# Repurposing Antidiabetic Drugs for Gangrene: A Mendelian Randomization and Text Mining Study

## TL;DR

This study finds that diabetes increases gangrene risk and identifies drugs that may help treat gangrene by combining genetic and text analysis.

## Contribution

The study identifies potential drugs for diabetes-induced gangrene using Mendelian randomization and bioinformatics analysis.

## Key findings

- Genetic susceptibility to type 1 diabetes increases gangrene risk (OR: 1.19, P=0.0134).
- Type 2 diabetes also increases gangrene risk (OR: 1.57, P=0.0269).
- 12 drugs targeting 6 genes were identified as potential therapies for diabetes-induced gangrene.

## Abstract

Objective: Gangrene has been a problem for many people with diabetes. Besides, the relationship and pathomechanism of diabetes-induced gangrene (DG) are still unclear. The aim of this study was to investigate the causal relationship between diabetes and gangrene through Mendelian randomization (MR) and to identify potential therapeutic agents using bioinformatics analysis.

Method: Summary data from genome-wide association studies (GWAS) were utilized to evaluate the connection between two types of diabetes and gangrene risk using a two-sample MR design. Single nucleotide polymorphisms (SNPs) that were significantly associated with diabetes were selected as instrumental variables, and their validity was verified by F-statistics and other methods. Next, we used text mining and protein-protein interaction (PPI) networks to filtrate significant genes for drug-gene interaction (DGI) to identify prospective medications for the therapy of DG.

Results: Through multiple methods analysis (IVW, MR-Egger and MR-PRESSO etc.), MR analysis showed that genetic susceptibility to type 1 diabetes was related to a higher risk of gangrene risk (OR: 1.19, 95% CI: 1.04-1.36, P-value: 0.0134), while type 2 diabetes mellitus (T2DM) could also increase the gangrene risk (OR: 1.57, 95% CI: 1.05-2.33, P-value: 0.0269). The outcomes of text mining disclosed 50 genes enriched in NOD-like receptor and RAGE signaling pathways commonly associated with both diabetes and gangrene for PPI analysis. Subsequent DGI analysis revealed six genes targeted by 12 drugs (DGI score > 5), presenting them as candidates for treating DG.

Conclusion: In conclusion, this study not only validates the causal effect of diabetes on gangrene risk but also identifies several potential therapeutic agents (CILAZAPRIL, RESATORVID, SILTUXIMAB, and OLOKIZUMAB) by integrating bioinformatics analysis, providing new directions for future clinical interventions.