Exploring causal correlations between inflammatory response-related genes and osteoporosis: a Multi-Omics Mendelian Randomization Study.

Abstract

Introduction: The relationship between the inflammatory response (IR) and osteoporosis (OP) has been the subject of extensive research; however, their genetic link remains unclear. This study used IR-related genes as instrumental variables (IVs) to represent IR, while summary data of OP served as the outcome to explore their genetic relationship.

Material and methods: IR-related genes were retrieved from the GeneCards database. OP transcriptome datasets were collected from the Gene Expression Omnibus (GEO) database and meta-analyzed to identify differentially expressed genes (DEGs) related to IR in OP. Genetic proxy instruments for IR-related genes were derived from studies of corresponding gene expression (n = 31,684) and DNA methylation (n = 1,980) quantitative trait loci (eQTLs and mQTLs), respectively. Aggregated data for OP (1,351 OP cases and 209,313 controls) were extracted from the largest genome-wide association study (GWAS) of OP. We integrated QTL data with OP GWAS data to estimate their genetic associations using summary data-based Mendelian randomization analysis (SMR). Additionally, Bayesian colocalization analysis was employed to reveal the potential relationships between IR gene expression and inflammatory factors, as well as various hormones. Finally, to further validate whether the statistical evidence provided in GWAS comprised true-positive findings, a replication study (1,955 cases and 278,169 controls) was conducted here through genotype-phenotype associations.

Results: A meta-analysis of four datasets identified 115 IR-related DEGs in OP out of 612 IR-related genes. Through SMR analysis, we found that the expression levels of two IR-related genes were associated with OP risk. Specifically, elevated gene expression levels of FAS (odds ratio (OR) = 1.094; 95% confidence interval (CI) = 0.892-1.341; false discovery rate (FDR) = 0.034) increased the risk of OP. Conversely, increased expression levels of CHUK decreased the risk of OP (OR = 0.518; 95% CI = 0.424-0.637; FDR = 0.039). Colocalization analysis identified potential interactions between the FAS gene and estradiol (PP.H4 = 0.95) as well as interleukin-1α (IL-1α) (PP.H4 = 0.65). Potential interactions were also observed between the CHUK gene and growth hormone (PP.H4 = 0.59) as well as macrophage inflammatory protein-1α (MIP-1α) (PP.H4 = 0.62). In addition, consistent results were observed in the replication study, further demonstrating the reliability of our findings.

Conclusions: This multi-omics integration study revealed a genetic link between IR and OP, as represented by IR-related genes, and provided new insights into the potential pathogenic mechanisms of OP. Additionally, these identified candidate genes offer avenues for future targeted functional studies aimed at developing appropriate therapeutic interventions and preventing OP.