Multi-Omics Mendelian Randomization Identifies Therapeutic Targets for Sjögren’s Disease with Clinical and Bayesian Validation

Jiale He, Kesong Li, Zilin Guo, Xinyao Zhou, Xiaopo Tang Department of Rheumatology, Guang’anmen Hospital, China Academy of Chinese Medical Sciences, Beijing, 100053, People’s Republic of ChinaCorrespondence: Xinyao Zhou, Email [email protected] Xiaopo Tang, Email [email protected]: There are currently no effective pharmacological treatments for Sjögren’s disease (SjD). Our study aims to identify potential therapeutic targets for the condition using druggable genome-wide Mendelian randomization (MR).Methods: Druggable genome data were obtained from the Drug-Gene Interaction Database (DGIdb) and the study by Finan et al. We then integrated these druggable genes with blood-derived cis-eQTL, cis-mQTL, and cis-pQTL datasets, each analyzed separately in two-sample MR analyses using SjD GWAS summary statistics as the outcome, applying a genome-wide MR approach focused on druggable targets. Bayesian colocalization was applied to validate shared causal genetic variants. Protein levels of prioritized genes were validated in clinical serum samples from SjD patients and controls using ELISA. Phenome-wide MR (Phe-MR) analysis was conducted across 1359 phenotypes from the UK Biobank to evaluate potential pleiotropic effects and safety profiles of the identified targets, assessed side effects and alternative indications of identified targets.Results: Fourteen druggable genes were identified, with eight (PLAT, SIRPB1, LAIR2, NEU1, SLC22A16, RAD52, PSPH, and CDH23) demonstrating consistent causal relationships with SjD. ELISA validation supported differential protein expression for these targets. Key findings include the protective role of NEU1 and PSPH in systemic immune regulation, the pathogenic impact of PLAT, SIRPB1, BRD2, and LAIR2 on inflammation, and the potential involvement of SLC22A16 and RAD52 in metabolic stress and immune activation. Existing pharmacological compounds targeting these genes, including Aminocaproic acid, Resveratrol, Levocarnitine, Imatinib, and Zinc chloride were identified as potential therapeutic candidates. No significant adverse effects were detected through Phe-MR analysis.Conclusion: Our research indicated PLAT, SIRPB1, LAIR2, NEU1, SLC22A16, RAD52, PSPH, and CDH23 may serve as promising targets for SjD, while the effectiveness of Aminocaproic acid, Resveratrol, Levocarnitine, Imatinib, and Zinc chloride for SjD requires further validation.Keywords: Sjögren’s disease, Mendelian randomization, druggable genome, Bayesian colocalization, therapeutic targets, Phe-MR analysis