Abstract: SA-OR52
Defining the Molecular Correlate of Arteriolar Hyalinosis in CKD Progression by Integration of Single-Cell Transcriptomic Analysis and Descriptor Scoring in KPMP and NEPTUNE
Session Information
- CKD: Cutting Edge of Mechanisms
November 06, 2021 | Location: Simulive, Virtual Only
Abstract Time: 04:30 PM - 06:00 PM
Category: CKD (Non-Dialysis)
- 2103 CKD (Non-Dialysis): Mechanisms
Authors
- Menon, Rajasree, University of Michigan Michigan Medicine, Ann Arbor, Michigan, United States
- Otto, Edgar A., University of Michigan Michigan Medicine, Ann Arbor, Michigan, United States
- Barisoni, Laura, Duke University, Durham, North Carolina, United States
- Hodgin, Jeffrey B., University of Michigan Michigan Medicine, Ann Arbor, Michigan, United States
- Kretzler, Matthias, University of Michigan Michigan Medicine, Ann Arbor, Michigan, United States
Group or Team Name
- for the Kidney Precision Medicine Project (KPMP)
Background
Single cell RNA sequencing generates transcriptomic data at cellular resolution allowing the identification of cell-type specific transcript expression. We performed an integrated analysis of single cell data with descriptors from histopathology analysis of biopsy samples of CKD and AKI patients.
Methods
As part of Kidney Precision Medicine Project (KPMP), single cell analysis from 12 AKI and 15 CKD patients proceeded per KPMP guidelines (including normalization, scaling, clustering and cell-specific marker identification). Top 5000 highly variable genes expressed in the endothelial cluster identified at the low cluster granularity were analyzed using weighted co-expression network analysis. Next, the co-expressed gene sets were associated with descriptors from the histopathology analysis. A composite score was generated using the expression levels of genes for the modules that significantly correlated with the descriptors. For validation purposes, similar composite scores were also generated from tubular interstitial gene expression data of NEPTUNE cohort .
Results
The unsupervised clustering identified kidney cell clusters including glomerular, tubular and immune cell types. The weighted co-expression network analysis of endothelial genes showed a gene module significantly associated (adjusted p < 0.02) with arteriolar hyalinosis, one of the descriptors from the histopathology analysis; the genes in this module were enriched in the arteriolar endothelial cluster identified by high resolution clustering. KPMP CKD patients with the top composite scores had baseline eGFR < 60 (ml/min/1.73m2). In NEPTUNE, the endothelial scores significantly associated with low eGFR (p < 0.0002) and the composite endpoint of CKD progression (< 40% reduction eGFR or ESRD) indicating poor prognosis for the samples with high endothelial scores (P < 0.0001). Pathway analysis showed adipocytokine signaling as the top enriched pathway for this gene set.
Conclusion
Using integrated analysis of single cell expression data with histopathology descriptors, we identified an arteriola endothelial gene set linking arteriolar hyalinosis to CKD progression.
Funding
- NIDDK Support