Abstract: PO0674
Animal Models Cannot Well Reflect the Transcriptomic Changes of Human Diabetic Nephropathy: A Comparative Study
Session Information
- Diabetic Kidney Disease: Basic - I
November 04, 2021 | Location: On-Demand, Virtual Only
Abstract Time: 10:00 AM - 12:00 PM
Category: Diabetic Kidney Disease
- 601 Diabetic Kidney Disease: Basic
Authors
- Zhou, Leting, Wuxi People's Hospital affiliated to Nanjing Medical University, Wuxi, China
- Ni, Li-Hua, Department of Nephrology, Zhongnan Hospital of Wuhan University, Wuhan, China
- Hu, Ling, Wuxi People's Hospital affiliated to Nanjing Medical University, Wuxi, China
- Zhu, Yu-Shan, Wuxi People's Hospital affiliated to Nanjing Medical University, Wuxi, China
- Zhang, Yue, Wuxi People's Hospital affiliated to Nanjing Medical University, Wuxi, China
- Wang, Sisi, Wuxi People's Hospital affiliated to Nanjing Medical University, Wuxi, China
- Wang, Liang, Wuxi People's Hospital affiliated to Nanjing Medical University, Wuxi, China
Background
Various mouse models have been developed and widely applied in investigating the pathogenesis of DN. Whether the models share the same underlying molecular changes with human DN is poorly understood. To this end, we performed a systematic analysis of the transcriptomes of the kidney tissues from patients with DN and various mouse models. To our knowledge, this is the largest analysis on this topic
Methods
This study included the bioinformatic analysis and in vivo validation. We comprehensively analyzed the genome-wide mRNA expression of kidney tissues collected from patients with biopsy-proven DN and widely used animal models(n=60). The bioiformatics workflow is shown in figure 1. Then, the expression levels of interested genes were further validated
Results
The transcriptomic profiles of all the animal models had poor correlation with those of patients with DN. However, we observed a much better correlation within species, regardless of the disease stages or modeling methods. In the GSVA analysis, we found the animal models shared similar pathological processes but could not well reflect the real circumstances in human DN. In enrichment analysis, we found the animal models shared the same pathways such as the accumulation of extracellular matrix and MAPK signaling with human DN. However, these models can not well mimic pathways such as cytokine signaling, vitamin D metabolism and SLC transporter disorders. Finally, the expression levels of the interested genes measured by the westernblot method showed good consistency with those generated by high throughput platforms
Conclusion
We found mouse models can not well reflect the transcriptomic changes of human DN in many aspects. We also provided a useful dataset to facilitate the translational research of DN
Funding
- Government Support – Non-U.S.