Abstract: TH-OR066
Murine Diabetic Nephropathy at Single Cell Resolution
Session Information
- Diabetic Kidney Disease: Discovery of Molecular Mechanisms
November 07, 2019 | Location: 207, Walter E. Washington Convention Center
Abstract Time: 06:06 PM - 06:18 PM
Category: Diabetic Kidney Disease
- 601 Diabetic Kidney Disease: Basic
Authors
- Wu, Haojia, Washington University School of Medicine, Saint Louis, Missouri, United States
- Gonzalez-Villalobos, Romer Andres, Janssen Research & Development, Cardiovascular Metabolism, Boston, Massachusetts, United States
- Yao, Xiang, Janssen Research & Development, Cardiovascular Metabolism, Boston, Massachusetts, United States
- Rankin, Matthew M., Janssen Research & Development, Cardiovascular Metabolism, Boston, Massachusetts, United States
- Kirita, Yuhei, Washington University School of Medicine, Saint Louis, Missouri, United States
- Breyer, Matthew, Janssen Research & Development, Cardiovascular Metabolism, Boston, Massachusetts, United States
- Humphreys, Benjamin D., Washington University School of Medicine, Saint Louis, Missouri, United States
Group or Team Name
- The Humphreys Lab
Background
Diabetic nephropathy (DN) is the most common cause of ESRD, but the transcriptional changes driving disease progression at the single cell level remain undefined. We hypothesized that single nucleus RNA-seq could provide insight into the cellular mechanisms of diabetic nephropathy.
Methods
We collected urine and kidney samples from control(n=2) and db/db(n=4) female mice at 17 weeks. We generated a total of 70,437 single nucleus transcriptomes and performed a comprehensive bioinformatics analysis.
Results
db/db mice had 255±35 µg/mg urine Alb:creatinine ratio compared to 24±6 in control. On average we detected 2,940 unique genes/nucleus. By unbiased clustering, we could identify 18 major cell types representing all major cell types, including macula densa, with differential expression of hundreds of genes across all clusters. Diabetic kidney had higher leukocyte infiltration including T cells, dendritic cells and macrophages. Diabetic macrophages upregulated the receptor Plxdc2, whose ligand PEDF ameliorates DN when administered exogenously. We generated a detailed diabetic glomerular intercellular communication map between podocytes, glomerular endothelium, mesangium and Cdh6+ parietal epithelia. This revealed podocyte – parietal epithelial cell signaling via b-catenin suppressive Igfbp4-Lrp6 signaling among others. We identified evidence for compensatory gene expression to combat podocyte damage, including downregulation of protein kinase C epsilon and upregulation of type 1 adenylate cyclase, proteins known to mediate susceptibility to proteinuria. Other upregulated podocyte genes included angiogenic EphA6 and Shroom3, a GWAS hit for CKD. Unexpectedly, diabetic stroma cells showed the largest gene expression changes including genes related to integrin linked kinase, cell adhesion and calcium signaling.
Conclusion
This is the first comprehensive single nucleus transcriptional atlas of a mouse model of DN. We demonstrate the utility of this approach by revealing (1) activated macrophage recruitment, (2) detailed diabetic glomerulus intercellular communication, (3) podocyte-specific expression of proteinuria susceptibility genes and (4) stromal cell activation.
Funding
- Commercial Support – Janssen Pharmaceuticals