Zhou Yanzhen, Li Jingmin, Chen Meihuan, Huang Hailong
College of Clinical Medicine for Obstetrics & Gynecology and Pediatrics, Fujian Medical University, Fuzhou, Fujian, China.
Fujian Maternity and Child Health Hospital, College of Clinical Medicine for Obstetrics & Gynecology and Pediatrics, Fujian Medical University, Fuzhou, Fujian, China.
Front Endocrinol (Lausanne). 2025 May 8;16:1545670. doi: 10.3389/fendo.2025.1545670. eCollection 2025.
Endometriosis is characterized by immune evasion and progressive inflammation. This study aimed to identify key genes related to immune and inflammation in endometriosis.
Differentially expressed genes between patients with and without endometriosis were identified from the GEO database. Furthermore, immune- and inflammation-related genes (IRGs) were identified by intersecting the differentially expressed genes with known immune and inflammatory genes. Functional analyses of the GO and KEGG pathways of these genes were performed. Subsequently, three machine learning models-LASSO regression, SVM-RFE, and Boruta-were conducted to identify the potential key genes in endometriosis. Finally, the expressions of key genes in endometriosis were verified in two validation cohorts using an online database and qRT-PCR, and their immunoregulatory properties were verified.
A total of 13 differentially expressed IRGs were identified. Using machine learning algorithms, five key genes were selected in the endometriosis: BST2, IL4R, INHBA, PTGER2, and MET. Furthermore, the three hub genes exhibited consistent trends across both training and validation datasets. The three keys also correlated with infiltrated immune cells, checkpoint genes, and immune factors in various degrees. Finally, validation analysis using the online database and qRT-PCR confirmed that MET expression aligned with outcomes from both training and validation datasets.
Three immune- and inflammation-related genes were identified as potential biomarkers of endometriosis, providing new insights into the molecular mechanisms underlying immune function in endometriosis. The immune-related function of MET, particularly its correlation with NK cell activity in endometriosis, will be the focus of future studies.
子宫内膜异位症的特征是免疫逃逸和进行性炎症。本研究旨在确定与子宫内膜异位症免疫和炎症相关的关键基因。
从基因表达综合数据库（GEO数据库）中识别子宫内膜异位症患者与非患者之间的差异表达基因。此外，通过将差异表达基因与已知的免疫和炎症基因进行交叉分析来确定免疫和炎症相关基因（IRGs）。对这些基因进行基因本体论（GO）和京都基因与基因组百科全书（KEGG）通路的功能分析。随后，采用三种机器学习模型——套索回归、支持向量机递归特征消除法（SVM-RFE）和Boruta算法，来识别子宫内膜异位症中的潜在关键基因。最后，使用在线数据库和定量逆转录聚合酶链反应（qRT-PCR）在两个验证队列中验证子宫内膜异位症中关键基因的表达，并验证其免疫调节特性。
共识别出13个差异表达的IRGs。使用机器学习算法，在子宫内膜异位症中选择了5个关键基因：骨髓基质细胞抗原2（BST2）、白细胞介素4受体（IL4R）、抑制素βA（INHBA）、前列腺素E受体2（PTGER2）和肝细胞生长因子受体（MET）。此外，这三个枢纽基因在训练集和验证数据集中均呈现出一致的趋势。这三个关键基因还在不同程度上与浸润免疫细胞、检查点基因和免疫因子相关。最后，使用在线数据库和qRT-PCR进行的验证分析证实，MET的表达与训练集和验证数据集的结果一致。
确定了三个免疫和炎症相关基因作为子宫内膜异位症的潜在生物标志物，为子宫内膜异位症免疫功能的分子机制提供了新见解。MET的免疫相关功能，特别是其与子宫内膜异位症中自然杀伤细胞活性的相关性，将是未来研究的重点。
