Metabolomic etiological insights into six diabetic complications: a large scale Mendelian randomization atlas with colocalization validation.

BACKGROUND AND

OBJECTIVE

Diabetic complications can significantly affect the quality of life and prognosis of patients with diabetes. This study employed a systematic approach to elucidate the causal relationship between serum metabolites and six prevalent diabetic complications using a Mendelian randomization (MR) strategy.

METHODS

Serum metabolite data were obtained from genome-wide association studies, and data on six diabetic complications were acquired from the FinnGen consortium. A two-sample MR approach was used to investigate the association between serum metabolites and common diabetic complications. Reverse MR analysis was conducted to investigate potential causal relationships between diabetic complications and serum metabolite levels. Sensitivity analyses were performed to evaluate the robustness of our findings. Analyses included inverse-variance-weighted, MR-Egger, linkage disequilibrium score regression, and colocalization approaches.

RESULTS

We identified 81 causal associations, highlighting the significance of serum metabolites in the context of diabetic complications. The results identified significant causal associations: Bilirubin degradation products were inversely linked to diabetic retinopathy, while androstane sulfate and N-succinyl-phenylalanine increased the risk of retinopathy. Caffeine metabolites and adenosine 5'-monophosphate-to-citrate ratios were positively associated with nephropathy. Reverse MR analysis confirmed unidirectional causality, and sensitivity tests ruled out pleiotropy. Colocalization analyses highlighted shared genetic loci, such as rs2991970, between metabolites and hypoglycemia.

CONCLUSION

These findings elucidate metabolite-specific pathways underlying diabetic complications and propose novel biomarkers for risk stratification. The limitations of this study include its European-centric nature and the lack of stratified covariates. This study highlights the value of integrating genetic and metabolomic data to enhance precision medicine in diabetes management.