The effects of overexpression of Cu2+/Zn2+ superoxide dismutase-1 (SOD-1) on indexes of renal injury were compared in 5-month-old nontransgenic (NTg) db/db mice and db/db mice hemizygous for the human SOD-1 transgene (SOD-Tg). Both diabetic groups exhibited similar hyperglycemia and weight gain. However, in NTg-db/db mice, albuminuria, glomerular accumulation of immunoreactive transforming growth factor-β, collagen α1(IV), nitrotyrosine, and mesangial matrix were all significantly increased compared with either nondiabetic mice or SOD-Tg-db/db. SOD-1 activity and reduced glutathione levels were higher, whereas malondialdehyde content was lower, in the renal cortex of SOD-Tg-db/db compared with NTg-db/db mice, consistent with a renal antioxidant effect in the transgenic mice. Inulin clearance (CIN) and urinary excretion of guanosine 3′,5′-cyclic monophosphate (UcGMP) were increased in SOD-Tg-db/db mice compared with corresponding values in nondiabetic mice or NTg-db/db mice. CIN and UcGMP were suppressed by Nω-nitro-l-arginine methyl ester in SOD-Tg-db/db but not in NTg-db/db mice, implying nitric oxide (NO) dependence of these increases and enhanced renal NO bioactivity in SOD-Tg-db/db. Studies of NO-responsive cGMP in isolated glomeruli supported greater quenching of NO in glomeruli from NTg-db/db compared with SOD-Tg-db/db mice. Evidence of increased NO responsiveness and the suppression of glomerular nitrotyrosine may both reflect reduced NO-superoxide interaction in SOD-Tg-db/db mice. The results implicate superoxide in the pathogenesis of diabetic nephropathy.

Oxidative and glyco-oxidative stress have been implicated in the pathogenesis of diabetic complications, including nephropathy (16). There is evidence for multiple pathways of increased generation of reactive oxygen species (ROS) in diabetes. These include enhanced glucose auto-oxidation; increased mitochondrial superoxide production; protein kinase C−dependent activation of NADPH oxidase by angiotensin-II, glucose, and other stimuli; uncoupled endothelial nitric oxide synthase (eNOS) activity; formation of advanced glycation products (AGEs); stimulation of cellular ROS production by extracellular AGEs via their receptors; and others (14). Markers of oxidative stress and reduced levels of antioxidants have been found in blood and/or tissues in both human and experimental diabetes, including kidney (110). In addition to the potential cytotoxic effects of increased ROS generation, there is increasing evidence that ROS function as physiological cellular second messengers via redox sensitive gene transcription factors and cell signaling pathways (1,2). Thus, increases in ROS in the diabetic environment may alter several redox-sensitive genes and/or cellular signaling pathways (1,2). With respect to diabetic nephropathy, studies in cultured mesangial cells support a role for increased superoxide production in signaling increases in matrix protein synthesis induced by either high ambient glucose concentrations or angiotensin-II (11,12), two factors strongly linked to renal injury in vivo in diabetes (13,14). Similarly, studies in cultured endothelial cells have demonstrated increased superoxide and eNOS expression in response to high glucose (15) and have implicated increased superoxide from mitochondria (2) and uncoupled eNOS (16) in the activation of signaling pathways linked to cell injury.

Consistent with a role for oxidative mechanisms in the pathogenesis of diabetic renal injury in vivo, dietary antioxidant supplementation with vitamin E, N-acetyl-cysteine, oxerutin, taurine, or lipoic acid have all been found to attenuate nephropathy in experimental models of diabetes (5,8,1618). Recently, we demonstrated that transgenic mice that overexpress human cytoplasmic Cu2+/Zn2+ superoxide dismutase (SOD)-1 are protected from early diabetic renal injury compared with their nontransgenic diabetic littermates in a model of streptozotocin (STZ)-induced type 1 diabetes (11). These observations provided the first in vivo evidence for the involvement of superoxide in the pathogenesis of diabetic nephropathy.

In the current studies, we assessed the effects of overexpression of SOD-1 on renal injury in an experimental model of type 2 diabetes and the possibility of altered superoxide interaction with nitric oxide (NO) in diabetic mice that overexpress SOD-1.

This study was conducted in conformance with guidelines for the use and care of animals of the Veterans Affairs Pittsburgh Healthcare system, whose animal facility is fully accredited by the Association for the Assessment and Accreditation of Laboratory Animal Care. To produce compound mutant mice with an SOD-1 +/− db/db genotype, hemizygous female SOD-1 transgenic mice (SOD-Tg), which express the human SOD-1 transgene driven by the human SOD-1 promoter (The Jackson Laboratory, Bar Harbor, ME), were bred to male mice heterozygous for the db mutation coupled to the misty (M) allele (C57BL-mdb mice kindly supplied by Dr. Streamson Chua, Molecular Biology/Molecular Genetic Core, NY Obesity Research Center, New York, NY). Both the SOD-Tg and db heterozygous breeders were on a C57BL/6J genetic background. Offspring from the F1 generation, which were heterozygous for the db mutation and also expressed the SOD-1 transgene, were then bred to mice heterozygous for the db mutation. The F2 generation contained the desired compound mutant (SOD-Tg, db/db) and the other genotypes used. Genotypes were determined by RT-PCR amplification of tail DNA as described below. The following four genotypes were studied using 10 age-matched, male mice in each group: 1) wild-type nontransgenic (NTg)-nondiabetic (ND), 2) NTg-db/db, 3) SOD-Tg-ND, and 4) SOD-Tg-db/db.

Animals were maintained in a 22°C room with a 12-h light/dark cycle and given free access to food (Purina pellet mouse chow) and water. Blood glucose levels were determined at monthly intervals on blood obtained from the retro-orbital venous plexus. Urine samples were collected in metabolic cages at age 5 months and frozen for subsequent determination of albumin and [14C]inulin when inulin clearance (CIN) was examined. CIN was determined in conscious unrestrained mice from the clearance of [14C]inulin given by osmotic minipumps (Durect, Cupertino, CA) implanted subcutaneously, as previously reported (5,11). CIN and fractional clearance of albumin (FCAlb) were calculated as previously described (5,11). At age 5 months, mice were weighed and then killed by exsanguination under isoflurane anesthesia. Their blood was collected in heparinized tubes and their kidneys were weighed and then processed for biochemical, histological, and immunohistochemical determinations, as described below or previously reported (5,11).

In separate studies using four additional experimental mouse groups analogous to those described above, CIN and 24-h urinary excretion of guanosine 3′,5′-cyclic monophosphate (UcGMP) and NO2/NO3 (UNOx) were determined before and on the 3rd day of oral administration of Nω-nitro-l-arginine methyl ester (l-NAME), a competitive inhibitor of NOS (5,11,19). Preliminary studies indicated that ingestion of ∼20 mg · kg body wt−1 · day−1 of l-NAME for 3 days yielded marked suppression of UNOx in NTg-db/db mice, and therefore this dosage was used in the current experiments. Average daily consumption of l-NAME per mouse was ∼23 mg/kg body wt and did not differ significantly among the four study groups. Mice were killed 1 week after completion of the l-NAME studies; their kidneys were removed and their glomeruli were harvested by differential sieving, as previously described (20). Glomeruli harvested from kidneys of five randomly selected mice from each study group were used for determination of SOD-1 activities.

Glomerular incubations.

After being isolated, glomeruli were washed with and then resuspended in Krebs-Ringer bicarbonate buffer containing 5 mmol/l glucose and 2 mmol/l 1-methyl-3-isobulylmethylxanthene, an inhibitor of phosphodiesterase activity. The suspension was then aliquoted (0.5 ml/well in 24-well plate) and incubated at 37°C for 30 min with shaking in a Plexiglas chamber in an atmosphere of 95%/5% CO2 in the presence or absence of carbamylcholine (CCh) or nitroprusside (NP), as indicated in results. Incubations were terminated by the addition of 0.5 ml sodium acetate (pH 6.0) at 95°C and cGMP extracted from glomeruli plus media.

cGMP was assayed in extracts of glomerular incubates and urine, as previously reported (20) using a commercial radioimmunoassay system (Amersham Pharmacia Biotech, Piscataway, NJ). The glomerular pellet was extracted in NaOH and used for determination of protein by the Bradford method (19).

RT-PCR.

DNA was extracted from tail snips using an Aquapure Genomic DNA Isolation Kit (Bio-Rad, Hercules, CA). The PCR protocol for detection of the human SOD-1 transgene was obtained from The Jackson Laboratories (Bar Harbor, ME). The primers used were CAT CAG CCC TAA TCC ATC TGA, forward, and CGC GAC TAA CAA TCA AAG TGA, reverse (21). Interleukin-2 (IL-2) was used as an internal control for human SOD-1. The primers used for IL-2 were CTA GGC CAC AGA ATT GAA AGA TCT, forward, and GTA GGT GGA AAT TCT AGC ATC ATC C, reverse. PCR amplification was performed with Taq polymerase (Invitrogen, Carlsbad, CA). The primers used amplified a 236-bp fragment of the human SOD-1 gene. Simultaneous amplification was performed with primers specific for a 324-bp fragment of the IL-2 gene as an internal control. A cycle number of 35 was chosen based on preliminary experiments to ensure that amplification was within the linear range.

The PCR protocol for detection of the db mutation has been previously described by Schreyer et al. (22). The primer sequences used were AGA ACG GAC ACT CTT TGA AGT CTC, forward, and CAT TCA AAC CAT AGT TTA GGT TTG TGT, reverse. The primers introduce a new Rsa I site into the mutant sequence. The PCR product is digested with Rsa for 1 h at 37°. The db mutant allele yields a 135-bp fragment that is not cut by Rsa I.

Additional assays.

Plasma and urinary albumin, GHb, UNOx, renal cortical activities of SOD-1, mitochondrial Mn2+ SOD, catalase and glutathione peroxidase (GPx), renal cortical malondialdehyde (TBARS), and reduced glutathione (GSH) content were determined as previously reported (5,11,19).

Immunohistochemical staining.

Immunohistochemical staining of renal cortical sections for transforming growth factor-β (TGF-β), collagen α1(IV), and human SOD-1 were performed using techniques and reagents previously described in detail (5,11). To assess nitrotyrosine content of glomeruli by immunohistochemistry, sections of formalin-fixed mouse renal cortex were pretreated for 30 min at 95° with Target Retrieval Solution (Dako, Carpinteria, CA). Sections were cooled and blocked with endogenous oxidation blocking solution (PBS plus 3% H202) for 10 min at room temperature. Sections were then blocked with Superblock (ScyTek, Logan, UT). The primary antibody, antinitrotyrosine (5 ug/ml, rabbit immunoaffinity purified IgG; Upstate Biotechnology, Lake Placid, NY) was then applied and the incubation was continued overnight at 0°C. Sections were incubated for 30 min with biotin-conjugated goat antirabbit IgG diluted 1/500 (The Jackson Laboratories) followed by a 10-min incubation with horseradish-conjugated streptavidin (Dako). Color was developed with diaminobenzidine. Nonspecific staining was assessed by prior exposure of the primary antinitrotyrosine antibody to 10 mmol/l nitrotyrosine for 1 h at room temperature and application of this pretreated antibody preparation to renal cortical sections.

Glomerular TGF-β, collagen α1(IV), and nitrotyrosine immunostaining were assessed with a SAMBA 4000 image analyzer (Image Products, Chantilly, VA) using designed software (Microsoft, Richmond, VA), as previously described (5,11). A total of 30 glomeruli per mouse were examined. Results are expressed as the proportional area of positive staining of the glomerular tuft (labeling index), as previously reported (5,11).

Glomerular morphometrics.

Glomerular volume (VG) was calculated from the cross-sectional area of the glomerular tuft as determined by light microscopy (23). The fractional area of the glomerular tuft occupied by mesangial matrix or mesangial cells was determined by a modification of the procedure reported previously (23). Briefly, a 1 × 1 cm grid was placed over each photomicrograph of periodic acid-Schiff stains of renal cortical sections. The number of boxes or fraction of boxes that contained portions of extracellular mesangial matrix or mesangial cells were counted and expressed as a proportion of the total number of boxes containing glomerular tuft. Measurements were made on 30 glomeruli per mouse at a magnification of ×257. For assessment of both quantitative immunohistochemical staining and glomerular morphometrics, all samples were randomized and examined in a coded fashion by a pathologist blinded to the treatment group from which the sample came.

Statistical analysis.

Significance of differences was determined by ANOVA followed by Fisher’s multiple comparison test using Statview software. Differences at P < 0.05 were considered significant (5,11).

As shown in Table 1, blood glucose levels, GHb levels, and body weight, but not kidney weight, were all significantly increased in the NTg-db/db and SOD-Tg-db/db mice compared with nondiabetic mice from the same colony. Values for these parameters were not different in NTg-db/db compared with the SOD-Tg-db/db groups. The CIN of NTg-ND, SOD-Tg-ND, and NTg-db/db did not differ significantly, although values were modestly increased in the latter group compared with nondiabetic mice. In contrast, CIN in SOD-Tg-db/db mice was clearly higher than in nondiabetic and NTg-db/db mice. Urinary albumin excretion (UAE) and FCAlb in the 5-month-old SOD-Tg-ND group were significantly lower than values in the NTg-ND mice. As previously reported (11), these differences may have been related to suppression of age-related increases in UAE in the SOD-Tg-ND compared with those of the NTg-ND mice. UAE and FCAlb in NTg-db/db mice were increased compared with values in either nondiabetic group. In SOD-Tg-db/db mice, UAE and FCAlb did not differ from values in SOD-Tg-ND mice and were significantly lower than those in the NTg-db/db mice.

Consistent with our earlier observations in SOD-Tg mice with or without STZ-induced diabetes (11), immunohistochemical staining of renal cortex for human SOD-1 was localized primarily to glomerular endothelial and tubular epithelial cells of both SOD-Tg-ND and SOD-Tg-db/db mice, with only faint staining of glomerular mesangial cells (not shown). No renal cortical staining for human SOD-1 protein was observed in either of the NTg groups.

As shown in Table 2, renal cortical SOD-1-specific activity was significantly higher in both the SOD-Tg-ND and the SOD-Tg-db/db mice compared with values in the corresponding nontransgenic groups. Renal cortical SOD-1 activity was comparable in the SOD-Tg-ND and SOD-Tg-db/db mice. Renal cortical SOD-1-specific activity of NTg-ND and NTg-db/db also did not differ significantly (Table 2).

Specific activities of renal cortical SOD-2 or catalase showed no significant differences among the four study groups (Table 2). However, renal cortical GPx activity was higher in both NTg-db/db and SOD-Tg-db/db mice compared with values in the corresponding nondiabetic mouse groups. Renal cortical TBARS content of NTg-db/db mice was significantly higher than values in either of the two nondiabetic mouse groups or the SOD-Tg-db/db mice. Renal cortical TBARS of the latter mice was somewhat higher than values in the SOD-Tg-ND mice, but this difference was not statistically significant. Conversely, renal cortical GSH content was lower in NTg-db/db mice than values in the SOD-Tg-db/db and the nondiabetic groups, and the levels in SOD-Tg-db/db did not differ significantly from levels in nondiabetic mice (Table 2).

As shown in Fig. 1, glomerular content of immunoreactive TGF-β, collagen α1(IV), and nitrotyrosine, expressed as the percent of the area of the glomerular tuft that stained positively, was significantly increased in NTg-db/db mice compared with corresponding values in either of the nondiabetic groups or in the SOD-Tg-db/db mice. Glomerular TGF-β and nitrotyrosine of SOD-Tg-db/db mice did not differ from corresponding values in either of the nondiabetic mice groups. However, glomerular content of collagen α1(IV) of SOD-Tg-db/db mice was higher than the value in the SOD-Tg-ND group (Fig. 1).

As shown in Fig. 2, the fractional area of the glomerular tuft occupied by the mesangial matrix was significantly higher in the NTg-db/db mice than in any of the other study groups. In contrast, the mesangial matrix area in the SOD-Tg-db/db mice did not differ detectably from values in either of the two nondiabetic groups. There were no significant differences in VG or the fractional area of the glomerular tuft occupied by mesangial cells among the study groups. No significant tubulointerstitial histological changes were observed in either of the diabetic groups compared with the nondiabetic mice.

As shown in Fig. 3, CIN and UcGMP were both higher in SOD-Tg-db/db mice than in the other groups. The administration of l-NAME for 3 days significantly suppressed CIN (Fig. 3, upper panel) and UcGMP (lower panel) in SOD-Tg-db/db mice, but not in the other groups. In the presence of l-NAME, CIN and UcGMP in SOD-Tg-db/db did not differ from corresponding values in the other groups. As shown in Table 3, administration of l-NAME did not significantly alter UNOx excretion in the nondiabetic mice. UNOx was higher in both diabetic groups compared with values in nondiabetic mice. By contrast, l-NAME markedly suppressed UNOx to similar absolute levels in the diabetic mice. However, values for UNOx in each diabetic group after l-NAME administration remained higher than the value in the corresponding nondiabetic group (Table 3).

As shown in Fig. 4, increases in cGMP accumulation in glomeruli from NTg-db/db mice in response to CCh or the lower concentration (0.1 mmol/l) of NP tested were significantly attenuated compared with corresponding responses of glomeruli from SOD-Tg-db/db or the nondiabetic groups. Conversely, the cGMP responses to CCh or 0.1 mmol/l NP of glomeruli from SOD-Tg-db/db were comparable to those of glomeruli from nondiabetic mice. NG-mono-methyl-l-arginine, a competitive inhibitor of NOS, blocked cGMP responses to CCh (Fig. 4), but not the increases induced by the exogenous NO donor NP (not shown), consistent with an action of CCh on cGMP-mediated (via endogenous NO) generation. Basal cGMP accumulation of glomerular incubates (no added test agent) and glomerular cGMP responses to a high (10 mmol/l) concentration of NP did not differ significantly among the study groups (Fig. 4). SOD-1-specific activity in glomeruli from SOD-Tg-db/db (2.2 ± 0.5 μg/mg protein) was significantly higher than in glomeruli from NTg-db/db (0.9 ± 0.2; P < 0.05). There were similar differences in glomerular SOD-1 between SOD-Tg-ND (2.5 ± 0.4) and NTg-ND (1.3 ± 0.3; P < 0.05). By contrast, glomerular SOD-2 activity did not differ in either diabetic or nondiabetic transgenic versus nontransgenic mice (not shown).

The mouse model of experimental diabetes used in our studies (db/db on a C57BL/6J background) is characterized by obesity, insulin resistance, sustained hyperinsulinemia, hyperleptinemia, and moderate hyperglycemia (24), characteristics that are analogous to those of many type 2 diabetic patients. Overexpression of SOD-1 activity in this model prevented or attenuated several indexes of diabetic renal injury, including albuminuria, glomerular accumulation of TGF-β, collagen α1(IV), and mesangial matrix expansion. These changes in 5-month-old NTg-db/db mice were not accompanied by a decline in renal function, and thus represented an early stage of nephropathy. In this regard, it is known that db/db mice on the C57BL/6J genetic background develop less severe hyperglycemia and renal injury than db/db mice on a C57BL/KsJ background (25,26). Of note in the SOD-Tg-db/db mice, the attenuation of markers of renal injury occurred despite the presence of glomerular hyperfiltration. This finding is analogous to those of earlier studies in the STZ-diabetic mouse, which overexpresses SOD-1 (11) as well as in STZ-diabetic rat models treated with antioxidants or ACE inhibitors as interventions (2729). It may be related to the fact that persistent increases in GFR in the presence of these agents are unassociated with glomerular capillary hypertension, as previously documented in the case of ACE inhibition (29). The renal protection observed in db/db mice that overexpressed SOD-1 was not attributable to effects on either the magnitude of hyperglycemia or weight gain (Table 1). It is also unlikely that the renal effects of SOD-1 overexpression in db/db mice were related to modification of the hyperinsulinemia that is characteristic of this model (24). Similar effects of SOD-1 overexpression were previously observed in the STZ mouse model of type 1 diabetes (11), which is characterized by weight loss and low plasma levels of insulin.

The renoprotective effect of SOD-1 overexpression in the db/db mice was associated with increased renal cortical SOD-1 enzymatic activity, lower TBARS, and higher GSH levels compared with corresponding values in renal cortex from NTg-db/db mice, findings consistent with a renal antioxidant effect. However, in the transgenic model used, SOD-1 activity is increased in multiple organs as well as in erythrocytes and leukocytes (30). Thus it is uncertain whether the changes in renal cortical TBARS and GSH reflect consequences of higher renal SOD-1, higher SOD-1 activity at one or more extrarenal sites, or a combination of these changes. It seems likely that superoxide generated in or entering the cell cytosol was the predominant target of SOD-1, given its subcellular distribution (31). However, an effect of SOD-1 on superoxide of mitochondrial origin cannot be completely excluded. Recent studies have demonstrated that SOD-1 is present not only in the cell cytosol but also in the intermembrane region of mitochondria, and may participate in the degradation of superoxide of mitochondrial origin that gain access to that region (32). The potential participation of SOD-1 in the dismutation of superoxide of mitochondrial origin is of interest in view of the work of Brownlee and colleagues (2,33). These studies implicate enhanced mitochondrial superoxide generation induced by high ambient concentrations of glucose, fatty acids, or leptin in vascular endothelial and other cells in the activation of multiple signaling pathways leading to cell injury (2,33). All of these potential mediators of mitochondrial superoxide generation are relevant to the db/db model (24) and may have contributed to renal oxidative stress.

SOD-1 converts superoxide to H2O2, a potent oxidizing agent. Thus, enhanced formation of H2O2 might be expected in mice overexpressing SOD-1 and is a potential concern. Of interest, however, is that an increase in cellular SOD-1 activity has not been uniformly observed to increase H2O2 levels. Indeed, overexpression of SOD-1 in some cells has been associated with a decrease in H2O2 levels (34). As noted by Liochev and Fridovich (35), the net effects of SOD increases on cellular H2O2 formation and levels depend on the cellular targets available to superoxide as well as the activities of antioxidant enzymes involved in H2O2 metabolism, predominantly catalase and GPx (36,37). No differences were found in the activity of renal cortical catalase among the four mouse groups in the current study. However, consistent with prior assessments of kidney from rats with experimental diabetes (37), renal cortical GPx was higher in both diabetic groups compared with controls (Table 2). This enzymatic change may have attenuated any increase in the levels of renal H2O2 in the diabetic mouse groups.

Overexpression of SOD-1 in db/db mice prevented accumulation of nitrotyrosine in glomeruli and was associated with both an NO-dependent increase in the glomerular filtration rate and preservation of glomerular cGMP responses to NO ex vivo. These findings suggest that enhanced dismutation of superoxide by SOD-1 may have suppressed the reaction of superoxide with NO with a resultant decrease in peroxynitrite generation and a concurrent increase in the bioavailability of NO. An increase in nitrotyrosine accumulation in kidney and other tissues has previously been described in experimental diabetes (6,3840). Accumulation of this product, at least in part, reflects formation of peroxynitrite (6,3840) from the reaction of NO with superoxide, and provides a stable footprint of peroxynitrite formation (6,3841). Increased peroxynitrite generation may lead to modification of proteins, nucleic acids, and lipids, with consequent cell dysfunction or death (41). Of note, studies in cultured vascular endothelial cells have indicated that eNOS per se is a target for inactivation by peroxynitrite and that inhibition of eNOS in these cells is prevented by overexpression of SOD-1, but not SOD-2 (42). Thus, overexpression of SOD-1 in glomerular endothelial cells in db/db mice may have enhanced NO production by these cells.

Similarly, the SOD-Tg-db/db mice displayed increases in CIN, UcGMP, and UNOx, which were dependent on NO, as reflected by the suppression of these increases by administration of l-NAME. NO-dependent increases in CIN and UcGMP imply enhanced production and/or bioavailability of NO in SOD-Tg-db/db compared with NTg-db/db mice. In the latter group, basal CIN and UcGMP were modestly, albeit insignificantly, increased. However, these parameters were not altered by l-NAME in NTg-db/db mice. In contrast, basal UNOx was clearly elevated in NTg-db/db mice and suppressed by l-NAME. UNOx has been used as an index of renal NO production (4245). Previous studies in nontransgenic STZ diabetic rats have found increases in UNOx and UcGMP associated with glomerular hyperfiltration, and have suggested a role for NO in mediating the latter two changes (19,45). In the current studies, the dissociation between the actions of l-NAME on UNOx versus CIN and UcGMP suggests that the increase in UNOx in NTg-db/db mice was not reflective of a corresponding enhancement in vivo of the renal bioactivity of NO. To the extent that increases in UNOx reflected changes in renal NO production in the diabetic mice, one possible explanation for the failure to observe concurrent NO-dependent increases in CIN and UcGMP in NTg-db/db mice is accelerated inactivation of NO in this group compared with the SOD-Tg-db/db mice. Consistent with this possibility, increased oxidant quenching of NO has been implicated in the impairment of endothelial dependent vasodilation that is characteristic of diabetes (46), an impairment that is reversed by antioxidants, including superoxide dismutase (46,47). Thus, in the SOD-Tg-db/db model used in the current study, inactivation of NO by superoxide may have been attenuated by enhanced SOD-1 activity.

Evidence for reduced bioactivity of NO in the NTg-db/db mice compared with either the SOD-Tg-db/db group or nondiabetic mice was also provided by ex vivo studies of the cGMP responses of isolated glomeruli to NO (Fig. 4). These observations suggest increased quenching of NO by glomeruli from the NTg-db/db mice, analogous to earlier findings in glomeruli from STZ-diabetic rats (19,48). Although the influence of an increase in SOD activity on NO depends on multiple variables and is complex (35), the current data imply that in glomeruli of SOD-Tg-db/db mice, the net effect of overexpression of SOD-1 was preservation of the biological action of NO relative to that observed in glomeruli from the nontransgenic diabetic mice. In addition to renal hemodynamic effects of NO (19,4345), studies in cultured glomerular mesangial cells support direct actions of NO to suppress activation of protein kinase C, increases in TGF-β, and accumulation of matrix protein induced by high concentrations of glucose, angiotensin II, or thromboxane-A2 (49,50). Thus, enhanced expression of NO actions in glomeruli of SOD-Tg-db/db mice relative to the NTg-db/db mice may be linked to the suppression of glomerular TGFβ1 and mesangial matrix accumulation observed in the transgenic group.

In summary, the current findings demonstrated that overexpression of SOD-1 prevents biochemical, functional, and structural changes induced in kidney in a model of type 2 diabetes. They imply a role for superoxide in vivo in the mediation of these changes, possibly in part via reduced superoxide interaction with NO in the SOD-Tg-db/db mouse.

FIG. 1.

Quantitative image analysis of immunohistochemical staining for glomerular TGF-β content (A), collagen α1 (IV) (B), and NO-tyrosine (C). Results are expressed as the percent area of positive staining of the glomerular tuft (labeling index). Data represent means ± SE of values from 10 mice in each group, 30 glomeruli per mouse. *P < 0.05 for db/db (▪) versus corresponding nondiabetic mice (□); $P < 0.05 for SOD-Tg mice vs. corresponding NTg.

FIG. 1.

Quantitative image analysis of immunohistochemical staining for glomerular TGF-β content (A), collagen α1 (IV) (B), and NO-tyrosine (C). Results are expressed as the percent area of positive staining of the glomerular tuft (labeling index). Data represent means ± SE of values from 10 mice in each group, 30 glomeruli per mouse. *P < 0.05 for db/db (▪) versus corresponding nondiabetic mice (□); $P < 0.05 for SOD-Tg mice vs. corresponding NTg.

FIG. 2.

The fractional area of the glomerular tuft occupied by mesangial matrix was expressed as the percent of the total area (mesangial matrix area/total glomerular tuft area × 100). Measurements were made using light microscopy and periodic acid-Schiff stains of renal cortical sections. Data represent means ± SE of determinations from 10 mice each group, 30 glomeruli per mouse. *P < 0.05 for db/db (▪) versus the corresponding nondiabetic group (□); $P < 0.05 for SOD-Tg mice versus corresponding nontransgenic (NTg) group.

FIG. 2.

The fractional area of the glomerular tuft occupied by mesangial matrix was expressed as the percent of the total area (mesangial matrix area/total glomerular tuft area × 100). Measurements were made using light microscopy and periodic acid-Schiff stains of renal cortical sections. Data represent means ± SE of determinations from 10 mice each group, 30 glomeruli per mouse. *P < 0.05 for db/db (▪) versus the corresponding nondiabetic group (□); $P < 0.05 for SOD-Tg mice versus corresponding nontransgenic (NTg) group.

FIG. 3.

Effects of treatment of 5 month-old-mice with l-NAME on CIN and UcGMP. Before and on the 3rd day of treatment, 24-h urine collections with mid-point plasma samples for [−14C]inulin were obtained with oral l-NAME (average dosage 19 mg · kg body wt−1 · day−1). Data represent means ± SE of determinations from 10 mice in each group. *P < 0.05 for values before (□) or after l-NAME (▪) in the db/db mouse groups versus values obtained under the same experimental conditions in the corresponding NTg or SOD-Tg nondiabetic groups; #P < 0.05 compared with value before l-NAME in the same group; $P < 0.05 for SOD-Tg-db/db (before or after l-NAME) versus corresponding value in NTg-db/db.

FIG. 3.

Effects of treatment of 5 month-old-mice with l-NAME on CIN and UcGMP. Before and on the 3rd day of treatment, 24-h urine collections with mid-point plasma samples for [−14C]inulin were obtained with oral l-NAME (average dosage 19 mg · kg body wt−1 · day−1). Data represent means ± SE of determinations from 10 mice in each group. *P < 0.05 for values before (□) or after l-NAME (▪) in the db/db mouse groups versus values obtained under the same experimental conditions in the corresponding NTg or SOD-Tg nondiabetic groups; #P < 0.05 compared with value before l-NAME in the same group; $P < 0.05 for SOD-Tg-db/db (before or after l-NAME) versus corresponding value in NTg-db/db.

FIG. 4.

Basal cGMP accumulation (no added test agent) and cGMP responses to 0.1 mmol/l CCh (▪), CCh plus 0.5 mmol/l NG-mono-methyl-l-arginine (), 0.1 mmol/l NP (;), or 10 mmol/l NP () in incubates of glomeruli isolated from the four study groups of mice shown. All incubations were conducted for 20 min in the presence of 2 mmol/l 1-methyl-3-isobulylmethylxanthene at 37°C; cGMP was extracted from media plus glomeruli at the termination of the incubations. Data represent means ± SE of determinations of glomerular from 10 mice in each group. *P < 0.05 compared with basal cGMP.

FIG. 4.

Basal cGMP accumulation (no added test agent) and cGMP responses to 0.1 mmol/l CCh (▪), CCh plus 0.5 mmol/l NG-mono-methyl-l-arginine (), 0.1 mmol/l NP (;), or 10 mmol/l NP () in incubates of glomeruli isolated from the four study groups of mice shown. All incubations were conducted for 20 min in the presence of 2 mmol/l 1-methyl-3-isobulylmethylxanthene at 37°C; cGMP was extracted from media plus glomeruli at the termination of the incubations. Data represent means ± SE of determinations of glomerular from 10 mice in each group. *P < 0.05 compared with basal cGMP.

TABLE 1

Characteristics of four study groups

Blood glucose (mmol/l)GHb (%)Body weight (g)Kidney weight (g)CIN (ml/min)UAE (μg/24 h)FCAlb (×105)
NTg-ND 8.8 ± 1.7 3.9 ± 0.7 29 ± 2 0.21 ± 0.02 0.23 ± 0.05 574 ± 95 8.4 ± 1.9 
NTg-db/db 17 ± 3* 6.8 ± 0.9* 53 ± 4* 0.25 ± 0.032 0.32 ± 0.07 1,689 ± 417* 16.3 ± 2.7* 
SOD-Tg-ND 9.3 ± 2.1 4.2 ± 0.8 33 ± 3 0.23 ± 0.03 0.25 ± 0.06 323 ± 68 4.1 ± 0.8 
SOD-Tg-db/db 18 ± 4* 7.0 ± 0.9* 56 ± 4* 0.26 ± 0.03 0.47 ± 0.10* 506 ± 151 3.4 ± 1.3 
Blood glucose (mmol/l)GHb (%)Body weight (g)Kidney weight (g)CIN (ml/min)UAE (μg/24 h)FCAlb (×105)
NTg-ND 8.8 ± 1.7 3.9 ± 0.7 29 ± 2 0.21 ± 0.02 0.23 ± 0.05 574 ± 95 8.4 ± 1.9 
NTg-db/db 17 ± 3* 6.8 ± 0.9* 53 ± 4* 0.25 ± 0.032 0.32 ± 0.07 1,689 ± 417* 16.3 ± 2.7* 
SOD-Tg-ND 9.3 ± 2.1 4.2 ± 0.8 33 ± 3 0.23 ± 0.03 0.25 ± 0.06 323 ± 68 4.1 ± 0.8 
SOD-Tg-db/db 18 ± 4* 7.0 ± 0.9* 56 ± 4* 0.26 ± 0.03 0.47 ± 0.10* 506 ± 151 3.4 ± 1.3 

Data are means ± SE of determinations from 10 mice in each group. Five sequential monthly determination of blood glucose values were obtained in the fed state. GHb is expressed as percent total hemoglobin. CIN, UAE, and FCAlb were determined at age 5 months during the week before the mice were killed.

*

P < 0.05 versus value in the corresponding nondiabetic group;

P < 0.05 versus value in corresponding nontransgenic group.

TABLE 2

Renal cortical antioxidant enzyme activities, TBARS, and GSH content

SOD-1 (μg/mg protein)SOD-2 (μg/mg protein)GPX (nmol · min−1 mg protein−1)Catalase (units/mg protein)TBARS (nmol/mg protein)GSH (μg/mg protein)
NTg-ND 2.7 ± 0.4 0.18 ± 0.03 201 ± 36 259 ± 49 0.61 ± 0.1 15 ± 2 
NTg-db/db 2.0 ± 0.3 0.22 ± 0.04 308 ± 43* 275 ± 56 1.39 ± 0.3* 9 ± 1* 
SOD-Tg-ND 4.9 ± 0.9 0.23 ± 0.04 217 ± 41 231 ± 43 0.53 ± 0.2 19 ± 2 
SOD-Tg-db/db 4.3 ± 0.8 0.25 ± 0.05 334 ± 52* 268 ± 4 0.80 ± 0.2 15 ± 2 
SOD-1 (μg/mg protein)SOD-2 (μg/mg protein)GPX (nmol · min−1 mg protein−1)Catalase (units/mg protein)TBARS (nmol/mg protein)GSH (μg/mg protein)
NTg-ND 2.7 ± 0.4 0.18 ± 0.03 201 ± 36 259 ± 49 0.61 ± 0.1 15 ± 2 
NTg-db/db 2.0 ± 0.3 0.22 ± 0.04 308 ± 43* 275 ± 56 1.39 ± 0.3* 9 ± 1* 
SOD-Tg-ND 4.9 ± 0.9 0.23 ± 0.04 217 ± 41 231 ± 43 0.53 ± 0.2 19 ± 2 
SOD-Tg-db/db 4.3 ± 0.8 0.25 ± 0.05 334 ± 52* 268 ± 4 0.80 ± 0.2 15 ± 2 

Data are means ± SE of determinations from 10 mice in each group.

*

P < 0.05 versus value in the corresponding nondiabetic group;

P < 0.05 versus value in corresponding nontransgenic group.

TABLE 3

Effects of l-NAME on UNOx

Before l-NAMEAfter l-NAME
NTg-ND 423 ± 97 298 ± 72 
NTg-db/db 1,269 ± 325* 607 ± 101* 
SOD-Tg-ND 586 ± 113 419 ± 83 
SOD-Tg-db/db 1,834 ± 389* 732 ± 130* 
Before l-NAMEAfter l-NAME
NTg-ND 423 ± 97 298 ± 72 
NTg-db/db 1,269 ± 325* 607 ± 101* 
SOD-Tg-ND 586 ± 113 419 ± 83 
SOD-Tg-db/db 1,834 ± 389* 732 ± 130* 

Data are means SE of determinations from 10 mice in each study group and represent nmol/24 h of UNOx.

*

P < 0.05 versus value in the corresponding non-diabetic group.

P < 0.05 versus corresponding value before treatment with l-NAME.

This work was supported by grants from the Department of Veterans Affairs and American Diabetes Association.

The authors are grateful for the excellent technical assistance of Julia Liachenko and Mark Barsic and to JoAnn Orbin for typing the manuscript and revisions.

1.
Evans JL, Goldfine ID, Maddux BA, Grodsky GM: Oxidative stress and stress-activated signaling pathways: a unifying hypothesis of type 2 diabetes.
Endocr Rev
23
:
599
–622,
2002
2.
Brownlee M: Biochemistry and molecular cell biology of diabetic complications.
Nature
414
:
813
–820,
2001
3.
Baynes J, Thorpe S: The role of oxidative stress in diabetic complications: a new perspective on an old paradigm.
Diabetes
48
:
1
–19,
1999
4.
Dickinson PF, Carrington AL, Frost GS, Boulton AJM: Neurovascular disease, antioxidants and glycation in diabetes.
Diabetes Metab Res Rev
18
:
260
–272,
2002
5.
Melhem MF, Craven PA, Liachenko J, DeRubertis FR: α-Lipoic acid attenuates hyperglycemia and prevents glomerular mesangial matrix expansion in diabetes.
J Am Soc Nephrol
13
:
108
–116,
2002
6.
Onozato ML, Tojo A, Goto A, Fujita T, Wilcox CS: Oxidative stress and nitric oxide synthase in rat diabetic nephropathy: effects of ACE-I and ARB.
Kidney Int
61
:
186
–194,
2002
7.
Kakimoto M, Inoguchi T, Sona T, Yu HY, Imamura M, Etoh T, Hashimoto T, Hajime N: Accumulation of 8-hydroxy-2′-deoxyguanosine and mitochondrial DNA deletion in kidney of diabetic rat.
Diabetes
51
:
1588
–1595,
2002
8.
Odetti P, Pesce C, Traverso N, Menini S, Maineri EP, Cosso L, Valentini S, Patriarca S, Cottalasso D, Marianari U, Pronzato MA: Comparative trial of N-acetyl-cystein, taurine, and oxerutin on skin and kidney damage in long-term experimental diabetes.
Diabetes
52
:
499
–505,
2003
9.
Schleicher E, Wagner E, Nerlich AG: Increased accumulation of the glycooxidation product N(epsilon)-(carboxymethyl)lysine in human tissues in diabetes and aging.
J Clin Invest
99
:
457
–468,
1997
10.
Paolisso G, D’Amore A, Balbi V, Volpe C, Galzerano D, Giugliano D, Sgambato S, Varricchio M, D’Onfrio F: Plasma vitamin C affects glucose homeostasis in healthy subjects and in non insulin dependent diabetes.
Am J Physiol
286
:
E261
–E286,
1994
11.
Craven PA, Melhem MF, Philips SL, DeRubertis FR: Overexpression of Cu2+/Zn2+ superoxide dismutase protects against early diabetic glomerular injury in transgenic mice.
Diabetes
50
:
2114
–2125,
2001
12.
Jaimes EA, Galceran JM, Raij L: Angiotensin II induces superoxide anion production by mesangial cells.
Kidney Int
54
:
775
–784,
1998
13.
U.K. Prospective Diabetes (UKPDS) Group: Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment risk of complications in patients with type 2 diabetes (UKPDS 33).
Lancet
352
:
837
–853,
1998
14.
Lewis EJ, Hunsicker LG, Bain RP, Rohde RD: The effect of angiotensin converting enzyme inhibition on diabetic nephropathy.
N Engl J Med
329
:
1456
–1462,
1993
15.
Cosentino F, Hishikawa K, Katusic ZS, Luscher TF: High glucose increases nitric oxide synthase expression and superoxide anion generation in human aortic endothelial cells.
Circulation
96
:
25
–28,
1997
16.
Koya D, Lee IK, Ishi H, Kanoh H, King GL: Prevention of glomerular dysfunction in diabetic rats by treatment with d-alpha-tocopherol.
J Am Soc Nephrol
8
:
426
–435,
1997
17.
Trachtman H, Futterweit S, Maesaka J, Ma C, Valderrama E, Fuchs A, Tarectean AA, Rao PS, Sturman JA, Boles TH, Fu M-X, Baynes J: Taurine ameliorates chronic streptozotocin induced diabetic nephropathy in rats.
Am J Physiol
269
:
F429
–F438,
1995
18.
Craven P, DeRubertis, Kagan VE, Melhem MF, Studer RK: Effects of dietary supplementation with vitamin C or E on albuminuria, glomerular TGFβ and size in diabetes.
J Am Soc Nephrol
8
:
1405
–1407,
1997
19.
Craven PA, DeRubertis FR, Melhem M: Nitric oxide in diabetic nephropathy.
Kidney Int
52 (Suppl.)
:
S46
–S53,
1997
20.
Craven PA, Studer RK, DeRubertis FR: Impaired nitric oxide-dependent cyclic guanosine monophosphate generation in glomeruli from diabetic rats: evidence for protein kinase C-mediated suppression of the cholinergic response.
J Clin Invest
93
:
311
–320,
1994
21.
Rosen DR, Siddigue T, Patterson D, Figlewicz DA, Sapp P, Hentati A, Donaldson D, Goto J, O’Regan JP, Deng HX: Mutations in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis.
Nature
362
:
59
–62,
1993
22.
Schreyer SA, Chua SC Jr, LeBoeuf RC: Obesity and diabetes in a TNF-α receptor-deficient mice.
J Clin Invest
102
:
402
–411,
1998
23.
Craven PA, Melhem MF, DeRubertis FR: Thromboxane in the pathogenesis of glomerular injury in diabetes.
Kidney Int
42
:
937
–946,
1992
24.
Kowalski TJ, Liu S-M, Leibel RL, Chua SC: Transgenic complementation of leptin-receptor deficiency. I. Rescue of the obesity/diabetes phenotype of LEPR-null mice expressing a LEPR-B transgene.
Diabetes
50
:
425
–435,
2001
25.
Hummel KP, Coleman DL: The influence of genetic background on the expression of mutations at the diabetes locus in the mouse. I. C57BL/KsJ and C57BL/6J strains.
Biochem Genet
7
:
1
–13,
1972
26.
Cohen MP, Sharma K, Jin Y, Hud E, Wu VY, Tomaszewski J, Ziyadeh FN: Prevention of diabetic nephropathy in db/db mice with glycated albumin antagonists.
J Clin Invest
95
:
2338
–2345,
1995
27.
Melhem MF, Craven PA, DeRubertis FR: Effects of dietary supplementation of α-lipoic acid on early glomerular injury in diabetes mellitus.
J Am Soc Nephrol
12
:
124
–133,
2001
28.
Trachtman H, Futterweit S, Maesaka J, Ma C, Valderrama E, Fuchs A, Tarectecan AA, Rao PS, Sturman JA, Boles TH, Fu MX, Baynes J: Taurine ameliorates chronic streptozotocin induced diabetic nephropathy in rats.
Am J Physiol
269
:
F429
–F38,
1995
29.
Zatz R, Dunn BR, Meyer TW, Anderson S, Rennie HG, Brenner BM: Prevention of diabetic glomerulopathy by pharmacological amelioration of glomerular capillary hypertension.
J Clin Invest
77
:
1925
–1930,
1986
30.
Epstein C, Avraham KB, Lovett M, Smith S, Elroy-Stein O, Rotman G, Bry C, Groner Y: Transgenic mice with increased Cu/Zn superoxide dismutase activity: animal model of dosage effects in Down syndrome.
Proc Natl Acad Sci U S A
84
:
8044
–8048,
1987
31.
Fridovich I: Superoxide radical and superoxide dismutases.
Annu Rev Biochem
64
:
97
–112,
1995
32.
Okado-Matsumoto A, Fridovich I: Subcellular distribution of superoxide dismutases (SOD) in rat liver.
J Biol Chem
276
:
38388
–38393,
2001
33.
Yamagishi S-I, Edelstein D, Du X-L, Kaneda Y, Guzman M, Brownlee M: Leptin induces mitochondrial superoxide production and monocyte chemoattractant protein-1 expression in aortic endothelial cells by increasing fatty acid oxidation via protein kinase A.
J Biol Chem
276
:
25096
–25100,
2001
34.
Teixeira H, Schumacher R, Meneghini R: Lower intracellular hydrogen peroxide levels in cells overexpressing CuZn-superoxide dismutase.
Proc Natl Acad Sci U S A
95
:
7872
–7875,
1998
35.
Liochev SI, Fridovich I: Superoxide and nitric oxide: consequences of varying rates of production and consumption: a theoretical treatment.
Free Radic Biol Med
33
:
137
–141,
2002
36.
Michiels C, Raes M, Toussaint O, Remacle J: Importance of Se-glutathione peroxidase, catalase, and Cu/Zn-SOD for cell survival against oxidative stress.
Free Radic Biol Med
17
:
235
–248,
1994
37.
Kakkar R, Mantha SV, Radhi J, Prasad K, Kalra J: Antioxidant defense system in diabetic kidney: a time course study.
Life Sci
60
:
667
–679,
1997
38.
Forbes JM, Cooper ME, Thallas V, Burns WC, Thomas MC, Brammar GC, Lee F, Grant SL, Burrell LA, Jerums G, Osica TM: Reduction of the accumulation of advanced glycation end products by ACE inhibition in experimental diabetic nephropathy.
Diabetes
51
:
3274
–3282,
2002
39.
Kowluru RA: Effect of reinstitution of good glycemic control on retinal oxidative stress and nitrative stress in diabetic rats.
Diabetes
52
:
818
–823,
2003
40.
Lyall F, Gibson JL, Greer IA, Brockman DE, Eis ALW, Myatt L: Increased nitrotyrosine in the diabetic placenta: evidence for oxidative stress.
Diabetes Care
21
:
1753
–1758,
1998
41.
Beckman JS, Koppenol WH: Nitric oxide, superoxide and peroxynitrite: the good, the bad and ugly.
Am J Physiology
271
:
C1424
–C1437,
1996
42.
Brennan LA, Wedgwood S, Bekker JM, Black SM: The overexpression of copper-zinc superoxide dismutase protects NOS II from nitric oxide-mediated inhibition.
DNA Cell Biol
21
:
827
–838,
2002
43.
Bank N, Aynedijan HS: Role of EDRF (nitric oxide) in diabetic renal hyperfiltration.
Kidney Int
43
:
1306
–1312,
1993
44.
Tolins JP, Shultz PJ, Raij L, Brown DM, Mauer SM: Abnormal renal hemodynamic response to reduced renal perfusion pressure in diabetic rats: role of NO.
Am J Physiol
265
:
F886
–F895,
1993
45.
Komers R, Allen TJ, Cooper ME: Role of endothelium-derived nitric oxide in the pathogenesis of the renal hemodynamic changes of experimental diabetes.
Diabetes
43
:
1190
–1197,
1994
46.
Bucala R, Tracey KJ, Cerami A: Advanced glycosylation products quench nitric oxide and mediate defective endothelium-dependent vasodilation in experimental diabetes.
J Clin Invest
87
:
432
–438,
1991
47.
Hattori Y, Kawasaki H, Abe K, Kanno M: Superoxide dismutase recovers altered endothelium-dependent relaxation in diabetic rat aorta.
Am J Physiol
261
:
H1086
–H1094,
1991
48.
Craven PA, Studer RK, DeRubertis FR: Impaired nitric oxide release by glomeruli from diabetic rats.
Metabolism
44
:
695
–698,
1995
49.
Craven PA, Studer RK, Felder J, Phillips S, DeRubertis FR: Nitric oxide suppresses protein kinase C signalled increases in transforming growth factor β and collagen synthesis induced by glucose in bovine mesangial cells.
Diabetes
46
:
671
–681,
1997
50.
Studer RK, DeRubertis FR, Craven PA: Nitric oxide suppresses increases in mesangial cell protein kinase C, transforming growth factor β and fibronectin synthesis induced by thromboxane.
J Am Soc Nephrol
7
:
999
–1005,
1996