中国组织工程研究

中国组织工程研究 ›› 2025, Vol. 29 ›› Issue (26): 5632-5641.doi: 10.12307/2025.628

• 组织构建相关数据分析 Date analysis of organization construction • 上一篇    下一篇

骨关节炎内质网应激关键基因的生物信息学筛选及实验验证

郝茂辰1，马  超2，刘  凯1，柳可心1，孟令婷1，王杏如1，王建忠2   

  1. 1内蒙古医科大学，内蒙古自治区呼和浩特市  010000；2内蒙古医科大学第二附属医院，内蒙古自治区呼和浩特市  010000
  • 收稿日期:2024-06-04 接受日期:2024-07-05 出版日期:2025-09-18 发布日期:2025-02-27
  • 通讯作者: 并列第一作者：马超，内蒙古医科大学第二附属医院，内蒙古自治区呼和浩特市 010000 通讯作者：王建忠，博士，主任医师，教授，内蒙古医科大学第二附属医院创伤外科副主任，内蒙古医科大学第二附属医院，内蒙古自治区呼和浩特市 010000
  • 作者简介:郝茂辰，男，1998年生，内蒙古自治区鄂尔多斯市人，内蒙古医科大学在读硕士，主要从事骨与关节疾病研究。
  • 基金资助:
    内蒙古自治区第三批首府地区公立医院高水平临床专科建设科技项目(2023SGGZ143)，项目负责人：王建忠；内蒙古医科大学重点项目(YKD2024ZD005)，项目负责人：王建忠；内蒙古自治区直属高校基本科研业务费项目(YKD2023ZY001)，项目负责人：刘凯；内蒙古自治区研究生科研创新项目(S20231189Z)，项目负责人：刘凯

Bioinformatics screening of key genes for endoplasmic reticulum stress in osteoarthritis and experimental validation

Hao Maochen1, Ma Chao2, Liu Kai1, Liu Kexin1, Meng Lingting1, Wang Xingru1, Wang Jianzhong2    

  1. 1Inner Mongolia Medical University, Hohhot 010000, Inner Mongolia Autonomous Region, China; 2The Second Affiliated Hospital of Inner Mongolia Medical University, Hohhot 010000, Inner Mongolia Autonomous Region, China
  • Received:2024-06-04 Accepted:2024-07-05 Online:2025-09-18 Published:2025-02-27
  • Contact: Ma Chao, The Second Affiliated Hospital of Inner Mongolia Medical University, Hohhot 010000, Inner Mongolia Autonomous Region, China Hao Maochen and Ma Chao contributed equally to this work. Corresponding author: Wang Jianzhong, PhD, Chief physician, Professor, The Second Affiliated Hospital of Inner Mongolia Medical University, Hohhot 010000, Inner Mongolia Autonomous Region, China
  • About author:Hao Maochen, Master’s candidate, Inner Mongolia Medical University, Hohhot 010000, Inner Mongolia Autonomous Region, China
  • Supported by:
    the Third-Batch Science and Technology Project of High-level Clinical Specialty Construction of Capital Region Public Hospitals in Inner Mongolia Autonomous Region, No. 2023SGGZ143 (to WJZ); Key Project of Inner Mongolia Medical University, No. YKD2024ZD005 (to WJZ); Basic Scientific Research Operational Fees Project of Colleges and Universities under the Direct Subsidiaries of the Inner Mongolia Autonomous Region, No. YKD2023ZY001 (to LK); Graduate Student Scientific Research Innovation Project in the Inner Mongolia Autonomous Region, No. S20231189Z (to LK)

PDF

267

摘要：


文题释义：
内质网应激：是细胞在应对内质网腔内错误折叠与未折叠蛋白聚集以及钙离子平衡紊乱等状况时，所激活的一系列保护性应激反应。
骨关节炎：是一种临床上最常见的以关节软骨退变和破坏为特征的慢性疾病，是导致关节疼痛、关节功能障碍和畸形以及肢体残疾的主要原因。 

背景：内质网应激和骨关节炎的发生、发展密切相关，但其关键基因及调控机制尚不明确。
目的：基于生物信息学筛选骨关节炎内质网应激关键基因，并加以细胞模型实验验证，以期从内质网应激角度为防治骨关节炎提供新策略。
方法：从GEO数据库下载骨关节炎相关数据集GSE55235，运用生物信息学分析方法筛选获取骨关节炎滑膜差异基因，与GeneCard数据库中内质网应激基因取交集得到骨关节炎内质网应激差异基因，随后对其进行GO/KEGG富集分析；构建蛋白-蛋白互作网络；在外部数据集验证其疾病预测效能。构建人原代关节滑膜成纤维细胞骨关节炎模型，对照组不作处理，实验组接受20 ng/mL脂多糖模拟骨关节炎滑膜细胞造模后进行实时荧光定量PCR实验验证各差异基因表达水平并进行免疫浸润分析。
结果与结论：①共获得骨关节炎内质网应激关键基因27个。②GO富集分析结果显示其主要富集在胶原蛋白代谢过程、趋化因子、抗原结合及免疫球蛋白受体结合等过程中。③KEGG分析表明其主要富集在类风湿性关节炎、松弛肽信号通路等通路中。④构建蛋白-蛋白互作网络，使用Cytoscape软件中Degree算法筛选得分最高的5个基因，包括基质金属蛋白酶1、肿瘤坏死因子配体超家族成员11(tumor necrosis factor ligand superfamily member 11，TNFSF11)、基质金属蛋白酶9、Ⅰ型胶原α1重组蛋白(collagen type I alpha 1，COL1A1)和趋化因子配体12(chemokine C-X-C motif ligand 12，CXCL12)。⑤免疫浸润分析结果显示，免疫细胞主要分布于M2型巨噬细胞，CXCL12与静息的肥大细胞呈显著正相关(r=0.70，P < 0.001)，与静息的记忆CD4+ T细胞呈显著负相关(r=-0.72，P < 0.001)；基质金属蛋白酶9与M0巨噬细胞呈显著正相关(r=0.94，P < 0.001)；COL1A1与静息的NK细胞(r=0.77，P < 0.001)和M0巨噬细胞(r=0.76，P < 0.001)呈显著正相关。⑥外部数据集GSE77298和GSE1919中进行受试者工作曲线分析表明5个关键基因具有较好的疾病预测价值。⑦体外细胞实验证明，骨关节炎细胞模型组中基质金属蛋白酶1，TNFSF11，CXCL12及基质金属蛋白酶9表达水平相比于对照组有显著差异。⑧结果显示，骨关节炎内质网应激关键基因基质金属蛋白酶1，TNFSF11，CXCL12及基质金属蛋白酶9通过胶原蛋白降解和免疫调控等环节影响着骨关节炎的发生发展，有望为骨关节炎的靶向治疗提供新见解。
https://orcid.org/0009-0007-6400-8668(郝茂辰)

中国组织工程研究杂志出版内容重点：组织构建；骨细胞；软骨细胞；细胞培养；成纤维细胞；血管内皮细胞；骨质疏松；组织工程

关键词: 骨关节炎, 内质网应激, WGCNA, 机器学习, 人滑膜细胞, 免疫浸润分析, 生物标志物, 实验验证, 细胞实验, 差异基因

Abstract: BACKGROUND: Endoplasmic reticulum stress is closely associated with the occurrence and progression of osteoarthritis, but the key genes and regulatory mechanisms remain unclear. 
OBJECTIVE: Utilizing bioinformatics to identify crucial endoplasmic reticulum stress-related genes in osteoarthritis, followed by experimental validation in cell models, aiming to offer new strategies for the prevention and treatment of osteoarthritis from the perspective of endoplasmic reticulum stress.
METHODS: Osteoarthritis-related dataset GSE55235 was downloaded from the GEO database. Differential genes in synovial tissue of osteoarthritis were obtained through WGCNA machine learning algorithm and intersected with endoplasmic reticulum stress-related genes from the GeneCard database to acquire differential endoplasmic reticulum stress-related genes in osteoarthritis (ERSDEGs). These genes underwent GO and KEGG enrichment analysis, construction of a protein-protein interaction network, and validation of diagnostic efficiency in external datasets. Human primary synovioblast model of osteoarthritis was constructed. The control group was not treated, and the experimental group received 20 ng/mL lipopolysaccharide to simulate osteoarthritic synoviocyte modeling. Real-time fluorescence quantitative PCR was then performed to validate the expression level of each differential gene followed by immune infiltration analysis.
RESULTS AND CONCLUSION: A total of 27 key endoplasmic reticulum stress-related genes in osteoarthritis were identified. GO enrichment analysis revealed that these genes were mainly enriched in collagen metabolism, chemokine, antigen binding, and immunoglobulin receptor binding processes. KEGG analysis indicated that they were mainly enriched in pathways such as rheumatoid arthritis and relaxin signaling pathways. The protein-protein interaction network was constructed, and the top five genes with the highest scores were identified using the Degree algorithm in Cytoscape software, including matrix metallopeptidase 1, tumor necrosis factor ligand superfamily member 11, matrix metallopeptidase 9, collagen type I alpha 1, and chemokine C-X-C motif ligand 12. Immune infiltration analysis showed that immune cells were mainly distributed in M2 macrophages, chemokine C-X-C motif ligand 12 showed a significant positive correlation with resting mast cells (r=0.70, P < 0.001) and a significant negative correlation with resting memory CD4+ T cells (r=-0.72, P < 0.001). Matrix metallopeptidase 9 showed a significant positive correlation with M0 macrophages (r=0.94, P < 0.001). Collagen type I alpha 1 was significantly positively correlated with resting NK cells (r=0.77, P < 0.001) and M0 macrophages (r=0.76, P < 0.001). Receiver operator characteristic curve analysis in external datasets GSE77298 and GSE1919 showed that the five key genes had good disease prediction value. In vitro cell experiments demonstrated significant differences in the expression levels of matrix metallopeptidase 1, tumor necrosis factor ligand superfamily member 11, matrix metallopeptidase 9, and chemokine C-X-C motif ligand 12 in the osteoarthritic cell model compared to the control group. These results showed that the key genes related to endoplasmic reticulum stress in osteoarthritis, including matrix metallopeptidase 1, tumor necrosis factor ligand superfamily member 11, matrix metallopeptidase 9, and chemokine C-X-C motif ligand 12, influence the occurrence and development of osteoarthritis through the links of collagen degradation and immune regulation, which are expected to provide new insights into the targeted treatment of osteoarthritis.

Key words: osteoarthritis, endoplasmic reticulum, WGCNA, machine learning, human synoviocytes, immune infiltration analysis, biomarkers, experimental validation, cell experiment, differential genes

中图分类号: