POS1025 COMBINATION OF BULK AND SINGLE CELL RNASEQ ANALYSES TO REVEAL MECHANISMS OF ABT-317 (JAK INHIBITOR) ON SYNOVIAL FIBROBLASTS
Y. Son,D. Korenfeld,B. Harvey,J. Wang,A. Suarez-Fueyo,D. Chang,T. Radstake,M. Ruzek
DOI: https://doi.org/10.1136/annrheumdis-2023-eular.532
2023-01-01
Abstract:Background Janus Kinase (JAK) inhibitors, including upadacitinib, are approved for clinical use, and show good therapeutic responses in Rheumatoid arthritis (RA) patients. In addition to impacts on cytokine signaling in immune cells, fibroblast-like synoviocytes (FLS) are also emerging as key cellular participants that contribute to the JAK- Signal transducer and activator of transcription (STAT) pathway guided cytokine-mediated interactions between FLS and immune cells. Despite the important role of FLS in RA pathogenesis, the mechanism of action of JAK inhibitors on FLS has not been extensively studied. Objectives The aim of this study is to characterize the molecular effects of a tool JAK1 selective inhibitor (ABT-317) on RA-FLS by performing bulk RNA-seq and single cell RNA seq (scRNA-seq) analysis on cell cultures following exposure to Staphylococcus enterotoxin B (SEB)-conditioned PBMC medium (CM) to mimic the RA inflammatory milieu +/- ABT-317. Methods RA-FLS cultures from 6 (Bulk RNA-seq) or 4 (ScRNA-seq) donors were pre-treated with various concentrations (0.1 and 1 µM) of ABT-317 with/without exposure to 30% SEB-conditioned PBMC medium. Total 36 Bulk RNA-seq library was prepared using NEB kit. EdgeR was used for differential expression gene (DEG) analysis, and DEGs (defined as padj < 0.1) were evaluated using Ingenuity Pathway Analysis (IPA). Total 24 ScRNA-seq libraries were prepared by 10x Genomics, and 87,108 cells were recovered. The data was analyzed with Cell Ranger, IPA, Seurat, and Harmony R packages. Gene Expression Omnibus (GEO) repository data sets (GSE12021 and GSE55457) from RA synovial tissue studies were used to compare with this study. Results The addition of ABT-317 to CM-stimulated FLS resulted in a downregulation of the same genes upregulated to the greatest degree by CM indicating that ABT-317 reverses a major portion of gene expression induced. IPA analysis demonstrated the expected impact of Jak inhibition on JAK/STAT and interferon signaling pathways that drive inflammation. Interestingly however, pathways unrelated to JAK signaling, such as the regulation of cell adhesion, complement activation and cytosolic DNA-sensing were also downregulated by ABT-317, suggesting that additional pathogenic mechanisms in synovial fibroblasts are also impacted by JAK inhibition. To determine whether specific fibroblast sub-populations are impacted by Jak inhibition, scRNA-seq was used to segregate FLS cells into distinctive clusters with and without CM and ABT-317 (1uM). We found that the pseudobulk data from scRNA-seq was highly correlated to the bulk RNA-seq data demonstrating consistency between our single cell and bulk datasets. We identified 9 clusters that only emerged following treatment of FLS with CM and upon addition of ABT-317 in a dose-dependent manner. ABT-317 reversed the elevated expression of cell migration related genes like ITGB8 after CM treatment in a specific cluster which suggests a potential for differential Jak responsiveness within subpopulations of fibroblasts. To establish the relevance of this in vitro data to patients, the relationship of FLS cultures stimulated with CM with public synovial tissue datasets from RA patients was explored. DE genes from the datasets (RA vs. OA) were positively correlated with the bulk and pseudobulk from the FLS culture scRNAseq (CM vs. UT), suggesting many of the same pathways are activated in vitro compared to patient synovial tissue. Conclusion We find that Jak inhibition is effective in downregulating several proinflammatory pathways induced by conditioned media stimulation in FLS cultures as well as additional mechanisms that could impact fibroblast pathogenesis in RA. These studies advance our understanding of Jak1 inhibitors and how they may contribute to clinical efficacy. REFERENCES: NIL. Acknowledgements: NIL. Disclosure of Interests Yuna Son Grant/research support from: Yes. I am an employee of AbbVie., Employee of: Yes. I am an employee of AbbVie., Daniel Korenfeld Grant/research support from: Employee of AbbVie, Employee of: Employee of AbbVie, Bohdan Harvey Shareholder of: Employee of AbbVie, Grant/research support from: Employee of AbbVie, Employee of: Employee of AbbVie, Jing Wang Shareholder of: Employee of AbbVie, Grant/research support from: Employee of AbbVie, Employee of: Employee of AbbVie, Abel Suarez-Fueyo Shareholder of: Employee of AbbVie, Grant/research support from: Employee of AbbVie, Employee of: Employee of AbbVie, Dan Chang Shareholder of: Employee of AbbVie, Speakers bureau: Employee of AbbVie, Grant/research support from: Employee of AbbVie, Employee of: Employee of AbbVie, Timothy Radstake Shareholder of: Employee of AbbVie, Speakers bureau: Employee of AbbVie, Grant/research support from: Employee of AbbVie, Employee of: Employee of AbbVie, Melanie Ruzek Shareholder of: Employee of AbbVie, Grant/research support from: Employee of AbbVie, Employee of: Employee of AbbVie.