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.

Diabetic nephropathy is the most common cause of ESRD in developed countries. A possible mechanism for the onset and progression of diabetic nephropathy is an ischemic insult to the diabetic kidney, resulting in tubulointerstitial fibrosis and reduced renal function. Hypoxia has been suggested as a possible final common pathway in end stage kidney injury,^1,2 and several studies have reported that sustained hyperglycemia will result in decreased intrarenal oxygenation.^3,4 Importantly, we have recently reported that tissue hypoxia, attributed to increased oxygen consumption (QO₂) secondary to increased mitochondria uncoupling, induced nephropathy independently of oxidative stress and hyperglycemia.⁵

In normal physiology, reduced oxygen tension (pO₂) promotes alterations in gene expression to counteract hypoxia and promote cell survival and adaptation. An adequate gene response is elementary to maintain sufficient tissue pO₂, and it is mainly accomplished by hypoxia-inducible transcription factors (HIFs).^6,7 HIFs consist of two subunits: an α-subunit that is continuously produced but rapidly degraded by prolyl hydroxylases in the presence of oxygen and a constitutively expressed β-subunit. In hypoxia, the α-subunit will accumulate because of the lack of oxygen and form transcriptionally active heterodimers with the β-subunit. These heterodimers bind to hypoxia-responsive elements in the DNA to induce transcription of hypoxia-responsive genes involved in angiogenesis, anaerobic metabolism, iron metabolism, oxidative phosphorylation, and antioxidant defense.^8,9 However, the HIF system has been shown to be inadequately activated in the hypoxic diabetic kidney.^10–12 Furthermore, the numerous reports of normal or close to normal hematocrit in experimental diabetes indirectly imply absent HIF activation.^3,4 Importantly, it has been reported that antioxidant treatment with the superoxide dismutase mimetic tempol reduced intrarenal hypoxia but paradoxically, increased HIF-1α staining, further highlighting the complexity of this issue.¹¹

In this study, we tested the hypothesis that chronic HIF activation protects the diabetic kidney from altered oxygen metabolism, kidney function, and histologic damage. The potential involvement of oxidative stress was investigated by studying the effects of acute administration of tempol.

Results

Baseline Characteristics and Effects of HIF Activation

All diabetic animals were hyperglycemic, gained less weight compared with controls (Table 1), and displayed normal blood gas and electrolyte status (Supplemental Table 1). Both hyperglycemia and weight gain were reduced in diabetic animals treated with CoCl₂. Diabetic kidneys were hypertrophic in the untreated group and displayed increased kidney-to-body weight ratio, whereas CoCl₂ to diabetic kidneys resulted in decreased kidney weight but did not decrease kidney-to-body weight ratio (Table 1).

Table 1.

Body weight, blood glucose, hematocrit, kidney weight, and kidney weight-to-body weight ratio in control and diabetic rats with and without chronic CoCl₂ treatment throughout the course of diabetes

Group	N	Body wt (g)	BG (mmol/L)	Hct (%)	KW (g)		KW/Body wt (×1000)
					Left	Right	Left	Right
Control	9	393±7	6.0±0.2	41±1	1.48±0.04	1.47±0.04	3.8±0.1	3.7±0.1
Control+CoCl2	9	393±10a	4.8±0.2a	47±1a,b	1.38±0.12a	1.25±0.05a	3.5±0.3a	3.2±0.1a
Diabetes	12	353±11b	26.8±1.7b	44±1b	2.23±0.09b	2.31±0.10b	6.3±0.2b	6.5±0.2b
Diabetes + CoCl2	10	270±8a,b	20.4±0.8a,b	59±1a,b	1.70±0.04a,b	1.70±0.04a,b	6.3±0.1b	6.3±0.2b
Two-way ANOVA
Type		P<0.001	P<0.001	P<0.001	P<0.001	P<0.001	P<0.001	P<0.001
Treatment		P<0.001	P=0.002	P<0.001	P<0.001	P<0.001	P=0.52	P=0.05
Interaction		P<0.001	P=0.03	P<0.001	P=0.01	P=0.01	P=0.60	P=0.39

All values are ±SEM. Type denotes control versus diabetes, and treatment denotes untreated versus CoCl₂. Body wt, body weight; BG, blood glucose; Hct, hematocrit; KW, kidney weight.

^aP<0.05 compared with the untreated diabetes group.

^bP<0.05 compared with the untreated control group.

Urinary flow was increased in the both diabetic groups compared with control animals, whereas mean arterial pressure (MAP) was higher in the untreated control group compared with all groups (Table 2).

Table 2.

Urinary flow and MAP in control and diabetic rats with and without chronic CoCl₂ treatment during baseline and after acute administration of tempol

Group	N	Urinary Flow (µl/min per kidney)		MAP (mmHg)
		Baseline	Tempol	Baseline	Tempol
Control	9	2.6±0.5	3.0±0.5	116±3	105±2a
Control+CoCl2	9	2.2±0.3	1.9±0.6	106±3b	92±2a,b
Diabetes	12	30.2±7.6b	21.6±4.2b	105±2b	97±1a
Diabetes+CoCl2	10	21.7±6.2b	15.7±4.8a,b	102±2b	93±4a,b
Two-way ANOVA
Type		P<0.001	P<0.001	P=0.01	P=0.17
Treatment		P=0.44	P=0.32	P=0.03	P<0.01
Interaction		P=0.48	P=0.50	P=0.15	P=0.08

All values are ±SEM. Type denotes control versus diabetes, and treatment denotes untreated versus CoCl₂.

^aP<0.05 compared with baseline within the same group.

^bP<0.05 compared with the untreated control group.

Both cortical and medullary pO₂ (Figure 1) were reduced in diabetic animals compared with controls, confirming the development of diabetes-induced intrarenal tissue hypoxia. Mitochondria from diabetic kidneys displayed increased leak respiration, the regulated fraction equally mediated by uncoupling protein-2 (UCP-2) and adenine nucleotide translocase (ANT) (Table 3). Measurements of mitochondrial membrane potential corroborated a functional role of the observed uncoupling through both UCP-2 and ANT (Table 4).

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.

Table 3.

Mitochondrial respiration (picomoles O₂ per second per milligram protein) during state 3 (active) and state 4 (rest), respiratory control ratio, and total leak respiration in the presence of ATP-synthase inhibitor oligomycin

Group	N	State 3	State 4	Respiratory Control Ratio	Total Leak Respiration	Leak Respiration through UCP-2	Leak Respiration through ANT	Total Regulated Leak Respiration
Control	10	121±19	23±3	5.1±0.3	28±2	0.2±0.3	1.6±0.3	1.8±0.4
Control+CoCl2	7–9	114±17a	19±3a	5.9±0.4a	23±2a	0.3±0.3a	1.5±0.3a	1.8±0.5a
Diabetes	9	265±29b	34±3b	7.8±0.6b	43±5b	5.4±0.6b	5.6±1.1b	11.3±1.6b
Diabetes+COCl2	9	167±11a	22±2a	7.4±0.3b	24±2a	0.2±0.2a	1.6±0.1a	1.8±0.2a
Two-way ANOVA
Type		P<0.001	P=0.01	P<0.001	P=0.02	P<0.001	P=0.001	P<0.001
Treatment		P=0.02	P=0.01	P=0.66	P<0.001	P<0.001	P<0.001	P<0.001
Interaction		P=0.04	P=0.19	P=0.18	P=0.03	P<0.001	P=0.001	P<0.001

Involvements of UCP and ANT were determined by the magnitude of inhibition by the specific inhibitors guanosine diphosphate and carboxyatractylate, respectively. All values are ±SEM. Type denotes control versus diabetes, and treatment denotes untreated versus CoCl₂.

^aP<0.05 compared with the untreated diabetes group.

^bP<0.05 compared with the untreated control group.

Table 4.

Mitochondrial membrane potential (fluorescence unit uptake per milligram protein) during state 4 respiration (glutamate), leak respiration (oligomycin), and inhibition of uncoupling proteins (guanosine diphosphate) and the adenine nucleotide translocase (carboxyatractylate) or both

Group	N	State 4	Leak Respiration	UCP Inhibition	ANT Inhibition	Complete Leak Inhibition
Glutamate		+	+	+	+	+
Oligomycin			+	+	+	+
Guanosine diphosphate				+		+
Carboxyatractylate					+	+
Control	9	5610±268	5731±481	5366±329	5910±331	5549±524
Control+CoCl2	8	5860±324	5596±264	5938±407	5625±250a	5359±372a
Diabetes	9	5848±417	5801±350	7325±337b	7734±466b	8737±374b
Diabetes+COCl2	9	5924±249	6029±391	5922±397	5901±343a	5593±395a
Two-way ANOVA
Type		P=0.64	P=0.52	P=0.01	P<0.01	P<0.001
Treatment		P=0.62	P=0.91	P=0.27	P<0.01	P<0.001
Interaction		P=0.79	P=0.64	P=0.01	P=0.04	P=0.001

All values are ±SEM. Type denotes control versus diabetes, and treatment denotes untreated versus CoCl₂. +, substrate present.

^aP<0.05 compared with the untreated diabetes group.

^bP<0.05 compared with the untreated control group.

Chronic HIF activation prevented intrarenal hypoxia in rats with diabetes, which was accomplished by conserving normal kidney QO₂ (Figure 2A) through prevention of mitochondrial leak respiration (Table 3) and normalizing decreased tubular sodium transport (TNa)/QO₂ (Figure 2B) and GFR (Figure 3A). HIF activation to controls did not alter any of these parameters.

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.

Filtration fraction was higher in both diabetic groups compared with controls (Figure 3B), whereas renal vascular resistance (RVR) (Figure 3C) and total renal blood flow (RBF) (Figure 4A) were similar in all groups. Cortical RBF, but not medullary RBF, was lower in both diabetic groups compared with controls (Figure 4, B and C).

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.

Diabetic kidneys presented with increased urinary protein excretion (Figure 5) and tubulointerstitial injury, the latter evident as increased periodic acid–Schiff (PAS) and pathologic tubulointerstitial vimentin staining (Figure 6). Chronic HIF activation resulted in reduced urinary protein excretion (Figure 5), tubulointerstitial injury (Figure 6), and increased tubular expression of copper- and zinc-containing superoxide dismutase (Cu/Zn-SOD) (Figure 7).

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.

Successful HIF activation was confirmed by increased gene transcripts of erythropoietin (EPO), vascular endothelial grown factor (VEGF), and hemeoxygenase-1 (HO-1) (Figure 8), resulting in increased hematocrit (Table 1).

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.

Effects of Acute Tempol Administration

Tempol reduced urine flow in both diabetic groups to the same extent, although it only reached statistical significance in the CoCl₂-treated group (Table 2). Tempol decreased MAP in all groups (Table 2). Importantly, tempol corrected the diabetes-induced intrarenal tissue hypoxia in untreated rats with diabetes, whereas it had no effect in any of the other groups (Figure 1). Tempol reduced RVR in both untreated groups (Figure 3C) but did not cause any other study-relevant systematic changes.

Discussion

In this study, we show that chronic pharmacologic HIF activation mitigates diabetes-induced alterations in renal oxygen metabolism and mitochondria function and that prevention of these alterations is accompanied by normalized renal function, reduced proteinuria, and improved tubulointerstitial injury. Normal QO₂ in the diabetic kidney after HIF activation is maintained by preventing the commonly occurring diabetes-induced glomerular hyperfiltration and mitochondrial leak respiration as well as reducing TNa/QO_2, that otherwise would increase the oxygen demand required for active tubular electrolyte transport. Mitochondrial uncoupling through UCP-2 is activated by oxidative stress,^13,14 and we have previously reported markedly increased UCP-2 activity in the diabetic kidney, resulting in increased mitochondrial QO₂.^15,16 Preventing increased uncoupled QO₂, therefore, seems feasible to minimize oxygen wasting in the diabetic kidney.

HIF is an important protective physiologic mechanism activated in conjunction with renal injury to counteract hypoxia and prevent damage.^8,9 However, we could not detect any compensatory HIF activation, which is indicated by the lack of increased HIF target genes, in the hypoxic diabetic kidney. It has been reported that hyperglycemia-induced covalent modification of p300 decreases the HIF-1α/p300 association required for normal HIF signaling.¹⁰ Interestingly, Rosenberger et al.¹¹ reported that antioxidant treatment increased medullary HIF expression in diabetic kidneys, although pO₂ also improved. Thus, it seems possible that oxidative stress may affect HIF-1α/p300 association and therefore, normal HIF signaling.

Cobalt is commonly referred to as a hypoxia mimetic because of its ability to inhibit the degradation of the HIF α-subunit by prolyl hydroxylases, resulting in accumulation and subsequent HIF activation.¹⁷ In this study, diabetes-induced glomerular hyperfiltration was normalized by HIF activation. In addition, HIF activation prevented the intrarenal tissue hypoxia by preventing excessive kidney QO₂ and reduced TNa/QO₂.

It has been shown that hypoxia can induce activation of post-translational collagen-modifying enzymes, leading to an extracellular matrix composition resistant to degradation.¹⁸ Diabetic animals displayed pronounced tubulointerstitial injury as indicated by increased pathologic tubulointerstitial vimentin and PAS scores. Although it has previously been shown that hypoxia-induced HIF activation can induce fibrosis,¹⁹ that study used mouse molecular genetics to activate HIF, resulting in supraphysiologic activation of HIF for a long time. In that study, however, pharmacologic HIF activation by CoCl₂ reduced fibrosis and tubular injury in the diabetic animals. The reason for this apparent discrepancy is largely beyond the scope of this study but may be because of differences in HIF downstream regulation by the extent and duration of HIF activation.²⁰ Downstream effectors investigated in this study were EPO, VEGF, HO-1, and Cu/Zn-SOD, which all were induced by CoCl₂. EPO is known to exert renoprotective, antiapoptotic effects, protecting peritubular capillaries and preventing glomerulosclerosis,^21,22 and VEGF may ameliorate kidney injury.²³ In addition, EPO induces HO-1 activation, offering an additional protection against oxidative stress.²⁴ Overexpression of the potent antioxidant Cu/Zn-SOD attenuated renal injury in the db/db mouse model of type 2 diabetes.²⁵

The finding that chronic HIF activation reduced proteinuria is in line with previous studies. Song et al.²⁶ recently showed that HIF activation halted progression of proteinuria in a remnant kidney rat model. Similarly, Ohtomo et al.²⁷ reported that HIF activation by cobalt reduced renal expressions of TGF-β, connective tissue growth factor, and NADPH oxidase as well as reduced proteinuria, tubulointerstitial damage, and peritubular capillary loss in experimental type 2 diabetes. Indeed, tubulointerstitial fibrosis is a reliable marker for progression of kidney disease, and it is, therefore, interesting that HIF activation ameliorated fibrosis development in the diabetic kidney. There are reports of a correlation between tubulointerstitial fibrosis and disease progression since the late 1960s.^28,29 Because tubulointerstitial damage is currently considered the most reliable marker of disease progression,³⁰ these results further strengthen the possibility that pharmacologic activation of the HIF system could offer protection against development of diabetic nephropathy.

The reduced plasma glucose levels in animals treated with CoCl₂ are likely caused by increased glucose uptake as previously shown.^31–33

Importantly, tempol corrected the diabetes-induced intrarenal tissue hypoxia in untreated rats with diabetes, possibly by O₂⁻ inactivation conserving kidney QO₂. Unfortunately, no confirming QO₂ measurements were possible after administration of tempol, because the tempol was found to interfere with the blood gas analysis. This limitation does not, however, weaken the main conclusion.

Acute tempol administration to diabetic rats improved both cortical and medullary pO₂, normalized kidney QO₂ and TNa/QO₂, and decreased RVR, whereas it had no or minor effects on either of the control groups or on the CoCl₂-treated rats with diabetes. These results imply that kidney function in the untreated diabetic rats is heavily influenced by increased oxidative stress and that chronic HIF activation prevents it. Acute administration of tempol decreased MAP, which is consistent with previous reports and likely caused by tempol-induced O₂⁻ inactivation increasing NO availability.^34–36 The finding that tempol reduces RVR is also consistent with previous findings,^34,35 which can be attributed to the direct BP-lowering effect as previously reported.³⁷ However, we did not detect any effect of tempol on RVR in CoCl₂-treated rats, despite similar MAP reduction, which may indicate altered vascular regulation after chronic HIF activation. However, tempol has previously been shown to reduce urine flow in diabetic animals.³⁸

The reduced cortical RBF in diabetic kidneys, despite unaltered total RBF, may relate to a technical issue when using the laser Doppler technique on kidneys with different stages of hypertrophy. It is also unlikely that augmented oxygen delivery, secondary to increased hematocrit caused by the HIF activation, is a major contributor to the mitigated intrarenal hypoxia, because intrarenal pO₂ in the diabetic kidney mainly is determined by QO₂.³⁹

In conclusion, this study shows that diabetes-induced alterations in renal oxygen metabolism are prevented by HIF activation by three different mechanisms. By normalizing GFR, which reduces tubular electrolyte load, preventing mitochondrial leak respiration, and maintaining effective TNa/QO₂, total kidney QO₂ and thus, tissue pO₂ can be maintained within the normal range in the diabetic kidney. Early pharmacologic activation of the HIF system to reduce hypoxia-induced kidney damage may prevent the onset or progression of diabetic nephropathy.

Concise Methods

All chemicals were from Sigma-Aldrich (St. Louis, MO) and of the highest grade available if not otherwise stated.

Animals, Induction of Diabetes, and Chronic Treatment

Age-matched male Sprague–Dawley rats weighing about 200–250 g (about 8 weeks of age) were purchased from B&K (Sollentuna, Sweden). Animals had free access to water and standard rat chow (0.3% Na, 0.8% K, 21% protein; R3; Ewos, Södertälje, Sweden) throughout the study. All experiments were performed in accordance with the National Institutes of Health guidelines for use and care of laboratory animals and approved by the Animal Care and Use Committee for Uppsala University. Diabetes was induced by an injection of streptozotocin (STZ; 55 mg/kg body wt) in the tail vein. Animals were considered diabetic if blood glucose concentrations increased to above 18 mmol/L within 24 hours after STZ injection and remained elevated. Blood glucose concentrations were determined with test reagent strips (FreeStyle; Abbott Laboratories, Alameda, CA) from blood samples obtained from the cut tip of the tail in all animals.

The control and STZ-injected animals were randomly divided into receiving either vehicle or CoCl₂ (20 mg·kg body wt⁻¹ per 24 hours⁻¹) in the drinking water for the entire duration of diabetes.²⁷ A detailed schematic drawing of the experimental design is depicted in Figure 9.

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.

Surgical Preparation for In Vivo Kidney Function

Four weeks after allocation to the study, the animals were anesthetized with an intraperitoneal injection of thiobutabarbital (Inactin; 120 mg/kg body wt for normoglycemic controls and 80 mg/kg for diabetic animals), placed on a thermo-controlled operating table at 37°C, and tracheotomized. Polyethylene catheters were placed in the right femoral vein for infusion of Ringer solution (5 ml·kg body wt⁻¹ per hour⁻¹ for normoglycemic controls and 10 ml·kg body wt⁻¹ per hour⁻¹ for diabetic animals), the right femoral artery for BP measurements (Statham P23dB; Statham Laboratories, Los Angeles, CA), and the left renal vein and carotid artery for blood samplings. The left ureter was catheterized to collect urine for subsequent analysis, and the urinary bladder was catheterized to allow urinary drainage. The left kidney was exposed by a left subcostal flank incision, immobilized in a plastic cup, and embedded in pieces of saline-soaked cotton wool, and the surface was covered with paraffin oil (Apoteksbolaget, Gothenburg, Sweden).

Simultaneous Measurements of GFR, RBF, Total Kidney QO₂, and Tissue pO₂

Animals were allowed a 30-minute recovery period after surgery followed by 40 minutes of baseline measurements of GFR, RBF, and QO₂. Thereafter, tempol (174-µmol/kg bolus+174-µmol·kg⁻¹ per min⁻¹ continuous infusion)⁴⁰ was acutely administered, and GFR and RBF were measured for an additional 40 minutes to determine the role of increased oxidative stress. Because tempol causes interference with the blood gas analysis, the effect of tempol on QO₂ and thus, tubular transport efficiency (TNa/QO₂) could not be studied.

GFR was estimated by clearance of ³H-inulin (185-kilobecquerel bolus+185 kilobecquerel·kg⁻¹ body wt per h⁻¹; American Radiolabeled Company, St. Louis, MO). The ³H activities in urine and plasma were measured using a standard liquid scintillation technique, and GFR was calculated according to GFR=U·V/P, where U and P denote the activity of ³H-inulin in the urine and plasma, respectively, and V denotes the urine flow in milliliters per minute.

Total RBF was measured using an ultrasound probe placed around the left renal artery (Transonic Systems, Ithaca, NY), and cortical and medullary RBFs were measured with laser Doppler flowmetry (PF 4001–2; Perimed, Stockholm, Sweden). Intrarenal pO₂ was measured in kidney cortex and medulla by oxygen microsensors (Oxy-10; Unisense A/S, Aarhus, Denmark) at the end of each experimental period. All measured parameters were recorded with a Power Lab instrument (AD Instruments, Hastings, UK). Blood gas parameters were analyzed (iSTAT; Abbott Laboratories, Princeton, NJ) on samples drawn from the left renal vein and carotid artery only at the end of the baseline period. Tempol was found to interfere with the analysis in a separate set of pilot experiments, and no attempt was, therefore, made to measure QO₂ after the acute tempol administration. Kidney weights were determined at the end of the experiments. Urine volumes were measured gravimetrically, urinary Na⁺ concentrations were measured by flame spectrophotometry (IL543; Instrumentation Lab, Milan, Italy), and urinary protein concentration was measured by DC Protein Assay (Bio-Rad, Hercules, CA).

Calculations of In Vivo Parameters

The filtration fraction was estimated as GFR/RBF·(1−hematocrit). RVR was calculated as MAP divided by total RBF. In vivo renal QO₂ (micromoles per minute⁻¹ per kidney⁻¹) was estimated from the arteriovenous difference in O₂ content (O_2ct) using a standard equation (O_2ct = [Hb·O₂ saturation·1.34+pO₂·0.003) multiplied by total RBF. TNa/QO₂ was calculated using TNa equals plasma Na⁺ concentration multiplied by GFR.

Tissue Collection and Analysis of Histologic Alterations and Gene Expressions

A second set of control and diabetic animals with and without chronic CoCl₂ administration (n=4–5/group) was used for tissue collection and subsequent analysis of histologic alterations and gene expressions. In brief, the animals were anesthetized with Inactin, and a catheter was placed in the carotid artery. Twenty milliliters ice-cold PBS was infused intra-arterially, and the renal vein was cut open to wash the kidneys before they were removed. The kidneys were dissected and either placed in methyl Carnoy’s fixative (methanol:chloroform:acetic acid at 6:3:1) or snap frozen using liquid nitrogen.

Estimation of Histologic Damage of the Kidney

Three-micrometer sections of the fixed paraffin-embedded kidney tissues were stained with the PAS reagent and counterstained with hematoxylin. Semiquantification of tubulointerstitial injury was performed in a blinded manner using 20 randomly selected fields of cortex per cross-section (×200). Tubulointerstitial injury was graded (0–4+) on the basis of the percentage of tubular cellularity, basement membrane thickening, accumulation of PAS-positive droplets, sloughing, or interstitial widening as follows: 0, no change; 1, <25% tubulointerstitial injury; 2, 25%–49% tubulointerstitial injury; 3, 50%–74% tubulointerstitial injury; and 4, 75%–100% tubulointerstitial injury.⁴¹

Quantification of Vimentin- and Cu/Zn-SOD–Positive Tubules

Indirect immunoperoxidase methods were used to identify vimentin with mouse mAb V9 (Dako, Carpinteria, CA) and Cu/Zn-SOD (representative target of HIF activation) with goat polyclonal Ab (Santa Cruz Biotechnology).^41,42 The numbers of vimentin- or Cu/Zn-SOD–positive tubules were expressed as the mean±SD of the number per low-power field (×100). This value was calculated from counts obtained for 25 randomly selected cortical areas for each rat kidney in midcoronal sections in a blinded manner.

RNA Isolation and Analysis of Gene Expressions

Total RNA was extracted from kidney homogenates with Isogen (Nippon Gene, Tokyo, Japan). To synthesize cDNA from total RNA, SuperScript II Reverse Transcription was used (Life Technologies, Rockville, MD). Renal mRNA levels were assessed by real-time quantitative PCR using SYBR Green PCR Reagent (Qiagen, Hilden, Germany) and the iCycler PCR System (Bio-Rad) according to the manufacturer’s instructions. Briefly, amplification reactions consisted of 1 µl cDNA, 12.5 µl Universal 2× PCR MasterMix (Qiagen), and 5 µL each specific primers. Primer concentrations in the final volume of 25 µL were 500 nmol/L. In control experiments with triplicates, no false positives were detected, and the variance between each of the replicates was within 5%. All PCR reactions were performed in triplicate. Threshold cycle (C_t) was used for relative quantification of the input target number. The amount of C_t for control samples was considered one (i.e., 2°). The numbers of C_ts for other samples were subtracted by cycles of control samples and recorded as C_t. The relative amount of amplified genes is given by 2^−ΔCt. The mRNA levels of target genes were normalized to levels of β-actin. PCR primers for EPO, VEGF, and HO-1 have been described previously.⁴³

Mitochondria Isolation and Functional Analysis

A third set of animals (n=8–10/group) was used to assess mitochondrial function. Mitochondria was isolated from kidney cortex, and mitochondrial oxygen consumption was recorded using high-resolution respirometry using an Oroboros O2K (Oroboros Instruments, Innsbruck, Austria) and corrected for protein concentration.¹⁵ Mitochondria function was determined as state 4 respiration in the presence of glutamate (10 mmol/L), state 3 respiration after the addition of 400 µmol/L ADP, and respiratory control ration calculated as state 3 over state 4 respiration. Mitochondrial uncoupling was determined as leak respiration (i.e., respiration in the presence of ATP-synthase inhibitor oligomycin [12 µg/mg protein]), and responsible mechanisms were evaluated as the sensitivity of leak respiration to guanosine diphosphate (GDP; 500 µmol/L; inhibitor of UCP) and carboxyatractylate (CAT; 0.5 µmol/L; inhibitor of ANT). The combination of GDP and CAT was used to estimate total regulated leak respiration.

Mitochondrial membrane potential was evaluated with tetramethyl rhodamine methylester in the presence of glutamate alone or combined with oligomycin, GDP, and CAT as previously described.⁴⁴

Statistical Analyses

Two-way ANOVA was used to compare baseline data between the two groups (control and rats with diabetes) receiving two different treatments (with and without CoCl₂ treatment), and when appropriate, it was followed by Bonferroni post hoc test. t Test for paired comparison was used when comparing data before and after acute administration in each group. Histologic score data were analyzed using nonparametric tests, whereas all other datasets were considered normally distributed and therefore, analyzed using parametric statistics. Statistical analysis was performed using GraphPad Prism for Windows (GraphPad Software Inc., San Diego, CA). For all comparisons, P<0.05 was considered statistically significant. All values are expressed as mean±SEM if not otherwise stated.

Disclosures

None.