Activation of hypoxia-inducible factors prevents diabetic nephropathy

Abstract

Hyperglycemia results in increased oxygen consumption and decreased oxygen tension in the kidney. We tested the hypothesis that activation of hypoxia-inducible factors (HIFs) protects against diabetes-induced alterations in oxygen metabolism and kidney function. Experimental groups consisted of control and streptozotocin-induced diabetic rats treated with or without chronic cobalt chloride to activate HIFs. We elucidated the involvement of oxidative stress by studying the effects of acute administration of the superoxide dismutase mimetic tempol. Compared with controls, diabetic rats displayed tissue hypoxia throughout the kidney, glomerular hyperfiltration, increased oxygen consumption, increased total mitochondrial leak respiration, and decreased tubular sodium transport efficiency. Diabetic kidneys showed proteinuria and tubulointerstitial damage. Cobalt chloride activated HIFs, prevented the diabetes-induced alterations in oxygen metabolism, mitochondrial leak respiration, and kidney function, and reduced proteinuria and tubulointerstitial damage. The beneficial effects of tempol were less pronounced after activation of HIFs, indicating improved oxidative stress status. In conclusion, activation of HIFs prevents diabetes-induced alteration in kidney oxygen metabolism by normalizing glomerular filtration, which reduces tubular electrolyte load, preventing mitochondrial leak respiration and improving tubular transport efficiency. These improvements could be related to reduced oxidative stress and account for the reduced proteinuria and tubulointerstitial damage. Thus, pharmacologic activation of the HIF system may prevent development of diabetic nephropathy.

Keywords: diabetic nephropathy; hypoxia; proteinuria.

Figure 1.

Diabetes-induced regional tissue hypoxia corrected by HIF activation. (A) Cortical and (B) medullary oxygen tension in control and diabetic rats with and without chronic CoCl₂ treatment during baseline and after acute administration of tempol (n=9–12/group). *P<0.05 compared with the corresponding period in the untreated control group. ^†P<0.05 compared with the corresponding period in the untreated diabetes group.

Figure 2.

Derranged kidney oxygen consumption and electrolyte transport efficiency corrected by HIF activation in diabetes. (A) Kidney oxygen consumption and (B) TNa/QO₂ in control and diabetic rats with and without chronic CoCl₂ treatment during baseline and after acute administration of tempol (n=9–12/group). *P<0.05 compared with the untreated control group. ^†P<0.05 compared with the untreated diabetes group.

Figure 3.

Normalization of GFR and renal vascular resistance in response to tempol by HIF activation in diabetes. (A) GFR, (B) filtration fraction, and (C) renal vascular resistance in control and diabetic rats with and without chronic CoCl₂ treatment during baseline and after acute administration of tempol (n=9–12/group). *P<0.05 compared with the corresponding period in the untreated control group. ^†P<0.05 compared with the corresponding period in the untreated diabetes group.

Figure 4.

No major effects of diabetes or HIF activation on total and regional renal blood flow. (A) Total, (B) cortical, and (C) medullary blood flow in control and diabetic rats with and without chronic CoCl₂ treatment during baseline and after acute administration of tempol (n=9–12/group). *P<0.05 compared with the corresponding period in the untreated control group.

Figure 5.

Correction of diabetes-induced proteinuria by HIF activation. Urinary protein excretion in control and diabetic rats with and without chronic CoCl₂ treatment (n=9–12/group). *P<0.05 compared with the untreated control group. ^†P<0.05 compared with the untreated diabetes group.

Figure 6.

HIF activation prevents histological damage in diabetes. (A) PAS and vimentin staining revealing more tubulointerstitial injury in untreated diabetic rats compared with diabetic rats chronically treated with CoCl₂. (B) Semiquantitative results of tubulointerstitial injury in control and diabetic rats with and without chronic CoCl₂ treatment. Quantification of vimentin-positive tubules in control and diabetic rats with and without chronic CoCl₂ treatment. No tubulointerstitial injury was observed in any of the control groups.

Figure 7.

HIF activation increases tubular staining of Cn/Zn-SOD. (A) Immunostaining and quantification of Cu/Zn-SOD expression in control and diabetic rats with and without chronic CoCl₂ treatment (n=5/group). (B) High magnification revealing preferential tubular staining for Cu/Zn-SOD.

Figure 8.

Substantial upregulation of HIF-regulated genes by CoCl₂. Gene expressions of the HIF-regulated genes EPO, VEGF, and HO-1 in control and diabetic rats with and without chronic CoCl₂ treatment (n=4–5/group). The results are shown as fold increases compared with expression levels of diabetic rats without CoCl₂ treatment.

Figure 9.

Experimental design. Diabetes was induced 4 weeks before acute experiments and animals randomly divided to receive either chronic CoCl₂, or no treatment. Results were compared to corresponding age-match control groups.