Zhang Yuanyuan, Wang Yu, Wang Shan, Zhang Huiping
The Department of Obstetrics and Gynecology, Affiliated Hospital of Yan'an University, Yan'an, Shaanxi, 716000, China.
Department of Obstetrics and Gynecology, Northwest University First Hospital, Xi'an, Shaanxi, 710043, China.
Reprod Sci. 2025 Aug 14. doi: 10.1007/s43032-025-01953-0.
Polycystic ovary syndrome (PCOS) is a common endocrine disorder in women. In recent years, endoplasmic reticulum (ER) stress has gained increasing attention in the pathogenesis of PCOS. This study aims to explore the potential role of ER stress in PCOS by constructing a predictive model based on ER stress-related genes, and further evaluate the characteristics of immune infiltration and screen potential drugs. Five algorithms, including Lasso, Support Vector Machine (SVM), Random Forest (RF), Gradient Boosting Algorithm (XGB), and Generalized Linear Model (GLM), were used to screen key genes associated with PCOS and endoplasmic reticulum (ER) stress. A predictive model was constructed to analyze its diagnostic value in PCOS. External validation of the model was conducted using different datasets to assess its predictive accuracy. Furthermore, immune infiltration analysis was performed to explore the relationship between ER stress-related genes and the immune microenvironment in PCOS, revealing their potential role in disease development through immune response regulation. Finally, molecular docking and drug screening platforms were utilized to identify potential drugs that can modulate the ER stress pathway, providing new drug targets for the clinical treatment of PCOS. Two downregulated genes, NQO1 and NPY, and three upregulated genes, TFEB, JUP, and ATF4, were identified in PCOS cases. The constructed nomogram model demonstrated that the area under the ROC curve for NQO1, TFEB, JUP, NPY, and ATF4 in the validation set were 0.629, 0.600, 0.629, 0.543, and 0.743, respectively, indicating that the PCOS diagnostic model built from these five hub genes has good reliability. Immune infiltration analysis revealed that the expression of the JUP gene was positively correlated with T lymphocyte infiltration, while the expression of TFEB and NPY was negatively correlated with T lymphocyte infiltration, suggesting their potential involvement in immune regulation in PCOS. Through molecular docking and drug screening, 66 potential drugs were identified, 18 of which are already approved for use, providing options for pharmacological treatment of PCOS. The results of this study suggest that endoplasmic reticulum (ER) stress-related genes play an important role in the pathogenesis and development of PCOS, and that accurate predictive models may provide new insights for early diagnosis of the disease. Immune infiltration analysis revealed the potential mechanisms of immune cell involvement in PCOS, while drug screening provides a theoretical basis for future targeted therapies for PCOS.
多囊卵巢综合征（PCOS）是女性常见的内分泌紊乱疾病。近年来，内质网（ER）应激在PCOS发病机制中的作用日益受到关注。本研究旨在通过构建基于内质网应激相关基因的预测模型，探讨内质网应激在PCOS中的潜在作用，并进一步评估免疫浸润特征及筛选潜在药物。采用包括套索回归（Lasso）、支持向量机（SVM）、随机森林（RF）、梯度提升算法（XGB）和广义线性模型（GLM）在内的五种算法，筛选与PCOS和内质网（ER）应激相关的关键基因。构建预测模型并分析其对PCOS的诊断价值。使用不同数据集对模型进行外部验证，以评估其预测准确性。此外，进行免疫浸润分析，以探讨内质网应激相关基因与PCOS免疫微环境之间的关系，揭示它们通过免疫反应调节在疾病发展中的潜在作用。最后，利用分子对接和药物筛选平台，识别可调节内质网应激途径的潜在药物，为PCOS的临床治疗提供新的药物靶点。在PCOS病例中鉴定出两个下调基因NQO1和NPY，以及三个上调基因TFEB、JUP和ATF4。构建的列线图模型显示，验证集中NQO1、TFEB、JUP、NPY和ATF4的ROC曲线下面积分别为0.629、0.600、0.629、0.543和0.743，表明由这五个核心基因构建的PCOS诊断模型具有良好的可靠性。免疫浸润分析显示，JUP基因的表达与T淋巴细胞浸润呈正相关，而TFEB和NPY的表达与T淋巴细胞浸润呈负相关，提示它们可能参与PCOS的免疫调节。通过分子对接和药物筛选，鉴定出66种潜在药物，其中18种已获批准使用，为PCOS的药物治疗提供了选择。本研究结果表明，内质网（ER）应激相关基因在PCOS的发病机制和发展中起重要作用，准确的预测模型可能为该疾病的早期诊断提供新的见解。免疫浸润分析揭示了免疫细胞参与PCOS的潜在机制，而药物筛选为未来PCOS的靶向治疗提供了理论依据。
