INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

THE DIVERSE FUNCTIONS of nitric oxide (NO) have been studied extensively in a number of different tissues. The potential cytotoxic effects of NO have been investigated in several cell types, including those in which NO may also play an important physiological role. NO is one of the proposed cytotoxic species produced by inflammatory cells, and several reports suggest that NO may be an important mediator of tissue injury. For example, in certain models of inflammation, inhibitors of NO synthase (NOS) have been shown to prevent vascular and tissue injury (17, 23). However, relatively little is known about the pathophysiological role of NO in proximal tubule cells.

NO is synthesized from L-arginine by NOS. Within the kidney, de novo synthesis of L-arginine occurs mainly in proximal tubules (5). NOS constitutes a family of enzymes that synthesise NO (27). Both constitutive NOS (cNOS) and inducible NOS (iNOS) have been identified in proximal tubules (15, 16). The amount of NO generated by cNOS is small but constant; in contrast, iNOS requires several hours for activation and once activated produces NO in large quantities for a prolonged period (11).

Iron is a potent cause of cytotoxicity and has been implicated as a key factor in tubular cell damage in a variety of acute and chronic renal insults (9, 21). It has been well documented that a dominant pathway for iron toxicity involves hydroxyl radical (· OH) production via the Haber Weiss-Fenton reactions (7). Iron catalyzes the production of · OH, which in turn causes lipid peroxidation and cytotoxicity, as demonstrated by malondialdehyde (MDA) production and lactate dehydrogenase (LDH) leakage, respectively. Iron-induced cytotoxicity and lipid peroxidation are closely linked, and the extent of lipid peroxidation parallels the degree of renal damage (21). However, with the use of various scavengers of reactive oxygen species, studies from fresh proximal tubule segments suggested that cytotoxicity and lipid peroxidation are not necessarily linked and that iron may cause cytotoxicity by mechanisms independent of · OH (31). Similarly, studies using hepatocytes exhibited a discordancy between the evolution of MDA production and of LDH leakage, implying that lipid peroxidation and enzyme leakage were induced by different pathways (18).

Therefore, we questioned whether such a discordancy between cell damage and lipid peroxidation might exist in cultured proximal tubular cells (which have recovered from the physical stress of the procedure for isolating proximal tubule segments) and, if so, whether this could be explained by the involvement of NO in iron-induced toxicity.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Materials. Collagenase (type II), Percoll, Dulbecco's modified Eagle's medium (DMEM), nutrient mixture F-12 Ham's medium, Krebs-Henseleit bicarbonate buffer, fetal calf serum (FCS), epidermal growth factor (EGF), insulin, pyruvic acid, -nicotinamide adenine dinucleotide (reduced form, -NADH), transferrin, catalase, superoxide dismutase (SOD), dimethylpyrroline N-oxide (DMPO), desferal (DFO), 4,6-dihydroxy-2-mercaptopyrimidine (2-thiobarbituric acid; TBA), malondialdehyde bis(dimethyl acetal) (MDA), butanol, pyridine, nitrilotriacetatic acid (NTA), ferric ammonium citrate, sodium nitroprusside (SNP), tetramethoxypropane, arginine, and aminoguanidine were purchased from Sigma-Aldrich (Sydney, Australia). Rat tail collagen I (type II) and nitrate reductase were purchased from Boehringer-Mannheim Australia (Sydney, Australia). Dowex AG50WX-8 (Na⁺ form) column was from Bio-Rad (Sydney, Australia), and L-[³H]arginine was from Amersham Life Science (Sydney, Australia). All other miscellaneous chemicals and solvents were of at least analytical reagent grade and were purchased from Ajax Chemicals (Sydney, Australia).

Animals. This study was approved by the Animal Care and Ethics Committee of the Western Sydney Area Health Service, and experiments conformed to standards of the National Health and Medical Research Council. Male Wistar rats weighing 200-250 g were housed under conditions of constant temperature (22°C) and humidity on a 12:12-h light-dark cycle (light on 0600-1800) with free access to commercial rat pellets (Allied Foods, Sydney, Australia) and water.

Isolation and primary culture of proximal tubule cells. Rat renal proximal tubule segments were isolated according to the method of Vinay et al. (26). Briefly, rats were anesthetized with ketamine (40 mg/kg ip) and xylazine (4 mg/kg ip), and kidneys were perfused via the aorta with Krebs-Henseleit buffer. After removal, kidneys were cut in half, and medullas were carefully dissected out. The cortices were cut into small pieces and digested in Krebs-Henseleit solution with 0.1% collagenase and 5% bovine serum albumin (BSA) at 37°C for 20 min. After digestion, the cell suspension was filtered through a 50 mesh and subjected to sedimentation [2,000 revolutions/min (rpm) for 2 min] on a Percoll gradient (32%). This procedure yielded a preparation primarily consisting of proximal tubule fragments (>95%) with ~90% viability as assessed by exclusion of the vital dye trypan blue. Tubule fragments were suspended in the mixture of DMEM and nutrient Ham's F-12 (1:1) medium supplemented with 10% FCS and plated onto plastic 30-mm culture dishes coated with collagen I. Twenty-four hours later, unattached tubules were washed away, and fresh medium was added without FCS but with 10 ng/ml EGF, 5 mg/ml insulin, 5 mg/ml transferrin, and 5 × 10⁸ M hydrocortisone. Thereafter, culture medium was changed every other day. Proximal tubule cells became confluent after 4 days of culture. The proximal origin of cultured cells confirmed by their expression of proximal brush border enzymes, ultrastructure, and formation of domes. Confluent monolayers were assayed in situ for the brush border enzyme alkaline phosphatase, a specific marker of proximal tubule cells, using a cytochemical assay (20). Virtually 100% of cells showed staining for alkaline phosphatase. Electron microscopy showed that as far as could be assessed all cells were of proximal tubular origin with microvilli and many mitochondria and lysosomes.

Assessment of cell toxicity. Cytotoxicity was assessed ultrastructurally by transmission electron microscopy and quantitated functionally using the release of LDH from cells into the medium. Ultrastructure was assessed by one observer, who was blinded to the culture conditions. LDH activity was determined by measuring the increase in absorbance of -NADH during the oxidation of lactate to pyruvate as described by Bergmeyer and Bernt (2) using a Cary 2300 Spectrophotometer (Varian Techtron, Melbourne, Australia). To determine total cellular LDH content, cultures were exposed to 0.1% (vol/vol) Triton X-100 for 20 min to lyse all cells without inhibiting the enzyme. None of the chemicals used interfered with the enzyme assay or decreased the total enzyme content. Results are expressed as the percent of LDH leakage.

Electron microscopy. Monolayers were fixed in situ with Karnovsky's fixative buffer and incubated for 1 h at 4°C. After being washed twice in 0.1 M 3-(N-morpholino)propanesulfonic acid (MOPS) buffer, cells were resuspended in 400 µl MOPS buffer and stored at 4°C until required. Cells were encapsulated in BSA to form blocks, and blocks were postfixed in 2% buffered osmium tetroxide, dehydrated through a graded ethanol series, and embedded in Spurr's epoxy resin. After polymerization (70°C for 10 h) ultrathin sections were stained with 2% ethanolic uranyl acetate and Reynold's lead citrate and examined in a Philips CM10 electron microscope at 80 kV.

Malondialdehyde assay. MDA, a lipid peroxidation product, was measured by the TBA as described previously (19) but with some modifications. Cells were washed three times with 0.01 M phosphate buffer and then transferred into 1 ml Eppendorf tubes. After centrifugation, cells were resuspended in 80 µl distilled water and lysed using an ultrasonic homogenizer. Aliquots were taken for MDA measurement (40 µl) and for protein measurement (40 µl). Forty-milliliter samples were added to an 80 µl mixture of 8.1% sodium dodecyl sulfate (SDS), 0.8% TBA, 20% acetic acid (2:15:15, vol/vol/vol), then heated in boiling water for 70 min. After being cooled with tap water, 20 µl distilled water and 100 µl of a mixture of N-butanol and pyridine (15:1, vol/vol) were added and vortexed. Samples were centrifuged at 4,000 rpm for 10 min, the organic layer was separated, and its absorbance at 532 nm was measured (DU-68 spectrophotometer; Beckman, Fullerton, CA), using MDA as a standard. Protein content was determined using Lowry's method with BSA as a standard (14). MDA values are expressed as nanomoles per milligram of protein.

/ measurement. was determined by a method described by Cattell et al. (3) with minor modifications. In brief, 550 µl of cell culture medium were incubated for 30 min at room temperature with 20 µl of nitrate reductase (90 mU/ml) and 30 µl of NADPH (1.5 mg/ml). The reaction was stopped by boiling for 15 min. Griess reagent [100 µl; 1:1 mixture of 2% sulfanilamide in 5% H₃PO₄ and 0.2% N-(1-naphthy)ethylenediamine dihydrochloride in water] was then added to the reduced samples and incubated at room temperature for 15 min. reacts with the Griess reagent to form a chromophore and its absorbance at 546 nm was measured in a Beckman DU-68 spectrophotometer. Standard curves (0-8 nmol/tube of sodium nitrate in 550 µl medium) were constructed for each experiment. Values are expressed as nanomoles per dish.

Nitric oxide synthase activity. NOS activity was assessed by measuring the conversion of L-arginine to L-citrulline by a method similar to that of Rengasamy and Johns (22). Cell pellet was reconstituted in 0.5 ml of cold homogenizing buffer [50 mM N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES), 0.5 mM EDTA, 1 mM dithiothreitol, 0.01% phenylmethylsulfonyl fluoride, pH 7.5] and sonicated. The homogenate was then centrifuged at 20,000 g for 1 h at 4°C, and the supernatant was passed over a 1 ml Dowex AG50WX-8 (Na⁺ form) column to remove endogenous L-arginine. Protein content was determined by Lowry's method. The NOS reaction mixture (400 µl) consisted of 320 µl of enzyme extract, 100 µM L-[³H]arginine (1 µCi/ml, 1 µCi = 37 kBq), 1 mM NADPH, and 2 mM CaCl₂. The reaction was carried out at 37°C and stopped at 60 min by removing 200 µl of the reaction mixture and adding it to 2 ml of HEPES/EDTA buffer [in mM: 20 HEPES, 2 EDTA, 2 ethylene glycol-bis(-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA), pH 5.5]. The mixture was then passed over the Dowex AG50WX-8 (Na⁺ form) column, and [³H]citrulline was eluted with an additional 2 ml of water. Radioactivity in the eluate was counted by a LKB Rackbeta (model 1215) liquid scintillation spectrometer (LKB, Turku, Finland). Results are expressed as nanomoles of [³H]citrulline produced per milligram of protein in 60 min.

Treatment. Cultured confluent proximal tubule cells were treated on day 5 of culture with NTA or NTA-Fe by addition of NTA or NTA-Fe to the fresh medium. NTA-Fe was prepared by adding NTA (a carrier of Fe) to ferric ammonium citrate in the molar ratio 2:1 with pH adjusted to 7.4. To investigate the effects of scavengers on iron-induced toxicity, each scavenger was added to the culture medium simultaneously with NTA-Fe. The dose of scavenger was determined by previous studies in cultured hepatocytes or fresh proximal tubule segments (24, 36). Similarly, experiments observing the role of NO in iron-induced toxicity were performed by adding L-arginine or aminoguanidine to the medium at the same time as NTA-Fe.

Statistical analysis. Results are expressed as means ± SD of 3-4 separate experiments in which values were determined in triplicate. Analysis of variance and Fisher's significant test method were used for comparisons among multiple means, and the Student's unpaired t-test was used for comparison between two means. P < 0.05 was considered significant.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Effect of iron on LDH leakage in proximal tubule cells in primary culture. LDH leakage in proximal tubule cells was constant (5-10%) after day 5 of primary culture. NTA, in a wide range of concentrations (50-800 µM), did not affect LDH leakage in comparison to control cells. However, NTA-Fe induced LDH leakage in both a dose- and time-dependent manner (Fig. 1). Between 4 and 24 h incubation of proximal tubule cells with NTA-Fe, LDH leakage rose dramatically. LDH leakage increased slightly or remained stable thereafter. A moderate concentration of NTA-Fe (400 µM) was chosen for the following experiments.

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Dose- and time-dependent leakage of lactate dehydrogenase (LDH) induced by nitrilotriacetate-Fe (NTA-Fe) in cultured proximal tubule cells. Varying doses of NTA-Fe (, 50 µM; , 200 µM; , 400 µM; , 600 µM; , 800 µM) or NTA (×, 800 µM) were added to cell culture medium on day 5 of cell culture. LDH was determined at time points indicated. Control () is normal medium without addition of NTA or NTA-Fe. Data are means ± SD, n = 9. SD bars are omitted where SD was very small.

Effect of iron on MDA production in proximal tubule cells in primary culture. In contrast to LDH leakage, after incubation of proximal tubule cells with 400 µM NTA-Fe, MDA production increased at an earlier time of incubation. MDA production was significantly increased at 1 h and peaked at 5 h of incubation (3.57 ± 0.52 and 20.09 ± 3.82 nmol/mg protein, respectively, P < 0.01 compared with control). MDA production declined thereafter to 5.07 ± 1.05 nmol/mg protein after 24 h (Fig. 2). MDA production was very low (1.68 ± 0.41 nmol/protein) in control cells. There was no significant difference in MDA production between control cells and NTA-treated cells.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Malondialdehyde bis(dimethyl acetal) (MDA) production induced by NTA-Fe in cultured proximal tubule cells. NTA-Fe (, 400 µM) or NTA (×, 800 µM) was added to cell culture medium on day 5. MDA was determined at time points indicated. Control () is no treatment. Data are means ± SD, n = 9. SD bars are omitted where SD was very small.

Changes in ultrastructural appearance in proximal tubule cells. After 24 h treatment of cells with 400 µM NTA-Fe, swollen mitochondria with loss of granules, focal vacuolation, and disorganized cistae were observed. There was a general increase in lysosomes containing electron dense material (Fig. 3A). In 800 µM NTA-treated cells, mitochondria showed no disruption of morphology, and prominent mitochondria granules were identified in most cells examined (Fig. 3B).

View larger version (133K): [in this window] [in a new window]   Fig. 3.   Changes in morphology of proximal tubule cells induced by NTA-Fe. On day 5 of primary culture, 400 µM NTA-Fe (A) or 800 µM NTA (B) was added to culture medium for 24 h. Cell morphology was examined using electron microscopy. Magnification: A, ×31,500; B, ×23,000.

NO production in iron-treated proximal tubule cells. NO production (/) was measured in control, NTA-treated, and NTA-Fe-treated cells after 2, 4, 8, 12, and 24 h treatment. There was no significant difference in / production between control and NTA-treated cells at each time point. However, / levels were significantly increased in cells treated with 400 µM NTA-Fe from 8 to 24 h (Table 1).

Effect of L-arginine on LDH leakage and NO production in control and iron-treated proximal tubule cells. L-Arginine, a substrate of NOS, was added to the medium of control cells and cells treated 400 µM NTA-Fe for 24 h. As shown in Fig. 4, 5 mM L-arginine alone did not affect LDH leakage or / production after 24 h of treatment. However, LDH leakage and / production induced by 400 µM NTA-Fe was significantly enhanced by 5 mM L-arginine. In contrast, D-arginine did not enhance iron-induced NO production and LDH leakage.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Effects of L-arginine on LDH leakage and nitric oxide (NO) production in control and NTA-Fe-treated proximal tubule cells. L-Arginine (5 mM) was added to culture medium with or without 400 µM NTA-Fe on day 5. LDH leakage and NO production were assayed 24 h later. Data are means ± SD, n = 9. * P < 0.01 compared with 400 µM NTA-Fe-treated cells.

Effect of aminoguanidine on iron-induced LDH leakage and NO production in proximal tubule cells. To investigate the effect of inhibiting NO production on iron-induced toxicity, aminoguanidine (10 mM), an inhibitor of NOS, was added to the medium simultaneously with 400 µM NTA-Fe for 24 h. As shown in Fig. 5, in the case of LDH leakage, NTA-Fe-induced toxicity was partially prevented and, in the case of / production, completely prevented by aminoguanidine. Aminoguanidine itself had no effect on either LDH leakage or / production.

View larger version (12K): [in this window] [in a new window]   Fig. 5.   Effect of aminoguanidine (AG) on LDH leakage and NO production in control and NTA-Fe-treated proximal tubule cells. Aminoguanidine (10 mM) was added to culture medium with or without 400 µM NTA-Fe on day 5. LDH leakage and NO production were assayed 24 h later. Data are means ± SD, n = 9. * P < 0.01 compared with 400 mM NTA-Fe-treated cells.

Effect of exogenous NO on LDH leakage and MDA production in proximal tubule cells. To examine for a direct effect of NO on proximal tubular cell toxicity, exogenous NO was generated by addition to the medium of SNP for 24 h. LDH leakage was increased by SNP in a dose-dependent manner (Table 2). However, MDA production was not altered by a wide range of concentrations of SNP (0.02-20 mM) (Table 2).

NOS activity induced by NTA-Fe in proximal tubule cells. To investigate whether NOS activity was induced by NTA-Fe in proximal tubule cells, NOS activity was determined after 24 h of treatment. The production of [³H]citrulline in 400 µM NTA-Fe-treated cells was significantly higher than in control cells (Fig. 6).

View larger version (10K): [in this window] [in a new window]   Fig. 6.   Iron-induced NO synthase (NOS) activity in proximal tubule cells in primary culture. Cells were treated with 400 µM NTA-Fe on day 5, NOS activity ([3H]citrulline) was measured after 24 h treatment. Data are means ± SD, n = 9. * P < 0.01 compared with NTA or control group.

Effects of scavengers on iron-induced changes in proximal tubule cells. Various scavengers of reactive oxygen species were added to day 5 cell culture medium simultaneously with 400 µM NTA-Fe for 24 h. LDH leakage, MDA production, and NO generation were measured after 24, 5, and 24 h, respectively. MDA production was inhibited by SOD (1,000 IU/ml), catalase (1,0,000 IU/ml), DMPO (32 mM), or DFO (200 µM). However, LDH leakage was reduced only by scavengers that also reduced NO production. Thus DMPO and DFO but not SOD or catalase decreased both LDH leakage and NO production (Table 3).

Effect of iron in lower concentration. The above experiments were performed at 400 µM NTA-Fe because the concentration of iron in proximal tubule cells (280 ± 10 nmol/mg protein, n = 4) was similar to that obtained in tubules in proteinuric renal disease, such as puromycin nephrosis (25). However, based on iron concentrations in proteinuric urine, others have examined lower concentrations of iron (100 µM) (24). Therefore, further studies were performed using this lower, clinically relevant concentration. Similar results were seen to those observed with 400 µM NTA-Fe. NTA-Fe (100 µM) increased LDH leakage (20.5 ± 0.7 vs. 10.2 ± 0.9%, P < 0.01 ), MDA production (11.32 ± 1.35 vs. 1.65 ± 0.11 nmol/mg protein, P < 0.01), and NO production (2.79 ± 0.14 vs. 1.64 ± 0.10 nmol/dish, P < 0.05) in comparison to control. DFO and DMPO reduced MDA production induced by 100 µM NTA-Fe (3.06 ± 0.11, 5.00 ± 0.28 vs. 11.32 ± 1.35 nmol/mg protein, P < 0.01), LDH leakage (12.2 ± 1.3, 12.3 ± 1.2 vs. 20.5 ± 0.7%, P < 0.01), and NO production (1.98 ± 0.08, 2.06 ± 0.12 vs. 2.79 ± 0.14 nmol/dish, P < 0.01). However, SOD and catalase decreased MDA production (4.91 ± 0.11, 6.10 ± 0.16 vs. 11.32 ± 1.35 nmol/mg protein, P < 0.01), had no effects on LDH (20.8 ± 1.7, 20.2 ± 1.1 vs. 20.5 ± 0.7%), and NO production (2.92 ± 0.05, 2.80 ± 0.18 vs. 2.79 ± 0.14 nmol/dish).

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

The present study clearly demonstrates (to our knowledge, for the first time) that NO is one of the mediators of iron-induced injury in cultured proximal tubule cells. The cytotoxicity induced by NO in this situation appears to be independent of lipid peroxidation. These findings may explain discordancy between lipid peroxidation and cytotoxicity in iron-induced injury.

Iron-induced toxicity has been observed in vivo in whole kidney (8) and in freshly isolated proximal tubule segments (31). The present study has confirmed the toxicity of iron in cultured proximal tubule cells. NO production was induced by 400 µM NTA-Fe, a concentration of iron shown to cause cytotoxicity. To determine whether NO was one of the factors contributing to iron-mediated injury or its production merely an epiphenomenon, NO levels were modulated with L-arginine and aminoguanidine, a substrate and inhibitor respectively of NOS. As expected, L-arginine enhanced, whereas aminoguanidine decreased, iron-induced NO production in cultured proximal tubule cells. These results suggested that NO production is, in fact, a contributing factor to iron-mediated cytotoxicity. Aminoguanidine completely inhibited NO production induced by NTA-Fe but only partially protected against LDH leakage, emphasizing that NO is only one of the mediators of iron-induced cytotoxicity. SNP has recently been shown to generate NO either in an enzymatic reaction or in a nonenzymatic reaction with cysteine, depending on the nitrogen oxidation state (10). The fact that SNP could induce LDH leakage in a dose-dependent manner further demonstrated the toxicity of NO in cultured tubular cells.

NO is synthesized from the amino acid L-arginine by NOS. When the conversion of L-[³H]arginine to L-[³H]citruline was measured, NOS activity was detected in untreated proximal tubule cells, and this activity was enhanced by 400 µM NTA-Fe. Data concerning the effect of iron on the regulation of NOS have been conflicting. Weiss et al. (28) reported that iron decreased NOS activity by 50% in a murine macrophage cell line, whereas another study showed that iron potentiated bacterial lipopolysaccharide-induced NO formation in liver, intestine, lung, heart, spleen, and kidney by the activation of iNOS (12). These conflicting results emphasize the importance of the system under examination and the danger of extrapolating results from one system to another. In the present study, iron increased NOS activity 20-fold, whereas NO production was increased twofold. A possible explanation for the difference between NOS activity and NO production is that the reaction is limited by the concentration of L-arginine in cell culture medium (~0.85 mM).

Two types of NOS (iNOS and cNOS) have been identified in proximal tubule cells. Cytokines and hypoxia have been shown to induce proximal tubule cell iNOS (15, 30). iNOS induction usually requires several hours of stimulation, and, once induced, its activity lasts longer and generates more NO than cNOS. In the current study, 8 h of exposure to iron was required to detect an increase in NO, and the amount of NO produced in response to iron suggested that iNOS was the species of NOS involved.

The promoter region of inducible murine macrophage NOS contains an element that binds activator protein-1 and nuclear factor B, and these transcription factors are activated in response to oxidative stress (29). Because iron stimulates the production of reactive oxygen species (ROS), oxygen radicals may serve as a common mechanism by which iron stimulates the induction of NOS. Studies using hepatocytes demonstrated that the generation of ROS by cytokines or other physiological processes plays a major role in NOS induction (6). The current study examined the effect of various scavengers of ROS on NO production. DMPO and DFO at least partially inhibited NO production, whereas catalase and SOD had no effect. The inhibition of NO production by the iron chelator DFO and by DMPO, an inhibitor of the · OH, suggested the involvement of both iron and · OH in NOS induction in iron-treated proximal tubule cells.

One of the pathways by which NO induces toxicity is its reaction with superoxide to form the intermediate anion peroxynitrite (ONOO). ONOO further reacts to form highly toxic · OH, which causes membrane lipid peroxidation and LDH leakage (1). However, in the present study, lipid peroxidation, at least as assessed by MDA production, was not demonstrable in NO-dependent toxicity induced by SNP. The fact that lipid peroxidation was not induced in the present study suggests that NO-mediated toxicity in cultured proximal tubule cells may involve pathways other than that leading to ONOO and · OH. For example, NO itself is cytotoxic by protein nitrosylation, as had been demonstrated in macrophages and brain cells (4, 13).

The effects of various scavengers of ROS on iron-induced LDH leakage and MDA production were examined in the present study. DMPO and DFO reduced both MDA production and LDH leakage, whereas SOD and catalase failed to reduce LDH leakage despite decreasing MDA production to a similar degree to DMPO. The fact that each scavenger inhibited MDA production could be explained by the involvement of each of the corresponding oxygen species in the stimulation of lipid peroxidation by iron (18). The reduction of LDH leakage by DMPO and DFO but not SOD and catalase could be explained by iron- and · OH-induced NO production and peroxidation-independent toxicity of NO.

In summary, NO production and NOS activity were increased in cultured proximal tubule cells by iron treatment. L-Arginine augmented and aminoguanidine decreased NO production and LDH leakage. We conclude that NO is one of the mediators of iron-induced toxicity. There are at least two pathways for proximal tubule cell injury by iron: one dependent on lipid peroxidation and one not. The latter is NO dependent and may account for the discordancy between the LDH leakage and MDA production in iron-induced cell injury.