Accumulation of 8-Hydroxy-2′-Deoxyguanosine and Mitochondrial DNA Deletion in Kidney of Diabetic Rats

Male Wister rats were bred under pathogen-free conditions at Kyushu University Animal Center, Fukuoka, Japan. Two experimental groups were formed: STZ-induced diabetic rats and age-matched control rats. The animals had free access to tap water and standard chow (Clea Japan, Tokyo).

At 8 weeks of age, the rats were injected intraperitoneally with STZ (Sigma, St. Louis, MO) at 80 mg/kg body wt. One day after injection, the development of diabetes was verified by the presence of hyperglycemia (plasma glucose level ≥300 mg/dl). The rats were used for experiments at 4 or 8 weeks after the onset of diabetes. This study was approved by the Committee of the Ethics on Animal Experiments, Graduate School of Medical Sciences, Kyushu University.

Two-milliliter samples of well-mixed 24-h urine were collected from rats. The samples were purged of air with a steam of nitrogen to prevent the artificial formation of 8-OHdG and stored frozen at −80°C until analyzed. Urine samples were centrifuged at 2,000g for 20 min, and after proper dilution, the supernatant was used for the determination of 8-OHdG by a competitive enzyme-linked immunosorbent assay (ELISA) kit (8-OHdG Check; Japan Institute for the Control of Aging, Fukuroi, Japan). The determination range was 0.5–200 ng/ml. The urinary 8-OHdG was expressed as total amounts excreted in 24 h.

At 8 weeks after the onset of diabetes, rats were anesthetized by intraperitoneal injection of pentobarbital (80 mg/kg) before performance of surgical procedures. The kidney was rapidly excised and separated into cortices and papillae. Then, the samples were frozen in liquid nitrogen and kept at −80°C until analyzed. On the day of assay, the samples were homogenized in 5 ml of 50 mmol/l Tris-HCl (pH 7.4) with 20 strokes of Dounce homogenizer. The homogenates were centrifuged at 800g for 10 min to precipitate nuclear fraction. The supernatant was again centrifuged at 7,000g for 10 min to yield mitochondrial fraction. One milliliter of solution containing 10 mmol/l EDTA, 10 mmol/l Tris-HCl (pH 8.0), 150 mmol/l NaCl, and 0.2% SDS was added to each fraction. After homogenization, the mixture was incubated at 56°C for 70 min with proteinase K (700 μg/ml) and then heated at 95°C for 10 min. DNA was extracted with equal volumes of phenol, chloroform, and isoamyl alcohol (25:24:1) and then with chloroform. DNA was precipitated with a one-tenth volume of 2 mol/l NaCl and two volumes of 70% ethanol at −20°C for 2 h and rinsed with ethanol. DNA was resuspended in 10 mmol/l Tris-HCl and 0.1 mmol/l EDTA (pH 8.0). Five microliters of 200 mmol/l sodium acetate buffer (pH 4.8) and 5 μg nuclease P1 (USB, Cleveland, OH) (1 mg/ml in 10 mmol/l sodium acetate) were added to 45-μl DNA samples. After purging with a nitrogen steam, the mixtures were incubated at 37°C for 1 h to digest the DNA to nucleotides. Then, 5 μl of 500 mmol/l Tris-HCl (pH 8.0), 10 mmol/l MgCl₂, and 0.6 units Escherichia coli alkaline phosphatase (Toyobo, Osaka, Japan) were added to the samples. After purging with a nitrogen steam, the mixtures were incubated at 37°C for 1 h to hydrolyze the nucleotides to nucleosides. The nucleoside samples were used for the determination of 8-OHdG by competitive ELISA kit as described above.

Total DNA was isolated from kidneys as described above at 4 or 8 weeks after the onset of diabetes. Primers for the PCR amplifications used are shown in Table 1. The primer sets for amplification of 4,834-bp deletion mtDNA were L782 and H1309, L768 and H1296, and L797 and H1309. The primer set for control amplification of wild-type mtDNA was L1576 and H1626. For amplification of a longer mtDNA fragment (5,881–15,142), a primer set of L588 and H1511 (29) was used. All primers were synthesized by Invitrogen (Tokyo). Sequence and numbering are based on the published rat mtDNA sequence (23,26).

PCRs contained 0.2 mmol/l dNTP, 0.3 μmol/l of each primer, 1.0 unit Taq polymerase (Toyobo), and various amounts (0.01–0.2 μg) of starting total DNA as template in a 50 μl reaction solution. The thermal cycling condition was started with one cycle at 94°C for 2 min. This was followed by 35 cycles at 94°C for 15 s, 60°C for 30 s, 68°C for 1 min, and 68°C for final extension for 5 min. Amplification was carried out in a Thermal Cycler system (Astec, Fukuoka, Japan). PCR products were electrophoresed on a 5% polyacrylamide gel and visualized by bistragreen staining. The number of mtDNA deletion products was determined for each sample by densitometry.

Using primer sets of L588 and H1511, multiple PCR products were amplified from mtDNA templates isolated from diabetic kidney but not from mtDNA templates isolated from control kidney (Fig. 1A). These results suggest that multiple mtDNA deletions might exist in diabetic kidney. Therefore, in the present study, the frequency of 4,834 mtDNA deletion, which was reported to be one of the most frequent deletions (26–28), was evaluated for quantitation. Three different sets of primers were used on all samples to ensure that the PCR products seen were not the result of primer misannealing. The 4,834 mtDNA deletion products shifted appropriately in size when different sets of primers were used, as shown in Fig. 1B. Levels of the 4,834-deleted mtDNA were expressed as the ratio of the signal intensity relative to that for the wild-type mtDNA.

Powdered frozen tissues from kidney, as described above, were homogenized on ice with a Polytron for 20 s in a buffer containing 50 mmol/l Tris-HCl (pH 7.4), 2% SDS, 1 mmol/l phenylmethylsulfonyl fluoride, and 10 μg/ml aprotinin and then homogenized again with 20 strokes of a Dounce homogenizer. The homogenates were centrifuged at 10,000g for 20 min to precipitate cell debris. The samples were separated on 12% SDS gel under reducing conditions and then immunoblotted with monoclonal anti-cytochrome C oxidase (COX) subunit I or anti-COX subunit IV antibody (Molecular Probe, Eugene, OR). The blots were then incubated with horseradish peroxidase-linked second antibody followed by chemiluminescence detection, according to the manufacture’s instructions (Pierce, Rockford, IL)

Starting at 8 weeks after the onset of diabetes, STZ-induced diabetic rats were treated with insulin. Intermediate-acting insulin (Novolin N; Novo Nordisk, Tokyo) was subcutaneously administered twice a day until the animals were killed. The doses of insulin to be given were determined each time from a sliding scale, according to blood glucose levels, to achieve strict glucose control. At 1 and 2 weeks after the induction of insulin treatment, the 8-OHdG levels in urine and kidney and the frequency of mtDNA deletion in kidney were measured.

All values shown are expressed as means ± SE. Statistical analysis was performed by ANOVA followed by Fisher’s comparison test.

The total amounts of urinary 8-OHdG excretion were significantly greater in STZ induced diabetic rats than in age-matched control rats at both 4 weeks (2,089 ± 259 ng/day, n = 6 vs. 793 ± 44 ng/day, n = 6, P < 0.01) and 8 weeks after the onset of diabetes (2,280 ± 230 ng/day, n = 6 vs. 632 ± 56 ng/day, n = 6, P < 0.001).

The levels of 8-OHdG in the mtDNA were significantly increased in both renal cortices and papillae of diabetic rats as compared with those from the control rats at 8 weeks after the onset of diabetes (cortices: 11.8 ± 1.1 ng/mg DNA, n = 6 vs. 4.2 ± 0.9 ng/mg DNA, P < 0.01; papillae: 19.2 ± 3.3 ng/mg DNA, n = 6 vs. 5.2 ± 2.1 ng/mg DNA, P < 0.01) (Fig. 2A). In contrast, the levels of 8-OHdG in nuclear DNA were not significantly increased in either renal cortices or papillae from diabetic rats (Fig. 2B).

PCR amplification showed that the 4,834 mtDNA deletion products from diabetic kidney were increased in a dose-dependent manner from 0.01 to 0.2 μg of template DNA, but those from control kidney had very low levels at the same doses of starting template DNA (Fig. 3). The mtDNA deletion products from diabetic kidney were increased in a cycle-dependent manner from 25 to 37 cycles as well (Fig. 3). The wild-type mtDNA products were also increased in a cycle dependency from 10 to 30 cycles (Fig. 3). Similar results were obtained on all samples.

To compare the frequency of mtDNA deletion between diabetic and control rats, the PCR products amplified in a condition of 0.2 μg template DNA and 35 cycles were quantified by densitometoric analysis. These semiquantitative analysis showed that the kidneys from diabetic rats had low levels of the mtDNA deletion products at 4 weeks after the onset of diabetes and that the difference between diabetic and control rats was not statistically significant. In contrast, at 8 weeks, the kidneys from diabetic rats exhibited markedly higher levels of mtDNA deletion products compared with those from control rats, as shown in Fig. 4.

STZ induced diabetic rats were treated with insulin, starting 8 weeks after the onset of diabetes. Glycemic control was achieved as follows: mean plasma glucose levels and fructosamine levels on the last day were 113.5 ± 6.7 mg/dl and 169.0 ± 1.2 μmol/l in control rats, 474.5 ± 33.9 mg/dl and 292.0 ± 15.32 μmol/l in nontreated diabetic rats, and 109.5 ± 2.0 mg/dl and 160.3 ± 2.5 μmol/l in insulin-treated diabetic rats, respectively. In nontreated diabetic rats, total amounts of urinary 8-OHdG were gradually increased, at least up to 10 weeks after the onset of diabetes, as shown in Fig. 5. Insulin treatment completely reversed the increased amounts of urinary 8-OHdG to control levels after 1 and 2 weeks.

In parallel, the levels of 8-OHdG in the mtDNA from kidney were also completely normalized to control levels after 2 weeks of insulin treatment (Figs. 6A and B). However, the increased frequency of deleted mtDNA in kidney of diabetic rats remained elevated to a similar degree as that in nontreated diabetic rats, even after 2 weeks of insulin treatment (Fig. 6C). Therefore, to exclude the effect of STZ on this deletion, insulin treatment was also performed between 4 and 8 weeks after the onset of diabetes. The institution of glycemic control with insulin treatment achieved near euglycemia between 4 and 8 weeks after the onset of diabetes and completely prevented the increased frequency of deleted mtDNA as well as increased the 8-OHdG levels in kidney of diabetic rats to control levels.

According to Western blot analysis, the amounts of mtDNA-encoded COX subunit I were significantly decreased (P < 0.01) in kidney from diabetic rats compared with those from control rats (Fig. 7). In contrast, the amounts of nuclear DNA-encoded COX subunit IV did not differ between diabetic and control rats (Fig. 7). In parallel with the changes of mtDNA deletion, increased amounts of COX subunit I in diabetic rats were completely normalized to control levels by preventive insulin treatment, and they remained elevated to a similar degree as those in nontreated diabetic rats after 2 weeks of interventive insulin treatment.

Enhanced oxidative stress has been considered to contribute to the pathological processes of diabetic complications, including nephropathy. However, the detailed biomolecular mechanism remains to be elucidated. In general, oxidative stress can affect nucleic acids and generate various modified bases in DNA. Among the latter, 8-OHdG is the most abundant and appears to play a crucial role in mutagenesis (30). Increased levels of 8-OHdG have been reported in urine (11,31), mononuclear cells (11,32), and skeletal muscles (12) of diabetic patients. Ha et al. (13) have also shown that the levels of 8-OHdG are increased in kidney tissues of STZ-induced diabetic rats. In the present study, we further revealed that levels of 8-OHdG were increased in mtDNA from both cortices and papillae of diabetic rats at 8 weeks after the onset of diabetes. In contrast, the levels of 8-OHdG in nuclear DNA were not significantly increased in either cortices or papillae of diabetic rats, consistent with the previous finding that mtDNA is more sensitive to oxidative stress compared with nuclear DNA. Changes in cortical mtDNA as well as papillary mtDNA might reflect those occurring in tubular cells, given the relative mass of the tubular cells versus glomerular cells in renal cortices. Changes in glomerular mtDNA remain to be elucidated. In parallel with the results of renal tissues, the amounts of urinary 8-OHdG excretion were markedly increased in diabetic rats, and those increases were reversed by insulin treatment. Although urinary 8-OHdG (33) and serum 8-hydroxy-guanine levels (34) are supposed to be markers of the total systemic oxidative stress in vivo, the relative contribution of renal 8-OHdG to urinary 8-OHdG should be evaluated in future studies.

As the possible consequence of increased formation of 8-OHdG, we showed the increase in mtDNA with a 4,834 deletion in kidney of diabetic rats. The kidneys of diabetic rats exhibited markedly higher levels of deleted mtDNA compared with those of control rats at 8 weeks after the onset of diabetes, but not at 4 weeks. These results suggest that deleted mtDNA may be accumulated with duration of diabetes in kidney tissues of diabetic rats. This specific large-scale deletion of mtDNA has been reported to exist in organs of aged rats (26–28). The age-associated large-scale deletions seen most often in humans are the 4,977- and 7,436-bp deletions, which are among the many mtDNA deletions first reported in the muscle of patients with Kearns-Sayer syndrome or chronic progressive external ophthalmoplegia syndrome (35–37). These deletions in rats and humans are flanked by two homologous repeats. These homologous repeats are supposed to induce such specific large-scale deletions by slip-mispairing during DNA replication (38). Because these deletions span a region that encodes five kinds of tRNA, respiratory enzyme subunits in complex I, complex IV, and complex V, the functions of the respiratory enzymes containing deleted mtDNA-encoded proteins subunits may decline gradually in the tissue cells as the deleted mtDNA are accumulated. The present results show that the proportion of deleted mtDNA to the total mtDNA reached ∼0.1% at 8 weeks after the onset of diabetes, using a kinetic PCR analysis. This level of deletion observed in kidney tissues of diabetic rats may still be too low to affect overall organ function. However, it is likely that this common deletion in mtDNA may be accompanied by other multiple deletions, as previously shown in aging tissues (39–41). According to Katsumata et al. (41), the total detection system for mtDNA deletions revealed that 235 types of deletions existed with various sizes in the heart of a 78-year-old human. The present study also suggested that multiple deletions might exist in kidney tissues of diabetic rats. Furthermore, the present study showed that the amounts of mtDNA-encoded COX subunit were significantly decreased in kidney tissues of diabetic rats as one of the possible consequences of mtDNA deletions. Thus, the defective respiratory chain induced by mtDNA mutations may generate more ROS and free radicals, which further enhance the oxidative damage to mtDNA as well as other biomolecules. Finally, such a vicious cycle of oxidative damage may result in deterioration in ATP synthesis and lead to cellular dysfunction and apoptosis (42,43). Oxidative stress and subsequent mtDNA mutations may contribute to the development of diabetic nephropathy. The correlation between cumulative effect of total multiple deletions and functional changes in kidney tissues of diabetes should be evaluated in future studies.

In addition, we examined the effect of intervention by insulin treatment on the increases in 8-OHdG level and deleted mtDNA in kidney tissues of diabetes. It is of great interest that intervention by insulin treatment starting at 8 weeks after the onset of diabetes rapidly normalized both increases in urinary 8-OHdG excretion and 8-OHdG levels in kidney of diabetic rats. This result clearly confirmed that the increases in urinary 8-OHdG excretion and 8-OHdG levels in mtDNA were caused by the diabetic state and not by a direct toxic effect of STZ. Hyperglycemia, a key clinical manifestation of diabetes, is supposed to generate ROS through various mechanisms, such as increased formation of AGEs (2), enhanced polyol pathway (6,7), increased superoxide release from mitochondria (44), and activation of NAD(P)H oxidase (45). Such rapid reversibility at least suggests that the contribution of AGEs to enhanced oxidative stress may not be very strong, considering that it would take longer to reverse the AGE level. The present results also suggest that repair of oxidative mtDNA damage induced by diabetes is very active. This finding is in agreement with recent reports (46,47) showing that the repair system of mtDNA damage may be more active than previously thought. The levels of 8-OHdG may represent the dynamic steady-state levels determined by the balance of oxidative attack and repair in mtDNA rather than simple accumulation of 8-OHdG. In contrast to 8-OHdG levels, an increased level of deleted mtDNA in diabetic kidney was not reversed by interventive insulin treatment. This result suggests that mtDNA mutations are not easily reversed by glycemic control after they have been established. This might be related to the phenomenon known as “the point of no return” observed in the progression of diabetic nephropathy. Further study regarding the effect of insulin treatment for a longer period is needed to prove this hypothesis.

In conclusion, our study demonstrated for the first time that the 4,834-bp deleted mtDNA was accumulated in parallel with an increased level of 8-OHdG in kidney of STZ-induced diabetic rats. This might be involved in the pathogenesis of diabetic nephropathy, and mtDNA might be a new therapeutic target for preventing the development of diabetic nephropathy.

This work was supported by a Grant-in-Aid for Scientific Research (no. 11671126) from the Ministry of Education, Science and Culture, Japan, and was supported in part by the Takeda Medical fund. This work was performed in part at Kyushu University Station for Collaborative Research.