通过单细胞RNA测序和机器学习对炎症微环境中滑膜和髌下脂肪垫来源的骨关节炎巨噬细胞的分子特征及诊断建模

Molecular features and diagnostic modeling of synovium- and IPFP-derived OA macrophages in the inflammatory microenvironment via scRNA-seq and machine learning.

作者信息

Lin Chao, Wan Yue, Xu Yong, Zou Qingsong, Li Xiaoxiao

机构信息

Wuxi School of Medicine, Jiangnan University, Wuxi, 214000, Jiangsu, China.

Department of Orthopaedics, Laboratory of Key Technology and Materials in Minimally Invasive Spine Surgery, Center for Spinal Minimally Invasive Research, Tongren Hospital, Hongqiao International Institute of Medicine, Shanghai Jiao Tong University School of Medicine, 1111 XianXia Road, Shanghai, 200336, China.

出版信息

J Orthop Surg Res. 2025 Apr 17;20(1):382. doi: 10.1186/s13018-025-05793-1.

DOI:10.1186/s13018-025-05793-1

PMID:40247403

原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC12004787/

Abstract

BACKGROUND

Osteoarthritis (OA) is the leading cause of degenerative joint disease, with total joint replacement as the only definitive cure. However, no disease-modifying therapy is currently available. Inflammation and fibrosis in the infrapatellar fat pad (IPFP) contribute to OA onset and progression. However, the cellular composition and molecular mechanisms in the IPFP microenvironment remain unclear. This study investigates the functions of OA-macrophages and their clinical significance.

METHODS

We analyzed single-cell RNA sequencing (scRNA-seq) data from normal and OA patients. Enrichment analysis revealed differences in biological pathways across cell types. Pseudotime and cell-cell communication analyses revealed the developmental trajectory and interactions of OA-macrophages with other cell types. Machine learning (ML) algorithms identified feature genes of OA-macrophages. An OAMGS diagnostic score was developed, and CIBERSORT was used to analyze immune infiltration and its association with immune cells. Rat OA and normal models were established, and feature gene expression was validated using immunofluorescence (IF) staining and quantitative reverse transcription PCR (RT-qPCR).

RESULTS

OA-macrophages play a central role in inflammation and fibrosis, enhancing leukocyte recruitment, chondrocyte apoptosis, and angiogenesis. They interact with chondrocytes, endothelial cells, and fibroblasts via CXCL and NF-κB signaling. High-dimensional weighted gene co-expression network analysis (hdWGCNA) identified 352 module genes linked to OA-macrophages. Machine learning developed a four-gene-based OAMGS score that accurately identifies OA-macrophages, with an AUC of 1 in the discovery cohort and 0.990 in an external cohort. Gene expression was validated in the OA model using RT-qPCR and IF.

CONCLUSION

This study identifies a macrophage subcluster elevated in OA patients. OA-macrophages play an immunoregulatory role and may serve as diagnostic markers. The OAMGS score, based on four genes, provides an accurate diagnostic tool and potential therapeutic target for OA.

摘要

背景

骨关节炎（OA）是退行性关节疾病的主要病因，全关节置换是唯一的根治方法。然而，目前尚无改善病情的治疗方法。髌下脂肪垫（IPFP）中的炎症和纤维化会导致OA的发生和发展。然而，IPFP微环境中的细胞组成和分子机制仍不清楚。本研究旨在探讨OA巨噬细胞的功能及其临床意义。

方法

我们分析了正常人和OA患者的单细胞RNA测序（scRNA-seq）数据。富集分析揭示了不同细胞类型之间生物途径的差异。伪时间和细胞间通讯分析揭示了OA巨噬细胞与其他细胞类型的发育轨迹和相互作用。机器学习（ML）算法确定了OA巨噬细胞的特征基因。开发了OAMGS诊断评分，并使用CIBERSORT分析免疫浸润及其与免疫细胞的关联。建立大鼠OA和正常模型，并使用免疫荧光（IF）染色和定量逆转录PCR（RT-qPCR）验证特征基因表达。

结果

OA巨噬细胞在炎症和纤维化中起核心作用，增强白细胞募集、软骨细胞凋亡和血管生成。它们通过CXCL和NF-κB信号通路与软骨细胞、内皮细胞和成纤维细胞相互作用。高维加权基因共表达网络分析（hdWGCNA）确定了352个与OA巨噬细胞相关的模块基因。机器学习开发了一种基于四个基因的OAMGS评分，可准确识别OA巨噬细胞，发现队列中的AUC为1，外部队列中的AUC为0.990。使用RT-qPCR和IF在OA模型中验证了基因表达。