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.
CONCLUSIONS: 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.
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.
CONCLUSIONS: 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.
摘要：
目的：使用单细胞转录组测序数据集和ConnectivityMap（CMap）数据库筛选，探索筛选治疗糖尿病肾病（DKD）的潜在药物的可行性。
方法：使用Seurat4.0分析DKD单核转录组测序数据集，以获得与DKD相关的特异性足细胞亚簇和差异表达基因（DEG）。随后针对CMap数据库对这些DEGs进行搜索以筛选候选药物。进行细胞和动物实验以评估前3种候选药物的功效。
结果：最初，我们分析了DKD单核转录组测序数据集，以获得内在的肾细胞，如足细胞，内皮细胞，系膜细胞,近端肾小管细胞,收集导管细胞和免疫细胞。足细胞进一步分为四个亚簇，其中DKD肾脏中POD_1细胞的比例明显高于对照肾脏（34.0％vs.3.4%）。使用POD_1子集群中标识的DEG搜索CMap数据库，还有毒品,包括Tozasertib,帕罗西汀,还有赛拉嗪,已获得。基于细胞的实验表明，tozasertib,帕罗西汀和赛拉嗪在0.01-50μM的浓度范围内没有明显的足细胞毒性。Tozasertib,帕罗西汀,和赛拉嗪都逆转了糖基化终产物（AGEs）诱导的足细胞标记水平的降低，但帕罗西汀的作用更为突出。动物实验表明，帕罗西汀使DKD模型小鼠尿液ALB/Cr水平降低约51.5%(115.7mg/gvs.238.8mg/g,P<0.05)。组织病理学评估显示帕罗西汀减弱了基底膜增厚，恢复足细胞足突起的数量,减少足部融合。此外,帕罗西汀还可以减轻肾小管间质纤维化。机械上,帕罗西汀抑制GRK2和NLRP3的表达,降低p65的磷酸化水平,恢复NRF2的表达,缓解炎症和氧化应激。
结论：这种基于单细胞转录组测序和CMap数据的策略可以促进临床DKD药物的鉴定和快速开发。帕罗西汀,通过这种策略筛选，具有出色的肾脏保护作用。
方法：使用Seurat4.0分析DKD单核转录组测序数据集，以获得与DKD相关的特异性足细胞亚簇和差异表达基因（DEG）。随后针对CMap数据库对这些DEGs进行搜索以筛选候选药物。进行细胞和动物实验以评估前3种候选药物的功效。
结果：最初，我们分析了DKD单核转录组测序数据集，以获得内在的肾细胞，如足细胞，内皮细胞，系膜细胞,近端肾小管细胞,收集导管细胞和免疫细胞。足细胞进一步分为四个亚簇，其中DKD肾脏中POD_1细胞的比例明显高于对照肾脏（34.0％vs.3.4%）。使用POD_1子集群中标识的DEG搜索CMap数据库，还有毒品,包括Tozasertib,帕罗西汀,还有赛拉嗪,已获得。基于细胞的实验表明，tozasertib,帕罗西汀和赛拉嗪在0.01-50μM的浓度范围内没有明显的足细胞毒性。Tozasertib,帕罗西汀,和赛拉嗪都逆转了糖基化终产物（AGEs）诱导的足细胞标记水平的降低，但帕罗西汀的作用更为突出。动物实验表明，帕罗西汀使DKD模型小鼠尿液ALB/Cr水平降低约51.5%(115.7mg/gvs.238.8mg/g,P<0.05)。组织病理学评估显示帕罗西汀减弱了基底膜增厚，恢复足细胞足突起的数量,减少足部融合。此外,帕罗西汀还可以减轻肾小管间质纤维化。机械上,帕罗西汀抑制GRK2和NLRP3的表达,降低p65的磷酸化水平,恢复NRF2的表达,缓解炎症和氧化应激。
结论：这种基于单细胞转录组测序和CMap数据的策略可以促进临床DKD药物的鉴定和快速开发。帕罗西汀,通过这种策略筛选，具有出色的肾脏保护作用。
