The balance between the production of reactive oxygen species (ROS) and the antioxidant defense system determines the degree of oxidative stress. An increased production of ROS, including superoxide anion (O2·−), contributes to the pathogenesis of renal injury in diabetes mellitus.1-3 Nicotinamide adenine dinucleotide phosphate reduced form (NADPH) is the major source of O2·− production.3,4 Protein kinase C (PKC) enhanced by de novo synthesis of diacylglycerol as well as by renal angiotensin II increases renal expression of NADPH oxidase, promoting the translocation of its cytosolic subunits, including p47phox, to the membrane subunits.2,5 Suppression of tissue angiotensin II with an angiotensin-converting enzyme inhibitor (ACEI) or an angiotensin receptor blocker (ARB) may inhibit the membrane translocation of p47phox, prevents the activation of NADPH oxidase in a similar way to apocynin, and decreases renal damage.5,6 Hyperglycemia itself also increases ROS via mitochondrial electron transport chain reaction and auto-oxidation of free glucose or Amadori product forming ROS.7 The antioxidant defense system, superoxide dismutase (SOD) converts O2·− to hydrogen peroxide (H2O2), and H2O2 is degraded to water and molecular oxygen by catalase and glutathione peroxidase. H2O2 can also be converted to hypochlorite by myeloperoxidase (MPO).8 As SOD activity and catalase activities are decreased in diabetes,9 the improvement of the antioxidant system including SOD could be a potential therapeutic target in diabetic nephropathy.
Tempol (4-hydroxy-2,2,6,6-tetramethyl-piperidine-1-oxyl) is a stable, metal-independent low-molecular weight SOD mimetic with an excellent cell-permeability. Tempol normalized blood pressure in hypertensive rats10 and dilated afferent arterioles of diabetic rabbits.11 Tempol protected the brain,12,13 the liver,14 and the kidney15 from ischemic damage. It also prevented atherosclerosis in the adrenomedullin knockout mice16 and inhibited hypertensive renal damage in the Dahl salt-sensitive model.17 Recently, it has been demonstrated that tempol also has a direct action on sympathetic nerve activity inhibition beyond SOD mimetic action.18 However, there is not enough evidence of the effect of tempol in diabetic nephropathy. In this study, we treated streptozotocin (STZ)-induced diabetic rats with tempol, evaluated the ROS producing and degrading systems and their products, and correlated them with renal damage.
MATERIALS AND METHODS
Female Sprague-Dawley rats weighing 180-250 g were housed in cages and fed standard pellet diet (Na+ content, 0.21 g/100 g) and tap water ad libitum. Diabetes was induced by a single tail vein injection of streptozotocin (STZ, 60 mg/kg body weight; Sigma Chemical Co., St. Louis, MO). Two weeks later, the diabetic rats were randomly divided into 2 groups matched for body weight and blood glucose: untreated DM group (DM, N = 6) and DM treated with tempol (DM + Tempol, N = 6, 200 mg/kg/d subcutaneously; Sigma Chemical Co., St. Louis, MO). Age-matched rats without STZ injection served as controls (N = 6). Twenty-four hour urine was collected using metabolic cages 8 weeks after STZ injection. Animals from each group were anesthetized with pentobarbital (50 mg/kg body weight intraperitoneally). The abdominal aorta was cannulated, and the mean blood pressure (MBP) was measured by a pressure transducer (Nihon Koden, Tokyo, Japan). Blood was collected, and the kidneys were perfused retrogradely with ice-cold phosphate-buffered saline (PBS). The right kidney was removed and cut into half for Western blot analysis (N = 6), cryosection (N = 3), or glomerular isolation (N = 3). The left kidney was perfused with periodate-lysine-paraformaldehyde (PLP) solution. Kidney slices for immunohistochemical analysis were immersed in PLP solution overnight at 4°C and embedded in wax (polyethylene glycol 400 distearate; Polysciences Inc., Warrington, PA).
Measurement of Hydrogen Peroxide, Hypochlorite, SOD, Creatinine, Protein, and Glucose
Production of peroxides including H2O2 and peroxynitrite was measured in whole kidney homogenates and in isolated glomeruli obtained by graded sieving using the 2′,7′-dichlorodihydrofluorescein-diacetate (DCFH-DA) method, as described previously.3 Briefly, samples were incubated with DCFH-DA (16 μg/mL final concentration; Molecular Probes, Eugene, OR) for 20 minutes at 37°C. DCFH-DA is oxidized by peroxides to the highly fluorescent compound, 2′,7′-dichlorofluorescein (DCF), which was measured with a spectrofluorometer using excitation/emission wavelength at 485/535 nm.
Hypochlorite anion was measured in whole kidney homogenates reacted with aminophenyl fluorescein (APF) and hydroxyphenyl fluorescein (HPF).19 APF produces fluorescence by the reaction with hydroxyl radical, peroxynitrite anion, and hypochlorite anion, whereas HPF only reacts with hydroxyl radical and peroxynitrite anion. Thus, subtraction of the fluorescent reaction of HPF from that of APF will indicate the reaction with hypochlorite anion. Samples were incubated with APF or HPF (1 μmol/L final concentration; Molecular Probes) for 30 minutes at 37°C, and the fluorescence produced by the reaction with APF or HPF was measured with a spectrofluorometer with excitation/emission wavelength at 490/515 nm.
SOD activity was measured by spectrophotometric assay kit (Oxis Research, Portland, OR). Urine protein was measured by the Bradford method (Bio-Rad, Richmond, VA), and corrected by urinary creatinine. Blood glucose was measured by Glutest E II (Kyoto Daiichi Kagaku, Kyoto, Japan). HbA1c was measured with the DCA 2000 plus system (Bayer Medical, Tokyo, Japan).3,20
Western Blot Analyses
As described in detail previously,3,6,21 the right kidney was removed immediately after perfusion with PBS and homogenized in 4 mL of ice-cold buffer containing 20 mmol/L Tris, at pH 7.2, 0.5 mmol/L ethylenediaminetetraacetic acid (EDTA), 0.5 mmol/L ethylene glycol-bis (β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid (EGTA), 20 μmol/L leupeptin, 10 mmol/L dithiothreitol, and 0.1 mmol/L phenylmethylsulfonyl fluoride. Homogenates were centrifuged and membrane fractions (25 μg of protein) were electrophoresed and blotted on to polyvinylidene fluoride membranes. The membranes were incubated overnight with monoclonal antibodies against p47phox, p22phox (Transduction Laboratories, Lexington, KY), and catalase (Sigma) at 1:1000 dilutions or a polyclonal antibody for myeloperoxidase (MPO, Santa Cruz Biotechnology, Santa Cruz, CA) at a 1:100 dilution. Then membranes were incubated with horseradish peroxidase (HRP)-conjugated secondary antibody (Dako, Glostrup, Denmark) at 1:1000 dilutions. HRP labeling was detected with 0.8 mmol/L diaminobenzidine (DAB; Dojindo Laboratories, Kumamoto, Japan), 0.01% H2O2, and 3 mmol/L NiCl2. The density of the bands was analyzed using NIH Image software.
RT-PCR for NADPH Oxidase p47phox and p22phox in the Kidney
Total RNA was extracted from four 4 μm sections from each animal using a kit for tissue mRNA extraction from paraffin-embedding samples (Isogen PB kit, Nippon Gene Co., LTD, Tokyo, Japan) and reverse transcription-polymerase chain reaction (RT-PCR) was performed using a Qiagen One Step RT-PCR kit (Qiagen K.K., Tokyo, Japan) for 30 cycles with a synthetic gene-specific primer for NADPH oxidase p47phox; sense primer 5′-ATACTTCAACGGCCTCATGG-3′ (325-344) and antisense primer 5′-CTGTTCCCGAACTCTTCTCG-3′(557-538) [AB002663]. The primer for p22phox was sense: 5′-TTGTTGCAGGAGTGCTCATC-3′(128-147) and antisense: 5′-CTGCCAGCAGGTAGATCACA-3′(373-354) [AJ295951]. Parallel amplification of rat glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was done for reference; sense primer 5′-GGTGATGCTGGTGCTGAGTA-3′(256-275) and antisense primer 5′-ACTGTGGTCATGAGCCCTTC-3′(527-508) [AB017801]. The intensity of each gene's band was expressed relative to the corresponding densities of the GAPDH bands from the same RNA samples.
Kidney slices were processed for immunohistochemistry using the labeled streptavidin biotin method, as described previously.3,22 Wax sections (2 μm) were incubated with polyclonal antibodies recognizing fibronectin (Chemicon International Inc., Temecula, CA) or TGF-β1 (Santa Cruz Biotechnology) at 1:100 dilutions, followed by incubation with biotinylated anti-rabbit IgG secondary antibody (Dako) at a 1:400 dilution, and then with HRP-conjugated streptavidin solution (Dako). HRP labeling was detected using a peroxide substrate solution containing 0.8 mmol/L DAB and 0.01% H2O2.
Wax sections were stained with periodic acid Schiff (PAS). The degree of mesangial matrix expansion was scored blind to the samples: grade 0, normal glomeruli; grade 1, mesangial expansion area up to 25%; grade 2, 25 to 50%; grade 3, 50 to 75%; grade 4, 75 to 100%.20,23 The glomerular matrix expansion index was defined as the average of all glomeruli in each section, and 5 rats were examined in each group.
All data are expressed as mean ± SE. The mean values were compared among the 3 groups using analysis of variance (ANOVA) followed by Fisher's protected least significant difference test. Probability values less than 0.05 were required for statistical significance.
As shown in the Table 1, blood glucose and HbA1c were significantly higher in diabetic rats than in controls. Tempol treatment showed an increased tendency in blood glucose and HbA1c; however, the difference was not significant. Mean blood pressure was comparable between DM and control, but it was reduced by tempol.
Expression of NADPH Oxidase Subunits
Western blot analyses of membrane fraction that was obtained by differential centrifugation of the kidney homogenates showed specific bands for p47phox that were increased in DM rats (Fig. 1). Tempol prevented the significant increase in the expression of p47phox protein in DM rats (Fig. 1). The messenger RNA for p47phox followed the change in its protein expression (Fig. 1). Both protein and mRNA expression of p22phox were unchanged among three groups (Fig. 1).
SOD Activity, Expression of Catalase, and Peroxide Production in the Kidney
O2·− generated by NADPH oxidase is converted to hydrogen peroxide by SOD and then degraded to water by catalase. SOD activity in the kidney was unchanged in DM rats compared with control (13.2 ± 0.37 vs. 14.1 ± 0.47 unit/mg protein; P = NS; Fig. 2A), and increased by tempol treatment (15.9 ± 1.09 unit/mg protein; P < 0.05 vs. DM; Fig. 2A). Western blot analyses showed that the expression of catalase in the kidney was suppressed in the DM rats, and tempol did not change catalase expression in DM rats (Fig. 2B).
Associated with the increased NADPH oxidase p47phox translocation and decreased catalase in DM rats, peroxide production detected by DCFH-DA was increased in both glomeruli and renal tubules in diabetic rat compared with control. The membrane-permeable SOD mimetic tempol decreased peroxide production in glomeruli, whereas it could not inhibit peroxide production in renal tubules (Fig. 3). This inhibitory effect of tempol in glomerular peroxide production was also confirmed using isolated glomeruli from each rat group (Fig. 3). The peroxide production in the whole kidney was significantly increased in DM rats (19.9 ± 1.1 vs. 15.3 ± 1.1 FI unit/mg protein; P < 0.05), but it was not suppressed by tempol treatment (19.3 ± 1.6; P = NS vs. DM), supporting that tempol failed to inhibit tubular production of peroxide.
Expression of MPO and Hypochlorite Production in the Kidney
Hydrogen peroxide also reacts with MPO and chloride and generates hypochlorite anion. Western blot analyses showed an increase in MPO expression in DM rats (Fig. 4). Treatment with tempol did not change MPO expression in DM rats (Fig. 4). Associated with the increased MPO in DM rats, their kidneys showed increased hypochlorite production (44.8 ± 3.4 vs. 28.1 ± 4.5 FI unit/mg protein; P < 0.01; Fig. 4), which was not reduced by tempol treatment (47.5 ± 2.0; P = NS vs. DM; Fig. 4).
Urinary Protein Excretion and Mesangial Matrix Expansion
Urinary protein excretion was increased in DM rats, and tempol failed to suppress proteinuria (Table 1). However, the enhanced mesangial expression of TGF-β and fibronectin and the resultant mesangial matrix expansion observed in DM rat was inhibited by tempol treatment (Fig. 5). The semi-quantitative evaluation of mesangial matrix expansion confirmed that increased mesangial matrix index in DM rat (1.36 ± 0.16 vs. 0.27 ± 0.05; P < 0.001; Fig. 6) was ameliorated by tempol treatment (0.77 ± 0.07; P < 0.001 vs. DM; Fig. 6).
O2·− formed by NADPH oxidase has an important role in the pathogenesis of diabetic nephropathy, and scavenging O2·− can be considered a reasonable therapeutic strategy. In this study, we demonstrated that scavenging superoxide with SOD mimetic tempol inhibited glomerular matrix expansion via suppression of TGF-β; at the same time, hydrogen peroxide produced by tempol was converted to hypochlorite by MPO in the diabetic kidney, causing proteinuria (Fig. 7). Therefore, it can be said that tempol treatment has a double-edged action.
One of the beneficial actions of tempol is that it decreases the translocation of the NADPH oxidase cytosolic component p47phox protein to the membrane components in the DM rat. It has been demonstrated that the phosphorylation and translocation of p47phox to the membrane components is essential for activation of NADPH oxidase.6,24,25 In the present study, p47phox in the membrane fraction of kidney homogenate was significantly increased in the kidney of DM rat, and tempol treatment reduced p47phox in the membrane fraction, suggesting decreased membrane translocation of p47phox and consequent decreased activation of NADPH oxidase. On the other hand, NADPH oxidase membrane component p22phox was not changed in the DM with or without tempol treatment. ROS has a positive feedback activation of NADPH oxidase, especially of p22phox expression,26 and tempol did not decrease ROS in whole kidney; therefore, tempol might fail to reduce p22phox expression.
Tempol did not block peroxide in the renal tubules; however, it suppressed glomerular peroxide production (Fig. 3), suggesting that tempol may easily permeate the podocytes and glomerular endothelial cells during glomerular filtration because of its low molecular weight (172 Dalton). It has been shown that high glucose in diabetes causes increase in ROS and TGF-β expression, and inhibition of ROS reduces TGF-β and extracellular matrix expansion.27 In our study, tempol scavenged superoxide in the glomeruli and reduced TGF-β and fibronectin production, ameliorating glomerular mesangial expansion. Tempol also prevented renal injury in the Dahl salt-sensitive hypertensive rat via reduction of oxidative stress and TGF-β.28
The collateral effect of tempol is related to the increased formation of hypochlorite in diabetic rat. O2·− is usually converted to hydrogen peroxide by SOD, and then to water and molecular oxygen by catalase (Fig. 7). However, there are several reports demonstrating that SOD and catalase expression are decreased in the kidney of DM9,29 and that TGF-β inhibits these antioxidative enzymes.30,31 In our study, TGF-β was increased and catalase was suppressed in the kidney of DM rats. Tempol treatment increased renal SOD activity and increased the formation of hydrogen peroxide that was converted to hypochlorite via enhanced MPO in DM rat. Recently, it has been demonstrated that vascular-bound MPO is increased in diabetes and could amplify diabetic vascular diseases.32 Tempol prevented the increase in NADPH oxidase; however, it did not reduce MPO expression and hypochlorite production in DM rat. Interestingly, hypochlorite has an important role in the glomerular damage, and the hypochlorite-modified protein was detected in the basement membrane and podocytes of the human membranous nephropathy33 and also in the kidney and urine of hyperlipidemic anti-Thy-1 nephritis model.34 Therefore, despite the reduction of superoxide production in the glomeruli, tempol could not inhibit the harmful hypochlorite that might have damaged the glomerular basement membrane, causing proteinuria in diabetic nephropathy.
There is a controversy about the effect of SOD mimetic in renal diseases, and its effect may be different according to the animal model. Tempol prevented urinary protein excretion and glomerulosclerosis with reduction of blood pressure in the Dahl salt-sensitive rats35 and in the aldosterone and high salt induced hypertensive rats.17 On the other hand, tempol could not prevent proteinuria in the hyperthyroid hypertensive rats, even though tempol reduced blood pressure.36 We also demonstrated that tempol failed to reduce proteinuria in diabetic rats, even though glomerular matrix expansion was suppressed. The controversial effects of tempol could be explained by the degrading pathways of hydrogen peroxide via catalase and glutathione peroxidase or via MPO. Thus, to protect the kidney from ROS injury, it is necessary not only to inhibit O2·− with tempol but also to suppress hypochlorite production.
It is interesting to compare the renoprotective effect of apocynin as an inhibitor of superoxide generation and tempol as a superoxide scavenger in diabetic nephropathy. We demonstrated that apocynin inhibited oxidative stress in the whole kidney and ameliorated proteinuria and glomerular mesangial matrix expansion,6 whereas tempol failed to inhibit oxidative stress in the whole kidney and failed to reduce proteinuria. It is possible that higher doses of tempol could be more effective. It has been reported that NADPH oxidase inhibitors, including apocynin and diphenyleneiodonium chloride, suppressed O2·− production by zymosan-stimulated polymorphonuclear leukocytes more effectively than high doses of SOD.37 Thus, it is possible that the NADPH oxidase inhibitor is more effective than SOD mimetic in diabetic nephropathy.
SOD mimetic, tempol, reduced peroxide production in glomeruli and reduced mesangial matrix expansion; however, tempol treatment failed to reduce proteinuria due to the increased hypochlorite formed by MPO.
1. Ha H, Lee HB. Oxidative stress in diabetic nephropathy: basic and clinical information. Curr Diab Rep
2. Kitada M, Koya D, Sugimoto T, et al. Translocation of glomerular p47phox and p67phox by protein kinase C-beta activation is required for oxidative stress in diabetic nephropathy. Diabetes
3. Onozato ML, Tojo A, Goto A, et al. Oxidative stress and nitric oxide synthase in rat diabetic nephropathy: effects of ACEI and ARB. Kidney Int
4. Chabrashvili T, Tojo A, Onozato ML, et al. Expression and cellular localization of classic NADPH oxidase subunits in the spontaneously hypertensive rat kidney. Hypertension
5. Onozato ML, Tojo A. Role of NADPH oxidase in hypertension and diabetic nephropathy. Curr Hypertens Rev
6. Asaba K, Tojo A, Onozato ML, et al. Effects of NADPH oxidase inhibitor in diabetic nephropathy. Kidney Int
7. Brownlee M. The pathobiology of diabetic complications: a unifying mechanism. Diabetes
8. Malle E, Buch T, Grone HJ. Myeloperoxidase in kidney disease. Kidney Int
9. Sindhu RK, Koo JR, Roberts CK, et al. Dysregulation of hepatic superoxide dismutase, catalase and glutathione peroxidase in diabetes: response to insulin and antioxidant therapies. Clin Exp Hypertens
10. Schnackenberg CG, Welch WJ, Wilcox CS. Normalization of blood pressure and renal vascular resistance in SHR with a membrane-permeable superoxide dismutase mimetic: role of nitric oxide. Hypertension
11. Schnackenberg CG, Wilcox CS. The SOD mimetic tempol
restores vasodilation in afferent arterioles of experimental diabetes. Kidney Int
12. Kwon TH, Chao DL, Malloy K, et al. Tempol
, a novel stable nitroxide, reduces brain damage and free radical production, after acute subdural hematoma in the rat. J Neurotrauma
13. Mehta SH, Webb RC, Ergul A, et al. Neuroprotection by tempol
in a model of iron-induced oxidative stress in acute ischemic stroke. Am J Physiol Regul Integr Comp Physiol
14. Sepodes B, Maio R, Pinto R, et al. Tempol
, an intracelullar free radical scavenger, reduces liver injury in hepatic ischemia-reperfusion in the rat. Transplant Proc
15. Patel NS, Chatterjee PK, Chatterjee BE, et al. TEMPONE reduces renal dysfunction and injury mediated by oxidative stress of the rat kidney. Free Radic Biol Med
16. Kawai J, Ando K, Tojo A, et al. Endogenous adrenomedullin protects against vascular response to injury in mice. Circulation
17. Nishiyama A, Yoshizumi M, Hitomi H, et al. The SOD mimetic tempol
ameliorates glomerular injury and reduces mitogen-activated protein kinase activity in Dahl salt-sensitive rats. J Am Soc Nephrol
18. Xu H, Fink GD, Galligan JJ. Tempol
lowers blood pressure and sympathetic nerve activity but not vascular O2−
in DOCA-salt rats. Hypertension
19. Setsukinai K, Urano Y, Kakinuma K, et al. Development of novel fluorescence probes that can reliably detect reactive oxygen species and distinguish specific species. J Biol Chem
20. Onozato ML, Tojo A, Goto A, et al. Radical scavenging effect of gliclazide in diabetic rats fed with a high cholesterol diet. Kidney Int
21. Tojo A, Bredt DS, Wilcox CS. Distribution of postsynaptic density proteins in rat kidney: relationship to neuronal nitric oxide synthase. Kidney Int
22. Tojo A, Onozato ML, Fukuda S, et al. Nitric oxide generated by nNOS in the macula densa regulates the afferent arteriolar diameter in rat kidney. Med Electron Microsc
23. Kobayashi N, Hara K, Tojo A, et al. Eplerenone shows renoprotective effect by reducing LOX-1-mediated adhesion molecule, PKCepsilon-MAPK-p90RSK, and Rho-kinase pathway. Hypertension
24. Koya D, Hayashi K, Kitada M, et al. Effects of antioxidants in diabetes-induced oxidative stress in the glomeruli of diabetic rats. J Am Soc Nephrol
25. Decoursey TE, Ligeti E. Regulation and termination of NADPH oxidase activity. Cell Mol Life Sci
26. Djordjevic T, Pogrebniak A, BelAiba RS, et al. The expression of the NADPH oxidase subunit p22phox is regulated by a redox-sensitive pathway in endothelial cells. Free Radic Biol Med
27. Lee HB, Yu MR, Yang Y, et al. Reactive oxygen species-regulated signaling pathways in diabetic nephropathy. J Am Soc Nephrol
28. Hisaki R, Fujita H, Saito F, et al. Tempol
attenuates the development of hypertensive renal injury in Dahl salt-sensitive rats. Am J Hypertens
29. Wohaieb SA, Godin DV. Alterations in free radical tissue-defense mechanisms in streptozocin-induced diabetes in rat. Effects of insulin treatment. Diabetes
30. Kayanoki Y, Fujii J, Suzuki K, et al. Suppression of antioxidative enzyme expression by transforming growth factor-beta 1 in rat hepatocytes. J Biol Chem
31. Reddi AS, Bollineni JS. Selenium-deficient diet induces renal oxidative stress and injury via TGF-beta1 in normal and diabetic rats. Kidney Int
32. Zhang C, Yang J, Jennings LK. Leukocyte-derived myeloperoxidase amplifies high-glucose-induced endothelial dysfunction through interaction with high-glucose-stimulated, vascular non-leukocyte-derived reactive oxygen species. Diabetes
33. Grone HJ, Grone EF, Malle E. Immunohistochemical detection of hypochlorite-modified proteins in glomeruli of human membranous glomerulonephritis. Lab Invest
34. Scheuer H, Gwinner W, Hohbach J, et al. Oxidant stress in hyperlipidemia-induced renal damage. Am J Physiol Renal Physiol
35. Meng S, Cason GW, Gannon AW, et al. Oxidative stress in Dahl salt-sensitive hypertension. Hypertension
36. Moreno JM, Rodriguez Gomez I, Wangensteen R, et al. Cardiac and renal antioxidant enzymes and effects of tempol
in hyperthyroid rats. Am J Physiol Endocrinol Metab
37. Dodd OJ, Pearse DB. Effect of the NADPH oxidase inhibitor apocynin on ischemia-reperfusion lung injury. Am J Physiol Heart Circ Physiol