Gao Kun, Huang Zhenyu, Liao Zhouwei, Wang Yanfei, Chen Dayu
Department of Orthopedics, The Fourth Clinical Medical College of Guangzhou University of Chinese Medicine, Shenzhen, China.
The Fourth Clinical Medical College of Guangzhou University of Chinese Medicine, Shenzhen, China.
J Int Med Res. 2025 Apr;53(4):3000605251333646. doi: 10.1177/03000605251333646. Epub 2025 Apr 27.
BackgroundKnee osteoarthritis is a debilitating disease with a complex pathogenesis. Synovitis, which refers to inflammation of the synovial membrane surrounding the joint, is believed to play an important role in the development and progression of knee osteoarthritis. To better understand the molecular mechanisms underlying knee osteoarthritis, we conducted a comprehensive analysis of gene expression in knee osteoarthritis synovium using machine learning.MethodsDifferentially expressed genes between knee osteoarthritis and control synovial tissues were analyzed using the GSE55235 dataset. We employed several machine learning algorithms, including least absolute shrinkage and selection operator and support vector machine-recursive feature elimination, to screen for key genes. Then, we validated the key genes using an external dataset (GSE51588) and an in vitro knee osteoarthritis animal model. CIBERSORT was used to compare immune cell infiltration levels between knee osteoarthritis and control synovial tissues and determine their relationship with the key genes. Finally, we performed a Connectivity Map analysis to screen for potential small-molecule compounds. Moreover, we conducted single-cell RNA sequencing analysis using knee joint tissues to annotate different subtypes of cells.ResultsA total of 930 differentially expressed genes were identified. Least absolute shrinkage and selection operator regression and support vector machine-recursive feature elimination identified and as key genes. The expression levels of both genes were further validated in the GSE51588 dataset as well as verified through an in vitro experiment involving a knee osteoarthritis mouse model. Multiple significant correlation pairs were found between the immune cell infiltration levels. We unveiled the genetic basis of knee osteoarthritis using genome-wide association study and specific signaling pathways through gene set enrichment analysis. The GeneCards database was used to obtain 3032 pathogenic genes associated with knee osteoarthritis, and we found that expression was significantly negatively correlated and expression was significantly positively correlated with interleukin-1β expression. We predicted several small-molecule compounds based on Connectivity Map analysis. Finally, single-cell RNA sequencing analysis revealed the expression levels of the two key genes in chondrocytes and tissue stem cells.Conclusion and may play key roles in the pathogenesis of knee osteoarthritis, exhibiting correlations with immune cell infiltration levels. These findings indicate that these genes have potential as therapeutic targets. However, further research and validation are necessary to confirm their exact roles and therapeutic potential in knee osteoarthritis.
背景
膝关节骨关节炎是一种致残性疾病，发病机制复杂。滑膜炎是指关节周围滑膜的炎症，被认为在膝关节骨关节炎的发生和发展中起重要作用。为了更好地理解膝关节骨关节炎的分子机制，我们使用机器学习对膝关节骨关节炎滑膜中的基因表达进行了全面分析。
方法
使用GSE55235数据集分析膝关节骨关节炎和对照滑膜组织之间的差异表达基因。我们采用了几种机器学习算法，包括最小绝对收缩和选择算子以及支持向量机递归特征消除，以筛选关键基因。然后，我们使用外部数据集（GSE51588）和体外膝关节骨关节炎动物模型对关键基因进行验证。使用CIBERSORT比较膝关节骨关节炎和对照滑膜组织之间的免疫细胞浸润水平，并确定它们与关键基因的关系。最后，我们进行了连通性图谱分析以筛选潜在的小分子化合物。此外，我们使用膝关节组织进行单细胞RNA测序分析，以注释不同的细胞亚型。
结果
共鉴定出930个差异表达基因。最小绝对收缩和选择算子回归以及支持向量机递归特征消除确定了[具体基因1]和[具体基因2]为关键基因。这两个基因的表达水平在GSE51588数据集中得到进一步验证，并通过涉及膝关节骨关节炎小鼠模型的体外实验得到证实。在免疫细胞浸润水平之间发现了多个显著的相关对。我们通过全基因组关联研究揭示了膝关节骨关节炎的遗传基础，并通过基因集富集分析确定了特定的信号通路。使用GeneCards数据库获得了3032个与膝关节骨关节炎相关的致病基因，我们发现[具体基因1]的表达与白细胞介素-1β的表达显著负相关，[具体基因2]的表达与白细胞介素-1β的表达显著正相关。基于连通性图谱分析，我们预测了几种小分子化合物。最后，单细胞RNA测序分析揭示了这两个关键基因在软骨细胞和组织干细胞中的表达水平。
结论
[具体基因1]和[具体基因2]可能在膝关节骨关节炎的发病机制中起关键作用，与免疫细胞浸润水平相关。这些发现表明这些基因具有作为治疗靶点的潜力。然而，需要进一步的研究和验证来证实它们在膝关节骨关节炎中的确切作用和治疗潜力。
