Abstract: FR-PO305
Identification of Kidney Cell-Type-Specific Regulatory Elements Reveals Genetic Variants, Genes, and Cell Types Influencing Kidney Function Heritability
Session Information
- Genetic Diseases: Models, Mechanisms, Treatments
November 04, 2022 | Location: Exhibit Hall, Orange County Convention Center‚ West Building
Abstract Time: 10:00 AM - 12:00 PM
Category: Genetic Diseases of the Kidneys
- 1102 Genetic Diseases of the Kidneys: Non-Cystic
Authors
- Loeb, Gabriel, University of California San Francisco, San Francisco, California, United States
- Kathail, Pooja, University of California Berkeley, Berkeley, California, United States
- Shuai, Richard W., University of California Berkeley, Berkeley, California, United States
- Cheung, Kennix-Kalen, University of California San Francisco, San Francisco, California, United States
- Chung, Ryan K., University of California Berkeley, Berkeley, California, United States
- Ioannidis, Nilah, University of California Berkeley, Berkeley, California, United States
- Reiter, Jeremy, University of California San Francisco, San Francisco, California, United States
Background
Kidney disease is highly heritable. Understanding how genetic variants influence kidney function can reveal molecular pathways underlying kidney disease. Genome-wide association studies have identified hundreds of genomic loci associated with kidney function. However, due to linkage disequilibrium and localization of most risk variants within noncoding regions, the cell types and genes affected by the vast majority of these risk loci remain unclear.
Methods
We used single-cell ATAC-sequencing of human kidneys, partitioned LD-score regression of kidney biomarker genome-wide association studies, functionally informed fine-mapping, and deep learning models that predict chromatin accessibility from sequence to functionalize the variants, genes, and cell types associated with decreased kidney function.
Results
We mapped chromatin accessibility in over 40,000 human kidney cells, allowing us to identify kidney regulatory elements genome-wide. Tubule epithelial cell type-specific regulatory elements demonstrated strong enrichment for kidney function heritability. Functionally informed fine-mapping revealed putative causal variants in both monogenic kidney disease genes and novel genes. A deep learning model trained to predict cell type-specific chromatin accessibility from sequence predicted the effect of variants on chromatin accessibility (AUROC = 0.77). We used this model and allele-specific ATAC-sequencing to identify kidney function variants that disrupt tubule epithelial chromatin accessibility, providing mechanistic insight into how these variants alter tubule gene expression to affect kidney function.
Conclusion
Enrichment of kidney function heritability within tubule epithelial cells suggests that perturbation of tubule gene expression is a major driver of kidney disease. Using functionally informed fine-mapping of kidney function variants, we identified regulatory elements and genes, which, when perturbed in tubule epithelial cells, affect human kidney function.
Funding
- NIDDK Support