Dapagliflozin alleviates renal inflammation and protects against diabetic kidney diseases, both dependent and independent of blood glucose levels

Abstract

Introduction: Diabetic kidney disease (DKD) has become the leading cause of end-stage renal disease worldwide. Therefore, efforts to understand DKD pathophysiology and prevent its development at the early phase are highly warranted.

Methods: Here, we analyzed kidneys from healthy mice, diabetic mice, and diabetic mice treated with the sodium-glucose cotransporter 2 inhibitor dapagliflozin using ATAC and RNA sequencing. The findings were verified at the protein levels and in cultured cells.

Results: Our combined method of ATAC and RNA sequencing revealed Csf2rb, Btla, and Isg15 as the key candidate genes associated with hyperglycemia, azotemia, and albuminuria. Their protein levels were altered together with multiple other inflammatory cytokines in the diabetic kidney, which was alleviated by dapagliflozin treatment. Cell culture of immortalized renal tubular cells and macrophages unraveled that dapagliflozin could directly effect on these cells in vitro as an anti-inflammatory agent independent of glucose concentrations. We further proved that dapagliflozin attenuated ischemia/reperfusion-induced chronic kidney injury and renal inflammation in mice.

Discussion: Overall, our data emphasize the importance of inflammatory factors to the pathogenesis of DKD, and provide valuable mechanistic insights into the renoprotective role of dapagliflozin.

Keywords: diabetic kidney disease; inflammation; macrophage; renal tubular cell; sodium-glucose cotransporter 2 inhibitors.

Figure 1

Dapagliflozin alleviated hyperglycemia and kidney injury in type 2 diabetic mice in vivo. (A) Body weight of normal (Ctrl), diabetic (Case), and dapagliflozin-treated diabetic (Dapa) 14-week-old mice. (B) Blood glucose levels of 14-week-old mice in the Ctrl, Case, and Dapa groups. (C) Serum creatinine levels of 14-week-old mice in the Ctrl, Case, and Dapa groups. (D) Blood urea nitrogen levels of 14-week-old mice in the Ctrl, Case, and Dapa groups. (E) Urinary albumin creatinine ratio of 14-week-old mice in the Ctrl, Case, and Dapa groups. (F) Periodic acid–Schiff (PAS)–stained sections of 14-week-old mouse kidneys in the Ctrl, Case, and Dapa groups. NS, not significant. *P < 0.05, **P < 0.01, ***P < 0.001, one-way ANOVA test. Bar = 20 μm.

Figure 2

RNA sequencing revealed differentially expressed genes and pathways in diabetic kidney disease and dapagliflozin treatment. (A) Heatmap indicating distinct gene expression patterns of normal (Con1-5) versus diabetic (Db1-6) mouse kidneys. (B) KEGG classification analysis suggested that DEGs between normal and diabetic mouse kidneys were enriched in the immune system (red). (C) KEGG enrichment analysis indicating the top 30 pathways enriched in DEGs between normal and diabetic mouse kidneys. (D) GO enrichment analysis indicating the top 30 pathways enriched in DEGs between normal and diabetic mouse kidneys, highlighting interferon-related pathways (red). (E) Heatmap indicating distinct gene expression patterns of diabetic (Db1-6) versus dapagliflozin-treated diabetic (Db7-12) mouse kidneys. (F) KEGG classification analysis suggesting DEGs between diabetic and dapagliflozin-treated diabetic mouse kidneys were enriched in the immune system (red). (G) KEGG enrichment analysis indicating the top 30 pathways enriched in DEGs between diabetic and dapagliflozin-treated diabetic mouse kidneys. (H) GO enrichment analysis indicating the top 30 pathways enriched in DEGs between diabetic and dapagliflozin-treated diabetic mouse kidneys, highlighting interferon-related pathways (red).

Figure 3

K-means analysis of RNA sequencing results. K-means analysis yielded 10 clusters with different expression patterns among the Ctrl, Case, and Dapa groups. Cluster 2 and 3 showed apparently different expression patterns among Ctrl, Case, and Dapa groups. (A) Expression patterns of genes in Cluster 2 with the top 10 enriched pathways listed according to GO terms. (B) Expression patterns of genes in Cluster 3 with the top 10 enriched pathways listed according to GO terms. (C) Expression patterns of genes in Cluster 2 with the top 10 enriched pathways listed according to the KEGG database. (D) Expression patterns of genes in Cluster 3 with the top 10 enriched pathways listed according to the KEGG database.

Figure 4

Combined ATAC and RNA sequencing revealed potential candidate genes for diabetic kidney disease. (A) Violin plot of RNA sequencing indicating Isg15, Csf2rb, and Btla as potential candidate genes that were upregulated in diabetic mouse kidneys and downregulated by dapagliflozin treatment (Isg15 and Csf2rb), or downregulated in diabetic mouse kidneys and upregulated by dapagliflozin treatment (Btla). (B) ATAC analysis indicating elevated chromatin accessibility of Isg15 in diabetic mouse kidneys that was reduced by dapagliflozin treatment, with corresponding quantifications in (C). (D) ATAC analysis indicating decreased chromatin accessibility of Btla in diabetic mouse kidneys that was augmented by dapagliflozin treatment, with corresponding quantifications in (E). (F) ATAC analysis indicating elevated chromatin accessibility of Csf2rb in diabetic mouse kidneys that was reduced by dapagliflozin treatment, with corresponding quantifications in (G).

Figure 5

Isg15, Csf2rb, and Btla expression levels were associated with clinical features of diabetic kidney disease. (A) Univariate analysis demonstrating a positive correlation between Isg15 expression levels and blood glucose levels. (B) Univariate analysis demonstrating a positive correlation between Isg15 expression levels and urinary albumin creatinine ratio. (C) Univariate analysis demonstrating a negative correlation between Btla expression levels and blood glucose levels. (D) Univariate analysis demonstrating a negative correlation between Btla expression levels and serum creatinine levels. (E) Univariate analysis demonstrating a negative correlation between Btla expression levels and urinary albumin creatinine ratio. (F) Univariate analysis demonstrating a positive correlation between Csf2rb expression levels and blood glucose levels. (G) Univariate analysis demonstrating a positive correlation between Csf2rb expression levels and serum urea nitrogen levels. (H) Univariate analysis demonstrating a positive correlation between Csf2rb expression levels and urinary albumin creatinine ratio.

Figure 6

Dapagliflozin regulated the expression of candidate genes and attenuated renal inflammation in type 2 diabetic mice in vivo. (A) Western blot analysis showing BTLA and ISG15 protein levels in normal (Ctrl), diabetic (Case), and dapagliflozin-treated diabetic (Dapa) mouse kidneys, with corresponding quantifications in (B). (C) Scatter plot indicating distinct protein expression levels of normal versus diabetic mouse kidneys. (D) GO enrichment analysis suggesting leukocyte activation and JAK-STAT pathway activation in diabetic mouse kidneys. (E) Scatter plot indicating distinct protein expression levels of diabetic versus dapagliflozin-treated diabetic mouse kidneys. (F) Heatmap showing different protein levels of inflammatory cytokines in diabetic (Db) and dapagliflozin-treated diabetic (Dapa) mouse kidneys. *P < 0.05, ***P < 0.001, one-way ANOVA test.

Figure 7

Dapagliflozin presented an anti-inflammatory and anti-fibrotic effect on human renal tubular cells independent of glucose concentrations in vitro. (A) QPCR analysis showing Btla and Isg15 mRNA levels in HK-2 cells cultured in normal-glucose (Ctrl), high-glucose (Case), and high-glucose with dapagliflozin (Dapa) medium. (B) Western blot analysis showing BTLA and ISG15 protein levels in HK-2 cells of Ctrl, Case, and Dapa groups, with corresponding quantifications in (C). (D) Scatter plot indicating distinct protein expression levels of HK-2 cells in Ctrl versus Case group. (E) GO enrichment analysis suggesting STAT pathway activation, inflammatory response, and leukocyte activation pathways in HK-2 cells in the high-glucose medium. (F) Scatter plot indicating distinct protein expression levels of HK-2 cells in Case versus Dapa group. (G) Heatmap showing different protein levels of inflammatory and fibrotic cytokines in HK-2 cells of Case and Dapa group. NS, not significant. *P < 0.05, **P < 0.01, ***P < 0.001, one-way ANOVA test.

Figure 8

A novel anti-inflammatory role of dapagliflozin in macrophages independent of glucose concentrations in vitro. (A) QPCR analysis showing Csf2rb, Btla, and Isg15 mRNA levels in RAW 264.7 cells cultured in normal-glucose (Ctrl), high-glucose (Case), and high-glucose with dapagliflozin (Dapa) medium. (B) Western blot analysis showing CSF2RB, BTLA, and ISG15 protein levels in macrophages of Ctrl, Case, and Dapa groups, with corresponding quantifications in (C). (D) Scatter plot indicating distinct protein expression levels of macrophages in Ctrl versus Case group. (E) GO enrichment analysis suggesting endothelial cell apoptosis and leukocyte chemotaxis pathways in macrophages in the high-glucose medium. (F) Scatter plot indicating distinct protein expression levels of macrophages in Case versus Dapa group. (G) QPCR analysis showing Tnf-α, Il-1β, iNos, Il-10, and Tgf-β mRNA levels in macrophages of Ctrl, Case, and Dapa group. NS, not significant. *P < 0.05, **P < 0.01, ***P < 0.001, one-way ANOVA test.

Figure 9

Dapagliflozin attenuated kidney injury and alleviated renal inflammation in an ischemia/reperfusion-induced mouse CKD model. (A) Serum creatinine levels of normal (Ctrl), dapagliflozin-treated normal (Ctrl+Dapa), ischemia/reperfusion-operated (I/R), and dapagliflozin-treated I/R (I/R+Dapa) 14-week-old mice. (B) Blood urea nitrogen levels of 14-week-old mice in the Ctrl, Ctrl+Dapa, I/R, and I/R+Dapa groups. (C) blood glucose levels of 14-week-old mice in the Ctrl, Ctrl+Dapa, I/R, and I/R+Dapa groups. (D) Western blot analysis showing BTLA and ISG15 protein levels in mouse kidneys in the Ctrl, Ctrl+Dapa, I/R, and I/R+Dapa groups, with corresponding quantifications in (E). NS, not significant. *P < 0.05, **P < 0.01, ***P < 0.001, one-way ANOVA test.