Combining single-cell RNA sequencing and population-based studies reveals hand osteoarthritis-associated chondrocyte subpopulations and pathways

Hand osteoarthritis (OA) is a prevalent and heterogeneous joint disorder, distinct from knee and hip OA in etiology and pathophysiology. Patients experience pain, functional limitations, and disability. Despite numerous clinical trials testing various drugs, there's currently no effective cure. A deeper understanding of the underlying mechanisms is crucial for developing effective prevention and treatment strategies. Single-cell RNA sequencing (scRNA-seq) offers a powerful tool to comprehensively characterize cellular and molecular profiles in both healthy and diseased tissues, identifying disease-dependent differences at single-cell resolution. This study combined scRNA-seq with population-based studies, including Mendelian randomization (MR), to investigate the molecular mechanisms of hand OA. MR utilizes genetic variants as proxies for exposure, minimizing confounding and reverse causation, enabling a robust assessment of causal relationships between genetic predisposition and disease risk. This combined approach facilitates translation from molecular to population levels, aiding in the development of effective prevention and treatment strategies for hand OA. The study aimed to perform scRNA-seq analysis on cartilage from osteoarthritic and non-osteoarthritic interphalangeal joints, compare alterations in cellular composition and gene expression, conduct an MR study using UK Biobank data to investigate causal associations between key genes and hand OA, and conduct a cross-sectional study using data from the Xiangya Osteoarthritis Study to examine associations between serum biomarkers and hand OA.

Existing literature highlights the heterogeneity of hand OA, emphasizing its distinct characteristics compared to other OA forms. Studies have documented the significant clinical burden of hand OA, comparable to rheumatoid arthritis. Previous clinical trials evaluating symptomatic slow-acting and anti-inflammatory drugs have yielded disappointing results, underscoring the need for improved understanding of disease mechanisms. Single-cell RNA sequencing has emerged as a valuable technique for detailed characterization of cellular and molecular profiles in various tissues, while Mendelian randomization offers a powerful approach for estimating causal effects, minimizing confounding biases. The combination of these approaches promises to advance the understanding of hand OA.

This study employed a multi-pronged approach combining single-cell RNA sequencing (scRNA-seq) with Mendelian randomization (MR) and a population-based cross-sectional study. Articular cartilage specimens from five donors (five osteoarthritic and five non-osteoarthritic samples from interphalangeal joints) underwent scRNA-seq analysis. Chondrocytes were isolated using a two-step enzymatic digestion protocol. ScRNA-seq data were processed using standard bioinformatic pipelines for quality control, dimensionality reduction, clustering, and differential gene expression analysis. Cell populations were annotated using known marker genes and compared between osteoarthritic and non-osteoarthritic samples. Functional analyses, including Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment, were performed to determine the biological processes and pathways associated with differentially expressed genes. CellChat analysis was used to identify intercellular communication patterns. Immunohistochemistry (IHC) was used to validate the spatial distribution of specific cell types and proteins. For validation, an MR study utilized data from UK Biobank (332,668 individuals) to assess the causal relationship between *FTH1* gene expression and hand OA risk. A cross-sectional analysis of the Xiangya Osteoarthritis Study (1241 participants) examined the association between serum ferritin levels (a biomarker of iron overload encoded by *FTH1*) and hand OA prevalence. Statistical analyses included Wilcoxon matched-pairs signed rank test, Wilcoxon rank-sum test, logistic regression, and generalized estimating equations (GEEs).

ScRNA-seq analysis of 105,142 cells identified 13 chondrocyte subpopulations, including a novel inflammatory chondrocyte (InflamC) subpopulation characterized by the expression of pro-inflammatory genes (*CCL2*, *CCL20*, *NOS2*, *MMP3*). Fibrocartilage chondrocytes (FCs) showed extensive gene expression changes in osteoarthritic cartilage, with upregulation of catabolic genes (*MMP2*, *MMP3*, *MMP13*) and inflammatory genes. Both InflamC and FCs showed a trend towards increased numbers in osteoarthritic cartilage. KEGG analysis revealed enrichment of the ferroptosis pathway in both InflamC and FCs from osteoarthritic cartilage, with elevated expression of iron overload-related genes, particularly *FTH1*. Mendelian randomization analysis using UK Biobank data demonstrated a significant causal association between genetically predicted higher *FTH1* mRNA expression and increased hand OA risk (odds ratio [OR] = 1.07, 95% confidence interval [CI]: 1.02–1.11, *P* = 0.005). The Xiangya Osteoarthritis Study showed a significant positive association between high serum ferritin levels and hand OA prevalence (*P*-for-trend = 0.037). Comparative analysis with existing scRNA-seq data on knee OA revealed distinct cellular and molecular differences, with ferroptosis and *FTH1* playing unique roles in hand OA.

This study provides the first single-cell resolution transcriptomic atlas of hand articular cartilage, identifying key chondrocyte subpopulations (InflamC and FCs) and implicating ferroptosis as a critical pathway in hand OA. The findings demonstrate the importance of chronic inflammation and iron overload in hand OA pathogenesis. The novel inflammatory chondrocyte subpopulation (InflamC) and its interactions with macrophages warrant further investigation for potential therapeutic targeting. The consistent upregulation of *FTH1* in both scRNA-seq and population-based studies strongly supports its role in hand OA. The validation using Mendelian randomization strengthened the causal inference, mitigating confounding effects. The positive association between serum ferritin levels and hand OA prevalence further supports the involvement of iron homeostasis dysregulation. These findings highlight the distinct pathophysiology of hand OA compared to knee or hip OA and suggest potential therapeutic avenues, such as ferroptosis inhibitors or iron chelators.

This study provides a comprehensive molecular atlas of hand articular cartilage at single-cell resolution, identifying inflammatory and fibrocartilage chondrocytes as key players in hand OA. The study implicates ferroptosis and iron overload as pivotal mechanisms in disease pathogenesis. The findings, validated through independent population-based studies, suggest potential therapeutic targets for hand OA, warranting further investigation into ferroptosis inhibitors or iron chelators. Future research could focus on expanding the study to include other joint components, validating findings in vivo using animal models, and conducting mechanistic studies to further explore the role of InflamC and ferroptosis.

The relatively small number of cartilage samples used in the scRNA-seq analysis limits the generalizability of the findings. The lack of an animal model for hand OA hinders in vivo validation of the findings. The study focused primarily on cartilage; including other joint components (synovium, subchondral bone, and synovial fluid) in future studies could provide a more comprehensive understanding of hand OA pathogenesis.