Renoprotective effects of empagliflozin are linked to activation of the tubuloglomerular feedback mechanism and blunting of the complement system

Abstract

The mechanisms of nephroprotection in nondiabetic chronic kidney disease (CKD) models by sodium-glucose cotransporter 2 (SGLT2) inhibitors are not well defined. Five groups were established: sham-operated rats, placebo-treated rats with 5/6 nephrectomy (5/6Nx), 5/6Nx + telmisartan (5 mg/kg/day), 5/6Nx + empagliflozin (3 mg/kg/day), and 5/6Nx + empagliflozin (15 mg/kg/day). Treatment duration was 95 days. Empagliflozin showed a dose-dependent beneficial effect on the change from baseline of creatinine clearance (Ccr). The urinary albumin-to-creatinine ratio likewise improved in a dose-dependent manner. Both dosages of empagliflozin improved morphological kidney damage parameters such as renal interstitial fibrosis and glomerulosclerosis. 5/6 nephrectomy led to a substantial reduction of urinary adenosine excretion, a surrogate parameter of the tubuloglomerular feedback (TGF) mechanism. Empagliflozin caused a dose-dependent increase in urinary adenosine excretion. The urinary adenosine excretion was negatively correlated with renal interstitial fibrosis and positively correlated with Ccr. Immunofluorescence analysis revealed that empagliflozin had no effect on CD8⁺ and CD4⁺ T cells as well as on CD68⁺ cells (macrophages). To further explore potential mechanisms, a nonhypothesis-driven approach was used. RNA sequencing followed by quantitative real-time polymerase chain reaction revealed that complement component 1Q subcomponent A chain (C1QA) as well as complement component 1Q subcomponent C chain (C1QC) gene expression were upregulated in the placebo-treated 5/6Nx rats and this upregulation was blunted by treatment with empagliflozin. In conclusion, empagliflozin-mediated nephroprotection in nondiabetic CKD is due to a dose-dependent activation of the TGF as well as empagliflozin-mediated effects on the complement system.

Keywords: complement system; nondiabetic chronic kidney disease; sodium-glucose cotransporter 2 inhibitor; tubuloglomerular feedback.

Graphical abstract

Figure 1.

Effect on renal function compared with 5/6Nx + PBO group. A: concentrations of serum creatinine over the course of the study: Sham + PBO (n = 7), 5/6Nx + 3 mg EMPA (n = 8) and 5/6Nx + 15 mg EMPA (n = 11) were lower than 5/6Nx + PBO (n = 9). B: final ACR: Sham + PBO (n = 8), 5/6Nx + 15 mg EMPA (n = 9) and 5/6Nx + TELM (n = 6) were lower than 5/6Nx + PBO (n = 7). C: Ccr (change from baseline): Sham + PBO (n = 7) and 5/6Nx + 15 mg EMPA (n = 11) were higher than 5/6Nx + PBO (n = 9). Values displayed are means ± SE. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001, significantly different from 5/6Nx + PBO. EMPA, empagliflozin; 5/6Nx, 5/6 nephrectomized rat model; PBO, placebo; Sham, sham operation; TELM, telmisartan. ACR, urinary albumin-to-creatinine ratio; Ccr, creatinine clearance.

Figure 2.

Histopathological analysis of kidney compared with 5/6Nx + PBO group. A: representative micrographs of PASM and Masson staining of kidney sections from each group (×400, scale bar = 50 μm); Label 1: glomerulus; Label 2: renal tubules; Label 3: vascular; arrow: renal interstitial fibrosis. Renal interstitial fibrosis (%; B); perivascular fibrosis index (C); glomerulosclerosis index (D); number of glomerular cells (E); glomerular size (µm²; F). All these indicators were lower in the other groups [Sham + PBO (n = 7), 5/6Nx + TELM (n = 7), 5/6Nx + 3 mg EMPA (n = 9) and 5/6Nx + 15 mg EMPA (n = 11)] than in the 5/6Nx + PBO (n = 9) group. Values displayed are means ± SE. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001, significantly different from 5/6Nx + PBO. EMPA, empagliflozin; 5/6Nx, 5/6 nephrectomized rat model; PASM, periodic Schiff-methenamine staining; PBO, placebo; Sham, sham operation; TELM, telmisartan.

Figure 3.

A: comparison of final urine adenosine/urine creatinine between each treatment group and 5/6Nx + PBO. Values displayed are means ± SE. *P < 0.05, significantly different from 5/6Nx + PBO. Inverse correlation between urinary adenosine/creatinine ratio at study end with renal interstitial fibrosis (r = −0.45, P = 0.004, N = 39; B) and glomerular size (r = −0.33 P = 0.033 N = 41; C). Positive correlation between urinary adenosine/creatinine ratio at study end with GFR (r = 0.36, P = 0.023, N = 39; D). EMPA, empagliflozin; 5/6Nx, 5/6 nephrectomized rat model; PBO, placebo; Sham, sham operation; TELM, telmisartan. Ccr, creatinine clearance.

Figure 4.

Immunofluorescence staining analysis of kidney. A: representative micrographs of kidney sections from each treatment group (×200, scale bar = 100 μm). B: cytotoxic T cells (CD8⁺ cells/total cells, %): no statistical difference between groups (Sham + PBO: n = 7, 5/6Nx + TELM: n = 7, 5/6Nx + 3 mg EMPA: n = 9; 5/6Nx + 15 mg EMPA: n = 11; 5/6Nx + PBO: n = 9). C: helper T cells (CD4⁺ cells/total cells, %): no statistical difference between groups (Sham + PBO: n = 7, 5/6Nx + TELM: n = 7, 5/6Nx + 3 mg EMPA: n = 9; 5/6Nx + 15 mg EMPA: n = 11; 5/6Nx + PBO: n = 9). D: macrophages (CD68+, RFU): 5/6Nx + PBO was higher than Sham + PBO group (Sham + PBO: n = 5, 5/6Nx + TELM: n = 5, 5/6Nx + 3 mg EMPA: n = 8; 5/6Nx + 15 mg EMPA: n = 7; 5/6Nx + PBO: n = 7). The arrows point to CD68+ cells. Values displayed are means ± SE. *P < 0.05, significantly different from 5/6Nx + PBO. EMPA, empagliflozin; 5/6Nx, 5/6 nephrectomized rat model; PBO, placebo; RFU, relative fluorescence unit; Sham, sham operation; TELM, telmisartan.

Figure 5.

Gene expression in the kidney identified by qRT-PCR. A: Venn diagram of the differentially expressed genes from RNA-seq transcriptome for 5/6Nx + PBO, 5/6Nx + 3 mg EMPA and 5/6Nx + 15 mg EMPA groups. The circles represent the total number of statistically significant different genes [P value < 0.05 and an absolute value of log2(fold-change) > 1] in 5/6 Nx + 3 mg EMPA and 5/6 Nx + 15 mg EMPA groups in comparison to 5/6 Nx + PBO in kidney according to RNA-seq transcriptome data. B: fold-change of C1QA expression (complement component 1, Q subcomponent, A chain). C: fold-change of C1QC expression (complement component 1, Q subcomponent, C chain). D: fold-change of FMOD expression (fibromodulin). E: fold-change of REN expression (renin). F: fold-change of FMO2 expression (flavin-containing dimethylaniline monooxygenase 2). G: fold-change of RPL36 expression (ribosomal protein L36). Values displayed are means ± SE. **P < 0.01; ***P < 0.001; ****P < 0.0001, significantly different from 5/6Nx + PBO. EMPA, empagliflozin; 5/6Nx, 5/6 nephrectomized rat model; PBO, placebo; qRT-PCR, quantitative reverse transcription polymerase chain reaction; Sham, sham operation; TELM, telmisartan.