Dapagliflozin was obtained from Apexbio (United States, ^#A5854), with a purity of ≥99%; TGF-β1was obtained from PeproTech (United States). The western blot antibodies used were Collagen 1 (Abcam, United States, ^#ab270993), α-SMA (Sigma, United States, ^#A5691), vimentin (Proteintech, United States, ^#10366-1-AP), VDAC1(Proteintech, United States, ^#55259-1-AP), FN (Proteintech, United States, ^#15613-1-AP), p-JNK (CST, United States, ^#4668), JNK (CST, United States, ^#9252), p-ERK (CST, United States, ^#4370), ERK (CST, United States, ^#4695), P-P38 (CST, United States, ^#4631), p38 (CST, United States, ^#8690), total oxphos complexes (Thermo, United States, AB-2533835), CPT1-α (Proteintech, United States, ^#15184-1-AP), 0.2% adenine diet (TP 1S002), and control diet (LAD 3001) obtained from Trophic Animal Feed High-tech Co., Ltd., China. The SOD activity kit (^#A001-3), GSH content assay kit (^#A006-2-1) and MDA content assay kit (^#A003-1) were purchased from the Nanjing Jiancheng Institute of Bioengineering.

Eight-week-old male C57BL/6J mice were obtained from Silaike Laboratory (Hunan, China) and fed in the animal experiment laboratory of Central South University with approval from the Animal Care and Use Committee of Central South University. After strict adaptive feeding for 1 week, mice were divided into four groups: normal diet with vehicle control (oral gavage, 0.1 mL/10 g, n = 6), 0.2% adenine diet with vehicle control (oral gavage, 0.1 mL/10 g, n = 8), 0.2% adenine diet with dapagliflozin therapy at a dose of 1 mg/kg/day (oral gavage, 0.1 mL/10 g, n = 8) (Wang et al., 2021), and 0.2% adenine diet with dapagliflozin therapy at a dose of 10 mg/kg/day (oral gavage, 0.1 mL/10 g, n = 8) (Huang et al., 2019). During this study, body weight and food consumption of the four groups were recorded. On day 21 of the experiment, the mice were anesthetized via intraperitonaeal injection of pentobarbital (1%, 50 mg/kg) to harvest the kidneys and blood for further study.

Dapagliflozin was fully dissolved in DMSO (Sigma Aldrich, United States, ^#D2650), according to the manufacturer’s instructions, and stored at −20°C. When needed, it was diluted in normal saline to a final concentration of DMSO of 1%. Dapagliflozin was administered when mice were fed a 0.2% adenine diet.

Blood samples obtained from the previous animal experiments were placed in an incubator at 37°C for 1 h. After centrifugation at 4°C and 3,000 rpm for 10 min, the serum was aspirated and sent to the laboratory of Xiangya Hospital of Central South University for detection of serum creatinine and urea nitrogen.

Tissue fixative was used to fix the isolated kidney, dehydrated, and embedded in paraffin. The paraffin sections were stained with hematoxylin-eosin to enable visualization of the kidneys pathological morphology, and the sections were stained with Masson’s trichrome to facilitate observation of the degree of renal fibrosis. Each kidney sample was randomly selected from 10 regions (200×) under a light microscope (Nikon, Ni Ci E100 E200, Japan) for semi quantitative analysis, and the grading was 0–4 (0: 0%; 1: <25%; 2: 26%–50%; 3: 51%–75%; 4: ≥76% of the damaged renal tubules), and ImageJ software (Image-Pro Plus 6.0 software, Media Cybernetics, Bethesda, MD, United States) was used for measurement and calculation (Ren et al., 2021). Frozen sections were prepared, and the kidney tissues were stained with Oil Red O dye for the observation of lipid deposition in the kidney tissues, the samples were examined using microscope (Nikon, Ni Ci E100 E200, Japan), and the images were analyzed using ImageJ software (Image-Pro Plus 6.0 software, Media Cybernetics, Bethesda, MD, United States) (Chen et al., 2021).

Total RNA in kidney tissue was extracted using TRIzol (Invitrogen, United States, ^#15596026), and the quality of total RNA was evaluated with an RNA6000 Nano LabChip kit (Agilent, California, United States) using the Agilent biological analyzer 2100 system. RNA sequencing was conducted by Beijing Berry Hekang Biotechnology Co., Ltd. EdgeR was used to analyze the significance of the differential expression. The actual analysis parameters used in this analysis were log₂ | FoldChange |>2.0 and a q value of <0.05. Gene Ontology (GO) software was used for GO enrichment analysis of differentially expressed genes, and the KOBAS (v3.0) software was used for Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis.

The kidney tissue was first fixed with electron microscope fixative (Pinofet, S191102), followed by 1% osmium tetroxide, and was then dehydrated in graded alcohol and acetone and prepared into ultrathin sections (60–80 nm) in pure 812 embedding agent (SPI, 90529-77-4). The sections were stained with uranyl acetate and lead citrate, dried. Using transmission electron microscope (Hitachi; HT7800/HT7700), the number and size of mitochondria were measured in 20 random unoverlapped proximal tubular cells of each sample (Xuan et al., 2021).

Human proximal renal tubular epithelial cells (HK2) were purchased from Zhongqiao Xinzhou (China, ZQ0313). DMEM/F12 (Gbico, United States, ^#11330032) containing 10% fetal bovine serum (FBS) (Gbico, United States, ^#10099-141) and 1% penicillin-streptomycin was used to culture the cells in a cell culture box containing 5% carbon dioxide at 37°C. When the cell density was above 90%, 0.05% tryptase was used to digest the cells. The cell suspension obtained was terminated with DMEM/F12 containing FBS. Firstly, the cells from the 3×10⁴/well were inoculated into a 96 well plate. When the cells adhered to the wall and grew to 70% confluence, different concentrations of dapagliflozin were added. After 24 h of action, the CCK8 kit (Apexbio, United States, ^#K1018) instructions were strictly followed to explore the effects of the drugs on normal cells. Cells were selected in good growth conditions and inoculated 5×10⁵ cells per well into a 12 well plate, then divided into a normal group, TGF-β1 (10 ng/mL) stimulation group, and dapagliflozin treatment group. TGF-β1 was given to both the stimulation group and the treatment group when the cell growth was 70%, and dapagliflozin was added to the treatment group. The principle behind selecting the treatment concentration was to select 1 or 2 concentrations that had no effect on normal cells. Cell proteins were obtained for subsequent analysis after 24 h of stimulation.

Sodium dodecyl sulfate (SDS) contained 1% protease inhibitor and 1% phosphatase inhibitor to lyse an appropriate amount of each kidney sample and cell sample. Cells were centrifuged at 13000 rpm at 4°C for 10 min. The supernatant was collected after the completion of lysis, and bicinchoninic acid (BCA) protein detection kit was used to quantify the obtained protein samples (Sharma et al., 2020). According to the indicators to be analyzed, 8%–12% polyacrylamide gel to electrophoresis was added to the obtained protein samples, which were subsequently transferred to a polyvinylidene fluoride (PVDF) membrane under a condition of 260 mA current for 90 min. The rapid sealing solution was applied for 15 min and the corresponding primary antibody was incubated at 4°C overnight. The next day, after incubation with the corresponding secondary antibody for 1 h, the membrane was imaged with a super-sensitivity developer or an ordinary sensitivity developer. ImageJ software was used to analyze the strip produced by the development.

Trizol was used to extract total RNA from kidney tissues. A Revert Aid First Strand cDNA Synthesis Kit (Thermo Scientific, MA, ^#K1622) was used to reverse-transcribe RNA into DNA. Real-time PCR was performed using the CFX96 Quantitative PCR Detection System (Bio-rad, United States, ^#1855200). The primer sequences used in this study are shown in Supplementary Table S1.

To detect SOD, GSH and MDA content in the kidney tissues of mice in each group, kidney tissues stored in liquid nitrogen with appropriate weight were weighed and ground into 10% tissue homogenate with normal saline. We strictly followed the instructions to detect the activity of SOD the content of GSH and MDA in the kidney tissue (Sharma et al., 2018b; Fernandes et al., 2018; Sharma et al., 2019).

Intracellular ROS was determined by 2′,7′-dichlorofluorescein diacetate (DCFH-DA) (Solarbio, ^#4091-99-0) according to the manufacturer’s instructions. Briefly, HK-2 cells were seeded in 24-well plates. After 24 h incubation, cells were pretreated with or without various concentrations of dapaliglozin (20 and 40 μM) and coincubated with or without H₂O₂ (300 μmol/L) for 24 h (Zhang et al., 2019). Then HK-2 cells were loaded with 5 μmol/L DCFH-DA for 30 min at 37°C in the dark and washed with PBS three times. The cells were digested for analysis using a microplate reader at 485/528 nm (Bio-tek Synergy HT, United States).

All data are expressed as mean ± standard deviation. Comparisons between groups were analyzed using one-way ANOVA, and comparisons between two groups were analyzed using the least significant difference test. p < 0.05 was considered statistically significant.

Serum levels of urea nitrogen (BUN) and creatinine in 0.2% adenine diet-fed mice were significantly higher than those in control mice, the serum levels of urea nitrogen and creatinine were significantly decreased, and dapagliflozin at a dosage of 10 mg/kg was better than the dosage of 1 mg/kg (Figure 1A). During the period of the experiment, we found that the mice treated with 0.2% adenine had a significantly reduced body weight, dapagliflozin treatment was effective in reducing their weight loss, and the dosage of 10 mg/kg was better than the dosage of 1 mg/kg (Figure 1B). Hematoxylin and eosin (HE) and Masson staining were used to evaluate tubulointerstitial injuries and interstitial extracellular matrix (ECM) deposition, respectively. HE and Masson staining showed that the tubules in the kidneys of mice fed with a 0.2% adenine diet were markedly dilated and atrophied, and ECM deposition was observed in the renal interstitium, whereas mice treated with dapagliflozin showed a partial but significant decrease in both renal tubular injury and interstitial ECM deposition (Figure 1C). The HE and Masson pathological scores were higher in the 0.2% adenine diet mice than control group. After dapagliflozin treatment, the pathological scores in dapagliflozin treatment group were lower than the 0.2% adenine diet mice, indicating that dapagliflozin alleviated tubulointerstitial injuries and interstitial ECM deposition, and the higher dosage (10 mg/kg) was better than the lower dosage (1 mg/kg) (Figure 1D).

As the effect of dapagliflozin at a dosage of 10 mg/kg was better than the dosage of 1 mg/kg, we used this dosage in further studies. Western blotting showed that proteins such as α-SMA, Vimentin and Collagen 1 were significantly upregulated in the kidneys of mice fed a 0.2% adenine diet, and significantly downregulated by 10 mg/kg dapagliflozin treatment (Figures 2A, B). Simultaneously, real-time Quantitative Polymerase Chain Reaction (PCR) confirmed that the expression of fibrotic genes, such as col1a1, col1a3, FN, α-SMA, and vimentin, was significantly upregulated in the kidneys of mice fed 0.2% adenine, and their expression was significantly downregulated by dapagliflozin treatment (Figure 2C).

RNA-seq analysis was performed to understand how dapagliflozin alleviates renal fibrosis. The volcano plot shows significantly different gene expression profiles among normal diet mice, 0.2% adenine mice, and dapagliflozin (10 mg/kg)-treated mice. Among these differentially expressed genes, 2459 genes were upregulated, and 1357 genes were downregulated in the kidneys of 0.2% adenine diet-fed mice compared with control mice (p < 0.05) (Figure 3A). However, dapagliflozin (10 mg/kg) reversed the changes in 494 downregulated and 323 upregulated genes (p < 0.05) (Figure 3B). Significant dapagliflozin-modulated genes are illustrated by a heatmap (Figure 3C). Genes related to fibrosis (col1a1, col1a3) and inflammation (TNF-α, IL-6, IL-1β) were upregulated in 0.2% adenine diet mice, and were downregulated by dapagliflozin (10 mg/kg) (Figure 3C). The heatmap also showed that mitochondrial metabolism and fatty acid oxidation (PDH, Acox1, and Acox2)-related genes were downregulated in 0.2% adenine diet-fed mice and upregulated by dapagliflozin (Figure 3C). Furthermore, GO (Figure 3D) and KEGG (Figures 3E, F) enrichment analyses suggested that these differentially expressed genes were involved in mitochondrial and lipid metabolism and inflammatory responses.

Transcriptomic GO and KEGG enrichment analyses suggested that both mitochondrial metabolism and fatty acid oxidation levels were significantly downregulated in the kidneys of mice fed a 0.2% adenine diet compared with mice in the control group. Mitochondrial metabolism and fatty acid oxidation are important in the occurrence and development of renal injury and fibrosis (Quadri et al., 2019; Console et al., 2020), therefore, we explored the effect of dapagliflozin on these two aspects. First, we examined mitochondrial morphology using electron microscopy. In comparison to control mice, mitochondria with degenerative changes, including disruption of the mitochondrial membrane and matrix swelling, were observed in 0.2% adenine diet-fed mice, and these changes were significantly improved by dapagliflozin treatment (Figure 4A). In addition to the structure of mitochondria, the area and number of mitochondria in renal tubular were decreased in 0.2% adenine diet-fed mice, and the treatment of dapagliflozin could maintain the area and number of mitochondria in renal tubular cell (Figure 4B). Furthermore, mitochondrial respiratory chain complexes regulate mitochondrial oxidative phosphorylation (OxPhos), which is an important part of mitochondrial energy metabolism, like the tricarboxylic acid (TCA) cycle (Edmond, 2009). In comparison to the control mice, the expression of all complexes was decreased in 0.2% adenine diet-fed mice, and dapagliflozin attenuated the reduction of all complexes (Figures 4C, D), partly due to the maintenance of the amount of mitochondria (Figure 4E). Moreover, we used real time PCR to confirm whether the mitochondrial metabolism-related genes, such as ND1, ND4, AKDGH, and PDH, the mitochondrial DNA that encode oxphos-related protein, were downregulated in 0.2% adenine diet mice, and dapagliflozin significantly upregulated the mRNA levels of these genes (Figure 4F). Except for up-regulating the expression of mitochondrial DNA, the RNA-Seq Transcriptomic results shown dapagliflozin also upregulated the expression of nuclear DNA (Supplementary Figure S1). Fatty acid oxidation (FAO) mainly occurs in mitochondria; therefore, mitochondrial dysfunction often leads to the downregulation of fatty acid oxidation (Console et al., 2020). Oil red staining of the mouse kidney tissue showed that lipid deposition in the renal tubules of the mouse kidney was obvious in the 0.2% adenine diet-fed mice compared with control mice, and dapagliflozin reduced lipid deposition in the tubules (Figure 4G). Given that CPT1-α is a key enzyme in fatty acid oxidation (Miguel et al., 2021), we investigated its expression further. Western blotting showed that the expression of CPT1-α was downregulated in the mice that were fed a 0.2% adenine diet, and it was significantly upregulated by dapagliflozin treatment (Figure 4H). The results of real time PCR showed that genes related to fatty acid oxidation, such as CPT1-α, PPAR-α, Acox1, and Acox2, were downregulated in the 0.2% adenine mice, and dapagliflozin attenuated these reductions (Figure 4I).

Inflammation and oxidative stress are important factors in the development of renal fibrosis (Meng et al., 2014; Rayego-Mateos and Valdivielso, 2020). The transcriptomic results also showed that inflammation was significantly activated in the mice fed with a 0.2% adenine diet, whereas it was alleviated when the mice were treated with dapagliflozin. Consistently, our real-time PCR results showed that the expression of inflammatory factors such as IL-1β, IL-6, cxcl-1, TNF-α, and MCP-1 was increased in the mice fed with a 0.2% adenine diet compared with control mice, and dapagliflozin treatment reduced their expression (Figure 5A). We also detected SOD activity, GSH and MDA contents in the renal tissue of mice in each group. The results showed that the activity of SOD and content GSH in the kidneys of mice fed with 0.2% adenine decreased, but significantly increased after treatment with dapagliflozin (Figure 5B). The content of MDA increased in the kidneys of mice fed with 0.2% adenine decreased, but decreased after treatment with dapagliflozin (Figure 5C).

TGF-β1 is considered a key driver of renal fibrosis (Meng et al., 2016). Downstream of TGF-β, the MAPK (ERK, JNK, and p38) signaling pathways are overactivated and promote the EMT process in renal fibrosis (Rhyu et al., 2005; Hung et al., 2016). Therefore, we explored whether dapagliflozin affected the TGF-β1/MAPK pathway. Real-time PCR confirmed that TGF-β1 was upregulated when the mice were fed a 0.2% adenine diet, and dapagliflozin reduced its expression (Figure 6A). Western blotting showed that the levels of p-ERK and p-P38 were upregulated in mice fed a 0.2% adenine diet and downregulated by dapagliflozin treatment (Figures 6B, C).

In order to determine the toxicity of dapagliflozin to normal cells, we incubated HK2 cells with dapagliflozin at differing concentrations (0–100 μM) for 24 h. Cell viability was tested using CCK8 reagent (Figure 7A). We then selected the concentrations (20 and 40 μM) that had no apparent effect on normal cell viability for further research. The results suggested that both concentrations of dapagliflozin could significantly inhibit the expression of FN and α-SMA in HK2 cells under stimulation by TGF-β1 (10 ng/mL), and a higher concentration (40 μM) had a more pronounced effect (Figures 7B, C). To further confirm the effects of dapagliflozin, we observed the effects of dapagliflozin in vitro model of kidney oxidative stress (H2O2-induced HK-2 cells). The results suggested the ROS level was increased in H2O2-induced HK-2 cells compared to the control group. Treatment with dapagliflozin (40 μM) lowered the level of ROS in HK-2 cells stimulated with H2O2 (Figure 7D).

As for the therapeutic group, HK2 cells were incubated with 20 and 40 μM dapagliflozin for 24 h, and 10 ng/mL TGF-β1 was administered to both the model group and the therapeutic groups for 30 min. The cell protein was then harvested for further detection. As seen in Figures 8A, B, the levels of p-JNK, p-ERK, and p-P38 were higher than those in the control group, whereas when treated with dapagliflozin, their expression was much lower, and a higher concentration (40 μM) resulted in a greater reduction in expression. We also found that the expression of mitochondrial respiratory chain complexes decreased in HK2 cells stimulated by TGF-β1, and dapagliflozin (40 μM) increased the level of the complexes (Figures 8C–E).