Vasohibin-1, a negative feedback regulator of angiogenesis, ameliorates renal alterations in a mouse model of diabetic nephropathy

Abstract

Objective: The involvement of proangiogenic factors such as vascular endothelial growth factor as well as the therapeutic efficacy of angiogenesis inhibitors in early diabetic nephropathy has been reported. Vasohibin-1 (VASH-1) is a unique endogenous angiogenesis inhibitor that is induced in endothelial cells by proangiogenic factors. We investigated the therapeutic efficacy of VASH-1 in an early diabetic nephropathy model.

Research design and methods: Streptozotocin- induced type 1 diabetic mice received intravenous injections of adenoviral vectors encoding VASH-1 (AdhVASH-1) or beta-gal (AdLacZ) every other week and were killed after 28 days.

Results: Treatment with AdhVASH-1 resulted in sustained increase in the protein levels of VASH-1 in the liver and sera, in the absence of any inflammatory alterations. AdhVASH-1 treatment significantly suppressed renal hypertrophy, glomerular hypertrophy, glomerular hyperfiltration, albuminuria, increase of the CD31(+) glomerular endothelial area, F4/80(+) monocyte/macrophage infiltration, the accumulation of type IV collagen, and mesangial matrix compared with AdLacZ-treated diabetic mice. Increase in the renal levels of transforming growth factor-beta1, monocyte chemoattractant protein-1, and receptor for advanced glycation end products in diabetic animals was significantly suppressed by AdhVASH-1 (real-time PCR and immunoblot). VASH-1 significantly suppressed the increase of transforming growth factor-beta, monocyte chemoattractant protein-1, and receptor for advanced glycation end products, induced by high ambient glucose in cultured mouse mesangial cells. Increased phosphorylation of VEGFR2 was suppressed in AdVASH-1-treated diabetic animals and in cultured glomerular endothelial cells. Endogenous mouse VASH-1 was localized to the mesangial and endothelial area in glomeruli of diabetic mice.

Conclusions: These results suggest the potential therapeutic efficacy of VASH-1 in treating early diabetic nephropathy potentially mediated via glomerular endothelial and mesangial cells.

FIG. 1.

A: Immunoblot analysis. Immunoblot for human VASH-1 and actin are shown. Each lane was loaded with 50 μg protein obtained from the serum samples or liver. The AdhVASH-1–injected diabetic mice exhibited significantly elevated serum VASH-1 (42 kDa) levels compared with the AdLacZ-injected diabetic mice (4 weeks). Similarly, enhanced protein levels of VASH-1 in the liver were observed in AdhVASH-1–treated mice compared with AdLacZ-treated diabetic mice. Immunoblots for actin are shown to confirm equal loading. B: Increase in KW-to-BW ratio induced by STZ was diminished in the AdhVASH-1–treated diabetic group. Kidney weight relative to body weight was determined before termination of the experiments. C: Increase in UACR induced by STZ was significantly suppressed by treatment with AdhVASH-1. Data obtained at 4 weeks after initiating treatment with Ad-LacZ or AdhVASH-1 is shown. D: Increase in Ccr induced by STZ was partially suppressed by AdhVASH-1. B–D: *P < 0.05 versus N. †P < 0.05 versus Ve or LacZ. E–H: Representative light microscopic appearance of glomeruli (periodic acid-Schiff staining, original magnification ×400) for nondiabetic control mice (E) and diabetic mice treated with either vehicle buffer (F), AdLacZ (G), or AdhVASH-1 (H). I: Increase in glomerular volume induced by STZ was diminished by treatment with AdhVASH-1. J: Mesangial matrix index was defined as the proportion of the glomerular tuft occupied by the mesangial matrix area (excluding nuclei). I and J: *P < 0.01 versus N. †P < 0.01 versus Ve or LacZ; n = 5 for each group. N, nondiabetic control; Ve, diabetic mice treated with vehicle buffer; LacZ, diabetic mice treated with AdLacZ (5 × 10⁹ vp/mice); Vas, diabetic mice treated with AdhVASH-1 (5 × 10⁹ vp/mice). Each column consists of means ± SE. (A high-quality digital representation of this figure is available in the online issue.)

FIG. 2.

A-E: Immunofluorescent staining of CD31, an endothelial cell marker. Distribution of CD31 was determined by indirect immunofluorescence technique in nondiabetic control mice (A) and diabetic mice treated with either vehicle buffer (B), AdLacZ (C), or AdhVASH-1 (D). E: Glomerular CD31⁺ endothelial area was quantitated. Increase in the CD31⁺ glomerular capillary area was significantly suppressed after treatment with AdhVASH-1. *P < 0.01 versus N. †P < 0.01 versus Ve or LacZ. F and G: Immunoblot analysis. Immunoblots for VEGF-A, phosphorylated VEGFR2 (pVEGFR2), total VEGFR2, and actin are shown. Each lane was loaded with 50 μg protein obtained from the renal cortex. Each band was scanned and subjected to densitometry. F (lower panels): Intensities of VEGF-A protein relative to actin are shown. *P < 0.01 versus N. H and I: Intensities of pVEGFR2 relative to total VEGFR2 (upper graph) and those of VEGFR2 relative to actin (lower graph) are shown. *P < 0.01 versus N. †P < 0.05 versus Ve or LacZ; n = 5 for each group. (A high-quality digital representation of this figure is available in the online issue.)

FIG. 3.

A-C: Immunoblot analysis (cultured hGECs). Immunoblots for pVEGFR2, VEGFR2, and actin are shown. A: Cells were stimulated with 1 nmol/l of VEGF for 2–15 min. Intensities of pVEGFR2 protein relative to total VEGFR2 are shown. *P < 0.01 versus control (0). B: Cells were stimulated with 1 nmol/l of VEGF for 5 min in the presence of recombinant VASH-1 (0–20 nmol/l). Intensities of pVEGFR2 protein relative to VEGFR2 are shown. *P < 0.01 versus control. †P < 0.05 versus VEGF(+)/VASH-1(0). C: Cells were cultured under normal glucose (5.5 mmol/l) or high glucose (25 mmol/l) for 24 h in the presence of recombinant VASH-1 (0–20 nmol/l). Intensities of pVEGFR2 protein relative to VEGFR2 are shown. *P < 0.01 versus normal glucose or normal glucose/Manni. †P < 0.05 versus high glucose/V0. Each lane was loaded with 15 μg protein obtained from hGECs. Normal glucose+Manni, normal d-glucose plus d-mannitol (19.5 mmol/l); V0, without VASH-1; V1, 1 nmol/l VASH-1; V10, 10 nmol/l VASH-1; V20, 20 nmol/l VASH-1.

FIG. 4.

A–D: Glomerular accumulation of type IV collagen was assessed by the indirect immunofluorescence method for nondiabetic control mice (A) and diabetic mice treated with either vehicle buffer (B), AdLacZ (C), or AdhVASH-1 (D). A–D: Original magnification ×200. E and F: Immunohistochemistry of F4/80⁺ monocyte/macrophage. Representative light microscopic appearances of glomerulus in diabetic mice treated with vehicle buffer stained in the presence of the primary antibodies (E) or normal rat IgG (F) are shown. F4/80⁺ cells were observed in diabetic mice (arrowheads; original magnification ×400). G: The amount of immunoreactive type IV collagen in glomeruli relative to the nondiabetic control group determined by computer image analysis is shown. H: The number of glomerular F4/80⁺ monocyte/macrophages is shown. Increase in the F4/80⁺ monocyte/macrophage number was significantly suppressed after treatment with AdhVASH-1; n = 5 for each group. *P < 0.01 versus N. †P < 0.01 versus Ve or LacZ. (A high-quality digital representation of this figure is available in the online issue.)

FIG. 5.

Immunoblot analysis and real-time PCR of TGF-β, MCP-1, and RAGE. A, D, and G: Immunoblots for TGF-β, MCP-1, RAGE, and actin are shown. Each lane was loaded with 50 μg protein obtained from the renal cortex. B: Intensities of TGF-β protein relative to actin are shown. E: Intensities of MCP-1 protein relative to actin are shown. H: Intensities of RAGE protein relative to actin are shown. C and F: The levels of TGF-β1 mRNA and MCP-1 mRNA detected by real-time PCR. Total RNA was extracted from the renal cortex and subjected to the examination using quantitative real-time PCR. C: The amount of TGF-β1 mRNA relative to 18s rRNA is shown. Results are expressed relative to nondiabetic control mice that were arbitrarily assigned a value of 1.0. F: The amount of MCP-1 mRNA relative to GAPDH is shown. B, C, E, F, and H: *P < 0.01 versus N. †P < 0.05 versus Ve or LacZ; n = 5 for each group.

FIG. 6.

A-E: Immunoblot analysis (cultured mesangial cells). A and D: Immunoblots for MCP-1, TGF-β, RAGE, and actin are shown. Each lane was loaded with 20 μg protein obtained from mouse mesangial cells. B: Intensities of MCP-1 protein relative to actin are shown. *P < 0.05 versus normal glucose (NG) or normal glucose/Manni. †P < 0.05 versus high glucose (HG)/V0. ‡P < 0.05 versus high glucose/V1. C: Intensities of TGF-β protein relative to actin are shown. *P < 0.05 versus normal glucose or normal glucose/Manni. †P < 0.05 versus high glucose/V0 or high glucose/V1. E: Intensities of RAGE protein relative to actin are shown. *P < 0.05 versus normal glucose or normal glucose/Manni. †P < 0.05 versus high glucose/V0. ‡P < 0.05 versus high glucose/V1.

FIG. 7.

Double immunofluorescent staining for endogenous mVASH-1, CD31, and α-SMA. A: Double immunofluorescent staining of mVASH-1 (green), CD31 (red), and merged images in the kidney from nondiabetic (upper panels) or diabetic (lower panels) mice. Although mVASH-1 was faintly observed in nondiabetic glomeruli, increased immunoreactivity for mVASH-1 was observed and partially colocalized with the CD31⁺ endothelial cells (arrowheads) in the control diabetic mice. B: Double immunofluorescent staining of mVASH-1 (green), α-SMA (red), and merged images in the kidney from nondiabetic mice (upper panels) or control diabetic animals (lower panels). In the nondiabetic kidney, immunoreactivity for α-SMA was observed in extraglomerular arterioles colocalized with mVASH-1 (arrows). Immunoreactivity of mVASH-1 partially colocalized with the α-SMA⁺ mesangial cells (arrowheads) in the control diabetic mice. Original magnification ×400. (A high-quality digital representation of this figure is available in the online issue.)