# The Identification of Biomarkers and Therapeutic Targets for Diabetic Kidney Disease by Integrating the Proteome with the Genome

## TL;DR

This study identifies two proteins, UMOD and SIRPA, that may contribute to diabetic kidney disease risk, offering potential new treatment targets.

## Contribution

The novel integration of proteomic and genomic data reveals causal protein markers for diabetic kidney disease.

## Key findings

- Two cardiometabolic proteins, UMOD and SIRPA, show causal roles in increasing DKD risk.
- UMOD is highly expressed in renal macrophages and could be a novel therapeutic target.
- SIRPA may be a target for drug repurposing in DKD treatment.

## Abstract

Background: The blood proteome is a major source of biomarkers and therapeutic targets. We conducted a proteome-wide Mendelian randomization (MR) study to identify cardiometabolic protein markers for diabetic kidney disease (DKD). Methods: We measured all 369 proteins in the Olink Explore 384 Cardiometabolic and Cardiometabolic panel of 500 patients with type 2 diabetes from 11 communities in Shanghai. Protein quantitative trait loci (pQTLs) were derived by coupling genomic and proteomic data. Cis-pQTLs identified for proteins were used as instrumental variables in MR analyses of DKD risk, and the outcome data were obtained from 8401 Japanese individuals with type 2 diabetes (2809 cases and 5592 controls). Replication MR analysis was performed in the UK Biobank Pharma Proteomics Project (UKB-PPP). Colocalization analysis and the Heidi test were used to examine whether the identified proteins and DKD shared causal variants. Results: Among the 369 proteins, we identified 66 independent cis-pQTLs for 64 proteins. MR analysis suggested that two cardiometabolic proteins (UMOD and SIRPA) may play a causal role in increasing DKD risk, with UMOD showing replication in UKB-PPP. Bayesian colocalization further supported the causal effects of these proteins. Additional analyses indicated that UMOD is highly expressed in renal macrophages. Further downstream analyses suggested that UMOD could be a potential novel target and that SIRPA could be a potential repurposing target for DKD; however, further validation is needed. Conclusions: By integrating proteomic and genetic data from patients with type 2 diabetes, we identified two protein biomarkers potentially associated with DKD risk. These findings provide insights into DKD pathophysiology and therapeutic target development, but further replication and functional studies are needed to confirm these associations.