical syndrome centered on cardiac pump dysfunction,characterized by diverse etiologies and complex disease progression,posing significant challenges to clinical treatment.This study aims to systematically collect and analyze transcriptomic data from cardiac tissues of patients with different types of heart failure from public databases to uncover their immunophenotypes,cellular infiltration characteristics,and molecular biomarkers.By conducting a thorough analysis of common biological pathways among different types of heart failure,this study seeks to develop a high-sensitivity diagnostic model based on these pathways,explore its potential role in the progression of heart failure,and investigate the specific biological functions of related genes.Finally,the study will validate the model’s effectiveness in a heart failure animal model.Methods:In this study,we analyzed datasets from the Gene Expression Omnibus(GEO)to identify heart failure-specific gene expression patterns.The training dataset included gene expression information from various cardiomyopathy types manifesting heart failure and healthy controls,specifically:179 dilated cardiomyopathy(DCM),111hypertrophic cardiomyopathy(HCM),223 ischemic cardiomyopathy(ICM),12 viral cardiomyopathy(VCM),and 202 normal controls(NC).The test set consisted of 166 DCM,28 HCM,and 166 NC samples.Differential gene expression analysis followed by enrichment using the Gene Ontology(GO)and Reactome databases revealed common biological pathways in heart failure.Diagnostic markers were refined using multiple machine learning techniques including random forest,Support Vector Machine-Recursive Feature Elimination(SVM-RFE),and Least Absolute Shrinkage and Selection Operator(LASSO)regression models to analyze these pathways and identify key diagnostic genes.The x Cell software was utilized to establish cellular infiltration patterns in heart failure samples and analyze differences in cell distribution,which were visualized through box plots and heat maps.Single-sample Gene Set Enrichment Analysis(ss GSEA)and correlation analysis further elucidated the mechanisms of identified biomarkers during heart failure progression.Lastly,a left ventricular heart failure model was established in rats through Transverse Aortic Constriction(TAC)surgery.M-mode echocardiography measured left ventricular ejection fraction and end-diastolic diameter to assess the severity of heart failure.Subsequently,q PCR was performed to test the expression levels of previously identified biomarkers,validating the reliability of the diagnostic model.Results:In this study,we identified 142 differentially expressed genes(DEGs)that were commonly upregulated in heart failure,and 181differentially expressed genes that are commonly downregulated.Through enrichment analysis,we further elucidated the crucial role of mast cell activation across various types of cardiomyopathy-induced heart failure.Utilizing LASSO regression and SVM-RFE methods,we selected five key biomarkers(FCER1A,KIT,NDRG1,FGR,and GATA2)related to mast cell activation,which demonstrated significant diagnostic capabilities in training and validation datasets for heart failure.Notably,the expression of the gene FCER1A was the most significantly altered among all identified genes.In the analysis of cellular infiltration characteristics,tissue samples from ischemic(ICM)and dilated cardiomyopathy(DCM)heart failures showed a decreased relative abundance of macrophages,while an increased relative abundance of fibroblasts,mast cells,and CD8~+T cells was observed.Importantly,the expression of FCER1A showed a negative correlation with the relative abundance of macrophages and a positive correlation with the relative abundance of fibroblasts in cardiac tissues.Conclusion:This study employed bioinformatics techniques to explore the role of mast cells in heart failure and identified FCER1A,KIT,NDRG1,FGR,and GATA2 as diagnostic biomarkers.Our analysis revealed that the expression of FCER1A was negatively correlated with the relative abundance of macrophages and positively correlated with the relative abundance of fibroblasts.These findings provide new insights into the role of mast cells in heart failure and suggest directions for future research.更多还原