Background: Rheumatoid arthritis (RA) and psoriatic arthritis (PsA) are inflammatory arthritides, but bone erosions and straight vasculature in synovial tissue are characteristic of RA, whereas bony growth and bushy blood vessels in synovial tissue are characteristic of PsA [1]. The aetiology of these phenotypes remains poorly understood. Understanding the molecular endotypes of synovial cells such as fibroblasts may provide insight into the pathogenesis of RA and PsA, their responses to standard anti-inflammatory treatments, and may identify novel therapeutic targets.

Objectives: This study sets out to explore the differential gene expression of synovial tissue cells in RA and PsA before and after intra-articular methylprednisolone injection during the post-treatment interval when symptom relief is expected. This study provides insight as to the technical feasibility to discriminate differences in single nucleus transcriptomic profiles of such small tissue biopsies.

Methods: Patients at Nuffield Orthopaedic Centre were enrolled in accordance with ethical approval. Ultrasound guided synovial tissue biopsy was performed using 16G Quick-Core biopsy needles which yield very small tissue fragments. Single nucleus RNA sequencing was performed on knee synovial tissue biopsies from 4 patients with RA and 4 patients with PsA, before and 4 weeks after 80 mg intra-articular methylprednisolone acetate injection (Figure 1a). For quality control (Figure 1b), droplets more than 3 median absolute deviations from the median were discarded. Filtering thresholds of nFeatures>300, nCounts>500, mitochondrial RNA<5%, decontX score<0.2, automated soupX and default scDblFinder settings were applied to all samples. Cell clusters were manually annotated using literature-based marker genes. Transcriptomic profiles were compared using DESeq2 on a per-cell type basis. Absolute Log2FoldChange>1 and adjusted p value <0.05 were considered significant. Gene set enrichment analysis was performed with g:GOSt.

Results: A total of 26,453 nuclei passed quality control with mean of 1,653 nuclei per sample. UMAP of cell cluster annotations is shown in Figure 1c. As we previously presented at EULAR, the downregulation of tissue inhibitor of metalloproteinase 1 in lining fibroblasts after intra-articular methylprednisolone treatment was consistent with known glucocorticoid effects [2]. When comparing PsA with RA in pre-treatment cells and post-treatment cells, differential gene expression analysis showed only the pre-treatment lining fibroblasts were demonstrably different between PsA and RA. In the comparison of PsA with RA in pre-treatment lining fibroblasts, 27 genes were detectably differentially expressed in RA and 56 genes were detectably differentially expressed in PsA (Figure 1d). Strikingly in pre-treatment lining fibroblasts, the differentially expressed genes with the lowest padj values were upregulation of matrix Gla protein ( MGP ) in RA and upregulation of bone morphogenetic protein-binding endothelial cell precursor-derived regulator ( BMPER ) in PsA. Differentially expressed genes were enriched in pathways involving system development, axonogenesis and morphogenesis.

Conclusions: Despite limited synovial tissue from this small pilot study, differential gene expression analysis in pre-treatment synovial lining fibroblasts was detected between PsA and RA patients before intra-articular methylprednisolone injection. Lining fibroblasts showed upregulation of MGP in RA compared to PsA. Previous research has also shown increased MGP in RA synovial tissue compared to osteoarthritis [3]. MGP upregulation may account for bony erosion and straight synovial tissue vasculature in RA, by inhibiting calcification and sprouting angiogenesis through bone morphogenetic proteins (BMP) inhibition. BMPER expression was upregulated in pre-treatment lining fibroblasts in PsA compared to RA. BMPER modulates BMP which may drive bony growth and bushy synovial tissue vasculature in PsA (1). Although lining fibroblast expression of MGP and BMPER may be novel endotypes of RA and PsA, respectively, they have opposing roles in BMP signalling which may account for the divergent phenotypes [3-5]. The paucity of differentially expressed genes in lining fibroblasts between RA and PsA post-treatment may suggest a treatment effect or be because of a very small sample size. Due to the limitations of this hypothesis-generating study with very few donors, these findings will need to be confirmed by future studies across a larger patient population.

Figure 1.

(a) Schematic overview of the study design. (b) Overview of the computational analysis. (c) UMAP depicting all annotated cell types in the single nucleus RNA sequencing data from pre- and post-treatment RA and PsA samples. (d) Differential gene expression in pre-treatment lining fibroblasts in RA and PsA.

REFERENCES: [1] Reece R, Canete J, Parsons W, Emery P, Veale D. Distinct vascular patterns of early synovitis in psoriatic, reactive, and rheumatoid arthritis. ARTHRITIS AND RHEUMATISM. 1999;42:1481-4.

[2] Law S, Cohen C, Coates LC, Snelling S, Mimpen J, Richards D. OP0279 SINGLE NUCLEI TRANSCRIPTOMIC STUDY OF INTRA-ARTICULAR METHYLPREDNISOLONE EFFECTS ON INFLAMMATORY ARTHRITIS SYNOVIUM. Annals of the Rheumatic Diseases. 2024;83(Suppl 1):15-6.

[3] Tieliwaerdi X, Aripova N, Duryee M, Jiang X, Klassen L, O’Dell J, et al. Matrix Gla Protein (MGP) Modified with Malondialdehyde/Acetaldehyde Is Increased in Rheumatoid Arthritis and Cardiovascular Patients. ARTHRITIS & RHEUMATOLOGY. 2020;72.

[4] Yao Y, Bennett B, Wang X, Rosenfeld M, Giachelli C, Lusis A, et al. Inhibition of Bone Morphogenetic Proteins Protects Against Atherosclerosis and Vascular Calcification. CIRCULATION RESEARCH. 2010;107(4):485-U102.

[4] Yao Y, Nowak S, Yochelis A, Garfinkel A, Bostroem K. Matrix GLA protein, an inhibitory morphogen in pulmonary vascular development. J Biol Chem. 2007;282(41):30131-42.

Acknowledgments: NIL.

Disclosure of Interests: None declared.