Abstract: TH-PO1097
PAR-CLIP Identification of Cell Type and Context-Specific miRNA/mRNA Interactions
Session Information
- Glomerular Diseases: Podocyte Biology - I
November 07, 2019 | Location: Exhibit Hall, Walter E. Washington Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Glomerular Diseases
- 1204 Podocyte Biology
Authors
- Walter, Debra L., University of Michigan, Ann Arbor, Michigan, United States
- Zhang, Huanqing, University of Michigan, Ann Arbor, Michigan, United States
- Zhang, Hongyu, University of Michigan, Ann Arbor, Michigan, United States
- O'Connor, Christopher Lund, University of Michigan, Ann Arbor, Michigan, United States
- Turner, David L., University of Michigan, Ann Arbor, Michigan, United States
- Bitzer, Markus, University of Michigan, Ann Arbor, Michigan, United States
Background
miRNA have been implicated as mediators of acute and chronic kidney diseases and as intervention targets. miRNA:mRNA interactions are highly complex and are poorly explored in kidney cells. We applied in vitro and in silico methods to identify miRNA:mRNA interactions in two cultured kidney epithelial cells and in response to TGFB1, a common damage-associated cytokine.
Methods
Immortalized human podocytes and proximal tubular (HK2) cells were treated with TGFB1 (20 ng/mL) or vehicle for 24 hours (approx. 1 mio cells per sample; n=2). Photoactivatable ribonucleoside-enhanced crosslinking and immunoprecipitation (PAR-CLIP) was used to identify miRNA:mRNA interactions. Target genes identified by PAR-CLIP were compared with miRNA targets identified by computational prediction algorithms. RNA-seq and small-RNA-seq were performed in parallel.
Results
PAR-CLIP identified 3,645 genes targeted by miRNAs in both cell types, while 117 and 24 genes were targeted only in podocytes or HK-2 cells, respectively. 3,744 genes were targeted in TGFB1 and vehicle-treated samples, while 39 and 3 genes were targeted only in cells exposed to TGFB1 or vehicle, respectively. Integration of cell type and treatment revealed that podocyte specific genes targeted by miRNAs after TGFB1 treatment, but not vehicle, were detected as targets in vehicle-, but not TGFB1-, treated HK2 cells, and vice versa. miR-21-5p had one cell type specific target in each podocytes and HK-2 cells, while miR-21-3p had only one gene specific target in HK2 cells. miRDB predicted 10-11%, Target Scan predicted 18-42% and RNA22 predicted 30-19% of PAR-CLIP identified genes per 5p and 3p arms, with limited overlap between different algorithms. RNA-seq identified 3,001 and 3,235 expressed genes in podocytes and HK2 cells, with 3,510 and 2,712 genes differentially expressed after TGFB1. 172 and 135 miRNAs were detected in podocytes and HK2 cells, with expression of 45 and 94 altered by TGFB1.
Conclusion
Together, we identified cell type and context specific targets for miRNAs, with the vast majority of miRNA targets detected in both epithelial cell types. However, miRNA target gene profiles may differ more in non-epithelial cells. Importantly, miRNA targets identified by PAR-CLIP are largely not predicted using computational algorithms.
Funding
- NIDDK Support