Single-cell eQTL Mendelian randomization reveals the monocyte-associated PRKCB–CTSL axis as a candidate regulatory pathway and potential therapeutic target in rheumatoid arthritis

Objectives Rheumatoid arthritis (RA) involves persistent immune dysregulation, but cell-specific genetically regulated pathways remain unclear. We integrated cell-specific genetic and multi-omics evidence to prioritize candidate regulatory pathways. Methods cis-eQTLs from 17 immune-cell populations were integrated with RA GWAS data by two-sample Mendelian randomization. Transcriptomics, WGCNA, and machine-learning feature selection identified convergent candidates. Single-cell and virtual perturbation analyses assessed cellular localization and downstream changes. Molecular docking and dynamics computationally prioritized compounds, and knockdown/rescue experiments evaluated the PRKCB–CTSL relationship. Results Eleven candidates had convergent genetic and transcriptomic support. PRKCB was retained by LASSO, SVM, and random forest and preferentially expressed in monocytes. Virtual perturbation implicated CTSL and antigen-processing/presentation pathways. PRKCB knockdown reduced CTSL expression and NF-κB phosphorylation, whereas CTSL restoration partially rescued NF-κB phosphorylation. Separately, structure-based analyses computationally prioritized ursodiol for the most favorable predicted PRKCB interaction among evaluated compounds. Conclusions Genetic, transcriptomic, single-cell, computational, and in vitro evidence supports monocyte-associated PRKCB–CTSL as a candidate regulatory pathway in RA. Further mechanistic and in vivo studies are required to establish its causal and pharmacological relevance. Key Points • Multi-omics analysis identified 11 key RA-associated targets. • RA was classified into two molecular subtypes with distinct immune features. • PRKCB was identified as the core target enriched in monocytes. • The PRKCB–CTSL axis may drive RA inflammation, and ursodiol showed therapeutic potential.