+
Skip to main content
BMC is moving to Springer Nature Link. Visit this journal in its new home.
Download PDF

Sodium-glucose cotransporter 2 inhibitors alleviate renal fibrosis in diabetic kidney disease by inhibiting Hmgcs2 and Btg2 in proximal tubular cells

Journal of Translational Medicine volume 23, Article number: 959 (2025) Cite this article

Abstract

Context

Sodium-glucose cotransporter 2 inhibitors (SGLT2i) have been shown to ameliorate renal fibrosis in diabetic kidney disease (DKD), but the mechanism has not been fully explored.

Methods

The single-cell sequencing (scRNA-seq) data were downloaded from the Gene Expression Omnibus (GEO) database, and we selected the tissue data from db/m mice, db/db mice and db/db mice with SGLT2i treatment. The results were also validated by immunofluorescent staining and western blot in vivo and in vitro, respectively.

Results

Our study demonstrates that SGLT2i directly ameliorated fibrosis of proximal tubular cells by downregulating 3-hydroxy-3-methylglutaryl-CoA synthase 2 (Hmgcs2) expression in S1 proximal tubular segment cells (PT_S1) and decreasing the number of proximal tubular cell cluster with B-cell translocation gene 2 (Btg2) highly expressed (Btg2_PT). In addition, SGLT2i could indirectly influence macrophages through cell-cell communication between epithelial cells and macrophages, specifically via the App-CD74 ligand-receptor pair, thus suppressing the inflammatory response in macrophages, ultimately contributing to the delay in DKD progression.

Conclusion

Our study found that Hmgcs2 and Btg2 are therapeutic targets for Sodium-glucose cotransporter 2 inhibitors to ameliorate kidney fibrosis in diabetic kidney disease. Single-cell sequencing technology provided a high resolution of this study at the cellular level.

Background

Diabetic kidney disease (DKD) is one of the major complications in diabetic patients and the most significant cause of end-stage kidney disease (ESKD) in diabetic patients [1]. DKD is initiated by diabetes-related disturbances in glucose metabolism, which then trigger other metabolic, hemodynamic, inflammatory, and fibrotic processes that contribute to disease progression [2]. Renal fibrosis is a final pathological change in DKD and is one of the most important mechanisms in the progression of diabetic nephropathy [3]. Therefore, finding drugs and targets to ameliorate renal fibrosis is important in the fight against DKD.

Sodium-glucose cotransporter 2 (SGLT2) inhibitors improve systemic glucose homeostasis by blocking glucose reabsorption in the proximal tubule. It has been reported that SGLT2i can provide reno-protective effects, including reducing albuminuria and lowering the estimated glomerular filtration rate in DKD [4,5,6]. Moreover, recent studies have found that SGLT2i may also delay renal fibrosis in DKD patients [6,7,8]. However, the mechanism of its ameliorative effect on renal fibrosis in diabetic nephropathy has not been fully clarified.

Recently, scRNA-seq technology has been proven capable of labeling and identifying cell types through clustering and is critical to understanding kidney homeostasis, disease, and regeneration [9,10,11]. However, studies of DKD using scRNA-seq and studies evaluating drug responses at the single-cell level are still limited. Here, we performed a study to explore the mechanisms by which SGLT2i treatment ameliorates renal fibrosis in mice with diabetic kidney disease through the analysis of scRNA-seq data and validated our conclusions by performing a series of animal experiments and cell experiments.

Results

Single-cell landscape of health and DKD kidneys with or without SGLT2i treatment

We analyzed the scRNA-seq data from the GEO database as described in the Methods. After quality control and filtering, a total of 104,836 single-cell transcriptomes were generated. Using unsupervised clustering (UMAP), 43 separate cell clusters were identified after combining all samples [Figure 1A]. These clusters were categorized into three major groups based on known marker gene expression: epithelial cells, immune cells, and stromal cells [Figure 1B]. These three groups were further subdivided into 9 distinct cell types, including 6 proximal tubular cell types (78,625 cells), 2 immune cell types (4,201 cells), and 1 endothelial cell type (234 cells) [Figure 1C]. Since Slc5a2, which encodes SGLT2, is expressed predominantly in the S1 proximal tubular segment [12]. We annotated one cell cluster expressing Slc5a2 in proximal tubular cells as “PT_S1” to investigate the effects on the S1 segment and other segments of proximal tubular cells separately. Another cluster was annotated as “mt_PT” due to its high expression of mitochondrial genes (e.g., mt-Nd4l). The remaining proximal tubular cells were collectively annotated as “PT” (proximal tubular cells). Marker genes for each cell cluster are shown in Fig. 1D. To evaluate major cell type changes during disease progression and treatment, the number of cells in each cluster was compared across the three groups. PT cells were the predominant component across all groups [Figure 1E]. Notably, the proportion of PT cells decreased in db/db mice compared with db/m mice but was restored in the SGLT2i treatment group. Conversely, the percentages of macrophages and T lymphocytes increased in db/db mice compared with db/m mice [Figure 1F]. These findings suggest that during DKD progression, the reduction of proximal tubular cells and increased immune cell infiltration may drive tissue inflammation, while SGLT2i treatment could partially reverse these pathological changes.

Fig. 1
figure 1

Single-cell map of kidney tissue in mice (A) 43 cell clusters obtained after clustering using UMAP, with each color representing a cell cluster. (B) A total of 3 supcelltypes were defined using markers for cell definition of 43 cell clusters. EP, epithelial cell; IMMUNE CELL, immune cell; STROMAL, stromal cell. (C) A total of 9 cell types were defined using markers for cell definition of 43 cell clusters. Colors and labels indicate different cell types based on marker gene expression. PT, proximal tubule cells; PT-S1, S1 proximal tubular segment; mt-PT, proximal renal tubular epithelial cell with mitochondrial gene highly expressed; DT, distal renal tubule epithelial cell; EC, endothelial cell; LOH, loop of henle; MC, medullary cell; T, T lymphocyte; MAC, macrophage. (D) Bubble plot shows the expression of marker genes for each cell type. The color of the dots represents the average log2 fold change (LogFC) value. (E) Bar plot shows the number of each cell type in different groups. (F) Plotline plot shows the change of the proportion of PT, PT-S1, mt-PT, T and MAC in different groups

Full size image

Role of Hmgcs2_PT_S1 in the amelioration of fibrosis by SGLTi2

We first divided PT_S1 cells into 7 subclusters using dimensionality reduction and typical markers in the S1-segment of renal tubular epithelial cells [Figure 2A, D]. Through the annotation of subclusters, we found that Hmgcs2_PT_S1 and S100a8_PT_S1 subclusters increased in db/db mice compared with db/m mice, and these changes were reversed in the SGLT2i treatment group [Figure 2B, C]. Hmgcs2 encode a mitochondrial enzyme belongs to the HMG-CoA synthase family [13, 14]. It is reported that mitochondrial Hmgcs2 accumulation disrupted mitochondrial function and further aggravated renal fibrosis in CKD [15]. High levels of S100 calcium binding protein A8 (S100A8) expression in tubular epithelial cells during diabetic condition activated the TLR4/NF-κB signal pathway which promoted the epithelial-mesenchymal transition (EMT) process. and finally led to RIF progression [16]. Therefore, we speculated that SGLT2i can protect mitochondrial function in renal tubular epithelial cells by reducing the number of Hmgcs2_PT_S1 and S100a8_PT_S1 cells, there by prevent the progression of fibrosis and EMT. In the process of renal fibrosis, renal tubular epithelial cells induce EMT, undergoing a phenotypic transition from the epithelial cell type to the mesenchymal-like cell type. We speculated that the Hmgcs2_PT_S1 and S100a8_PT_S1 subclusters were characterized by EMT [17]. To further validate our speculation, we examine the EMT score on PT_S1 subclusters using the gene set variation analysis (GSVA) [gene sets in Appendix 1]. We found that Hmgcs2_PT_S1 and S100a8_PT_S1 subclusters were with higher scores [Figure 2E, F]. In addition, the EMT score was increased in db/db mice compared with db/m mice which was reversed by SGLT2i treatment [Figure 2G]. These finding suggested that SGLT2i prevent PT cells fibrosis mainly by reducing the number of Hmgcs2_PT_S1 and S100a8_PT_S1 subclusters, thereby inhibiting the progression of EMT and improve DKD. Next, for a deeper understanding of the differentiation potential of the Hmgcs2_PT_S1 and S100a8_PT_S1 subclusters, we conducted the assessment of stemness in PT_S1. The result showed Hmgcs2_PT_S1 and S100a8_PT_S1 subclusters exhibited higher stemness and differentiation potential [Figure 2H]. We also performed the Pseudo-Temporal analysis in PT_S1 to further investigate the differentiation stages of the Hmgcs2_PT_S1 and S100a8_PT_S1 subclusters. Cela1_PT_S1 was chosen as the beginning of differentiation, the result revealed that Hmgcs2_PT_S1 were in the terminal stage of differentiation and both of Hmgcs2_PT_S1 and S100a8_PT_S1 subclusters were directly differentiated from Cela1_PT_S1. Additionally, S100a8_PT_S1 and Hmgcs2_PT_S1 subclusters had similar gene expression trajectories [Figure 2I]. In summary, we proposed that the disease state could promote the differentiation process of proximal tubular cells in the S1-segment from Cela1_PT_S1 to S100a8_PT_S1 and ultimately to Hmgcs2_PT_S1, and this process could be inhibited by the treatment of SGLT2i. To further explore the impact of SGLT2i on the Hmgcs2_PT_S1 subcluster, we conducted the Gene Regulatory Network (GRN) analysis to study the gene regulatory mechanisms of these subclusters. It showed that the Hmgcs2_PT_S1 was uniquely regulated by early growth response 1 (Egr1) in the M5 module. We identified the binding motif by searching the JASPAR database [Figure 2J]. Since downregulation of Egf1 alleviates renal tubulointerstitial fibrosis [18, 19]we speculated that SGLT2i may reduce Hmgcs2 expression in PT_S1 by indirectly regulating Egf1. But we eliminated the speculation through scCor correlation analysis, we found that Slc5a2 had a direct positive correlation (p < 0.001) with the expression of Hmgcs2 while no correlation was observed with Egr1[Figure 2K]. To explore if the effect of SGLT2 on Hmgcs2_PT_S1 is mediated through intermediate signaling pathways, we performed KEGG enrichment analyses of differential gene expression in Hmgcs2_PT_S1, and the result showed that the effect may be mediated through MAPK signaling pathways[Figure 2L]. Consequently, we concluded that SGLT2i inhibit the expression of Hmgcs2 by suppressing the expression of Slc5a2, thereby decreasing the number of Hmgcs2_PT_S1 and preventing fibrosis in PT_S1 proximal tubular cells.

Fig. 2
figure 2

SGLT2i downregulate the expression of Hmgcs2 in PT_S1 cells to prevent the fibrosis (A) PT_S1 cells were integrated into a single dataset and clustered using UMAP. Labels indicate different cell types. (B) Bar plot shows the Number of identified cell types in each sample. (C) Plotline plot shows the change of the proportion of Hmgcs2_PT_S1, Napsa PT_S1, Cela1 PT_S1and S100a8_PT_S1 in different groups. (D) Expression of selected marker genes for each cell type projected on UMAP. (E-F) The gene set variation analysis (GSVA) shows the EMT score and EMT expression on PT_S1 subclusters. (G) Bar plot shows the EMT expression in different groups. (H) Assessment of stemness in PT_S1. (I) Pseudo-Temporal analysis in PT_S1. (J) The Gene Regulatory Network (GRN) analysis showed that the Hmgcs2_PT_S1 was uniquely regulated by Egr1 in the M5 module. (K) scCor correlation analysis showed that Slc5a2 had a direct positive correlation with the expression of Hmgcs2.( p < 0.001) (L) KEGG enrichment analysis showed that the effect of SGLT2 on Hmgcs2_PT_S1 may be mediated through MAPK signaling pathways

Full size image

Btg2_PT was identified as key subtype in fibrotic changes in PT cells

In proximal tubular cells beyond the S1 segment(PT), we found that SGLT2i could similarly inhibit the progression of fibrosis by decreasing the number of fibrosis-promoting cell clusters. Through clustering and annotation, we divided the PT cluster into seven groups [Figure 3A, D]. In PT cluster, the proportion of cells that expressing B-cell translocation gene 2 (Btg2) gene was significantly elevated in db/db mice compared with db/m mice and significantly decreased in the SGLT2i treatment group. We annotated this group of cells expressing Btg2 as “Btg2_PT” [Figure 3B, C]. Btg2 mediates focal segmental glomerulosclerosis by promoting podocyte injury including EMT and fibrosis via a Smad3-dependent mechanism. Podocyte‐specific deletion of Btg2 can inhibit EMT markers such as α‐SMA and vimentin while restoring epithelial marker E‐cadherin [20]. We speculated that Btg2 plays a critical role in the progression of fibrosis in renal tubular epithelial cells and SGLT2i could decrease the number of Btg2_PT in the PT cluster to suppress EMT and inhibit the progression of tissue fibrosis. To further validate our speculation, we performed EMT scores and statistical analyses on Btg2_PT and other types of PT cells (named Group PT) using GSVA, respectively. The results showed that EMT-related genes were mainly expressed in Btg2_PT and the differences were statistically significant (CI95% [-1.3%, -1.2%], p < 0.001) [Figure 3E, F]. In addition, the EMT score was increased in db/db mice compared with db/m mice and was reversed by the SGLT2i treatment [Figure 3G], suggesting that SGLT2i mainly decreased the number of Btg2_PT in the PT cluster to suppress fibrotic changes in proximal tubular cells. We initially speculated that SGLT2i prevented fibrosis in proximal tubular cells by reducing the expression of Btg2 in PT cells. However, scCor analysis found no correlation between Slc5a2 and Btg2, suggesting an indirect mechanism. Then we speculated that SGLT2i primarily inhibit the differentiation of Btg2_PT to suppress the fibrosis. We performed a prediction analysis of differentiation potential in Btg2_PT to infer their stemness. In the PT cells, as shown in the Fig. 3H, the subclusters with a higher stemness index included Slco1a1_PT and Btg2_PT. The number of Slco1a1_PT significantly decreased in the db group compared with the control group, which was consistent with the finding from bulk RNA-seq study [21]. While no significant changes were observed in the SGLT2i treatment group. The down-regulating of solute carrier organic anion transporter family, member 1a1(Slco1a1) in primary hepatocytes meditates inflammation [22]. This suggested that epithelial cell inflammatory injury caused by DKD primarily occurred in the Slco1a1_PT, and SGLT2i did not specifically mitigate the inflammatory injury in these cells. Notably, we found that the number of Btg2_PT was significantly increased in db/db group while significantly decreasing in the SGLT2i treatment group. As previously mentioned, Btg2 was related to fibrosis. Therefore, we speculated that these cells had the potential to differentiate from epithelial cells to fibrotic cells. To validate this, we conducted the Pseudo-Temporal analysis which Slco1a1_PT was chosen as the beginning point of differentiation and the result revealed that Btg2_PT was in the terminal stage of differentiation. [Figure 3I, J] This suggested that the DKD disease state could promote the differentiation from Gatm_PT to Btg2_PT, and SGLT2i treatment could inhibit this differentiation process. To further explore the impact of SGLT2i on the Btg2_PT subcluster, we conducted the Gene Regulatory Network (GRN) analysis to study the gene regulatory mechanisms of these subclusters, it showed that Btg2_PT was uniquely regulated by predicted gene 45,871 (Gm45871) in the M2 module [Figure 3K]. However, the result of scCor correlation analysis showed that Slc5a2 had no correlation with Gm45871. To explore if the effect of SGLT2 on Btg2_PT is mediated through intermediate signaling pathways, we performed KEGG enrichment analyses of differential gene expression in Btg2_PT, and the result showed that the effect may be mediated through MAPK signaling pathways [Figure 2L]. In summary, our results revealed that in other renal tubular epithelial cells besides the S1-segment, SGLT2i primarily reduced the number of fibrosis-promoting subcluster by inhibiting cell differentiation towards Btg2_PT to suppress fibrotic changes in proximal tubular cells.

Fig. 3
figure 3

SGLT2i decrease the number of Btg2_PT to suppress fibrotic changes in proximal tubular cells. (A) PT cells were integrated into a single dataset and clustered by UMAP. Labels indicate different cell types. (B) Bar plot shows the Number of identified cell types in each sample. (C) Plotline plot shows the change of the proportion of Btg2_PT, Gatm_PT, Col27a1_PT, Slco1a1_PT and Apoe_PT in different groups. (D) Expression of selected marker genes for each cell type projected on UMAP. (E-F) The gene set variation analysis (GSVA) shows the EMT score on PT subclusters. (CI95% [-1.3%, -1.2%], p < 0.001). (G) Violin plot shows the EMT expression in different groups. (H) Assessment of stemness in PT. (I) Pseudo-Temporal analysis in PT. (J) Expression of top genes at different stages of differentiation (K) The Gene Regulatory Network (GRN) analysis showed that the Hmgcs2_PT_S1 was uniquely regulated by Egr1 in the M5 module. (L) KEGG enrichment analysis showed that the effect of SGLT2 on Btg2_PT may be mediated through MAPK signaling pathways *p < 0.05, **p < 0.01, and ***p < 0.001

Full size image

The anti-inflammatory effects exerted by SGLT2i

Our scRNA-Seq analysis of immune cells revealed that SGLT2i may reduce the secretion of inflammatory cytokines and inactive the Toll-like receptor signaling pathway. These effects inhibited the progression of inflammation and alleviated inflammatory damage, which could inhibit the progression of DKD. Since we observed the proportion of macrophages and T lymphocytes increased in db/db mice compared with db/m mice, but decreased in SGLT2i treatment group [Figure 1E], we speculated that SGLT2i could inhibit immune responses caused by diabetic nephropathy. We divided T cells into 5 subclusters using dimensionality reduction and typical markers [Figure 4A, B]. Through the annotation of subclusters, we found the proportion of Rgs_T increased in db group but decreased in SGLT2i treatment group [Figure 4C, D]. Regulator of G-protein signaling 1 (Rgs1) silencing inhibits the inflammatory response through suppressing the inflammatory cytokines produced by Toll-like receptors [23, 24]. Therefore, we concluded that SGLT2i decreases the number of Rgs1_T to inhibit the inflammatory response. Moreover, we divided MAC cells into 4 subclusters using dimensionality reduction and typical markers [Figure 4E, F]. Through the annotation of subclusters, we found that the proportion of the Ctss_MAC subcluster increased in the db group, while it decreased in the SGLT2i treatment group [Figure 4G, H]. Cathepsin S (Ctss) is reportedly an autocrine factor that stimulates cell activation and subsequently regulates the release of proinflammatory cytokines and chemokines, thereby exerting pro-inflammatory effects and exacerbating inflammatory damage [25,26,27,28,29,30,31]. Therefore, we concluded that SGLT2i may suppress the release of proinflammatory cytokines and chemokines by reducing the number of Ctss_MAC in macrophages, consequently inhibiting the inflammatory process and the progression of DKD. In summary, we concluded that SGLT2i has anti-inflammatory effects on tissue by inhibiting the immune response of macrophages and T lymphocytes, thereby impeding the progression of DKD.

Fig. 4
figure 4

SGLT2i has anti-inflammatory effects on tissue by suppressing the immune response of macrophages and T lymphocytes. (A) T cells were integrated into a single dataset and clustered using UMAP. Labels indicate different cell types. (B) Expression of selected marker genes for each cell type projected on UMAP. (C) Bar plot shows the Number of identified cell types in each sample. (D) Plotline plot shows the change of the proportion of identified cell types in different groups. (E) Macrophages were integrated into a single dataset and clustered using UMAP. Labels indicate different cell types. (F) Expression of selected marker genes for each cell type projected on UMAP. (G) Bar plot shows the Number of identified cell types in each sample. (H) Plotline plot shows the change of the proportion of identified cell types in different groups

Full size image

Cell-cell communication existed between proximal renal tubular cells and immune cells via the App-CD74 ligand-receptor pair

To explore the interactions between epithelial cells and immune cells We performed a cell-cell communication analysis involving four cell types: PT, PT_S1, macrophages and T lymphocytes. In the cell interaction weight diagram, thicker lines represent a higher number of interactions, indicating that the interaction weight/strength between two cell types is stronger. The results showed that interactions existed between epithelial cells and immune cells and the App-CD74 ligand-receptor pair exhibit higher activity and potential between Hmgcs2_PT_S1, Btg2_PT and Ctss_MAC [Figure 5A-D]. It was noticed that the App–CD74 ligand–receptor pair showed higher activity and possibility between failed repair PT cells and the immune microenvironment than successful repair, and CD74 molecule (CD74) was significantly highly expressed in immune cells during AKI–CKD progression [32]. In experimental diabetic nephropathy and glomerulonephritis, CD74 expression is increased in tubular cells [33]. In mice, cellular damage is exacerbated by the expression of Cd74 in epithelial cells [32]. Since the PT_S1 and PT clusters primarily expressed amyloid beta precursor protein (App) [Figure 5E], we performed scCor correlation analysis and found that App had a direct positive correlation with Slc5a2 (p < 0.001) and fibrosis (CI95% [2.33e-03, -0.02], p < 0.01) in PT and PT_S1[Figure 5F]. The results showed that SGLT2i may inhibit App expression in PT_S1 and PT cells by suppressing Slc5a2 expression to alleviate fibrosis in proximal tubular cells. Since Ctss_MAC mainly expressed CD74[Figure 5E], we extracted Ctss_MAC for correlation analysis. The results showed that CD74 had a direct negative correlation with Slc5a2 (p < 0.05) and inflammation (CI95% [-0.41, -0.35], p < 0.001) in Ctss_MAC[Figure 5G]. As a result, we concluded that SGLT2i alleviates fibrosis of proximal tubular cells through inhibiting Slc5a2 and App expression in PT_S1 and PT cells, which subsequently decreased Ctss_MAC with Cd74 highly expressed through cell-cell communication to perform anti-inflammatory effects.

Fig. 5
figure 5

SGLT2i indirectly influences macrophages through cell-cell communication between proximal tubular cells and macrophages via the App-CD74 ligand-receptor pair (A) Cell communication analysis showed that interactions existed between epithelial cells and immune cells (B-D) the App-CD74 ligand-receptor pair exhibit higher activity and potential between Hmgcs2_PT_S1, Btg2_PT and Ctss_MAC. (E) The expression of App and CD74 in PT, PT_S1, macrophages and T lymphocytes clusters. (F) scCor correlation analysis showed that App had a direct positive correlation with fibrosis (CI95% [2.33e-03, -0.02], p < 0.01)and the expression of Slc5a2 (p < 0.001) in PT and PT_S1. (G) scCor correlation analysis showed that Cd74 had a direct negative correlation with inflammation (CI95% [-0.41, -0.35], p < 0.001) and the expression of Slc5a2 (p < 0.05)in Ctss_MAC

Full size image

Validation of Hmgcs2 and Btg2 in the amelioration of renal fibrosis by SGLT2i in mice

We performed animal experiments as described in the methods and validated the expression of Hmgcs2 and Btg2 in the PT_S1 and PT clusters. We divided 10 weeks old male mice into three groups: db/db (BKS.Cg-Dock7m+/+Leprdb/Nju), db/db + SGLT2i and db/m. The mice from db/db group were treated with vehicle control (phosphate-buffered saline [PBS]) (n = 3) and the mice from db/db + SGLT2i group were treated with 3 mg/kg/day of dapagliflozin (n = 3). Results showed that blood glucose in the db/db group was consistently higher compared with db/m group but was decreased after the interevent of SGLT2i. [Figure 6A] There was no significant trend in body weight change during administration. [Figure 6B] We tested the UPCR values of all the mice at the 10 weeks of age, and the result showed that the db/db mice had met the diagnostic criteria for diabetic nephropathy at the 10 weeks of age (CI95% [14.29, 74.07], p < 0.05) [Figure 6C]. We performed fluorescence staining of kidney tissue sections which confirmed our conclusions. The results showed that the expression of Hmgcs2 and Btg2 of the proximal tubular cells were increased in the db/db mice compared with db/m mice and significantly decreased in the db/db + SGLT2i mice [Figure 6D]. In addition, Hematoxylin-eosin staining of renal tissues showed that the renal glomeruli and tubules of mice in the db/m group were clear in structure and not enlarged in size. While for mice in the db/db group, the glomeruli were enlarged in size and there was proliferation of membranous cells in the zones of the glomerular tethering membranes, and an increase in the number of intraglomerular cells. The above pathological changes were effectively improved in the db/db + SGLT2i group after SGLT2i intervention in mice. Immunohistochemistry results showed that the expression of α-SMA was increased in the db/db group compared with db/m mice and significantly decreased in the SGLT2i treatment group [Figure 6E].

Fig. 6
figure 6

Validation of Hmgcs2 and Btg2 on mice (A) Glucose of mice from different groups. (B) Weight of mice from different groups. (C) UPCR of mice from different groups. (CI95% [14.29, 74.07], p < 0.05) (D) Representative immunofluorescence images from each group for LRP2(PT marker, red) and SGLT2 (PT-S1 marker, green), and BTG2/HMGCS2 (yellow) (E) Representative HE and immunohistochemistry of α-SMA images from each group. *p < 0.05, **p < 0.01, and ***p < 0.001

Full size image

Validation of HMGCS2 and BTG2 in human kidney samples and cell lines

We collected human renal slices as described in the methods and validated the expression of HMGCS2 and BTG2 in the proximal tubular cells by immunofluorescence. The results showed that the expression of HMGCS2 and BTG2 of the proximal tubular cells were increased in the DKD compared with Control. [Figure 7A] In addition, to verify the expression of HMGCS2 and BTG2 in human renal proximal tubular cells in vitro, we performed experiments using Human Renal Proximal Tubular Cells (hRPTC) and Human Kidney-2 (HK-2) cells with high glucose concentration to simulate high glucose environment, and we divided the cells into three groups, the low glucose treated group (LG), the high glucose treated group (HG) and the SGLT2i treated group (HG + SGLT2i). Finally, the expression of the expression of HMGCS2 and BTG2 were detected by western blot experiment. The results showed that in both hRPTC [Figure 7B] and HK-2 cells [Figure 7C], the protein expression level of HMGCS2 and BTG2 were elevated in the HG group compared with the LG group, but their expression levels decreased after the intervened of SGLT2i. The differences were statistically significant. In order to explore whether SGLT2i directly interacts with these genes, we designed the experiment of knocking down SLC5A2 in human proximal tubular cells HK-2 cultured in vitro, and to detect the RNA expression of HMGCS2 and BTG2 by RT-qPCR. The results showed that down-regulation of SLC5A2 (CI95% [-1.03, -0.87]) in HK-2 cells could directly up-regulate the RNA expression levels of HMGCS2 (CI95% [2.75, 7.55]) and BTG2(CI95% [1.06, 3.19]). The differences were statistically significant [Figure 7D].

Fig. 7
figure 7

Validation of HMGCS2 and BTG2 on human (A) Representative immunofluorescence images from human renal slices for LRP2(PT marker, red) and SGLT2 (PT-S1 marker, green), and BTG2/HMGCS2 (yellow) (B) Representative western blot images from each group for Hmgcs2, Btg2 and Gapdh in hRPTC. (C) Representative western blot images from each group for Hmgcs2, Btg2 and Gapdh in HK-2 cells. (D) The RT-qPCR results showed that down-regulation of SLC5A2(CI95% [-1.03, -0.87]) in HK-2 cells could directly up-regulate the RNA expression levels of HMGCS2 (CI95% [2.75, 7.55]) and BTG2(CI95% [1.06, 3.19]). n = 3/group. Means (± S.E.) of n = 3 independent experiments. *p < 0.05, **p < 0.01, and ***p < 0.001

Full size image

Discussion

Renal fibrosis is a hallmark of progressive DKD and involves multiple cell types and signaling pathways. Understanding the development of different cell types in kidney is critical to understanding kidney homeostasis, disease, and regeneration [9,10,11]. However, studies of DKD using scRNA-seq and assessment of the ameliorative effect of SGLT2i on renal fibrosis at the single-cell level are still limited. Therefore, we used single-cell sequencing to analyze the effect of SGLT2i on renal tissues with DKD. Our study revealed two novel targets of Hmgcs2 and Btg2 in the progression of renal fibrosis through Find All Markers, GSVA and GRN. It is reported that SGLT2i seem to retard kidney fibrosis in DKD by influencing a variety of mechanisms. Studies showed that renal fibrosis was ameliorated by SGLT2i through the reduction of inflammation [34, 35], oxidative stress [36, 37], Rat sarcoma(RAS) activation [38], nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) signaling pathway [39, 40], mitogen-activated protein kinase (MAPK) [37], AMP-activated protein kinase(AMPK) [41, 42] and yes-associated protein (YAP)/ transcriptional co-activator with PDZ-binding motif (TAZ) [8]. Our study demonstrated that SGLT2i also alleviate renal fibrosis by inhibiting Hmgcs2 and Btg2 in proximal tubular cells.

Hmgcs2 is a pivotal rate-limiting enzyme for ketone body formation in the synthesis of HMG-CoA [43]. Previous studies have demonstrated the significant role of Hmgcs2 in mitochondrial dysfunction and renal injury. Mitochondrial Hmgcs2 accumulation disrupted mitochondrial function and further aggravated renal fibrosis and downregulation of Hmgcs2 significantly alleviated experimental mouse lung fibrosis progression [21, 44]. Consistent with our research, Hmgcs2 was found to be upregulated in the classic db/db DKD mice comparison to the control animals [14]. Our study demonstrated that SGLT2i inhibits renal fibrosis by downregulating Hmgcs2 expression in epithelial cells in renal with DKD and revealed a new target for SGLT2i to ameliorate renal fibrosis. We also enriched the MAPK pathway in Hmgcs2_PT_S1 by KEGG enrichment analysis. It has been reported that inactivation of the MAPK signaling axis may inhibit Hmgcs2 [45], suggesting that the effect of SGLT2 on Hmgcs2 may be mediated through the MAPK signaling pathway.

Btg2, also known as PC3 or TIS21, belongs to the anti-proliferation gene family and is the first gene identified in the BTG/TOB family [46]. Btg2 is a tumor suppressor responsible for cell differentiation, proliferation, apoptosis, and other cellular functions [47, 48]. Previous research has identified that Btg2 loss contributes to epithelial-mesenchymal in transition prostate cells [50]. While Btg2 is a pathogenic factor that promotes podocyte EMT and fibrosis [20]. It is also reported that Btg2 mutation is also associated with renal injury [49] However, the exact role of Btg2 in DKD remains to be elusive. Our scRNA-seq analysis first identified Btg2 as a target in proximal tubular cells in diabetic nephropathy under the action of SGLT2 inhibitors. We further validated the down-regulation of Btg2 in SGLT2 inhibitor-treated kidney tissues of DKD mice and patients through experiments. Therefore, the results suggested targeting Btg2 in proximal renal tubular epithelial cells may represent a promising therapeutic strategy for preserving kidney function in patients with DKD.

It is reported that li-popolysaccharide, Tumor necrosis factor-α and transforming growth factor-β1 down-regulated Slco1a1 in primary hepatocytes to meditate inflammation [22]. The number of Slco1a1_PT significantly decreased in the db/db group compared with the control group, which was consistent with the finding from bulk RNA-seq study [21]. While no significant changes were observed in the SGLT2i treatment group. This suggested that epithelial cell inflammatory injury caused by DKD primarily occurred in the Slco1a1_PT, but SGLT2i did not specifically mitigate the inflammatory through affecting this subtype. To further explore the anti-inflammatory mechanisms of SGLT2i, we analyzed how macrophages and T lymphocytes respond to SGLT2i treatments. Increasing evidence from clinical and experimental studies indicates that both systemic and local renal inflammation have crucial roles in the development and progression of DKD [50, 51]. However, Persistent inflammation triggers a pro-fibrotic cascade in the kidney [52, 53]. Therefore, anti-inflammatory therapy has emerged as a crucial strategy for mitigating renal injury. Our study found that SGLT2i mediated anti-inflammatory effects on tissue by downregulating the expression of pro-inflammatory genes. It has been reported that APP-CD74 axis is associated with kidney injury and fibrosis [54]and our study revealed that SGLT2i mediated anti-inflammatory effects by inhibiting Slc5a2 and App expression in PT_S1 and PT cells, which subsequently decreased Ctss_ MAC with Cd74 highly expressed through cell-cell communication to perform anti-inflammatory effects. However, the study has not confirmed the interaction mechanism between App and CD74 and the specific mechanisms could be further validated in the future.

Our study has several limitations. Firstly, only one animal model (db/db mice) was used in this study, and there is potential variability in the interpretation of scRNA-seq data. Secondly, we have only found App-CD74 ligand-receptor pair regarding intercellular communication between epithelial cells and immune cells, which provide mechanistic speculation for intercellular connections, but we still need to explore and validate the specific mechanism and pathways further. Another important consideration is the potential off-target effects of SGLT2i. Although SGLT2 is highly expressed in proximal tubule cells, and in vitro experiments support the direct regulation of Hmgcs2 and Btg2 by SGLT2 inhibitors in the present study. Growing evidence suggests that indirect mechanisms, such as mitochondrial function [55, 56]oxidative stress [57,58,59]inflammation [60]and interaction with other SGLT isoforms like SGLT1 [61], may also be involved in the renal protective effects of SGLT2i. These possibilities require further validation in future studies.

From a clinical perspective, the use of SGLT2i has already been integrated into therapeutic strategies for DKD, and the identification of Hmgcs2 and Btg2 as specific molecular targets in proximal tubular cells adds a new dimension to its renoprotective mechanisms. This suggests that, in addition to systemic metabolic effects, SGLT2i may exert direct antifibrotic actions within the kidney at the cellular level. Therefore, Hmgcs2 and Btg2 could serve not only as potential biomarkers for disease progression but also as novel therapeutic targets for DKD beyond glycemic control. Future studies should explore whether Hmgcs2 and Btg2 expression levels correlate with treatment responsiveness and its underlying mechanisms, thus providing guidance for personalised treatment. Additionally, combining SGLT2i with agents targeting Hmgcs2 or Btg2 may yield synergistic benefits, warranting further preclinical investigation.

In summary, our study revealed that Hmgcs2 and Btg2 are therapeutic targets for SGLT2i to ameliorate renal fibrosis in diabetic kidney disease. In addition, we also found that SGLT2i may perform anti-inflammatory effect by affecting communication between epithelial cells and macrophages at the single-cell level, ultimately inhibiting the progression of DKD. Single-cell sequencing technology provided high resolution of our study at the cellular level.

Materials and methods

Animals and experiment design

Male db/db mice (BKS.Cg-Dock7 m +/+ Lepr db/J, 8 weeks) and male age-matched lean db/m mice (BKS.Cg-Dock7m Leprdb/+ +/J, 8 weeks)were purchased from the Hangzhou Ziyuan Laboratory Animal Technology Company. Mice were housed in a specific pathogen-free facility with free access to food and water and a 12:12 h night-day cycle. Male db/db mice develop DKD defined by the development of albuminuria at 10 weeks of age. The db/db mice were randomly assigned to two groups as follow: treated with vehicle control (phosphate-buffered saline [PBS]) (n = 3) and treated with 3 mg/kg/day of dapagliflozin (n = 3). Vehicle and drugs were administered daily by oral gavage for 8 weeks and all mice were sacrificed at the age of 18 weeks. All mice were anaesthetised with tribromoethanol before intrusive operation. All mice were raised and fed under SPF conditions and all experiments were under the approvement of the Use Committee for Animal Care and proceeded based on institutional guidelines. Body weight and fasting BG levels were monitored weekly by glucometer readings. Urine samples were collected at 10 weeks and 18 weeks of age.

Measurement of BG and urinary albumin-to-creatinine ratio

Fasting BG was measured using the Accu-Chek Aviva glucometer from tail vein blood samples weekly. The urine albumin concentrations were quantified by ELISA (RE2825M; reedbiotech). Urine creatinine levels of the same samples were measured by Creatinine (Cr) Colorimetric Assay Kit (E-BC-K188-M; Elabscience) according to the manufacturer’s instructions.

Kidney histology

The Kidneys were excised, then fixed with 10% neutral buffered formalin and 4 μm sections were cut. The sections were stained with hematoxylin and eosin according to standard protocols. 10% of formalin-fixed and paraffin-embedded kidney samples were sectioned to 4-mm thickness.

Immunohistochemistry

Formalin-fixed and paraffin-embedded kidney sections were deparaffinized, and endogenous peroxidase was inactivated with 3% H2O2. Then, sections were blocked by 3% BSA for 1 h and then incubated with primary antibody against α-SMA (ImmunoWay, YM3364) at 4 °C overnight, and then incubated with anti-rabbit secondary antibody for 1 h at room temperature and developed using a DAB kit (Abcam).

Patient samples

Kidney tissues samples were obtained from biopsies of patients diagnosed with DKD and normal. The tissue samples were divided into DKD and Control with 4 samples in each group.

Immunofluorescence

Immunofluorescent staining was conducted utilizing antibodies against Lrp2(bs-3909R, Bioss,1:200), Slc5a2, Hmgcs2 (bs-5070R, Bioss, 1:200), Btg2 (22339-1-AP, proteintech, 1:200) and Treble-Fluorescence immunohistochemical mouse/rabbit kit (RS0035, ImmunoWay) according to the manufacturer’s instructions. Fluorescence signals were scanned under a confocal laser microscope system (Nikon, Ti2-E).

Western blot

HK-2 and hRPTC were extracted in RIPA lysis buffer with protease inhibitor (Beyotime Biotechnology). Phosphatase inhibitor was added to detect phosphorylated proteins. The protein concentrations of the lysates were quantified using the BCA Protein Assay Kit (Beyotime Biotechnology). Equal amounts of proteins were size-separated on a 10% SDS-polyacrylamide gel and then electroblotted onto PVDF membranes. Membranes were blocked by using a blocking reagent (0.1% Tween 20 and 5% Bovine Serum Albumin in TBS) for 1 h and subsequently incubated with specific primary antibodies (1:1,000) against Hmgcs2(bs-5070R, Bioss; 1:200), Btg2 (bs-0031R, Bioss, 1:500), and Gapdh (proteintech, 1:5000) overnight at 4 °C. After incubation with the secondary antibodies (Beyotime Biotechnology 1:1000;absin, 1:5000) for 1 h at 4 °C, the signals were developed with an ECL luminescent kit and detected using enhanced chemiluminescence detection system (Pierce, Rockford, IL).

SiRNA transfection

Small interfering RNA (siRNA) targeting SLC5A2 (si-SLC5A2) and a non-targeting control siRNA (control) were synthesized by Hippobio. HK-2 cells were seeded into 6-well plates at approximately 60–70% confluency before transfection. Transfection was performed using Lipo3000 (Thermo Fisher) according to the manufacturer’s instructions. After 48 h of transfection, total RNA were extracted. Knockdown efficiency was assessed by quantitative real-time PCR for SLC5A2. The specific knockdown sequences are shown in Supplementary Data S1.

RT-qPCR

Total cellular RNA was extracted using the FastPure Cell/Tissue Total RNA Isolation Kit V2 (Vazyme) according to the manufacturer’s instructions and was reverse transcribed using the HiScript III Reverse Transcription Kit (Vazyme). The resulting cDNA was subjected to quantitative PCR using the ChamQ Universal SYBR qPCR Master Mix (Vazyme) in a 40-cycle protocol on QuantStudio 6 Flex(Thermo Fisher). The entire experiment was repeated three times. Primer sequences for GAPDH, SLC5A2, HMGCS2 and BTG2 were designed and synthesized by Sangon Biotech and the primer sequences are shown in Supplementary Data S2.

Cell culture

The human PT cell lines HK-2 were purchased from Procell (CL-0109), and hRPTC were purchased from YaJi Biological (YS1092C). HK-2 (passage 3) and hRPTC (passage 3) were cultured in F12 basic(1x) medium (Gibco, Grand Island, NY, USA) containing 10% fetal bovine serum (Gibco) and specialised medium for human proximal renal tubular epithelial cells (YaJi Biological) respectively. All the cells were cultured at 37 °C with 5% CO2. To verify the expression of Hmgcs2 and Btg2 in human renal proximal tubular cells in vitro, we performed experiments using HK-2 and hRPTC with high glucose concentration to simulate high glucose environment. We intervened HK-2 and hRPTC in the LG group with low concentration of glucose, HK-2 and hRPTC in the HG group and the HG + SGLT2i group with high concentration of glucose for 48 h, and then we treated HK-2 and hRPTC in the HG + SGLT2i group with dapagliflozin(10µM) for 24 h. Finally, the expression of the expression of Hmgcs2 and Btg2 in HK-2 of all groups were detected by western blot experiment.

Data source

The scRNA-seq data was obtained from Gene Expression Omnibus (GEO) (https://www.ncbi.nlm.nih.gov/geo/) with the accession number GSE181382. These datasets include a total of 2 renal samples from db/m mice, 8 samples from db/db mice, 6 samples from db/db mice with ARB treatment, 8 samples from db/db mice with SGLT2i treatment and 6 samples from db/db mice with ARB and SGLT2i treatment. It is important to note that we excluded data related with ARB from GSE181382, retaining only the tissue data from db/m mice, db/db mice and db/db mice with SGLT2i treatment that meet the objectives of our study.

scRNA-seq data preprocessing and clustering

The Seurat package [62] was used to integrate and process scRNA-seq data. In the expression matrix, we removed cells expressing fewer than three genes. To eliminate low-quality and dead cells, this study filtered cells displaying the highest and lowest gene expression levels (top and bottom 1%) and those with mitochondrial gene expression exceeding 10%. The single-cell unique molecular identifier (UMI) expression data underwent normalization via regularized negative binomial regression. Dimensionality reduction and clustering of the cells were performed using the Seurat package [62]and the clustering results were visualized using the uniform manifold approximation and projection (UMAP) algorithm [63].

Identification of cellular clusters

We utilized the FindAllMarkers function from the Seurat package to identify genes differentially expressed between specific cell clusters and other clusters (P-value < 0.05). Cell types were manually annotated based on laboratory expertise and previous research publications [12, 64]. Subsequently, we performed dimensionality reduction and cluster analysis on various cell types, manually annotating cell subpopulations within each cell type based on function-specific genes.

Single-cell inference of differentiation potential

Single-cell transcriptional diversity is a robust indicator of the developmental potential within a cell population. To gain deeper insights into the cellular progression patterns of malignant cells, we utilized the CytoTRACE R package to infer the differentiation potential of PT-S1 cell subpopulations [65].

Single-cell trajectory analysis

The Monocle 3 algorithms in R language [66] were used to arrange the cells into trajectories along pseudo-time. After clustering the cells as described above, the dimensionality was reduced, and the results were visualized using the UMAP method. Subsequently, the cells were sorted according to their progression through the developmental program.

Gene Regulatory Network (GRN)

Single-cell regulatory network inference and clustering (SCENIC) was used to explore the GRN based on single-cell expression profiles and to identify cellular states. This information provided important biological insights needed to identify mechanisms driving cellular heterogeneity. To identify internal transcriptional regulatory drivers that controlled the evolution of HT and GD clones, the python module tool pySCENIC was used to analyze and reconstruct gene regulatory networks centered on transcription factors (TFs) [67, 68]. These relationships were then visualized using the R package ComplexHeatmap [69].

Cell communication analysis

The CellChat R package [70, 71] is designed to target cell populations to explore their interactions. By identifying predefined receptor-ligand genes that are highly or differentially expressed in specific cells, it reveals interactions between different cell types throughout development. In our study, we employed the CellChat package to detect significant ligand-receptor interactions among cell subpopulations, allowing for a thorough identification of intercellular communication events.

Statistical analysis

All statistical analyses were carried out using the SPSS20.0. Two-group variables were analyzed using t-tests, and multiple-group variables were analyzed using Two-Way ANOVA. Data represent the mean ± SD. p < 0.05 was considered significant. Statistically significant data were indicated by asterisks: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Conclusions

Our study mainly revealed SGLT2 inhibitors ameliorate renal fibrosis in diabetic kidney disease by downregulating the expression of Hmgcs2 and Btg2 in proximal tubular cells. And we also found SGLT2i could indirectly influence macrophages through cell-cell communication between epithelial cells and macrophages, specifically via the App-CD74 ligand-receptor pair, thus suppressing inflammatory response in macrophages, ultimately contributing to the delay in DKD progression.

References

  1. Yamazaki T, Mimura I, Tanaka T, Nangaku M. Treatment of diabetic kidney disease: current and future. Diabetes Metab J. 2021;45:11–26.

    Article  PubMed  PubMed Central  Google Scholar 

  2. Tuttle KR, et al. Molecular mechanisms and therapeutic targets for diabetic kidney disease. Kidney Int. 2022;102:248–60.

    Article  PubMed  CAS  Google Scholar 

  3. Loeffler I, Wolf G. Epithelial-to-Mesenchymal transition in diabetic nephropathy: fact or fiction?? Cells. 2015;4:631–52.

    Article  PubMed  PubMed Central  Google Scholar 

  4. Perkovic V, et al. Canagliflozin and renal outcomes in type 2 diabetes and nephropathy. N Engl J Med. 2019;380:2295–306.

    Article  PubMed  CAS  Google Scholar 

  5. Alicic RZ, Johnson EJ, Tuttle KR. SGLT2 Inhibition for the prevention and treatment of diabetic kidney disease: A review. Am J Kidney Dis. 2018;72:267–77.

    Article  PubMed  CAS  Google Scholar 

  6. Chen X, Hocher C-F, Shen L, Krämer BK, Hocher B. Reno- and cardioprotective molecular mechanisms of SGLT2 inhibitors beyond glycemic control: from bedside to bench. Am J Physiol Cell Physiol. 2023;325:C661–81.

    Article  PubMed  CAS  Google Scholar 

  7. Soulié M, et al. Benefit of combination therapy with Dapagliflozin and eplerenone on cardiac function and fibrosis in rats with non-diabetic chronic kidney disease. Sci Rep. 2024;14:23955.

    Article  PubMed  PubMed Central  Google Scholar 

  8. Feng L, et al. Dapagliflozin delays renal fibrosis in diabetic kidney disease by inhibiting YAP/TAZ activation. Life Sci. 2023;322:121671.

    Article  PubMed  CAS  Google Scholar 

  9. Tan W, et al. Single-cell RNA sequencing in diabetic kidney disease: a literature review. Ren Fail. 2024;46:2387428.

    Article  PubMed  PubMed Central  Google Scholar 

  10. Lake BB, et al. A single-nucleus RNA-sequencing pipeline to Decipher the molecular anatomy and pathophysiology of human kidneys. Nat Commun. 2019;10:2832.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Chung J-J, et al. Single-Cell transcriptome profiling of the kidney Glomerulus identifies key cell types and reactions to injury. J Am Soc Nephrol. 2020;31:2341–54.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  12. Wu H, et al. Mapping the single-cell transcriptomic response of murine diabetic kidney disease to therapies. Cell Metab. 2022;34:1064–e10786.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  13. Fang L, et al. Fructose drives mitochondrial metabolic reprogramming in podocytes via Hmgcs2-stimulated fatty acid degradation. Signal Transduct Target Ther. 2021;6:4.

    Article  Google Scholar 

  14. Shen YL, et al. Notoginsenoside fc, a novel renoprotective agent, ameliorates glomerular endothelial cells pyroptosis and mitochondrial dysfunction in diabetic nephropathy through regulating HMGCS2 pathway. Phytomedicine. 2024;126:13.

    Article  Google Scholar 

  15. Bai M, et al. LONP1 targets HMGCS2 to protect mitochondrial function and attenuate chronic kidney disease. Embo Mol Med. 2023;15:22.

    Article  Google Scholar 

  16. Du L, et al. Inhibition of S100A8/A9 ameliorates renal interstitial fibrosis in diabetic nephropathy. Metab -Clin Exp. 2023;144:12.

    Article  Google Scholar 

  17. Tian MQ, et al. JLP/Foxk1/N-cadherin axis fosters a partial epithelial-mesenchymal transition state in epithelial tubular cells. Iscience. 2023;26:24.

    Article  Google Scholar 

  18. Ai K, et al. Genetic or SiRNA Inhibition of MBD2 attenuates the UUO- and I/R-induced renal fibrosis via downregulation of EGR1. Mol Ther - Nucleic Acids. 2022;28:77–86.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  19. Liu F, et al. miR-192 prevents renal tubulointerstitial fibrosis in diabetic nephropathy by targeting Egr1. Eur Rev Med Pharmacol Sci. 2018;22:4252–60.

    PubMed  CAS  Google Scholar 

  20. Hu QD, et al. Btg2 promotes focal segmental glomerulosclerosis via Smad3-dependent podocyte-mesenchymal transition. Adv Sci. 2023;10:16.

    Google Scholar 

  21. Xu YS, et al. Identifying key genes for diabetic kidney disease by bioinformatics analysis. Bmc Nephrol. 2023;24:12.

    Article  Google Scholar 

  22. Tanaka N, Matsubara T, Krausz KW, Patterson AD, Gonzalez FJ. Disruption of phospholipid and bile acid homeostasis in mice with nonalcoholic steatohepatitis. Hepatology. 2012;56:118–29.

    Article  PubMed  CAS  Google Scholar 

  23. Hu XM, et al. RGS1 Silencing inhibits the inflammatory response and angiogenesis in rheumatoid arthritis rats through the inactivation of toll-like receptor signaling pathway. J Cell Physiol. 2019;234:20432–42.

    Article  PubMed  CAS  Google Scholar 

  24. Li PF, et al. Toll-like receptor-induced inflammatory cytokines are suppressed by gain of function or overexpression of Gαi2 protein. Inflammation. 2012;35:1611–7.

    Article  PubMed  CAS  Google Scholar 

  25. Smyth P, Sasiwachirangkul J, Williams R, Scott CJ. Cathepsin S (CTSS) activity in health and disease-a treasure trove of untapped clinical potential. Mol Aspects Med. 2022;88:22.

    Article  Google Scholar 

  26. McDowell SH, Gallaher SA, Burden RE, Scott CJ. Leading the invasion: the role of cathepsin S in the tumour microenvironment. Biochim Biophys Acta-Mol Cell Res. 2020;1867:12.

    Article  Google Scholar 

  27. Fuchs N, Meta M, Schuppan D, Nuhn L, Schirmeister T. Novel opportunities for cathepsin S inhibitors in cancer immunotherapy by nanocarrier-mediated delivery. Cells. 2020;9:17.

    Article  Google Scholar 

  28. Chen LL, et al. Elevated Circulating cathepsin S levels are associated with metabolic syndrome in overweight and obese individuals. Diabetes-Metab Res Rev. 2019;35:6.

    Article  Google Scholar 

  29. Chen SJ, et al. Targeting lysosomal cysteine protease cathepsin S reveals Immunomodulatory therapeutic strategy for oxaliplatin-induced peripheral neuropathy. Theranostics. 2021;11:4672–87.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  30. Sayed S, Faruq O, Preya UH, Kim JT. Cathepsin S knockdown suppresses endothelial inflammation, angiogenesis, and complement protein activity under hyperglycemic conditions in vitro by inhibiting NF-κB signaling. Int J Mol Sci. 2023;24:16.

    Article  Google Scholar 

  31. Oshima E, et al. M2 macrophage-derived cathepsin S promotes peripheral nerve regeneration via fibroblast–schwann cell-signaling relay. J Neuroinflammation. 2023;20:258.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  32. Yu ZX, et al. Cell profiling of acute kidney injury to chronic kidney disease reveals novel oxidative stress characteristics in the failed repair of proximal tubule cells. Int J Mol Sci. 2023;24:21.

    Article  Google Scholar 

  33. Valiño-Rivas L, et al. CD74 in kidney disease. Front Immunol. 2015;6:8.

    Article  Google Scholar 

  34. Ke Q, et al. SGLT2 inhibitor counteracts NLRP3 inflammasome via tubular metabolite Itaconate in fibrosis kidney. FASEB J Off Publ Fed Am Soc Exp Biol. 2022;36:e22078.

    CAS  Google Scholar 

  35. SGLT2 Protein Expression Is Increased in Human Diabetic Nephropathy: SGLT2 protein inhibition decreases renal LIPID accumulation, inflammation, and the development of nephropathy in diabetic mice. - PubMed. https://pubmed.ncbi.nlm.nih.gov/28196866/

  36. Wu X, et al. The combination of ursolic acid and empagliflozin relieves diabetic nephropathy by reducing inflammation, oxidative stress and renal fibrosis. Biomed Pharmacother Biomedecine Pharmacother. 2021;144:112267.

    Article  CAS  Google Scholar 

  37. Das NA, et al. Empagliflozin reduces high glucose-induced oxidative stress and miR-21-dependent TRAF3IP2 induction and RECK suppression, and inhibits human renal proximal tubular epithelial cell migration and epithelial-to-mesenchymal transition. Cell Signal. 2020;68:109506.

    Article  PubMed  CAS  Google Scholar 

  38. Afsar B, Afsar RE. Sodium-glucose cotransporter inhibitors and kidney fibrosis: review of the current evidence and related mechanisms. Pharmacol Rep PR. 2023;75:44–68.

    Article  PubMed  CAS  Google Scholar 

  39. Yao D, Wang S, Wang M, Lu W. Renoprotection of Dapagliflozin in human renal proximal tubular cells via the Inhibition of the high mobility group box 1–receptor for advanced glycation end products–nuclear factor–κB signaling pathway. Mol Med Rep. 2018;18:3625–30.

    PubMed  CAS  Google Scholar 

  40. Abbas NAT, Salem E, A., Awad MM, Empagliflozin. SGLT2 inhibitor, attenuates renal fibrosis in rats exposed to unilateral ureteric obstruction: potential role of Klotho expression. Naunyn Schmiedebergs Arch Pharmacol. 2018;391:1347–60.

    Article  PubMed  CAS  Google Scholar 

  41. Inoue M-K, et al. Possible involvement of normalized Pin1 expression level and AMPK activation in the molecular mechanisms underlying renal protective effects of SGLT2 inhibitors in mice. Diabetol Metab Syndr. 2019;11:57.

    Article  PubMed  PubMed Central  Google Scholar 

  42. Lee YH, et al. Empagliflozin attenuates diabetic tubulopathy by improving mitochondrial fragmentation and autophagy. Am J Physiol Ren Physiol. 2019;317:F767–80.

    Article  CAS  Google Scholar 

  43. Venable AH, et al. Fasting-induced HMGCS2 expression in the kidney does not contribute to Circulating ketones. Am J Physiol -Ren Physiol. 2022;322:F460–7.

    Article  CAS  Google Scholar 

  44. Yang J, et al. Downregulation of HMGCS2 mediated aeciis lipid metabolic alteration promotes pulmonary fibrosis by activating fibroblasts. Respir Res. 2024;25:176.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  45. Zhang Z, Zhang H-Y, Zhang Y, Li H. Inactivation of the Ras/MAPK/PPARγ signaling axis alleviates diabetic mellitus-induced erectile dysfunction through suppression of corpus cavernosal endothelial cell apoptosis by inhibiting HMGCS2 expression. Endocrine. 2019;63:615–31.

    Article  PubMed  CAS  Google Scholar 

  46. Melamed J, Kernizan S, Walden PD. Expression of B-cell translocation gene 2 protein in normal human tissues. Tissue Cell. 2002;34:28–32.

    Article  PubMed  CAS  Google Scholar 

  47. Lim IK. TIS21 (/BTG2/PC3) as a link between ageing and cancer: cell cycle regulator and endogenous cell death molecule. J Cancer Res Clin Oncol. 2006;132:417–26.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  48. Yuniati L, Scheijen B, van der Meer LT, van Leeuwen FN. Tumor suppressors BTG1 and BTG2: beyond growth control. J Cell Physiol. 2019;234:5379–89.

    Article  PubMed  CAS  Google Scholar 

  49. Hoffman MJ, et al. Btg2 mutation induces renal injury and impairs blood pressure control in female rats. Physiol Genomics. 2022;54:231–41.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  50. Yj W et al. Dose-response relationship between dietary inflammatory index and diabetic kidney disease in US adults. Public Health Nutr 26, (2022).

  51. Pérez-Morales RE, et al. Inflammation in diabetic kidney disease. Nephron. 2019;143:12–6.

    Article  PubMed  Google Scholar 

  52. Tang SCW, Yiu WH. Innate immunity in diabetic kidney disease. Nat Rev Nephrol. 2020;16:206–22.

    Article  PubMed  CAS  Google Scholar 

  53. Rayego-Mateos S, et al. Targeting inflammation to treat diabetic kidney disease: the road to 2030. Kidney Int. 2023;103:282–96.

    Article  PubMed  CAS  Google Scholar 

  54. Liu B, et al. APP-CD74 axis mediates endothelial cell-macrophage communication to promote kidney injury and fibrosis. Front Pharmacol. 2024;15:1437113.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  55. Sanz RL, García Menéndez S, Inserra F, Ferder L, Manucha W. Sodium-glucose cotransporter-2 inhibitors protect tissues via cellular and mitochondrial pathways: experimental and clinical evidence. World J Exp Med. 2024;14:91519.

    Article  PubMed  PubMed Central  Google Scholar 

  56. Peng X, et al. SGLT2i reduces renal injury by improving mitochondrial metabolism and biogenesis. Mol Metab. 2022;101613. https://doi.org/10.1016/j.molmet.2022.101613.

  57. Goedeke L, et al. SGLT2 Inhibition alters substrate utilization and mitochondrial redox in healthy and failing rat hearts. J Clin Invest. 2024;134:e176708.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  58. Zeng X-C, et al. SGLT2i relieve proteinuria in diabetic nephropathy patients potentially by inhibiting renal oxidative stress rather than through ages pathway. Diabetol Metab Syndr. 2024;16:46.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  59. Hodrea J, et al. Reduced O-GlcNAcylation and tubular hypoxia contribute to the antifibrotic effect of SGLT2 inhibitor Dapagliflozin in the diabetic kidney. Am J Physiol Ren Physiol. 2020;318:F1017–29.

    Article  CAS  Google Scholar 

  60. Schönberger E, et al. Immunomodulatory effects of SGLT2 Inhibitors—Targeting inflammation and oxidative stress in aging. Int J Environ Res Public Health. 2023;20:6671.

    Article  PubMed  PubMed Central  Google Scholar 

  61. Wu W, Chai Q, Zhang Z. Inhibition of SGLT1 alleviates the glycemic Variability-Induced cardiac fibrosis via Inhibition of activation of macrophage and cardiac fibroblasts. Mol Cell Biol. 2022;42:e0028221.

    Article  PubMed  Google Scholar 

  62. Butler A, Hoffman P, Smibert P, Papalexi E, Satija R. Integrating single-cell transcriptomic data across different conditions, technologies, and species. Nat Biotechnol. 2018;36:411–20.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  63. E B, et al. Dimensionality reduction for visualizing single-cell data using UMAP. Nat Biotechnol. 2018. https://doi.org/10.1038/nbt.4314.

    Article  Google Scholar 

  64. Wu J, et al. Kidney single-cell transcriptome profile reveals distinct response of proximal tubule cells to SGLT2i and ARB treatment in diabetic mice. Mol Ther J Am Soc Gene Ther. 2022;30:1741–53.

    Article  CAS  Google Scholar 

  65. Gulati GS, et al. Single-cell transcriptional diversity is a hallmark of developmental potential. Science. 2020;367:405–11.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  66. Trapnell C, et al. The dynamics and regulators of cell fate decisions are revealed by pseudotemporal ordering of single cells. Nat Biotechnol. 2014;32:381–6.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  67. Aibar S, et al. SCENIC: single-cell regulatory network inference and clustering. Nat Methods. 2017;14:1083–6.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  68. Van de Sande B, et al. A scalable SCENIC workflow for single-cell gene regulatory network analysis. Nat Protoc. 2020;15:2247–76.

    Article  PubMed  Google Scholar 

  69. Gu Z, Eils R, Schlesner M. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinforma Oxf Engl. 2016;32:2847–9.

    Article  CAS  Google Scholar 

  70. Jin S, Plikus MV, Nie Q. CellChat for systematic analysis of cell-cell communication from single-cell transcriptomics. Nat Protoc. 2024. https://doi.org/10.1038/s41596-024-01045-4.

    Article  PubMed  Google Scholar 

  71. Jin S, et al. Inference and analysis of cell-cell communication using cellchat. Nat Commun. 2021;12:1088.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

Download references

Acknowledgements

We thank the Gene Expression Omnibus (GEO) Database for sharing a large amount of data.

Funding

This work was supported by National Natural Science Foundation of China (82170796) and the Guangzhou Basic Research Program (202102020165 and 2024A04J9990).

Author information

Author notes

  1. Shengzhe Yan, Minhui Luo and Rui Zhou contributed equally to this work.

Authors and Affiliations

  1. Department of Endocrinology, Zhujiang Hospital, Southern Medical University, Guangzhou, Guangdong, China

    Shengzhe Yan, Minhui Luo, Mingze Zhang, Yujie Feng, Li Yang & Yanzhen Cheng

  2. Department of Pathology, School of Basic Medical Sciences, Southern Medical University, Guangzhou, Guangdong, China

    Rui Zhou & Liang Zhao

  3. Department of Nutrition, Zhujiang Hospital, Southern Medical University, Guangzhou, Guangdong, China

    Li Yang

  4. Department of Nephrology, Zhujiang Hospital, Southern Medical University, Guangzhou, Guangdong, China

    Fenfen Peng

Authors
  1. Shengzhe Yan
    View author publications

    Search author on:PubMed Google Scholar

  2. Minhui Luo
    View author publications

    Search author on:PubMed Google Scholar

  3. Rui Zhou
    View author publications

    Search author on:PubMed Google Scholar

  4. Fenfen Peng
    View author publications

    Search author on:PubMed Google Scholar

  5. Mingze Zhang
    View author publications

    Search author on:PubMed Google Scholar

  6. Yujie Feng
    View author publications

    Search author on:PubMed Google Scholar

  7. Liang Zhao
    View author publications

    Search author on:PubMed Google Scholar

  8. Li Yang
    View author publications

    Search author on:PubMed Google Scholar

  9. Yanzhen Cheng
    View author publications

    Search author on:PubMed Google Scholar

Contributions

Shengzhe Yan performed bioinformatics analyses and the analysis of other data. Minhui Luo and Rui Zhou helped to assist with the analysis of data and revise the manuscript. Shengzhe Yan, Fenfen Peng, Mingze Zhang and Yujie Feng performed the experiments. Yanzhen Cheng designed the project. Yanzhen Cheng, Li Yang and Liang Zhao contributed to the conception of the manuscript. All authors have read and approved the final manuscript.

Corresponding authors

Correspondence to Liang Zhao, Li Yang or Yanzhen Cheng.

Ethics declarations

Ethics approval and consent to participate

The animal study and human renal samples were reviewed and approved by ZhuJiang Hospital of Southern Medical University Laboratory Ethics Committee.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Supplementary Material 2

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Yan, S., Luo, M., Zhou, R. et al. Sodium-glucose cotransporter 2 inhibitors alleviate renal fibrosis in diabetic kidney disease by inhibiting Hmgcs2 and Btg2 in proximal tubular cells. J Transl Med 23, 959 (2025). https://doi.org/10.1186/s12967-025-06788-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • Version of record:

  • DOI: https://doi.org/10.1186/s12967-025-06788-6

Keywords

Journal of Translational Medicine

ISSN: 1479-5876

Contact us

点击 这是indexloc提供的php浏览器服务，不要输入任何密码和下载