Metabolome-Wide Mendelian Randomization to Evaluate the Causal Association Between Blood Metabolites and Rosacea

Abstract

Background Rosacea is a facial skin condition characterized by inflammation and redness. Metabolic dysfunction has emerged as a significant contributor to the pathogenesis and progression of rosacea. However, the precise causal impact of blood metabolites on the development of rosacea remains uncertain. Methods Utilizing a genome-wide association dataset, we conducted two-sample Mendelian randomization (MR) analyses to investigate the causal relationship between 486 blood metabolites and rosacea. Our study included two distinct rosacea datasets, each representing different phenotypic characteristics. One dataset comprised cases identified by International Classification of Diseases (ICD) 10 diagnosis codes for rosacea (ROSA), consisting of four subtypes: Perioral dermatitis, Rhinophyma, Other rosacea, and Unspecified rosacea. The other dataset included cases defined by ICD10 diagnosis codes for Other and Unspecified rosacea (OUR), encompassing two subtypes: Other rosacea and Unspecified rosacea. Causality assessment was primarily conducted using the random inverse variance weighted (IVW) method, complemented by MR-Egger and weighted median methods. Sensitivity analyses were performed employing the Cochran’s Q test, MR-Egger intercept test, MR-PRESSO, and leave-one-out analysis. Reverse MR, linkage disequilibrium regression score (LDSC), and colocalization analyses were conducted to address potential issues of reverse causation, genetic correlation, and linkage disequilibrium (LD). Additionally, multivariable Mendelian randomization (MVMR) analysis was employed to evaluate the independent effects of metabolites on rosacea while accounting for potential confounders. Furthermore, metabolic pathway analysis was performed using the web-based platform MetaboAnalyst 5.0. Statistical analyses were conducted using R software, and the STROBE-MR checklist was utilized to guide the reporting of our MR study. Results Our findings identified seven metabolites with causal effects on ROSA and 14 metabolites with causal effects on OUR. Reverse MR analysis provided no evidence supporting causal effects of rosacea on these metabolites. Multivariable MR analysis established the independent causal effects of various metabolites on rosacea. Colocalization analysis unveiled a presence of shared genetic variants occurring concurrently in both metabolites and rosacea. Moreover, analysis of metabolic pathways indicated the potential involvement of the arginine and proline metabolism pathway, as well as the caffeine metabolism pathway, in the underlying mechanism of rosacea pathogenesis. Conclusion Our study provides a comprehensive atlas that elucidates the causal relationships between plasma metabolites and rosacea. Furthermore, we have identified two pivotal metabolic pathways implicated in the pathogenesis of rosacea. These findings offer insights into potential predictive biomarkers and therapeutic targets for the treatment of rosacea.