Identifying the Key Mitochondria‐Related Genes in COPD by Integrating Machine Learning and Bioinformatics Analyses

Abstract

Background: Chronic obstructive pulmonary disease (COPD), a prevalent chronic respiratory disorder with high morbidity and mortality, is closely associated with mitochondrial dysfunction and immune dysregulation; however, the underlying mechanisms remain unclear. Aims: The aim of this study is to identify mitochondrial hub genes and evaluate their diagnostic potential in COPD. Methods: This study combined bioinformatics and experimental methods to investigate mitochondria-related differentially expressed genes (MitoDEGs) in COPD pathogenesis. Two GEO datasets (GSE38974/GSE8545) were analyzed to identify MitoDEGs, which were functionally characterized and refined via machine learning (LASSO/SVM-RFE). Key genes were further validated using quantitative real-time polymerase chain reaction (qRT-PCR) and Western blot in bronchial epithelial cells and COPD mouse model lung tissues. Immune infiltration analysis revealed connections between MitoDEGs and immune dysregulation in COPD that were experimentally confirmed using immunohistochemistry (IHC) and immunofluorescence (IF). Results: as core hub genes. Immune profiling revealed significantly increased M0 macrophage infiltration and reduced activated NK cells in COPD. BAX and DLST expression was positively correlated with M0 macrophages but negatively with activated NK cells, a finding corroborated by IHC and IF assays. Conclusions: as potential mitochondrial dysfunction biomarkers in COPD, linking their roles to immune cell infiltration. This study provides novel insights into cigarette smoke-induced COPD pathogenesis and underscores the diagnostic utility of targeting mitochondrial-immune interactions.