Causal links between programmed cell death and hypertrophic scars: Integrative analysis of multi-omics Mendelian randomization and preliminary experimental validation

Objective: This study aims to explore the causal relationship between programmed cell death (PCD) genes and the formation of hypertrophic scars (HS) using integrative multi-omics analysis (including DNA methylation, gene expression, and protein abundance) alongside preliminary experimental validation.

Methods: We leveraged publicly available databases (eQTL Gen, UKB-PPP, and FinnGen) to obtain quantitative trait loci (QTLs) data of DNA methylation, gene expression and protein abundance. We employed Mendelian randomization (MR) approaches to uncover causal relationships and validate robustness. The methods used included inverse variance weighted (IVW) analysis, false discovery rate (FDR), Cochran's Q test, I² statistic, MR-Egger regression, MR-PRESSO, leave-one-out method, co-localization analysis, and Steiger filtering test. Then, the multi-omic MR results were integrated and three tiers of genes were identified. Further, the tier 1 genes were chosen to perform drug prediction in DSigDB and molecular docking analyses with Autodock Vina. Lastly, the effects of the selected genes and drugs in HS were validated at both the tissue and cellular levels.

Results: Through integrating multi-omics data, we identified one tier 1 gene (GLB1), twelve tier 2 genes (including DAPK2, AP4E1, ARSA, CTSF, MSH6, NEDD4, PDK1, PELI3, RB1, UNC13D, CTSC, and GZMB), and two tier 3 genes (NOS3 and ITGA6), all of which show varying associations with HS. Particularly, the GLB1(cg05120113) was causal associated with HS risk in DNA methylation (OR=1.0972, 95 % CI: 1.0532-1.1430, FDR=0.0163), gene expression (OR=1.2923, 95 % CI: 1.1816-1.4135, FDR<0.001) and protein abundance (OR=1.5430, 95 % CI: 1.3296-1.7905, FDR<0.001). The candidate drugs for GLB1 included Fulvestrant (adjusted P = 0.046, Affinity=-8.8 kcal/mol) and Cyperquat (adjusted P = 0.036, Affinity=-6.2 kcal/mol). Further, the GLB1 expression and inhibitory effect of Fulvestrant were validated in HS tissues and HSFs. Additionally, significant changes in the mRNA and protein expression levels of fibrosis-related markers, including TGF-β1 and α-SMA, were observed in HSFs.

Findings: This study provides robust evidence for the causal involvement of PCD genes in HS formation and identified GLB1 along with 14 Other potential genes. Fulvestrant demonstrated therapeutic potential for HS by modulating fibrosis-related pathways in fibroblasts.

Fibrosis; Fulvestrant; Hypertrophic scar; Mendelian randomization; Multiple omics.