Phosphodiesterase-5 inhibition preserves renal hemodynamics and function in mice with diabetic kidney disease by modulating miR-22 and BMP7

Abstract

Diabetic Nephropathy (DN) is the leading cause of end-stage renal disease. Preclinical and experimental studies show that PDE5 inhibitors (PDE5is) exert protective effects in DN improving perivascular inflammation. Using a mouse model of diabetic kidney injury we investigated the protective proprieties of PDE5is on renal hemodynamics and the molecular mechanisms involved. PDE5i treatment prevented the development of DN-related hypertension (P < 0.001), the increase of urine albumin creatinine ratio (P < 0.01), the fall in glomerular filtration rate (P < 0.001), and improved renal resistive index (P < 0.001) and kidney microcirculation. Moreover PDE5i attenuated the rise of nephropathy biomarkers, soluble urokinase-type plasminogen activator receptor, suPAR and neutrophil gelatinase-associated lipocalin, NGAL. In treated animals, blood vessel perfusion was improved and vascular leakage reduced, suggesting preserved renal endothelium integrity, as confirmed by higher capillary density, number of CD31⁺ cells and pericyte coverage. Analysis of the mechanisms involved revealed the induction of bone morphogenetic protein-7 (BMP7) expression, a critical regulator of angiogenesis and kidney homeostasis, through a PDE5i-dependent downregulation of miR-22. In conclusion PDE5i slows the progression of DN in mice, improving hemodynamic parameters and vessel integrity. Regulation of miR-22/BMP7, an unknown mechanism of PDE5is in nephrovascular protection, might represent a novel therapeutic option for treatment of diabetic complications.

Figure 1

LS mean change from the baseline in: (A) plasma glucose, (B) urine glucose, (C) weight, (D) triglycerides, (E) MAP, (F) HR in CTRL, SILD, STZ and STZ + SILD mice. Error bars represent upper and lower 95% CI; *P < 0.05, ***P < 0.001 vs. CTRL; ^#P < 0.05, ^###P < 0.001 vs. SILD; ^•••P < 0.001 vs. STZ.

Figure 2

LS mean change from the baseline in: (A) creatinine, (B) ACR in CTRL, SILD, STZ and STZ + SILD mice. Error bars represent upper and lower 95% CI; **P < 0.01, ***P < 0.001 vs. CTRL; ^##P < 0.01, ^###P < 0.001 vs. SILD; ^••P < 0.01, ^•••P < 0.001 vs. STZ. (C) quantification (mean ± SE) of ACR over time course, in STZ and STZ + SILD groups, ^••P < 0.05, ^••P < 0.01 between groups. LS mean change from the baseline of (D) GFR, and (E) suPAR, in CTRL, SILD, STZ and STZ + SILD mice. Error bars represent upper and lower 95% CI; *P < 0.05, ***P < 0.001 vs. CTRL; ^###P < 0.001 vs. SILD; ^•••P < 0.001 vs. STZ. (F) relative expression of NGAL, mean ± SD, in CTRL, SILD, STZ and STZ + SILD mice. *P < 0.05 vs. CTRL; ^#P < 0.05 vs. SILD; ^•P < 0.05 vs. STZ.

Figure 3

(A) representation of RDU image, RRI measurement (on the right) and volume measurement (on the left); (B) LS mean change from baseline of RRI in CTRL, SILD, STZ and STZ + SILD mice. Error bars represent upper and lower 95% CI; ***P < 0.001 vs. CTRL; ^#P < 0.05, ^###P < 0.001 vs. SILD; ^•••P < 0.001 vs. STZ. (C) correlation between mean change from the baseline in GFR (Y axis) and in RRI (X axis), r = −0.766, P < 0.001; (D) LS mean change from baseline of renal volume in CTRL, SILD, STZ and STZ + SILD mice. Error bars represent upper and lower 95% CI; ^•P < 0.05 vs. STZ. (E) kidney weight/tibial length ratio, mean ± SD, in CTRL, SILD, STZ and STZ + SILD mice. *P < 0.05, vs. CTRL; ^#P < 0.05 vs. SILD; ^•P < 0.05 vs. STZ.

Figure 4

(A) representative photomicrographs of PAS-stained kidneys of STZ and STZ + SILD mice,bar represent 60 μm; (B) glomerular diameters, mean ± SD, in STZ and STZ + SILD mice. ^••P < 0.01; (C) correlation between glomerular diameter (Y axis) and RRI (X axis) at the end of observation period, r = −0.712, P = 0.002; (D) relative quantification, mean ± SD, of mesangial matrix, mesangial proliferation, vessel dilatation in STZ and STZ + SILD mice, ^•P < 0.05; (E) quantification, mean ± SD, of Evan’s Blue dye/mg tissue in CTRL, SILD, STZ and STZ + SILD mice, **P < 0.01 vs. CTRL; ^##P < 0.01 vs. SILD; ^••P < 0.01 vs. STZ. (F) rapresentation of vascular permeability express as degree of extravasation (+) by FITC dextran in CTRL, SILD, STZ and STZ + SILD mice, bar represent 30 μm.

Figure 5

(A) Immunofluorescence of CD31⁺ areas in CTRL, SILD, STZ and STZ + SILD mice, bar represent 60 μm; (B) Quantification, mean ± SD, of capillary density, expressed as percentage of CD31⁺ area in CTRL, SILD, STZ and STZ + SILD mice. *P < 0.05 vs. CTRL; ^##P < 0.01 vs. SILD; ^•P < 0.05 vs. STZ; (C) Immunofluorescence of CD31⁺ (green) and NG2⁺ (red) areas in CTRL, SILD, STZ and STZ + SILD mice, bar represent 60 μm; (D) Quantification, mean ± SD, of pericytes coverage, expressed as percentage of ratio NG2⁺/CD31⁺ area in CTRL, SILD, STZ and STZ + SILD mice. **P < 0.01 vs. CTRL; ^##P < 0.01 vs. SILD; ^••P < 0.01 vs. STZ; (E) Representative gating of CD31⁺CD45⁻ cells. (F) Percentage of CD31⁺CD45⁻ EC in kidney, before and after STZ induction in CTRL, SILD, STZ and STZ + SILD mice. **P < 0.01 between STZ and CTRL, SILD and STZ + SILD groups.

Figure 6

(A) Relative quantification, mean ± SD, of BMP7/18S by qPCR in kidney of CTRL, SILD, STZ and STZ + SILD mice. *P < 0.05 vs. CTRL, ^#P < 0.05 vs. SILD, ^•P < 0.05 vs. STZ; (B) Relative quantification, mean ± SD, of TGFβ/18S by qPCR in kidney of CTRL, SILD, STZ and STZ + SILD mice. *P < 0.05 vs. CTRL, ^#P < 0.05 vs. SILD, ^•P < 0.05 vs. STZ; (C) Relative quantification, mean ± SD, of miR-22/U6B by qPCR in kidney of CTRL, SILD, STZ and STZ + SILD mice. *P < 0.05 and ***P < 0.001 vs. CTRL, ^###P < 0.001 vs. SILD, ^•••P < 0.001 vs. STZ; (D) Relative quantification, mean ± SD, of miR-22/U6B by qPCR in HUVEC untreated, treated with sildenafil, PKG-PDE5-transduced, PGK-PDE5-transduced and treated with sildenafil. **P < 0.01 vs. CTRL, ^###P < 0.001 vs. SILD, ^•••P < 0.001 vs. PGK-PDE5.