Systematic druggable genome-wide Mendelian randomization identifies therapeutic targets for calcium pyrophosphate deposition disease.

BACKGROUND

There are no effective pharmacological treatments for calcium pyrophosphate deposition (CPPD) disease. We aim to identify potential therapeutic targets for CPPD disease by integrating various publicly available datasets.

METHODS

We integrated data from druggable genomes, cis-expression quantitative trait loci (cis-eQTL)/cis-protein quantitative trait loci (cis-pQTL) in human blood and skeletal muscle tissues, and genome-wide association study (GWAS) summary statistics related to CPPD disease. This integration aimed to investigate potential causal relationships between drug target genes and CPPD disease using Mendelian randomization (MR) method. To validate these causal links, we performed sensitivity analyses and Bayesian colocalization. Furthermore, we applied phenome-wide Mendelian randomization (Phe-MR) approach to assess potential side effects or alternative therapeutic indications for the identified drug targets. Finally, we explored actionable drugs for these target genes using existing databases.

RESULTS

MR analysis identified four druggable genes-ULK3, CSK, PPIE, and NFKB1-that may have potential causal associations with CPPD disease in human blood or skeletal muscle tissue. These genes were further confirmed by Bayesian colocalization (PPH4 > 80%). The upregulation of ULK3 (blood eQTL, beta: 0.62, 95% CI: 1.129-3.091, P = 0.015; muscle eQTL, beta: 0.93, 95% CI: 1.055-6.146, P = 0.037; blood pQTL, beta: 2.51, 95% CI: 1.199-126.169, P = 0.034), CSK (blood eQTL, beta: 0.69, 95% CI: 1.094-3.650, P = 0.024; muscle eQTL, beta: 0.98, 95% CI: 1.058-6.707, P = 0.037; blood pQTL, beta: 3.41, 95% CI: 1.509-606.864, P = 0.025), and PPIE (blood eQTL, beta: 0.52, 95% CI: 1.218-2.321, P = 0.001; muscle eQTL, beta: 0.85, 95% CI: 1.384-3.962, P = 0.001; blood pQTL, beta: 1.06, 95% CI: 1.094-7.565, P = 0.032) in blood and skeletal muscle tissues was positively associated with the risk of CPPD disease. Conversely, NFKB1 showed an inverse relationship with CPPD disease risk (blood eQTL, beta: - 0.99, 95% CI: 0.147-0.935, P = 0.035; muscle eQTL, beta: - 1.95, 95% CI: 0.046-0.437, P = 0.001; blood pQTL, beta: - 3.17, 95% CI: 0.002-0.762, P = 0.032). The Phe-MR analysis indicated that the four potential therapeutic targets for CPPD disease had no significant adverse effects.

CONCLUSIONS

Our research indicated that ULK3, CSK, PPIE, and NFKB1 may be promising targets for CPPD disease. Key Points • ULK3, CSK, PPIE, and NFKB1 may be drug target genes in CPPD disease.