Screening of potential drugs for the treatment of diabetic kidney disease using single-cell transcriptome sequencing and connectivity map data
Yi Li,Shaohui Gao,Zhaochen Guo,Zige Chen,Yihan Wei,Yutong Li,Yani Ba,Zhihong Liu,Hao Bao
DOI: https://doi.org/10.1016/j.bbrc.2024.150263
2024-09-17
Abstract:Objective: To explore the feasibility of screening potential drugs for the treatment of diabetic kidney disease (DKD) using a single-cell transcriptome sequencing dataset and Connectivity Map (CMap) database screening. Methods: A DKD single-nucleus transcriptome sequencing dataset was analyzed using Seurat 4.0 to obtain specific podocyte subclusters and differentially expressed genes (DEGs) related to DKD. These DEGs were subsequently subjected to a search against the CMap database to screen for drug candidates. Cell and animal experiments were conducted to evaluate the efficacy of the top 3 drug candidates. Results: Initially, we analyzed the DKD single-nucleus transcriptome sequencing dataset to obtain intrinsic renal cells such as podocytes, endothelial cells, mesangial cells, proximal tubular cells, collecting duct cells and immune cells. Podocytes were further divided into four subclusters, among which the proportion of POD_1 podcytes was significantly greater in DKD kidneys than in control kidneys (34.0 % vs. 3.4 %). The CMap database was searched using the identified DEGs in the POD_1 subcluster, and the drugs, including tozasertib, paroxetine, and xylazine, were obtained. Cell-based experiments showed that tozasertib, paroxetine and xylazine had no significant podocyte toxicity in the concentration range of 0.01-50 μM. Tozasertib, paroxetine, and xylazine all reversed the advanced glycation end products (AGEs)-induced decrease in podocyte marker levels, but the effect of paroxetine was more prominent. Animal experiments showed that paroxetine decreased urine ALB/Cr levels in DKD model mice by approximately 51.5 % (115.7 mg/g vs. 238.8 mg/g, P < 0.05). Histopathological assessment revealed that paroxetine attenuated basement membrane thickening, restored the number of foot processes of podocytes, and reduced foot process fusion. In addition, paroxetine also attenuated renal tubular-interstitial fibrosis. Mechanistically, paroxetine inhibited the expression of GRK2 and NLRP3, decreased the phosphorylation level of p65, restored NRF2 expression, and relieved inflammation and oxidative stress. Conclusion: This strategy based on single-cell transcriptome sequencing and CMap data can facilitate the identification and aid the rapid development of clinical DKD drugs. Paroxetine, screened by this strategy, has excellent renoprotective effects.