Abstract: TH-PO021
A Systematic Approach to Map Downstream Effects of Proteins on the Paired Plasma and Urine Metabolome
Session Information
- Augmented Intelligence for Prediction and Image Analysis
October 24, 2024 | Location: Exhibit Hall, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Augmented Intelligence, Digital Health, and Data Science
- 300 Augmented Intelligence, Digital Health, and Data Science
Authors
- Steinbrenner, Inga, Institute of Genetic Epidemiology, Faculty of Medicine and Medical Center - University of Freiburg, Freiburg, Germany
- Kottgen, Anna, Institute of Genetic Epidemiology, Faculty of Medicine and Medical Center - University of Freiburg, Freiburg, Germany
- Schlosser, Pascal, Institute of Genetic Epidemiology, Faculty of Medicine and Medical Center - University of Freiburg, Freiburg, Germany
Background
Proteins mediate and catalyze many metabolic functions, including enzymatic reactions or transport processes. The kidneys operate at the interface of plasma and urine by clearing molecular waste from the body while retaining valuable solutes, underscoring the importance of understanding the connection between transporters and enzymes and their substrates across biofluids. Using quantitative readouts of the proteome and metabolome, we employed hypothesis-generating genetic causal inference methods to study their molecular relations.
Methods
We conducted proteome-wide association studies (PWAS, 1,303 proteins) for the levels of 1,296 plasma and 1,399 urine metabolites as a causal inference approach, based on summary statistics from genome-wide association studies of metabolite levels (N= 7,213 and 5,023, respectively). Downstream analyses included statistical colocalization to control for the risk of genetic confounding due to linkage disequilibrium and a phenome-wide PWAS to highlight (patho-)physiological effects for the proteins with a significant metabolite association with over 4,000 phenotypes in the UK Biobank.
Results
We identified 284 putative causal effects of proteins on metabolite levels, arising from 61 unique proteins and 176 unique metabolite-biofluid combinations and balanced between plasma and urine (141 vs. 143). Colocalization analyses prioritized 169/284 findings (posterior probability of a shared causal variant > 80%). For 31 of the 61 proteins, 326 putative causal effects on health outcomes were identified (p<1.8e-7). For example, the kidney membrane enzyme DPEP1 displayed the strongest association with urine prolylglycine (p-value=1.1e-292) and was associated with 13 metabolite levels and 25 health outcomes including higher hemoglobin concentration and less frequently diagnosed high blood pressure. Another example were UGT1A1 and UGT1A6, enzymes that transform small lipophilic molecules into water-soluble, excretable metabolites. Accordingly, they were among the proteins with the highest number of associated metabolites and linked to risk of gallstone disease.
Conclusion
This hypothesis-generating in vivo causal inference screen of the proteome and metabolome highlights dozens of enzymes and their corresponding metabolites and clinical correlates.