Mitochondria-derived reactive oxygen species contribute to synergistic interaction of diabetes and hypertension in causing chronic kidney injury

Abstract

Diabetes (DM) and hypertension (HTN) are major risk factors for chronic kidney injury, together accounting for >70% of end-stage renal disease. The combination of DM and HTN significantly accelerates development of renal injury; however, the underlying mechanisms of this synergy are still poorly understood. This study assessed whether mitochondria (MT) dysfunction is essential in developing renal injury in a rat model with combined DM and HTN. Type 1 DM was induced in Wistar rats by streptozotocin. HTN was induced 6 wk later by interrenal aorta constriction (AC) between the renal arteries, so that right kidneys were exposed to HTN while left kidneys were exposed to normotension. Kidneys exposed to DM or HTN alone had only mild glomerular injury and urinary albumin excretion (UAE). In contrast, kidneys exposed to DM + 8-wk HTN had significantly increased UAE and glomerular structural damage with reduced glomerular filtration rate. Marked increases in MT-derived reactive oxygen species (ROS) were also observed in right kidneys exposed to HTN + DM. We further tested whether treatment with MT-targeted antioxidant (MitoTEMPO) after the onset of HTN attenuates renal injury in rats with DM + HTN. The results showed that kidneys in DM + AC + MitoTEMPO rats had lower UAE, less glomerular damage, and preserved MT function compared with untreated DM + AC rats. Our study indicates that MT-derived ROS play a major role in promoting kidney dysfunction when DM is combined with HTN. Preserving MT function might be a potential therapeutic approach to halt the development of renal injury when DM coexists with HTN.

NEW & NOTEWORTHY Our study shows that the combination of diabetes (DM) and hypertension (HTN) promoted renal dysfunction, albuminuria, and mitochondrial dysfunction, whereas DM or HTN alone caused only mild kidney injury. Treatment with the specific mitochondria-targeted antioxidant Mito-TEMPO greatly ameliorated renal injury in rats with both DM + HTN. These results suggest that targeting mitochondria-derived ROS may be a potential therapeutic strategy to attenuate the development of kidney injury induced by the combination of DM and HTN.

INTRODUCTION

The prevalence of diabetes (DM) has increased dramatically in the United States during the past several decades (1). Hypertension (HTN) is twice as prevalent in people with DM as in the general population, with ∼25% of young patients with type 1 DM and 50–85% of older patients with type 2 DM affected (2–4). HTN significantly increases the incidence of end-stage renal disease (ESRD), and a crucial therapeutic target for patients with DM and HTN is to reduce blood pressure (BP) to less than 130/80 mmHg (5, 6). However, clinical and experimental studies also suggest that current treatment for chronic kidney disease (CKD) only slows, rather than halts, progression of nephropathy to ESRD (7, 8). Therefore, there is an urgent need to identify the mechanisms responsible for diabetic-HTN nephropathy and develop new, more effective therapeutic targets.

We previously reported that moderate HTN promoted significant kidney injury in type 2 diabetic Goto-Kakizaki (GK) rats (9). To further understand the mechanisms by which HTN interacts with hyperglycemia in DM to promote kidney injury and to avoid complicating effects of obesity and hyperinsulinemia associated with type 2 DM models, we used a streptozotocin (STZ)-induced type 1 DM model and inter-renal aorta constriction (AC)-induced HTN in this study to examine the differential and synergistic impact of hyperglycemia and high BP in causing kidney injury. Notably, both kidneys in our model were exposed to the same levels of hyperglycemia, circulating hormones, and neural influences. However, the left kidney below the inter-renal AC experienced normal to slightly reduced BP, whereas the right kidney above the AC was subjected to elevated BP. Consequently, we could compare the structural and functional differences between the left and right kidneys in the same animal to ascertain the direct individual impacts of high blood glucose or HTN and the combined effects of high glucose and HTN.

Mitochondria (MT) are the “powerhouses” of cells, responsible for synthesizing ATP and supplying energy to cells. They are also involved in many other cellular functions, including generating reactive oxygen species (ROS) and regulating cell apoptosis (10). Given the kidney’s high energy and oxygen requirements, MT and kidney function are intimately linked. Hence, dysfunctional MT may be central to the development and progression of diabetic CKD (11). The primary pathophysiological changes associated with MT dysfunction include decreased ATP-linked respiration and increased MT-derived ROS production. Although these changes have been observed in kidneys of experimental animal models of CKD and in patients with CKD (10), investigations into MT function at different stages of CKD associated with DM and HTN have been lacking.

In this study, we sought to determine 1) whether hyperglycemia in type 1 DM and moderately increased BP synergistically promote renal MT dysfunction and the development of kidney injury and 2) whether MT-derived ROS play a significant role in kidney injury induced by the combination of DM and HTN.

MATERIALS AND METHODS

Animals

The experimental procedures described in this study followed the National Institutes of Health Guide for the Care and Use of Laboratory Animals. They had been approved by the Institutional Animal Care and Use Committee of the University of Mississippi Medical Center. Wistar rats were purchased from Charles River, US. Rats had free access to standard chow (Harlan Laboratories, Inc., IN) and were housed in individual cages maintained at 23 ± 2°C with a 12:12-h light-dark cycle. Rats were randomly divided into four groups in this study: group 1, type 1 DM only; group 2, HTN only; group 3, DM + HTN; and group 4, DM + HTN with MitoTEMPO treatment (Table 1).

Table 1.

Experimental groups

Group	Model	Procedure	Kidney Phenotype
DM	DM only	STZ injection and then sham surgery	Both kidneys exposed to DM, no HTN
AC	HTN only	Saline injection and then AC surgery	Right kidney exposed to HTN and left kidney exposed to normal/slightly reduced blood pressure
DM + AC	DM + HTN	STZ injection and then AC surgery	Right kidney exposed to DM + HTN and left kidney exposed to DM with normal/slightly reduced blood pressure
DM + AC + MitoTEMPO	DM + HTN with mitochondria-target ROS scavenger	STZ injection and then AC surgery followed by drug treatment	The same as DM + AC with MitoTEMPO treatment

AC, aorta constriction; DM, diabetes; HTN, hypertension; ROS, reactive oxygen species; STZ, streptozotocin.

Experimental Protocol

Type 1 DM was induced in male Wistar rats at 13 wk of age by a single injection of 40 mg/kg STZ (Sigma-Aldrich, MO) through the penile dorsal vein. Rats with blood glucose levels stabilized around 400 mg/dL were used for the following studies. At 16 wk of age, BP telemetry transmitters were implanted in the rats, followed by 7 days of recovery. Then, baseline mean arterial pressure (MAP), heart rate (HR), food intake, urine output, 24-h urinary albumin excretion (UAE), and blood glucose were measured. At 19 wk of age, inter-renal AC or sham surgeries were performed, followed by another 7 days of recovery. At 21 wk of age, an osmotic mini-pump (model 2ML4, Azet) was placed subcutaneously to deliver mitoTEMPO (Cayman Chemical) at 0.45 mg/kg/day to DM + AC rats. This dose has been previously shown to inhibit vascular ROS generation in a rat model of angiotensin II-induced hypertension (12). A second mini-pump with mitoTEMPO replaced the first mini-pump 4 wk after the first mini-pump implantation. MAP, HR, food intake, urine output, UAE, and blood glucose were measured for three consecutive days at 3 and 8 wk after AC or sham surgery. At 28 wk of age, kidney glomerular filtration rate (GFR) and single kidney GFR and UAE from each kidney were measured. Both kidneys were harvested and used immediately for MT respiration studies or stored for histology staining and electron microscopy scanning (Fig. 1A). Control nondiabetic animals were injected with saline vehicle at 13 wk of age and followed the same surgical protocol as STZ-diabetic rats. A limitation of our study is that only male rats were used and that we did not investigate potential sex differences. STZ is commonly used in male rodents to induce type-1 DM as female rodents are partially protected from pancreatic β cell injury (13).

Figure 1.

A: time course of animal protocol in male Wistar rats from 13 to 28 wk of age. B: mean arterial pressure measured by telemetry in DM, AC, DM+AC, and DM+AC+MitoTEMPO groups of rats before AC or sham surgery, and at 3 and 8 wk after surgery (n = 6 or 7 animals/group, *P < 0.05 when comparing DM vs. other groups by Tukey’s post hoc test at 3 and 8 wk after AC or sham surgery). C: heart rate measured in DM, AC, DM+AC, and DM+AC+MitoTEMPO groups of rats before AC or sham surgery, and at 3 and 8 wk after surgery (n = 6 or 7 animals/group, *P < 0.05 when comparing AC vs. other groups by Tukey’s post hoc test. #P < 0.05 when comparing DM+AC+MitoTEMPO with all other groups by Tukey’s post hoc test at 8 wk after AC or sham surgery). AC, aorta constriction; DM, diabetes.

Implantation of the Telemetry Transmitter

At 16 wk of age, rats were anesthetized with 2% isoflurane and a telemetry probe (HD-S10, Data Sciences International, MN) was implanted in the left carotid artery and advanced into the aorta. Seven to ten days after recovery from surgery, MAP and HR were measured by telemetry, 24 h/day for 3–5 consecutive days, using computerized methods for data collection. Daily MAP and HR were obtained from the average of 12:12-h light:dark recording using a sampling rate of 1 Hz and duration of 30 s every 10 min.

Aorta Constriction Surgery to Induce HTN

For the AC surgery, rats were anesthetized with 2% isoflurane and placed in the right lateral position and body temperature was maintained at 37°C with a heating pad. A ∼2–3-cm skin incision was made in the left lateral flank. The left kidney was retracted to allow visualization of the aorta between the renal arteries. A 3-0 silk suture was snared and pulled around the aorta. A bent 22-gauge needle was placed next to the aorta and the suture was snugly tied around the needle and the aorta. Following ligation, the needle was removed and the muscles and skin were closed. The suture band between two renal arteries was gauged to cause mild reductions of BP in the left kidney after HTN developed and moderate increases (∼35 mmHg) in systemic arterial pressure and perfusion pressure of the right kidney. With this method, both kidneys are exposed to the same levels of plasma glucose and electrolytes, circulating hormones, and neural influences but have different BPs.

To determine the BP gradient above and below the AC, a femoral artery catheterization was performed under anesthesia on the last day of the experiment. After cannulation, the catheter was connected to PowerLab data acquisition system (ADInstruments, CO) to record BP. Simultaneously, BP was measured from the telemetry catheter that had previously been implanted in the common carotid artery and advanced into the aortic arch above the AC. The pressure gradients across the AC averaged 30 ± 5 mmHg.

Separate Kidney GFR and Urine Collection in Left and Right Kidneys

Rats were anesthetized with 1% Inactin (Sigma) at 70–100 mg/kg and Ketamine at 50 mg/kg. A catheter was inserted into the femoral vein to infuse saline with FITC-inulin (0.1%, Sigma-Aldrich, St. Louis, MO) at the rate of 5.0 mL/h. To collect urine from right kidney, the right ureter was exposed and a tapered catheter (RenaPulse RT 040, Braintree Scientific, Inc. Braintree, MA) was inserted into the middle part of the ureter. A flanged catheter was inserted into the bladder to collect urine from the left kidney. GFR of each kidney was measured around 11 AM during the day for over 2 h using FITC-inulin clearance as previously described (9). Blood samples (100 µL) were collected at 60 and 90 min of the infusion period and GFR from each kidney was calculated from the average rate of inulin clearance. The formula for the single kidney GFR is as follows: Inulin clearance (GFR) = urinary FITC-inulin excretion rate/steady-state blood FITC-inulin concentration. GFRs from left and right kidneys were normalized to the respective kidney weights.

Blood and Urine Biochemistry Measurements

Fasting blood glucose levels were measured around 2 PM during the day using glucose strips (ReliOn Prime Blood Glucose Test Strips). Urine albumin levels in urine collected for 3 days in rats placed in metabolic cages or urine collected during the separate kidney GFR measurements were determined with ELISA (Crystal Chem Inc., IL).

High-Resolution Respirometry Measurement of MT Respiration in Permeabilized Kidney Samples

MT respiratory rate and ROS production were measured simultaneously in saponin-permeabilized renal cortex slices by high-resolution respirometry (Oroboros instruments-Oxygraph-2k, Innsbruck, Austria), according to substrate-uncoupler-inhibitor-titration (SUIT) protocol as previously described (14, 15). Briefly, representative cortices of the left and right kidneys (around 1.5–2.0 mg per slice) were incubated with 100 μg/mL saponin added to relaxation and preservation (RP) solution (7.2 mM K₂EGTA, 2.8 mM CaK₂EGTA, 20 mM imidazole, 0.5 M dithiothreitol, 20 mM taurine, 5.7 mM ATP, 14.3 mM phosphocreatine, 6.6 mM MgCl₂, 50 mM K-MES, pH 7.1) for shaking at 50 rpm for 40 min, followed by 3x washes with MiRO5 buffer (60 mM K-lactobionate, 0.5 mM EGTA, 3 mM MgCl₂, 20 mM taurine, 10 mM KH₂PO4, 20 mM HEPES, 110 mM sucrose, and 1 g/L BSA, pH 7.1) for 10 min each. Kidney slices were then weighed and transferred to the Oxygraph-2k chamber with MiRO5 solution at 37°C. Oxygen was added to the solution in each chamber to increase oxygen concentration above 350 μM. After oxygen concentration stabilized, MT respiration rate was measured as follows: addition of glutamate (final concentration, 10 mM) and malate (2 mM) as complex I substrates; addition of ADP (5 mM) to activate oxidative phosphorylation (OXPHOS); cytochrome c (10 μM) addition to assess the intactness of the outer MT membrane. Absence of an increase in respiration rate after adding cytochrome c indicates a high-quality preparation; succinate (20 mM), a complex II substrate addition to further stimulate respiration by activating convergent electron flow from Comple I + II into the Q junction, a step referred as maximum OXPHOS capacity; addition of ATP synthase inhibitor, oligomycin (5 μM), to measure the ATP-linked respiration rate. Inhibition of respiration with complex I inhibitor rotenone (1 μM) and complex III inhibitor antimycin A (5 μM) were added to measure non-OXPHOS respiration. Substrate and inhibitor concentrations were selected based on previous experimental determination of doses needed to stimulate or reduce the respiration rate maximally. Finally, data analysis was performed using DATLAB 4.2 software (Oroboros), and tissue respiration was expressed as oxygen flux (pmol/s/mg).

Simultaneous Measurement of Hydrogen Peroxide Fluxes as an Indicator of MT Superoxide Generation in Permeabilized Kidney Cortex Samples

MT superoxide flux was measured simultaneously with respirometry in the O2k-Fluorometer using the H₂O₂-sensitive probe Amplex Ultra Red (Invitrogen). Amplex UltraRed (AmR, 10 µM), 1 U/mL horse radish peroxidase (HRP) and 5 U/mL superoxide dismutase were added to the MiRO5 solution. The reaction product between AmR and H₂O₂, catalyzed by HRP, is fluorescent. Volume-specific H₂O₂ fluxes were calculated in real time by the DatLab software from the time derivative of the fluorescent signal over time (16).

Renal Histology

Paraffin-embedded sections (5 µm) were prepared from kidneys fixed in 10% phosphate-buffered formalin. Periodic acid–Schiff (PAS) stain was used to analyze renal morphology changes. Sections were scored in a blinded, semiquantitative manner using an established scoring scale. For each animal, at least 10 high-power (×400) fields were examined. The percentages of glomeruli that displayed glomerular capillary wall thickening, mesangial expansion, nodular sclerosis and global glomerulosclerosis were scored as follows: 0 = none, 1 = <10%, 2 = 10–25%, 3 = 25–50%, 4 = 50–75%, and 5 = >75%.

Transmission Electron Microscopy

Renal cortical tissues were cut into small pieces and rapidly immersed in tissue fixative buffer. After thin sectioned (70 nm in thickness) and applied on copper grids, the stained grid was loaded in a JOEL JEM1400 transmission electron microscopy (TEM) with an ANT camera system. At least five sections from each sample were examined under TEM. The entire sections were thoroughly viewed at low magnification (×300) for integrity and quality of stained tissues. Details of ultrastructural alterations were further investigated at high magnifications.

Statistical Analysis

The number of animals used was based on our previous studies and power analysis to predict a 20% difference among groups. Data are expressed as mean ± SE. A P value of <0.05 indicates a significant difference. Significant differences between two groups were determined by Student’s t test. Significant differences between multiple groups were determined by one-way ANOVA followed by Tukey’s post hoc test. Histological scoring was assessed using a nonparametric Kruskal–Walls test followed by Dunn’s multiple comparison test. Individual statistical analyses are described in the figure legends.

RESULTS

Metabolic Characteristics of Wistar Rats in Different Groups at Baseline and After AC Surgery

Significantly increased blood glucose levels confirmed the type 1 DM phenotype of Wistar rats after STZ injection. As shown in Table 2, blood glucose levels in rats of DM and DM + AC groups were significantly higher than in rats of AC groups that only received saline injection (419 ± 20 in DM and 418 ± 18 in DM + AC vs. 88 ± 4 mg/dL in AC). Rats in DM + AC+MitoTEMPO group had similar blood glucose levels compared with untreated DM + AC rats at weeks 3 and 8 after AC surgery, suggesting that MitoTEMPO treatment did not change blood glucose levels throughout the 6-wk treatment period.

Table 2.

Blood glucose levels, body weight, food intake in different groups of Wistar rats before and at 3 and 8 wk after AC or sham surgery

	Timepoints	DM	AC	DM + AC	DM + AC + MitoTEMPO
Blood glucose levels, mg/dL	Before AC	415 ± 29	93 ± 3*	399 ± 25	361 ± 15
	3 wk after AC or sham	426 ± 14	97 ± 3*	384 ± 14	396 ± 27
	8 wk after AC or sham	419 ± 20	88 ± 4*	418 ± 18	378 ± 22
Body weight, g	Before AC	413 ± 20	465 ± 19	426 ± 11	462 ± 10
	4 wk after AC or sham	434 ± 17	496 ± 21	453 ± 6	484 ± 22
	8 wk after AC or sham	442 ± 16	535 ± 22*	465 ± 10	483 ± 20
Food intake weight, g	Before AC	51 ± 1	28 ± 1*	45 ± 3	45 ± 5
	3 wk after AC or sham	50 ± 3	29 ± 1*	51 ± 4	52 ± 5
	8 wk after AC or sham	56 ± 3	28 ± 1*	56 ± 3	52 ± 4

Values are means ± SE; n = 6–8 animals per group. AC, aorta constriction; DM, diabetes; HTN, hypertension. *P < 0.05 compared with DM, DM + AC, and DM + AC + MitoTEMPO groups of rats by one-way ANOVA and Tukey’s multiple comparisons test at each timepoint.

We also measured body weight and food intake in different groups of Wistar rats at baseline and at 3 and 8 wk after AC surgery (Table 2). Rats in DM and DM + AC groups had significantly decreased body weight and increased daily food intake compared with rats in AC group. DM + AC+MitoTEMPO rats have similar body weight and food intake compared with DM + AC rats. Overall, type 1 DM was created with similar hyperglycemia and metabolic phenotypes in DM, DM + AC, and DM + AC+MitoTEMPO groups of rats. MitoTEMPO treatment for 6 wk did not significantly affect body weight, food intake, and blood glucose levels in DM + AC+MitoTEMPO rats (Table 2).

Changes in Blood Pressure and Heart Rate in Different Groups of Rats Before and After AC or Sham Surgery

At baseline, mean arterial pressure (MAP) showed no significant differences among all groups of rats. Following AC surgery, there was a substantial increase in MAP in the AC and DM + AC groups. As shown in Fig. 1B, MAP increased from an initial value of 107 ± 2 mmHg at baseline to 128 ± 7 mmHg at 3 wk after AC surgery, and further rose to 139 ± 3 mmHg in the DM + AC rat group. In rats from the DM + AC+MitoTEMPO group, MAP at the 3-wk post-AC surgery was similar to that of the DM + AC group. However, at the 8-wk post-AC surgery, there was a trend toward lower MAP in the DM + AC+MitoTEMPO group compared with the DM + AC group, although the difference was not statistically significant (131 ± 2 vs. 139 ± 3 mmHg, P = 0.09). In rats with type 1 DM in the DM, DM + AC, and DM + AC+MitoTEMPO groups, HR was significantly lower than HR in rats with normal glycemia in the AC group at all time points (Fig. 1C). Notably, HR in the rats of the DM + AC+MitoTEMPO group was significantly higher than in rats from DM + AC and DM groups at 8 wk after AC surgery.

MitoTEMPO Treatment Attenuated Increased Total 24-h UAE in Rats Exposed to DM + HTN

Twenty-four-hour urine output was significantly higher in all groups of rats with DM than in nondiabetic AC rats at baseline and after AC surgery. There were no significant differences in urine output in DM, DM + AC, and DM + AC+MitoTEMPO groups at 3 and 8 wk after AC or sham surgery (Fig. 2A).

Figure 2.

Twenty-four-hour urine output, total, and single kidney urinary albumin excretion in different groups of rats before, 3 and after 8 wk after AC or sham surgery. A: 24-h urine output in DM, AC, DM+AC and DM+AC+MitoTEMPO groups of rats (n = 5 or 6 animals/group; *P < 0.05 when comparing urinary albumin excretion in AC vs. all other groups at the same timepoint by Tukey’s post hoc test). B: 24-h urinary albumin excretion in DM, AC, DM+AC, and DM+AC+MitoTEMPO groups of rats (n = 5 or 6 animals/group, P < 0.05 when comparing urinary albumin excretion in DM+AC rats with all other groups at 8 wk after AC or sham surgery by Tukey’s post hoc test). C: single left and right kidney urinary albumin excretion in DM, AC, DM+AC, and DM+AC+MitoTEMPO groups rats at 8 wk after AC or sham surgery (n = 5 or 6 animals/group; *P < 0.05 when comparing the right kidneys in DM+AC rats with the right kidneys of other groups of rats by Tukey’s post hoc test). AC, aorta constriction; DM, diabetes.

During the baseline period, UAE was similar in all groups of rats. However, in DM rats, UAE increased slowly over time, reaching a significant level of 10.8 ± 3.7 mg/24 h at 13 wk after STZ injection. Similarly, hypertensive AC rats also exhibited a slow increase in UAE, which reached 7.2 ± 2.4 mg/24 h after 8 wk of HTN. Most notably, the combination of DM and HTN resulted in a synergistic increase in 24-h UAE, reaching 32.2 ± 10.1 mg/24 h in DM + AC rats after 8 wk (Fig. 2B). In contrast, DM + AC rats that received MitoTEMPO treatment showed only a slight increase in UAE, which was significantly lower than that observed in DM + AC rats at 3 wk and 8 wk after AC (8.0 ± 1.8 mg/24 h in DM + AC+MitoTEMPO vs. 32.2 ± 10.1 mg/24 h in WT-DM + AC at 8 wk after AC) (Fig. 2B).

To further investigate the specific kidney responsible for the substantial increase in UAE, we collected urine from the left and right kidneys separately in the same rats from all groups at 8 wk after AC or sham surgery. Figure 3C shows that UAE was markedly elevated in the right kidneys of DM + AC rats (3.3 ± 0.5 µg/min), which were exposed to both hyperglycemia and high BP, in contrast to the left kidneys of the same animals (1.4 ± 0.4 µg/min), which were exposed only to hyperglycemia. In addition, both kidneys in DM-only rats, exposed only to hyperglycemia, exhibited similar UAE compared with left kidney of DM + AC rats. Both left and right kidneys of DM + AC+MitoTEMPO rats showed significantly lower UAE compared with untreated DM + AC rats (0.5 ± 0.2 vs. 1.4 ± 0.4 in left kidneys and 1.9 ± 0.2 vs. 3.3 ± 0.5 µg/min in right kidneys, respectively). These findings suggest that the increased urinary albumin in DM + AC rats primarily originated from the right kidneys exposed to elevated BP and hyperglycemia. MitoTEMPO treatment mitigated increased UAE induced by high glucose combined with high BP.

Figure 3.

Kidney weight and single kidney GFR in different groups of rats at 8 wk after AC or sham surgery. Single left and right kidney GFR (A), left and right kidney weight (B), and GFR normalized to the kidney weight (C) in DM, AC, DM + AC, and DM + AC + MitoTEMPO groups of rats at 8 wk after AC or sham surgery are shown (n = 5 or 6 animals/group; *P < 0.05 when comparing the right kidneys in DM+AC rats and with the right kidney from other groups by Tukey’s post hoc test). AC, aorta constriction; DM, diabetes; GFR, glomerular filtration rate.

MitoTEMPO Treatment Attenuated Declines in GFR in Kidneys Exposed to DM + HTN

To assess the interaction of high BP and hyperglycemia on kidney function, we conducted measurements of single left and right kidney GFR in all rat groups at 8 wk following AC or sham surgery. Despite hyperglycemia, which usually increases GFR, the average GFR in the right kidneys of DM + AC rats was lower than in the left kidneys exposed to lower BP in the same rats (1.2 ± 0.1 mL/min in right kidney vs. 1.7 ± 0.1 mL/min in left kidney, P < 0.05) and significantly lower than GFR in the right kidneys of normotensive DM-only rats (1.2 ± 0.1 vs. 2.1 ± 0.2 mL/min, P < 0.05) (Fig. 3A). These findings suggest that increased BP may rapidly promote the decline of GFR in rats with DM. The average GFR in the right kidneys of DM + AC+MitoTEMPO rats was significantly higher than that in the right kidneys of DM + AC rats (2.7 ± 0.1 vs. 1.2 ± 0.1 mL/min) (Fig. 3A). This observation indicates that MitoTEMPO treatment may substantially mitigate the decline of GFR in kidneys exposed to both hyperglycemia and HTN.

At the end of the in vivo studies, we collected and weighed left and right kidneys isolated from each group. In the DM group of rats, the weights of the left and right kidneys were similar and significantly heavier than the kidneys from rats in AC-only groups (Fig. 3B), which is consistent with a hyperfiltration state of the kidneys due to DM. We also observed that in rats from AC, DM + AC, and DM + AC+MitoTEMPO groups, the right kidneys were heavier than the left kidneys (Fig. 3B), suggesting a higher BP-induced kidney hypertrophy. After normalizing the single kidney GFR with kidney weight, the results (Fig. 3C) paralleled those observed in Fig. 3A.

MitoTEMPO Treatment Attenuated MT Dysfunction and Reduced MT-Derived ROS Generation in Kidneys Exposed to DM + HTN

To examine MT function in kidneys, we used Oroboros respirometry to measure MT oxygen consumption rates in the kidney samples isolated from the cortex of left and right kidneys. We assessed the maximal oxidative phosphorylation (OXPHOS) capacity induced by Complex I and Complex II substrates in the presence of ADP as an indicator of MT respiration capacity. ATP-linked respiration rate was also measured as an indicator of MT respiration efficiency by measuring the changes in O₂ consumption rate after adding the ATP synthase inhibitor oligomycin. The results revealed that the right kidneys from DM + AC rats exhibited significantly reduced oxygen consumption rates and ATP-linked respiration compared with kidney samples isolated from the right kidneys of DM-only and AC-only rats (Fig. 4, A and B). Notably, the right kidneys from DM + AC+MitoTEMPO rats showed significantly higher MT respiration rates and ATP-linked respiration (Fig. 4, A and B) when compared with right kidney samples of DM + AC rats.

Figure 4.

Mitochondrial oxygen consumption rate and superoxide generation measured by Oroboros respirometry in the renal cortex of DM, AC, DM+AC, and DM+AC+MitoTEMPO groups of rats. A: maximal mitochondrial oxygen consumption rate. B: ATP-linked oxygen consumption rates. C: simultaneous measurement of superoxide generation in DM, AC, DM+AC, and DM+AC+MitoTEMPO groups of rats at 8 wk after AC or sham surgery (n = 6 or 7 animals/group; *P < 0.05 when comparing the right kidneys in DM+AC rats and with the right kidney from other groups of rats by Tukey’s post hoc test). AC, aorta constriction; DM, diabetes.

In addition, we simultaneously measured MT H₂O₂ flux as an indicator of MT-derived ROS generation, and the results indicated a significantly higher H₂O₂ flux in the right kidneys of DM + AC rats than right kidneys of DM rats (Fig. 4C). The right kidneys of DM + AC+MitoTEMPO rats showed a significant reduction in H₂O₂ flux rate compared with right kidneys of untreated DM + AC rats (Fig. 4C).

MitoTEMPO Treatment Attenuated Renal Glomerular Structural Changes in Kidneys Exposed to DM + HTN

As shown in Fig. 5, the right kidneys of DM + AC rats (Fig. 5D), which are exposed to both hyperglycemia and high BP, exhibited more severe renal damage compared with the left kidneys within the same group of rats (Fig. 5C), as well as the right kidneys of DM rats (Fig. 5A) and right kidneys of hypertensive AC rats (Fig. 5B). These morphological changes were characterized by thickening of Bowman’s capsule and glomerular capillary walls, increased glomerular mesangial matrix, increased glomerular sclerosis, and tubular cell injury (Fig. 5, F–I). In contrast, the right kidneys of DM + AC+MitoTEMPO rats displayed substantially less damage than untreated DM + AC rats. As shown in Fig. 5J, the overall kidney injury score in the right kidneys of DM + AC+MitoTEMPO rats was significantly lower than right kidneys of untreated DM + AC rats.

Figure 5.

Renal morphological changes in different groups of rats at 8 wk after AC or sham surgery. Representative images of PAS staining of kidney slices from the right kidney of DM rats (A), right kidney of WT-AC rats (B), left kidney of DM+AC rats (C), right kidney of DM+AC rats (D), and right kidney of DM+AC+MitoTEMPO rats (E). Glomerular and tubular injury can be observed in diabetic-hypertensive right kidneys in DM+AC rats, as shown in images F–I in magnified view. F: glomerular sclerosis (arrow) and tubular cell injury (arrowhead). G: mesangial matrix expansion (arrow). H: glomerular capillary collapse (arrowhead) and thickened tubular basement membranes (arrow). I: protein cast in Bowman’s space and tubules (arrow). J: renal injury was scored based on the overall renal morphological changes in DM, AC, DM+AC, and DM+AC+MitoTEMPO groups of rats (n = 6 or 7 animals/group, *P < 0.05 when comparing the right kidney in DM+AC rats with the right kidney from other groups of rats by Kruskal nonparametric test followed by Dunn’s multiple comparisons test). AC, aorta constriction; DM, diabetes.

Changes in glomerular ultrastructure were evaluated by electron microscopy of the right kidneys of four groups of rats. Right kidneys of DM + AC rats exposed to the DM and HTN exhibited disrupted glomerular ultrastructure, including detachment of the endothelial layer in the glomerular capillary, thickening of the glomerular basement membrane, effacement of podocytes, and fusion of podocyte foot processes (Fig. 6, C and G). In contrast, glomeruli from MitoTEMPO-treated DM + AC+MitoTEMPO rats displayed preserved ultrastructural features, as illustrated in Fig. 6, D and H, which were similar to glomeruli from DM and AC-only groups. We also examined ultrastructural changes in left kidneys isolated from different groups of rats, and results showed similar and relatively normal glomerular basement membrane and podocyte phenotype (Fig. 7).

Figure 6.

Representative images of electron microscopy scans in right kidneys of different groups of rats at 8 wk after AC or sham surgery (A–D, magnification ×12,700; E–H, magnification ×38,100). A and E: right kidney of DM rats. B and F: right kidney of AC rats. C and G: right kidney of DM+AC rats. D and H: right kidney of DM+AC+MitoTEMPO rats. CL, capillary lumen; US, urine space; asterisk indicates basement membrane; arrow indicates foot process effacement; arrowhead indicates disrupted endothelium. AC, aorta constriction; DM, diabetes.

Figure 7.

Representative images of electron microscopy scans in left kidneys of the same rats in Fig. 6 at 8 wk after AC or sham surgery (A–D, magnification ×12,700). A: left kidney of DM rats. B: left kidney of AC rats. C: left kidney of DM+AC rats. D: left kidney of DM+AC+MitoTEMPO rats. CL, capillary lumen; US, urine space; asterisk indicates basement membrane. AC, aorta constriction; DM, diabetes.

DISCUSSION

A major finding of our study is that the combination of DM with moderate HTN promoted rapid development of renal injury, increased albuminuria, reduced GFR, impaired MT function, and increased MT ROS generation. Mito-TEMPO, an MT-targeted superoxide dismutase mimetic, significantly ameliorated these changes. In contrast to kidneys exposed to DM + HTN, kidneys exposed to DM for 15 wk or moderate HTN alone for 8 wk had minimal albuminuria and renal injury.

Interaction of DM and HTN Promotes Renal Injury

Studies in humans indicate that in the early stages of diabetic nephropathy, before the development of marked increases in BP, there is glomerular hyperfiltration and albuminuria but mild renal injury (17). Previous studies also suggest that the superimposition of HTN on type 1 or type 2 DM produces much more severe kidney injury than DM alone (9, 18). Each 10 mmHg increase in mean systolic BP was associated with a 15% increase in the hazard ratio for developing micro- and macroalbuminuria and impaired kidney function (19). In fact, increased BP may be a prerequisite for the rapid progression of diabetic nephropathy.

Our current study showed that either DM or HTN alone caused only minimal albuminuria and renal injury. However, the coexistence of HTN and hyperglycemia synergistically promoted renal injury as reflected by increases in 24-h UAE, reduced GFR, reduced MT ATP-linked respiration, and substantial increases in indices of histological renal injury. In contrast, the left kidneys of DM + AC rats, which were exposed to hyperglycemia and normal BP, had only mild renal dysfunction and injury, similar to the kidneys in control DM rats. These findings are consistent with our previous study showing that in GK type 2 DM rats, the superimposition of HTN on DM caused severe kidney injury, whereas DM alone or HTN alone caused minimal kidney injury (16). Thus, increased blood glucose in DM appears to interact synergistically with high BP to cause chronic renal injury in type 1 and type 2 DM.

Hemodynamic Factors Could Contribute to Kidney Injury Caused by DM + HTN

Although the mechanisms by which DM and HTN interact to cause renal injury are not fully understood, hemodynamic factors could play a major role. Our model for inducing HTN by AC between the renal arteries closely resembles the classic 2 K-1C HTN model. Previous studies in mice and rats (20–22) have demonstrated significant increases in intrarenal renin and angiotensin II (Ang II) levels in the clipped kidney of 2 K-1C model compared with the contralateral unclipped kidney, indicating intrarenal renin-angiotensin system (RAS) activation in the underperfused kidney, which contributes to elevated circulating blood levels of Ang II in both kidneys. In the present study, the left kidney with normal or slightly reduced perfusion pressure had much less injury than the contralateral right kidney exposed to higher perfusion pressure but presumably less activation of the intrarenal RAS. Although we did not assess renal levels of Ang II, both kidneys were exposed to the same levels of circulating Ang II. These findings suggest that increased blood pressure plays a key role in kidney injury associated with DM + HTN and that underperfused kidneys with normal or slightly reduced perfusion pressure are protected from injury despite exposure to activation of the intrarenal RAS activation as well as DM. There is compelling evidence that hyperglycemia impairs normal autoregulation of GFR and renal blood flow, which may contribute to hyperglycemia-induced glomerular hyperfiltration (23, 24). To the extent that renal autoregulation is impaired, increases in systemic arterial pressure would be transmitted to the glomerular capillaries in diabetic kidneys to a greater extent than in normal kidneys, and could lead to kidney injury and subsequent decline in GFR after an initial period of hyperfiltration (25).

In the present study, we found that GFR was significantly reduced and UAE was increased in the right kidneys (exposed to elevated BP) of DM + AC rats after only 8 wk HTN. This rapid decrease of GFR in the hypertensive kidneys of DM + AC rats is similar to the decline observed in patients with diabetes who first undergo glomerular hyperfiltration followed by reductions of GFR that are associated with nephron injury preceding a further decline of GFR to subnormal levels as diabetic nephropathy progresses (26). In contrast, the left kidney (protected from high BP) of DM + AC rats had minimal injury despite being exposed to DM for 15 wk. Therefore, one potential mechanism for the synergy of HTN and DM to promote kidney injury may be the greater mechanical stress on the glomerular capillaries due to glomerular hyperfiltration and greater transmission of increases in systemic arterial pressure to glomeruli. Thus, long-lasting intraglomerular mechanical challenges may promote glomerular injury in endothelial cells and podocytes, ultimately causing podocyte loss in advanced CKD with overt proteinuria (27, 28). However, how different cells in the glomerulus respond to metabolic stress in DM, mechanical stress in HTN, and the potential synergistic effects of DM + HTN for development of glomerular injury in CKD are still not well understood.

Another important observation from our study is that even with significantly increased systemic arterial pressure in the right kidneys of AC-only rats, they did not exhibit severe increases in UAE or structural changes and maintained normal MT function. Although HTN over long periods may damage the small blood vessels in the kidneys, severe damage may not occur when normal autoregulation and increased preglomerular resistance protect the glomeruli from injury, or major injury may require years rather than weeks to develop (29).

Role of MT Dysfunction in the Development of Diabetic-Hypertensive Renal Injury

DM is a state of nutrient excess with higher glucose and free fatty acids (FFAs) in the circulation but impaired glucose utilization (30). MT actively adapts to these metabolic changes to balance energy supply with energy demand by increasing MT biogenesis and oxidative capacity (31). However, this adaptation during early stages of DM is not harmless as it is accompanied by a reduction in MT respiratory efficiency with increased proton leakage, uncoupling, and generation of ROS, which may make the kidneys more vulnerable when extra energy demands are superimposed, such as when HTN-induced mechanical stress is also present. Factors like advanced glycation end products (AGEs) and the fructose 6-phosphate/hexosamine pathway may contribute to diabetic podocyte injury (30). Also, a chronic increase in MT dysfunction is critical in triggering cellular apoptosis. In our study, we used Oroboros respirometry to simultaneously measure oxygen consumption alongside ROS generation at different stages of MT respiration. We observed the highest influx of H₂O₂ during the stage of maximal respiration stimulated by both CI and CII substrates in the kidneys exposed to the DM + HTN. These data suggest that an elevation in ROS generation accompanies the increased MT respiration rate, typically indicative of impaired CI and III functions in the MT electron transfer chain. Our studies also demonstrate that the maximal OXPHOS capacity and ATP-linked respiration in kidneys exposed solely to hyperglycemia or HTN remained unchanged. This result may be attributed to their relatively normal GFR and normal tubular reabsorption. However, in the case of DM + AC rats, the right kidney exhibited a significantly lower GFR, reduced MT oxygen consumption rates, and increased ROS generation. These results support the rationale of targeting the MT ROS as a central hub of the pathogenesis of kidney injury in a diabetic-hypertensive environment.

MT-Targeted ROS Scavenger as a Potential Therapeutic Approach for Renal Injury

Mito-TEMPO has recently emerged as an MT-targeted antioxidant. This compound combines the antioxidant piperidine nitroxide (TEMPO) with the lipophilic cation triphenylphosphonium (TPP+). TEMPO acts as a superoxide dismutase mimetic, effectively dismutating superoxide during its catalytic cycle, whereas TPP+ is a membrane-permeant cation that accumulates several 100-fold within MT due to the MT membrane potential (32). This unique combination results in an MT-targeted compound with potent superoxide scavenging capabilities.

MitoTEMPO has been documented to reduce mitochondria-derived reactive oxygen species (MT-ROS) generation in animal models of diabetic nephropathy (33–35), including db/db mice (36), Ang II infusion hypertension models (12, 29), and DBA/2J diabetic mice (37). Glomerular endothelial cells and podocytes have been identified as contributors to glomerular injury associated with increased MT-ROS production. Our study provides evidence for the important role of MT-ROS in mediating the synergistic effects of DM and HTN in inducing glomerular injury. Our electron microscopy examination demonstrated that injury to endothelial cells and podocytes was evident in the right kidneys of DM + AC rats, and MitoTEMPO treatment can markedly attenuate podocyte and endothelial cell injury in rats with DM + HTN. Therefore, generation of MT-ROS appears to be a causative factor in development of kidney injury resulting from the combination of DM and HTN as administration of mitoTEMPO, which inhibits MT-ROS generation, protects the kidneys from injury. Moreover, this protection persisted without significant changes in blood glucose and blood pressure levels following the treatment.

DM and HTN are systemic diseases known to exert adverse effects on various target organs, including the heart and vasculature. Previous studies have demonstrated increased MT-ROS formation in cardiomyocytes and vascular endothelial cells in humans with DM as well as in animal models of DM (38, 39). Consequently, the protection from kidney injury observed after systemic administration of MitoTEMPO in our studies could conceivably result from inhibition of MT-ROS formation in multiple tissues, including the kidneys. However, this seems unlikely since both kidneys were exposed to the same circulating blood carrying the same potential products of MT-derived ROS from the other tissues of the body and yet only the kidneys exposed to HTN developed substantial kidney injury.

There are still important questions that remain unanswered and require further investigation. One of our primary areas of interest is the molecular mechanisms behind the synergistic effect of hyperglycemia and HTN in increasing ROS production in MT. It is well-established that impaired complex I and III activities play a significant role in MT ROS generation. However, a cause-effect relationship between MT ROS generation and impaired complex activity remains unclear. Additional research will be needed to uncover the molecular pathways that transmit mechanical stress from the cell membrane caused by HTN and renal vasodilation, thereby inducing alterations in MT function and ultimately resulting in increased MT ROS generation. In addition, it is crucial to explore how various cell types, including glomerular endothelial cells, mesangial cells, and podocytes within the glomerulus, respond differently to increased ROS production. Better understanding the molecular pathways involved in the pathogenesis of diabetic nephropathy in the context of coexisting HTN is critical to improved treatment and prevention of dominant causes of CKD and ESRD.

Perspectives

Our results indicate that even mild HTN and DM can interact to promote renal dysfunction, albuminuria, decreased MT OXPHOS, increased ROS generation, and glomerular injury. Administration of an MT-targeted ROS scavenger attenuated renal dysfunction and reduced ROS generation in kidneys exposed to hyperglycemia and high BP. However, the molecular pathways that transduce HTN-induced mechanical forces and interactions with hyperglycemia to promote MT dysfunction and the specific responses of different glomerular cell types are still not fully understood and warrant further investigation. An important implication of our study is that simultaneous tight control of HTN and hyperglycemia may be required to slow or prevent the progression of diabetic nephropathy. Targeting MT and preventing oxidative stress by overproduction of ROS may be an important therapeutic strategy for slowing diabetic hypertensive nephropathy.

DATA AVAILABILITY

Data will be made available upon reasonable request.

GRANTS

This work was supported by National Institutes of Health Grants R00DK113280, R01DK121411, 1R01HL163076, P20GM104357, P30GM149404, and U54GM115428.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Z.W., Y.F., and J.E.H. conceived and designed research; Z.W. and Y.F. performed experiments; Z.W. and Y.F. analyzed data; Z.W. and Y.F. interpreted results of experiments; Z.W. and Y.F. prepared figures; Z.W. drafted manuscript; Z.W., Y.F., A.A.d., J.M.d., A.M., A.C.M.O., X.L., J.S., and J.E.H. edited and revised manuscript; Z.W. and J.E.H. approved final version of manuscript.