Employing plasma proteins in proteomic Mendelian randomization analysis to identify therapeutic targets for duodenal...

Proteomics serves as a primary source of therapeutic targets. In this study, we performed a Mendelian randomization (MR) analysis within the proteomic scope to identify candidate protein markers and potential therapeutic targets for duodenal ulcer (DU). This study utilized MR and co-localization analysis within the proteomic framework. Data on 2088 plasma proteins were carefully collected from a study that detected 4907 protein quantitative trait loci. The genetic association data for DU were sourced from the UK Biobank, which encompassed 1908 cases and 461,025 controls. MR used single nucleotide polymorphisms as a genetic tool to estimate the causal effects of exposure on outcomes, in order to screen candidate proteins associated with DU. Meanwhile, Bayesian co-localization analysis is used to determine the probability of shared causal genetic variation between features. Additionally, 2-step MR was employed to quantify the proportion of protein-mediated risk factors for DU. Finally, protein-protein interaction analysis was conducted to elucidate the potential link between proteins and drugs currently used for treating DU. Using the Drug Signature Database, potential targeted drugs for druggable proteins were explored. We identified 11 plasma proteins that were significantly associated with DU. Elevated levels of FLT4, IGSF3, IL6ST, EPHB4, DPEP2, SEMA6A, and IL1R1 were found to have a risk-conferring effect. Conversely, increased levels of REG1B, GOLM1, FAM3D, and QSOX2 exhibited a protective effect. Notably, none of these 11 proteins demonstrated evidence of reverse causality. Bayesian co-localization analysis indicated that REG1B, FLT4, GOLM1, EPHB4, and FAM3D shared the same genetic variations as those associated with DUs. Additionally, the protein target IL1R1, which is related to DU drugs, and 6 pharmaceutically relevant proteins, namely REG1B, IL6ST, FLT4, DPEP2, QSOX2, and EPHB4, were identified. Our research found that REG1B, FLT4, IGSF3, IL6ST, GOLM1, EPHB4, DPEP2, FAM3D, QSOX2, SEMA6A, and IL1R are associated with DU. Among them, IL1R1, REG1B, IL6ST, FLT4, DPEP2, QSOX2, and EPHB4 may become drug targets for further clinical research on DU. Targeting these proteins during drug development may provide a preferred and cost-effective approach for treating DU.