Potential Drug Targets for Migraine Identified Through Mendelian Randomization Analysis of the Human Plasma Proteome

Migraine is a common chronic condition with a clear need for innovative treatment approaches. Protein quantity trait loci (pQTLs), which specifically regulate protein expression levels, are the closest and most stable QTLs to the phenotypes under study. In this context, we employ plasma proteomes data to conduct Mendelian randomization and colocalization analysis to identify potential drug targets for migraine. Plasma proteome data were sourced from the Decode, UKB-PPP, and Interval databases. We selected cis-SNPs associated with plasma protein levels at a genome-wide significance level (p < 5 × 10^-8) as instrumental variables for Mendelian randomization analysis. Using the IEU OpenGWAS database (NCase = 13,971, NControl = 470,627) as an outcome, we initially explored the causal relationship with migraine through summary-data-based Mendelian randomization (SMR). We conducted a multi-tiered screening, including two-sample MR, HEIDI test, and Bayesian colocalization to confirm potential targets related to migraine. Additionally, we included GWAS data from the UK Biobank (NCase = 13,597, NControl = 449,336) for further robustness validation. Our initial SMR analysis identified 11 plasma proteins with potential causal links to migraine. Further investigation through two-sample Mendelian randomization using inverse variance weighted analysis (IVW) and sensitivity analysis yielded 8 significant proteins. These included 6 double-positive proteins (CALCB, COLEC11, SEMA6C, SERPING1, TBC1D17, UBE2L6) and 2 single-positive proteins (HYOU1, SERPINA1), all meeting criteria in the HEIDI test (P_HEIDI > 0.05) and colocalization analysis (PP.H₄ > 0.7). External validation using an alternate migraine GWAS dataset (non-cancer illness code, self-reported: migraine ukb-b-16868) confirmed the findings for 5 proteins. Both HEIDI test and Bayesian colocalization suggested robust positive results for CALCB and TBC1D17. Specifically, CALCB is associated with an increased risk of migraine, whereas TBC1D17 is associated with a reduced risk. Our research provides strong evidence for two proteins and suggestive evidence for three additional proteins as potential targets for migraine treatment. Notably, CALCB is already under development as a therapeutic target for migraines, lending credibility to the promising drug development prospects for TBC1D17. Future studies are needed to confirm the causal role of TBC1D17 and explore the underlying mechanisms in migraine.