Ke Da, He Xian, Li Wenzhe, Wu Hongyan, Sun Yaling, Tan Jie, Wang Ya
Department of Endocrinology, The First Affiliated Hospital of Yangtze University, Jingzhou First People's Hospital, Jingzhou, Hubei, China.
Department of Hubei Provincial Clinical Research Center for Personalized Diagnosis and Treatment of Cancer, The First Affiliated Hospital of Yangtze University, Jingzhou First People's Hospital, Jingzhou, Hubei, China.
Front Endocrinol (Lausanne). 2025 Sep 2;16:1570332. doi: 10.3389/fendo.2025.1570332. eCollection 2025.
Diabetes Mellitus (DM) is a complex metabolic disorder characterized by hyperglycemia, primarily arising from insufficient insulin secretion or the development of insulin resistance. Estrogen plays a significant role in regulating the occurrence and progression of DM. This study aims to investigate the role of estrogen-related genes in diabetes, focusing on identifying potential biomarkers and therapeutic targets for the disease.
We initially obtained gene expression datasets related to type 2 diabetes mellitus (T2DM) from the GEO database. A systematic and coherent series of methodologies was then implemented in a structured manner. First, Principal Component Analysis (PCA) was employed for preliminary data exploration and dimensionality reduction. Next, we identified Differentially Expressed Genes (DEGs). Subsequently, we conducted Weighted Gene Co-expression Network Analysis (WGCNA) to uncover gene modules associated with DM. This was followed by Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses to explore the biological functions and pathways associated with the identified genes. To enhance the precision of biomarker identification, we applied three distinct machine learning algorithms, including Least Absolute Shrinkage and Selection Operator (LASSO), Support Vector Machine-Recursive Feature Elimination (SVM-RFE), and Random Forest (RF), for further refined selection. This comprehensive approach ultimately identified the estrogen-related gene IER3 as a promising biomarker for DM. Furthermore, correlation analyses focusing on immune cell infiltration were conducted to clarify the immunological role of IER3 in DM.
Our findings revealed a significant downregulation of IER3 in DM patients, accompanied by an AUC value of 0.723 in the diagnostic curve ROC, indicating its considerable diagnostic and prognostic potential for DM. Furthermore, the expression levels of IER3 exhibited a strong correlation with variations in the proportions of diverse immune cell types, suggesting that it may play a pivotal role in the immunoregulatory mechanisms underlying DM.
In conclusion, our findings reveal that the estrogen-related gene IER3 is significantly downregulated in patients with DM, highlighting its potential as a diagnostic and prognostic marker for the disease. Therefore, IER3 may serve as a promising biomarker and therapeutic target for DM.
糖尿病（DM）是一种复杂的代谢紊乱疾病，其特征为高血糖，主要源于胰岛素分泌不足或胰岛素抵抗的发展。雌激素在调节糖尿病的发生和发展中起着重要作用。本研究旨在探讨雌激素相关基因在糖尿病中的作用，重点是识别该疾病的潜在生物标志物和治疗靶点。
我们最初从基因表达综合数据库（GEO）获得了与2型糖尿病（T2DM）相关的基因表达数据集。然后以结构化方式实施了一系列系统且连贯的方法。首先，主成分分析（PCA）用于初步数据探索和降维。接下来，我们鉴定了差异表达基因（DEG）。随后，我们进行了加权基因共表达网络分析（WGCNA）以揭示与糖尿病相关的基因模块。接着进行基因本体（GO）和京都基因与基因组百科全书（KEGG）富集分析，以探索与已鉴定基因相关的生物学功能和途径。为提高生物标志物识别的准确性，我们应用了三种不同的机器学习算法，包括最小绝对收缩和选择算子（LASSO）、支持向量机递归特征消除（SVM-RFE）和随机森林（RF）进行进一步精细筛选。这种综合方法最终确定雌激素相关基因IER3是糖尿病的一个有前景的生物标志物。此外，进行了聚焦免疫细胞浸润的相关性分析，以阐明IER3在糖尿病中的免疫作用。
我们的研究结果显示，糖尿病患者中IER3显著下调，诊断曲线ROC中的AUC值为0.723，表明其对糖尿病具有相当大的诊断和预后潜力。此外，IER3的表达水平与多种免疫细胞类型比例的变化呈强相关，表明它可能在糖尿病潜在的免疫调节机制中起关键作用。
总之，我们的研究结果表明，雌激素相关基因IER3在糖尿病患者中显著下调，突出了其作为该疾病诊断和预后标志物的潜力。因此，IER3可能是糖尿病的一个有前景的生物标志物和治疗靶点。
