The role of superoxide anion and hydrogen peroxide in glomerular injury induced by puromycin aminonucleoside in rats

Toll-like Receptor 9 Can be Activated by Endogenous Mitochondrial DNA to Induce Podocyte Apoptosis

Toll-like receptor 9 (TLR9) senses bacterial DNA characteristic of unmethylated CpG motifs to induce innate immune response. TLR9 is de novo expressed in podocytes of some patients with glomerular diseases, but its role in podocyte injury remains undetermined. Since TLR9 activates p38 MAPK and NFkB that are known to mediate podocyte apoptosis, we hypothesized that TLR9 induces podocyte apoptosis in glomerular diseases. We treated immortalized podocytes with puromycin aminonucleosides (PAN) and observed podocyte apoptosis, accompanied by TLR9 upregulation. Prevention of TLR9 upregulation by siRNA significantly attenuated NFκB p65 or p38 activity and apoptosis, demonstrating that TLR9 mediates podocyte apoptosis. We next showed that endogenous mitochondrial DNA (mtDNA), whose CpG motifs are also unmethylated, is the ligand for TLR9, because PAN induced mtDNA accumulation in endolysosomes where TLR9 is localized, overexpression of endolysosomal DNase 2 attenuated PAN-induced p38 or p65 activity and podocyte apoptosis, and DNase 2 silencing was sufficient to activate p38 or p65 and induce apoptosis. In PAN-treated rats, TLR9 was upregulated in the podocytes, accompanied by increase of apoptosis markers. Thus, de novo expressed TLR9 may utilize endogenous mtDNA as the ligand to facilitate podocyte apoptosis, a novel mechanism underlying podocyte injury in glomerular diseases.

Contribution of endogenously produced reactive oxygen species to the activation of podocyte NLRP3 inflammasomes in hyperhomocysteinemia

Hyperhomocysteinemia (hHcys) is an important pathogenic factor contributing to the progression of end-stage renal disease. Recent studies have demonstrated the implication of nicotinamide adenine dinucleotide phosphate oxidase-mediated NLRP3 inflammasome activation in the development of podocyte injury and glomerular sclerosis during hHcys. However, it remains unknown which reactive oxygen species (ROS) are responsible for this activation of NLRP3 inflammasomes and how such action of ROS is controlled. This study tested the contribution of common endogenous ROS including superoxide (O2(-)), hydrogen peroxide (H2O2), peroxynitrite (ONOO(-)), and hydroxyl radical (OH) to the activation of NLRP3 inflammasomes in mouse podocytes and glomeruli. In vitro, confocal microscopy and size-exclusion chromatography demonstrated that dismutation of O2(-) by 4-hydroxy-2,2,6,6-tetramethylpiperidine 1-oxyl (Tempol) and decomposition of H2O2 by catalase prevented Hcys-induced aggregation of NLRP3 inflammasome proteins and inhibited Hcys-induced caspase-1 activation and IL-1β production in mouse podocytes. However, scavenging of ONOO(-) or OH had no significant effect on either Hcys-induced NLRP3 inflammasome formation or activation. In vivo, scavenging of O2(-) by Tempol and removal of H2O2 by catalase substantially inhibited NLRP3 inflammasome formation and activation in glomeruli of hHcys mice as shown by reduced colocalization of NLRP3 with ASC or caspase-1 and inhibition of caspase-1 activation and IL-1β production. Furthermore, Tempol and catalase significantly attenuated hHcys-induced glomerular injury. In conclusion, endogenously produced O2(-) and H2O2 primarily contribute to NLRP3 inflammasome formation and activation in mouse glomeruli resulting in glomerular injury or consequent sclerosis during hHcys.

Impaired glomerular and tubular antioxidative defense mechanisms in nephrotic syndrome

The molecular mechanisms behind acquired nephrotic syndrome (NS) are still largely unknown. One possible explanation for the development of proteinuria is oxidative damage to the glomerular cells. Our hypothesis was that the oxidative defense is weakened in NS, and we focused on measurements of the oxidative-antioxidative status in the glomerular and tubular parts of the nephron. Gene expression was analyzed in renal biopsies from patients with NS. In addition, to compare the acute and chronic phases of the disease, we studied puromycin-treated rats. In the biopsy material, the expression of enzymes involved in the antioxidative defense was higher in the tubulointerstitial compartment than in the glomerular cells. Real-time PCR analysis revealed a decreased glomerular expression in nephrotic kidneys for the antioxidant enzymes catalase and glutathione peroxidase-3, and -4. The tubular gene expression was downregulated for catalase, glutathione peroxidase-3, and thioredoxin reductase-1 and -2. The altered gene expression was accompanied by increased lipid peroxidation in urine. In rats, serum concentrations of ascorbyl-free radicals, measured with electron spin resonance, were elevated in the acute phase of the disease, suggesting increased oxidative stress in the circulation. In addition, we saw an increase in the plasma antioxidant capacity combined with a decreased oxidation of proteins in sera from nephrotic rats, but not from humans. In conclusion, there is a marked downregulation of several antioxidative enzymes in nephrotic kidneys, especially in glomerular structures. Our data suggest that oxidative damage to glomerular cells may contribute significantly to the course and prognosis of nephrotic syndrome.

Over-expression of adenosine deaminase in mouse podocytes does not reverse puromycin aminonucleoside resistance

Background

Edema in nephrotic syndrome results from renal retention of sodium and alteration of the permeability properties of capillaries. Nephrotic syndrome induced by puromycin aminonucleoside (PAN) in rats reproduces the biological and clinical signs of the human disease, and has been widely used to identify the cellular mechanisms of sodium retention. Unfortunately, mice do not develop nephrotic syndrome in response to PAN, and we still lack a good mouse model of the disease in which the genetic tools necessary for further characterizing the pathophysiological pathway could be used. Mouse resistance to PAN has been attributed to a defect in glomerular adenosine deaminase (ADA), which metabolizes PAN. We therefore attempted to develop a mouse line sensitive to PAN through induction of normal adenosine metabolism in their podocytes.

Methods

A mouse line expressing functional ADA under the control of the podocyte-specific podocin promoter was generated by transgenesis. The effect of PAN on urinary excretion of sodium and proteins was compared in rats and in mice over-expressing ADA and in littermates.

Results

We confirmed that expression of ADA mRNAs was much lower in wild type mouse than in rat glomerulus. Transgenic mice expressed ADA specifically in the glomerulus, and their ADA activity was of the same order of magnitude as in rats. Nonetheless, ADA transgenic mice remained insensitive to PAN treatment in terms of both proteinuria and sodium retention.

Conclusions

Along with previous results, this study shows that adenosine deaminase is necessary but not sufficient to confer PAN sensitivity to podocytes. ADA transgenic mice could be used as a background strain for further transgenesis.

Alterations in fatty acid utilization and an impaired antioxidant defense mechanism are early events in podocyte injury: a proteomic analysis

Ultrastructural alterations of podocytes are closely associated with loss of glomerular filtration function. In the present study, we explored changes at the proteome level that paralleled the disturbances of podocyte architecture in the early stages of puromycin aminonucleoside (PA) nephrosis in vivo. Using two-dimensional fluorescence difference gel electrophoresis and vacuum matrix-assisted laser desorption/ionization mass spectrometry combined with postsource decay fragment ion analysis and high-energy collision-induced dissociation tandem mass spectrometry, 23 differentially expressed protein spots, corresponding to 16 glomerular proteins that are involved in various cellular functions, were unambiguously identified, and a subset was corroborated by Western blot analysis. The majority of these proteins were primarily related to fatty acid metabolism and redox regulation. Key enzymes of the mitochondrial beta-oxidation pathway and antioxidant enzymes were consistently down-regulated in PA nephrosis. These changes were paralleled by increased expression levels of CD36. PA treatment of murine podocytes in culture resembled these specific protein changes in vitro. In this cell system, the modulatory effects of albumin-bound fatty acids on the expression levels of Mn-superoxide dismutase in response to PA were demonstrated as well. Taken together, these results indicate that a disrupted fatty acid metabolism in concert with an impaired antioxidant defense mechanism in podocytes may play a role in the early stages of PA-induced lesions in podocytes.

Oxidants in chronic kidney disease

Chronic kidney disease is a worldwide public health problem that affects approximately 10% of the US adult population and is associated with a high prevalence of cardiovascular disease and high economic cost. Chronic renal insufficiency, once established, tends to progress to end-stage kidney disease, suggesting some common mechanisms for ultimately causing scarring and further nephron loss. This review defines the term reactive oxygen metabolites (ROM), or oxidants, and presents the available experimental evidence in support of the role of oxidants in diabetic and nondiabetic glomerular disease and their role in tubulointerstitial damage that accompanies progression. It concludes by reviewing the limited human data that provide some proof of concept that the observations in experimental models may be relevant to human disease.

Puromycin aminonucleoside induces oxidant-dependent DNA damage in podocytes in vitro and in vivo

A decline in podocyte number correlates with progression to glomerulosclerosis. A mechanism underlying reduced podocyte number is the podocyte's relative inability to proliferate in response to injury. Injury by the podocyte toxin puromycin aminonucleoside (PA) is mediated via reactive oxygen species (ROS). The precise role of ROS in the pathogenesis of PA-induced glomerulosclerosis remains to be determined. We sought to examine whether PA-induced ROS caused podocyte DNA damage, possibly accounting for the podocyte's inability to proliferate in response to PA. In vitro, podocytes were exposed to PA, with or without the radical scavenger 1,3-dimethyl-2-thiourea (DMTU). In vivo, male Sprague-Dawley rats were divided into experimental groups (n = 6/group/time point): PA, PA with DMTU, and control, killed at days 1.5, 3, or 7. DNA damage was measured by DNA precipitation, apurinic/apyrimidinic site, Comet, and 8-hydroxydeoxyguanosine assays. Cell cycle checkpoint protein upregulation (by immunostaining and Western blotting), histopathology, and biochemical parameters were examined. DNA damage was increased in cultured podocytes that received PA, but not PA with DMTU. PA exposure activated specific cell cycle checkpoint proteins, with attenuation by DMTU. DNA repair enzymes were activated, providing evidence for attempted DNA repair. The PA-treated animals developed worse proteinuria and histopathologic disease and exhibited more DNA damage than the DMTU pretreated group. No significant apoptosis was detected by terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling staining. A mechanism underlying the lack of podocyte proliferation following PA-induced injury in vitro and in vivo may be ROS-mediated DNA damage, with upregulation of specific cell cycle checkpoints leading to cell cycle arrest.

Transglutaminase-catalyzed site-specific glycosidation of catalase with aminated dextran

An enzymatic approach, based on a transglutaminase-catalyzed coupling reaction, was investigated to modify bovine liver catalase with an end-group aminated dextran derivative. We demonstrated that catalase activity increased after enzymatic glycosidation and that the conjugate was 3.8-fold more stable to thermal inactivation at 55 degrees C and 2-fold more resistant to proteolytic degradation by trypsin. Moreover, the transglutaminase-mediated modification also improved the pharmacokinetics behavior of catalase, increasing 2.5-fold its plasma half-life time and reducing 3-fold the total clearance after its i.v. administration in rats.

Induction of antioxidant enzymes in murine podocytes precedes injury by puromycin aminonucleoside

BACKGROUND

An imbalance between the generation of reactive oxygen species (ROS) and antioxidant defense mechanisms has been suggested to play an important role in podocyte injury in nephrotic syndrome. Experimental nephrotic syndrome induced by injection of puromycin aminonucleoside (PAN) into rats is a well-established model of nephrotic syndrome, and can be largely prevented by pretreatment with antioxidant enzymes (AOE), suggesting that podocyte injury may be mediated by ROS.

METHODS

To test the hypothesis that PAN-induced podocyte injury is modulated in part by podocyte antioxidant defenses, we analyzed AOE activities, lipid peroxidation products, and relative ROS levels in podocytes using our recently reported in vitro model of PAN-induced podocyte injury.

RESULTS

PAN treatment induced early increases in both podocyte hydrogen peroxide and superoxide and later increases in lipid peroxidation products. Compared to baseline activities, PAN also induced significant changes in the major cellular AOE activities (maximum increases of 151% for catalase, 134% for superoxide dismutase, and 220% for glutathione peroxidase vs. time-matched controls). These changes largely preceded the development of extensive podocyte process retraction and actin filament disruption, which was maximal at 7 days.

CONCLUSION

These results demonstrate that (1) PAN treatment induces significant early changes in podocyte ROS, (2) podocytes can mount an antioxidant defense against oxidant stress, and (3) this protective response is initiated prior to the development of extensive oxidant-induced podocyte structural injury. These findings suggest that enhancement of podocyte AOE activities represent a potential therapeutic target to protect from or ameliorate podocyte injury during nephrotic syndrome.

Puromycin aminonucleoside induces apoptosis and increases HNE in cultured glomerular epithelial cells(1)

Puromycin aminonucleoside induces apoptosis and increases 4-hydroxy-2-nonenal (HNE) in cultured glomerular epithelial cells. We have previously reported the detachment of cultured glomerular epithelial cells (GECs) from their substrata by puromycin aminonucleoside (PAN) treatment. In this study we explored whether or not apoptosis was involved in the mechanisms of the detachment. DNA fragmentation on gel electrophoresis was clearly shown by 10(-3) M PAN treatment of GECs. Nuclear staining by Hoechst 33342 indicated the greatest number of apoptotic cells at 10(-3) M PAN for 48 h treatment. Similarly, TUNEL methods revealed maximal apoptotic cells at 10(-3) M PAN for 48 h treatment. Caspase-3 (like) protease activity increased at 10(-3) M PAN, and decreased at 2 x 10(-3) M PAN for 48 h treatment as well as at 10(-3) M PAN for 60 h treatment. Pretreatment with 2'-deoxycoformycin (DCF), inhibitor of adenosine deaminase, abolished these effects of PAN on cultured GECs. PAN treatment increased HNE, a lipid peroxide adduct, modified protein in cultured GECs, which was also prevented by pretreatment by DCF. These results for the first time indicate that the PAN-induced detachment of GECs from culture substrata is mediated at least in part through apoptosis via oxidative stresses by adenosine deaminase activity.

Puromycin aminonucleoside induces glomerular epithelial cell apoptosis

Glomerular epithelial cell (GEC) injury has been considered to play an important role in puromycin aminonucleoside (PAN)-induced nephrosis. We studied the effect of PAN on rat as well as human GEC apoptosis. Morphogic evaluation of GEC apoptosis and necrosis was carried out by staining with H-33342 and propidium iodide. GEC apoptosis was further confirmed by DNA fragmentation assay (by both agarose gel electrophoresis and end-labeling). To determine the dose- and time-response effect of PAN, GECs were treated with variable concentrations of PAN (10 to 500 microg/ml) for variable time periods (6 to 48 h). To determine the role of gene synthesis, we studied the effect of actinomycin D (a transcriptional inhibitor) on PAN-induced GEC apoptosis. To determine the role of free radicals, we evaluated the effect of superoxide dismutase (SOD), dimethylthiourea (DMTU), and catalase on PAN-induced GEC apoptosis. PAN induced GEC apoptosis in a dose- and time-dependent manner. PAN at a high concentration (PAN, 100 microg/ml) also induced a moderate degree of GEC necrosis. In DNA fragmentation assays PAN-treated GECs showed the classic ladder pattern. PAN-induced GEC apoptosis was partly attenuated with free radical scavengers, such as SOD, DMTU, and catalase. In addition, actinomycin D attenuated PAN-induced GEC apoptosis. PAN induces GEC apoptosis, which may be mediated through the generation of reactive oxygen species.

Lack of a role for inducible nitric oxide synthase in an experimental model of nephrotic syndrome

Puromycin aminonucleoside (PAN) administration in rats produces an experimental model of nephrotic syndrome characterized by glomerular epithelial cell injury and proteinuria. The purpose of this study was to examine the role of nitric oxide (NO) in this model of minimal change glomerular disease. Aminoguanidine (AG) was used to inhibit inducible nitric oxide synthase (iNOS). Sprague-Dawley rats were divided into Control (N = 9), PAN (N = 14), AG (N = 2), and PAN + AG (N = 12) treatment groups. Control animals received saline (i.v. ), PAN animals received PAN (75 mg/kg, i.v.), and PAN + AG animals received PAN plus AG (50 mg/kg, i.p., twice daily). AG animals received a saline injection (i.v.) on day 0 in the place of PAN and then AG on the same schedule as the PAN + AG group. Animals were kept in metabolic cages, and urinary protein excretion and nitrite (NO(2)(-)) excretion were measured daily. PAN administration increased urinary NO(2)(-) excretion by day 2, and levels remained elevated through day 7. AG prevented this PAN-induced increase in urinary NO(2)(-) excretion. Plasma nitrate (NO(3)(-)) and NO(2)(-) (NOx) concentrations were also increased in the PAN and PAN + AG groups. iNOS protein expression was not detected in either the glomeruli or the cortex at day 7. Proteinuria developed in PAN animals on day 4 and increased steadily through day 7. PAN + AG animals showed a pattern similar to that of the PAN group. These results indicated that in contrast to models of proliferative glomerulonephritis, NO formation during PAN-induced nephrotic syndrome is increased but does not participate in the development of glomerular injury as measured by proteinuria.

Tissue distribution of alpha-tocopherol in nephrotic rats

1. Reactive oxygen species are involved in the pathogenesis of puromycin aminonucleoside (PAN) nephrosis and alpha-tocopherol is one of the major anti-oxidants in the body. 2. In the present study, we measured the levels of alpha-tocopherol by high-performance liquid chromatography in the plasma and in nine tissues of control and nephrotic rats obtained 10 days after either 0.9% saline solution or PAN injection, respectively. 3. In nephrotic rats, alpha-tocopherol levels increased four-fold in plasma; however, the molar ratio of alpha-tocopherol/ cholesterol remained unchanged, suggesting that the increase in alpha-tocopherol content was attributable to an increase in plasma lipid concentration. 4. In nephrotic rats, the alpha-tocopherol/cholesterol ratio increased 1.33-fold in adrenal glands and 1.34-fold in the testis, but remained unchanged in heart, spleen, liver, kidney lung, brain and muscle. 5. These data suggest that, in PAN nephrotic rats, there are alterations in the distribution of alpha-tocopherol and there is no deficiency of alpha-tocopherol in plasma or tissues.

Antioxidants in the prevention of renal disease

In view of the role of oxidative processes in inflicting damage that leads to glomerulosclerosis and renal medullary interstitial fibrosis, more attention could be paid to the use of antioxidant food constituents and the usage of drugs with recognized antioxidant potential. In any case atherosclerosis is an important component of chronic renal diseases. There is a wide choice of foods and drugs that could confer benefit. Supplementation with vitamins E and C, use of soy protein diets and drinking green tea could be sufficient to confer remarkable improvements.

Effect of the in vivo catalase inhibition on aminonucleoside nephrosis

Reactive oxygen species have been involved in the pathophysiology of puromycin aminonucleoside (PAN)-nephrosis. The role of H2O2 in these rats may be studied modulating the amount or activity of catalase, which breakdowns H2O2 to water and oxygen. To explore the role of H2O2 in this experimental model, we studied the effect of the in vivo catalase inhibiton with 3-amino-1,2,4-triazole (ATZ) on the course of PAN-nephrosis. Four groups of rats were studied: control rats (CT group), PAN-injected rats (PAN group), ATZ-injected rats (ATZ group), and ATZ- and PAN-injected rats (ATZPAN group). Rats were placed in metabolic cages to collect 24 h urine along the study, ATZ (1 g/kg) was given 24 h before PAN injection (75 mg/kg), and the proteinuria was measured on days 0, 2, 4, 6, 8, and 10. Proteinuria started before (day 4) and was significantly higher on days 6, 8, and 10 in the ATZPAN group than in the PAN group. On day 10, hypercholesterolemia was significantly higher in the ATZPAN group than in the PAN group. These data indicate that the in vivo catalase inhibition magnifies PAN-nephrosis, suggesting that H2O2 is produced in vivo and involved in the renal damage in this experimental disease.

The concept of glomerular self-defense

The balance between local offense factors and defense machinery determines the fate of tissue injury: progression or resolution. In glomerular research, the most interest has been on the offensive side, for example, the roles of leukocytes, platelets, complement, cytokines, eicosanoids, and oxygen radical intermediates. There has been little focus on the defensive side, which is responsible for the attenuation and resolution of disease. The aim of this review is to address possible mechanisms of local defense that may be exerted during glomerular injury. Cytokine inhibitors, proteinase inhibitors, complement regulatory proteins, anti-inflammatory cytokines, anti-inflammatory eicosanoids, antithrombotic molecules, and extracellular matrix proteins can participate in the extracellular and/or cell surface defense. Heat shock proteins, antioxidants, protein phosphatases, and cyclin kinase inhibitors may contribute to the intracellular defense. This article outlines how the glomerulus, when faced with injurious cells or exposed to pathogenic mediators, defends itself via the intrinsic machinery that is brought into play in resident glomerular cells.

Dipyridamole and dilazep suppress oxygen radicals in puromycin aminonucleoside nephrosis rats

BACKGROUND

Reactive oxygen species (ROS) are involved in the pathophysiology of puromycin aminonucleoside (PAN) nephrosis. To elucidate further the role of radicals in PAN nephrosis and the to determine the particular radical species scavenged by dipyridamole (DPM) and dilazep (DZ), we applied chemiluminescence and electron spin resonance (ESR) techniques.

METHODS

Chemiluminescence of glomeruli, which were isolated on day 7 from rats injected with 100 mg kg-1 PAN, was measured with or without scavengers. The inhibitory effects of DPM and DZ on hydroxyl radical adduct formation in the Fenton's reaction were evaluated using ESR.

RESULTS

Chemiluminescence was greater in glomeruli from rats with PAN nephrosis than in the the glomeruli of control rats. This increase was suppressed by superoxide dismutase, catalase, dimethylthiourea and also by DPM and DZ. ESR indicated that DPM and DZ inhibited hydroxyl radical adduct formation with a second-order rate constant of 2.9 x 10(10) and 1.6 x 10(10) (mol L(-1) s(-1) respectively, similar to that of dimethylthiourea.

CONCLUSION

DPM and DZ scavenge hydroxyl radicals, thereby alleviating PAN nephrosis.

An adenosine deaminase inhibitor prevents puromycin aminonucleoside nephrotoxicity

Puromycin aminonucleoside (PAN) toxicity was totally inhibited in the rat in vivo and in cultured glomerular epithelial cells (GECs) in vitro using the adenosine deaminase (ADA) inhibitor, 2'-deoxycoformycin (DCF). DCF completely inhibited ADA activity in glomeruli and protected against the development of PAN nephrosis; the 24-h urinary protein excretion of treated rats compared with controls (PAN rats) 9 days after PAN injection was 16 +/- 2 mg and 524 +/- 55 mg, respectively (p < .01). Morphological examination also demonstrated that the glomerular epithelial cells were protected against PAN-induced damage. Furthermore, when DCF was added to the first passage of GECs simultaneously with PAN, the adenosine triphosphate contents of remnant GECs on culture substrata increased in a dose-dependent manner, and PA toxicity was completely inhibited by 10(-4) M DCF. The order of ADA activity in glomeruli from various species was as follows: rat > monkey > guinea pig > dog > rabbit > mouse. High activity of ADA in the glomerulus was limited to species in which PAN induced nephrosis. Additionally, DCF increased glomerular cyclic AMP contents, resulting from enhanced adenosine accumulation in the pericellular space. These results indicate that the pathogenesis of PAN toxicity is closely related to adenosine metabolism and that ADA plays a key role in this model. Furthermore, we speculate that DCF contributes to the inhibition of reactive oxygen metabolites by decreasing the substrate of xanthine oxidase and/or increasing pericellular adenosine accumulation.

Effect of dietary antioxidants on puromycin aminonucleoside nephrotic syndrome

Several studies indicate the pathophysiological importance of reactive oxygen species in rats with nephrotic syndrome induced by puromycin aminonucleoside, an experimental model of the human minimal change disease. The role of reactive oxygen species in these rats was further evaluated, examining the effect of dietary deficiency and supplementation of antioxidants (vitamin E and selenium) on biochemical and renal ultrastructural alterations induced by puromycin aminonucleoside. Male Wistar rats, weaned at 3 weeks, were placed on diets normal, deficient or supplemented in vitamin E and selenium for 4 weeks. At the end of this period, rats were divided in two groups: control (sacrificed without any further treatment) and nephrotic (injected with puromycin aminonucleoside and sacrificed 7 and 22 days later). In control rats, the dietary deficiency or supplementation of antioxidants resulted in no significative differences in renal function, proteinuria or kidney ultrastructure. However, kidney lipoperoxidation, kidney glutathione peroxidase activity and circulating levels of vitamin E changed according to the amount of antioxidants in the diet. Seven days after the injection of puromycin aminonucleoside, rats fed normal, deficient or supplemented diets, developed nephrotic syndrome. However, proteinuria, hypoproteinemia, renal dysfunction and ultrastructural alterations were higher in rats fed a deficient diet. In contrast, proteinuria and kidney ultrastructural alterations were lower in rats fed a supplemented diet. Kidney lipoperoxidation and glutathione peroxidase activity increased on day 7 in rats fed a normal or a deficient diet, but not in rats fed a supplemented diet. This study shows that nephrotic syndrome induced by puromycin aminonucleoside in rats is modified by dietary antioxidants (vitamin E and selenium). Dietary supplementation ameliorates it and dietary deficiency exacerbates it.(ABSTRACT TRUNCATED AT 250 WORDS)

Increased glomerular and urinary malondialdehyde in puromycin aminonucleoside-induced proteinuria in rats

Puromycin aminonucleoside (PAN)-induced proteinuria in rats may be mediated by reactive oxygen metabolites (ROM), which are injurious to several cell components including membrane lipids. Increased malondialdehyde (MDA) production is indicative of lipid peroxidation. We examined if MDA content of glomeruli and its urinary excretion were increased in rats administered PAN. Of three groups of 8 Sprague-Dawley rats each, group 1 served a control, group 2 animals received a single intravenous injection of PAN (5 mg/100 g body weight) and group 3 animals PAN with intraperitoneal injections of dimethylthiourea (DMTU), a free radical scavenger of oxidants such as hydroxyl radicals, for 4 days. The rats were sacrificed on day 8 after PAN injection. Increasing proteinuria, starting on day 4, developed in animals in group 2 but not in the others. The glomerular MDA (nmol/mg protein) in group 2 animals was 2.93 +/- 1.91, significantly higher than 0.87 +/- 0.63 and 1.26 +/- 0.76 in groups 1 and 3, respectively. Urinary levels of MDA markedly increased in group 2 rats on day 3 and remained high thereafter, but no such increase occurred in the control animals and those administered PAN with DMTU; the latter was thus protective against PAN toxicity. Our observations support the view that ROM are involved in PAN-induced glomerular injury and that increased urinary MDA excretion can be a marker of ROM-mediated lipid peroxidation.

Antioxidants attenuate endotoxin-induced acute renal failure in rats

Acute kidney dysfunction, manifested by a reduction in renal blood flow and in the glomerular filtration rate, is a common finding in septic shock. The pathogenetic mechanisms responsible for the renal dysfunction observed in the endotoxemic murine model are not completely understood. In this study, an attempt was made to halt the progressive renal dysfunction in the rats by administration of the antioxidants dimethylthiourea (DMTU) (50 mg/100 g) and superoxide dismutase (SOD) (0.4 mg/100 g) before endotoxin infusion (0.5 mg/100 g), or by inducing endotoxin tolerance. Renal function, assessed by creatinine, inulin, and p-aminohippuric acid clearance, nicotinamide adenine dinucleotide, and electrolyte reabsorption, was measured 4 hours after the endotoxin infusion. Renal function declined in all rats throughout the study period. However, the reduction in renal function was markedly slower in endotoxemic rats administered DMTU and SOD compared with untreated rats. Similar results were found following induction of endotoxin tolerance. These data suggest that DMTU, SOD, and endotoxin tolerance may be potentially beneficial in halting progressive renal damage associated with endotoxemia.

Polymorphonuclear leukocytes increase glomerular albumin permeability via hypohalous acid

Acute glomerulonephritis is characterized by the presence of neutrophils within glomeruli and the generation of reactive oxygen species (ROS) by activated polymorphonuclear leukocytes (PMNs). Hydrogen peroxide (H2O2) and other ROS including hypothalous acids have been implicated in PMN mediated injury. To determine the role of specific ROS in PMN mediated glomerular injury, isolated rat glomeruli were incubated for 30 minutes at 37 degrees C with H2O2, with H2O2 and myeloperoxidase, or with activated PMNs. Scavengers of ROS were included in some experiments. PMNs were harvested from rat peritoneal cavity and activated with phorbol myristate acetate (PMA). Glomerular albumin permeability (Palbumin) was calculated from the volume response to an oncotic gradient. Palbumin of glomeruli incubated with H2O2 (10(-3) or 10(-1) M) was not increased, while Palbumin after incubation with H2O2 and MPO was markedly increased (0.94 +/- 0.004). Palbumin after incubation with PMA, or with non-activated PMNs was not different from that of control glomeruli, Palbumin of the glomeruli incubated with activated PMNs increased (0.85 +/- 0.01, P < 0.001). This increase in Palbumin was inhibited by superoxide dismutase, catalase, or taurine (Palbumin = 0.035 +/- 0.06, -0.39 +/- 0.10, 0.028 +/- 0.06, respectively) and ameliorated by sodium azide (Palbumin = 0.21 +/- 0.03). In contrast, dimethyl sulfoxide did not prevent the increase in Palbumin (Palbumin = 0.92 +/- 0.01). Our results show that the hypohalous acid derived from that of H2O2-MPO-halide system is capable of increasing Palbumin. We conclude that hypohalous acid may be the primary mediator of the immediate increase in glomerular protein permeability induced by PMNs.

Tissue lipoperoxidation and glutathione peroxidase activity in puromycin aminonucleoside injected rats

1. Lipoperoxidation (LPx) and glutathione peroxidase (GPx) activity were measured in kidney, liver, heart, lung, brain and testis from control and puromycin aminonucleoside (PAN) injected rats on days 1-6, 8, 10, 16 and 22 after vehicle or PAN injection. 2. PAN-injected rats developed proteinuria on day 3. 3. In PAN-injected rats: (a) LPx increased in kidney, liver, lung, brain and testis before day 3 and in heart on day 3; (b) GPx activity increased in kidney, liver, heart, lung and testis and diminished in brain on day 3 or after.

Reactive oxygen species in puromycin aminonucleoside nephrosis: in vitro studies

We examined the role of reactive oxygen species (ROS) in puromycin aminonucleoside (PAN)-induced changes to glomerular epithelial cells (GECs) in vitro. Levels of superoxide anion (O2.-), hydrogen peroxide (H2O2) and hydroxyl radical (HO.) were measured in rat kidney-slice cultures containing PAN with or without antioxidants (allopurinol, probucol and alpha-tocopherol/ascorbic acid). GEC morphology was assessed after three days of culture using transmission (TEM) and scanning (SEM) electron microscopy. The effects of hypoxanthine on GEC ultrastructure was also assessed. O2.-, H2O2 and HO. were generated when PAN was added to kidney-slice cultures in Medium 199. TEM morphometry revealed that incubation with PAN (100 micrograms/ml) significantly (P < 0.05 at least) retarded the loss of GEC foot processes normally seen in vitro. When the hydrophobic antioxidants probucol or alpha-tocopherol/ascorbic acid, which scavenged/inhibited generation of O2.-, H2O2 and HO., were added to cultures containing PAN, the effect of PAN on foot processes was abolished. The TEM appearance of GECs now resembled that seen in control cultures. On the other hand, SEM revealed that probucol and alpha-tocopherol/ascorbic acid provided no protection against the changes induced by PAN in GEC cell bodies or major processes. Allopurinol provided no protection against the changes induced by PAN in GEC cell bodies, major processes or foot processes. The addition of hypoxanthine to kidney-slice cultures did not result in the generation of O2.-, H2O2 or HO., or alter GEC ultrastructure. These findings indicate that ROS play a role in PAN-induced alterations to GEC foot process architecture in vitro. However, the xanthine oxidase pathway does not appear to play a major role in generating ROS from PAN in vitro.

Fc alpha R expression on polymorphonuclear leukocyte and superoxide generation in IgA nephropathy

Superoxide (O2-) production and Fc alpha R antigen expression of circulating polymorphonuclear leukocytes (PMNL) isolated from patients with IgA nephropathy (IgAN) and non-IgA mesangial proliferative glomerulonephritis (PGN) and healthy volunteers were investigated to establish their biological importance in the immunopathogenesis of mesangial proliferative glomerulonephritis. PMNL from both patient groups showed increased O2- production when stimulated with N-formyl methionyl leucyl phenylalanine (FMLP) and phorbol myristate acetate (PMA). The increased O2- generation demonstrated a positive correlation with the degree of proteinuria. Aggregated IgA caused enhanced O2- production only in patients with IgAN who also showed a significant correlation with proteinuria. Increased expression of Fc alpha R on circulating PMNL was observed in IgAN patients as determined by flow cytometric analysis. The amount of Fc alpha R on PMNL was positively correlated with O2- generation triggered with IgA aggregates. These results suggest that: 1. Circulating PMNL may potentially be participating in the pathogenesis of glomerular injury in mesangial proliferative glomerulonephritis, and 2. IgA aggregates/immune complexes may contribute to the immunopathogenesis of IgAN through augmenting the Fc alpha receptor-mediated generation of superoxide anion.

Inhibition of active oxygen generation by dipyridamole in human polymorphonuclear leukocytes

The effect of dipyridamole on active oxygen generation by human polymorphonuclear leukocytes (PMN) was investigated. Dipyridamole inhibited the production of oxidative metabolites from human PMN stimulated by opsonized zymosan and formyl-methionyl-leucyl-phenylalanine dose and time dependently. To determine whether dipyridamole directly inhibits the production of oxygen metabolites by human PMN, human PMN were preincubated with dipyridamole washed prior to stimulation. Dipyridamole was found to directly inhibit human PMN from generated active oxygen metabolites at therapeutic concentrations. Dipyridamole may possibly be a potential scavenger of active oxygen metabolites since it inhibited active oxygen metabolite production from human PMN very rapidly. Dipyridamole was also found to directly affect the scavenging of active oxygen metabolites generated by opsonized zymosan-stimulated human PMN at therapeutic concentrations. This action of dipyridamole was also noted to be exerted against hydroxyl radicals and superoxide anions produced biochemically by an electron spin resonance spectrometer. It thus follows that dipyridamole may inhibit human PMN active oxygen metabolite generation and affect directly the scavenging of active oxygen metabolites at therapeutic concentrations.

The role of reactive oxygen species in animal models of glomerular disease

Reactive oxygen metabolites have been demonstrated to play a pathobiologic role in a number of experimental models of both immune and nonimmune glomerular injury. Using scavenging substances, enzyme inhibitors, and transition metal chelators, all of the major reactive oxygen metabolites have been implicated in glomerular injury. In addition, in neutrophil-dependent models, interaction between neutrophil-derived myeloperoxidase and halide anions has been shown to be involved in glomerular damage, as well as halogenation of the glomerular basement membrane (GBM). Finally, recent attention has focused on the role of cytokines (perhaps elaborated by infiltrating monocytes/macrophages) in stimulating mesangial cells (MC) to produce reactive oxygen species. Theoretically, this pathobiologic sequence could further enhance an inflammatory state within the glomerular tuft and enable the propagation of initial glomerular injury, which may be associated with an increase in monocyte infiltration into the mesangium, to glomerulosclerosis in experimental models that manifest this transition.

Oxidative metabolism of polymorphonuclear leukocytes (PMN) in patients with IgA nephropathy

The production of hydrogen peroxide (H2O2) by neutrophilic polymorphonuclear leukocytes (PMN) after stimulation and the infiltration of PMN in glomeruli were determined in 20 patients with primary IgA nephropathy. The H2O2 production of PMN after the stimulation was measured with a spectrophotometer using horseradish peroxidase as substrate. The results were as follows: 1) when PMN were pretreated with cytochalasin B, H2O2 production after stimulation with heat-aggregated IgG (IgG) or serum-treated zymosan (STZ) was significantly higher in patients with IgA nephropathy than in controls, and 2) there was an increased amount of PMN localized in glomeruli in patients with IgA nephropathy using immunofluorescence of monoclonal anti-PMN antibody. It appeared that the increased renal infiltration of PMN which have a high potential for production of reactive oxygen species might induce the glomerular injuries in patients with IgA nephropathy.

Roles of active oxygen species in glomerular epithelial cell injury in vitro caused by puromycin aminonucleoside

The mechanism of puromycin aminonucleoside (PAN)-induced nephrosis has not yet been well defined. In the present study, we examined the protective effect of active oxygen scavengers on the PAN-induced injury of cultured rat glomerular epithelial cells (GECs) and the generation of active oxygen species in PAN-treated GECs. When exposed to PAN (greater than or equal to 25 micrograms/ml), cellular damage occurred in a time- and dose-dependent manner as evaluated by both the LDH release and MTT colorimetric assays. Concomitant addition of either the hydrogen peroxide (H2O2) scavenger, catalase, or the iron chelating agent, deferoxamine, to the culture medium caused a striking reduction of cellular injury. This suggested a role for H2O2 and for hydroxyl radicals (OH.) generated via the iron-catalyzed breakdown of H2O2 in PAN nephrosis. Using the scopoletin fluorescence assay, the release of H2O2 into the culture medium by GECs exposed to PAN (greater than or equal to 50 micrograms/ml) was shown to increase dose-dependently (greater than or equal to 57 +/- 11 pmol/4.4 x 10(6) cells per h, P less than 0.01) as compared with control cells (14 +/- 2 pmol/4.4 x 10(6) cells per h). These results strongly suggested that active oxygen species, especially H2O2 and OH., might play an important role in PAN-induced GEC injury in vitro as well as in vivo.

Complement activation retards resolution of acute ischemic renal failure in the rat

We investigated the role of complement activation on the resolution of acute ischemic renal failure in the rat. Acute renal failure was induced by clamping of the renal arteries of Sprague-Dawley rats for 45 minutes (Day 0). On subsequent days, groups of rats with acute renal failure were exposed to daily zymosan infusion (an activator of the complement system), or to blood incubated with cuprophane (CUP) or polyacrylonitrile (PAN) dialysis membranes. We serially measured the change in BUN daily, glomerular filtration rate and 24-hour proteinuria on Day 3 and Day 5 following ischemia. On Day 6, the animals were sacrificed and their kidneys examined histologically. Zymosan and cuprophane exposed rats had a significant delay in the recovery of renal failure, reduced glomerular filtration rate, and histologically had more neutrophil infiltration than control or PAN exposed animals. To investigate the potential pathophysiology of these observations, we assessed the response of zymosan-exposed rats to infusion of deferoxamine (DFO), a potent inhibitor of hydroxyl radical formation (OH.). Infusion of DFO prior to zymosan significantly improved recovery of renal function. We also measured urinary thromboxane B2 levels in these groups of rats. While the groups of rats exposed to zymosan had the highest levels of thromboxane B2, these levels were not different between the groups exposed to zymosan alone, or to zymosan and DFO. These observations suggest a role for hydroxyl radicals in the prolongation of renal failure in this model. Taken together, these findings may have implications for the dialytic intervention in patients with acute renal failure.

Effect of diabetes mellitus induced by streptozotocin on renal superoxide dismutases in the rat. A radioimmunoassay and immunohistochemical study

Two forms of superoxide dismutase, CuZn-SOD and MnSOD, have been investigated in the kidneys of streptozotocin-induced diabetic rats using both radio-immunoassay and immunoenzyme staining. The rats were killed 2, 8 and 12 weeks after the induction of diabetes mellitus and the kidneys excised. Two weeks after the induction of diabetes, the kidneys were hypertrophied because of the proliferation of renal tubular epithelium. However, the total CuZnSOD content of the kidneys did not increase and, because of the epithelial proliferation, the CuZnSOD concentration in each proximal tubular cell was decreased. Armanni-Ebstein lesions were found in the distal tubules 8 and 12 weeks after the induction of diabetes. The cells in these lesions were intensely stained for CuZnSOD, suggesting an adaptive response to the enhanced oxidative stress. The MnSOD staining in the thick ascending limbs of Henle's loops was enhanced in the diabetic kidneys, while that in the cortical tubules was unaltered. MnSOD was assumed to increase in response to hypermetabolism associated with the proliferation of renal tubules. This was most marked in the cells which were rich in mitochondria, again suggesting an adaptive response to enhanced oxidative stress induced by diabetes mellitus. The glomeruli of both the diabetic and control groups were not stained for SODs, and no significant microscopic change was found even 12 weeks after the induction of diabetes mellitus.

Reactive oxygen molecules, oxidant injury and renal disease

Oxidant injury has been implicated in the pathogenesis of inflammatory, metabolic and toxic insults, in ischemic-reperfusion injury, and in carcinogenesis, aging and atherosclerosis. Oxidant injury is initiated by free radicals and reactive oxygen molecules which are generated by activated neutrophils, monocytes, and mesangial cells, during normal and abnormal metabolic processes, and from the metabolism of exogenous drugs and toxins. When cells and organs are exposed to oxidant stress, several different antioxidant defense mechanisms operate to prevent or limit oxidant injury. When antioxidant defense mechanisms are decreased, or when the generation of reactive oxygen molecules is increased, oxidant injury results from the shift in the oxidant/antioxidant balance. Oxidant-induced alterations of proteins, membranes, DNA, and basement membranes leads to cell and organ dysfunction. Several renal diseases including glomerulonephritis, vasculitis, toxic nephropathies, pyelonephritis, acute renal failure, and others are likely to be mediated at least in part by oxidant injury. In the future, mechanisms to decrease the generation of reactive oxygen molecules and/or antioxidant therapy may develop into new avenues of therapeutic intervention.

Glucocorticoid activates glomerular antioxidant enzymes and protects glomeruli from oxidant injuries

We examined the effect of glucocorticoid on intrinsic glomerular antioxidant enzyme (AOE) activities. Munich-Wistar rats were treated with daily i.p. injection of vehicle or methylprednisolone [MP, 15 mg/kg body wt, (MP15)] either for three days or nine days. Glomeruli isolated from rats given MP15 had significantly higher activities of total (T-) and manganese (Mn-) superoxide dismutase (SOD), glutathione peroxidase (GSH-Px) and catalase than vehicle-treated rats (P less than 0.05). MP15-treated rats were subjected to intrarenal arterial infusion of hydrogen peroxide (35 mumol over 1 hr). Values for urinary protein excretion rate (UprV) after hydrogen peroxide infusion were markedly lower in rats pretreated with MP15 for both three days and nine days than in untreated rats (109 +/- 18 and 55 +/- 24 vs. 416 +/- 73 micrograms/min, respectively, both P less than 0.005). To test whether the same therapeutic intervention attenuates reactive oxygen species (ROS)-mediated glomerular injury in another model, rats given a single i.v. dose of puromycin aminonucleoside (PAN) (50 mg/kg body wt) were treated with daily i.p. injection of vehicle or MP15. Two days after PAN administration, when compared to vehicle-treated controls, PAN rats given MP15 had significantly higher activities of Mn-SOD, GSH-Px and catalase. After eight days of PAN injection, T- and Mn-SOD activities were, likewise, significantly higher in MP15- than vehicle-treated PAN rats. PAN rats given MP15 also had substantially less proteinuria, compared to PAN rats given vehicle alone, UprV averaging 32.3 +/- 9.4 versus 159.0 +/- 13.8 mg/24 hr (P less than 0.05). Elevated glomerular malondialdehyde (MDA) level characteristic of PAN rats was absent in rats treated with MP15. Moreover, epithelial foot process fusion and cell vacuolization seen in vehicle-treated PAN rats were markedly attenuated in MP15-treated PAN rats. These data indicate that the mechanism for therapeutic effect of glucocorticoids on ROS-mediated renal injuries includes an enhancement of endogenous glomerular AOE activities, which attenuates lipid peroxidation of glomerular tissue.

Puromycin aminonucleoside metabolism by glomeruli and glomerular epithelial cells in vitro

Two puromycin aminonucleoside (PAN) excretion products were purified by HPLC from urine of PAN-treated rats and characterized by nuclear magnetic resonance as N6-dimethyl-3'amino-3'deoxyadenosine (DA-Ado) and N6-methyl-3'amino-3'deoxyadenosine (MA-Ado), respectively, the former corresponding to unmodified PAN. DA-Ado was not a substrate for adenosine deaminase (ADA), purine nucleoside phosphorylase (PNP) or xanthine oxidase (XO), while MA-Ado was consecutively converted into hypoxanthine by a mixture of ADA and PNP. A different rate of transformation of DA-Ado and MA-Ado into hypoxanthine by isolated glomeruli was observed and was higher for the monomethylated analogue by a factor of 3 (79% vs. 21%); this was ascribed to the rate-limiting level of a demethylase activity acting on DA-Ado. Furthermore, DA-Ado was not transformed by glomerular epithelial cells in culture, while a little amount of MA-Ado was converted into hypoxanthine after six hours of incubation. In spite of this different metabolic behavior, the same order of cytotoxicity on glomerular epithelial cells in culture was observed for MA-Ado, DA-Ado and commercial PAN. All these molecules induced a dose response inhibition of [3H]thymidine incorporation into DNA after exposure for two hours and a marked alteration of cell viability which was not inhibited by free radical scavengers and deferoxamine. This study provides the first evidence for a glomerular metabolism of PAN and its urinary metabolite MA-Ado involving their transformation via the purine cycle enzymes.(ABSTRACT TRUNCATED AT 250 WORDS)

Reactive oxygen metabolites in endotoxin-induced acute renal failure in rats

Based on recent reports that reactive oxygen metabolites may play a role in endotoxin-induced injury in other tissues, we postulated that reactive oxygen metabolites may be important mediators of endotoxin-induced acute renal failure. Superoxide dismutase, a scavenger of superoxide, or catalase, which destroys hydrogen peroxide, did not protect against endotoxin-induced renal failure. Similarly, neither the hydroxyl radical scavenger dimethylthiourea nor the iron chelator deferoxamine (which presumably would act by preventing the generation of hydroxyl radical via the iron-catalyzed Haber-Weiss reaction) prevented the endotoxin-induced fall in renal function. In separate experiments, we found no increase in renal cortical lipid peroxidation (a marker of reactive oxygen metabolite-mediated tissue injury) in endotoxin-treated rats, providing further evidence against a role for reactive oxygen metabolites in endotoxin-induced renal injury. Finally, using the aminotriazole-induced inhibition of catalase (a measure of in vivo changes in the hydrogen peroxide generation) we found no evidence of enhanced hydrogen peroxide generation in the renal cortex in endotoxin-treated rats. Taken together, the data from these three separate experimental approaches suggest that reactive oxygen metabolites are not important mediators of endotoxin-induced acute renal failure.

Intraglomerular platelet aggregation and experimental glomerulonephritis

Oxygen free radical production inhibits ADPase-mediated antithrombotic action. Different forms of experimental glomerulonephritis (GN) are characterized by early glomerular influx of inflammatory cells and thrombus formation. The causal relationship of these inflammatory events is obscure. Previous studies have shown that glomerular ADPase in the rat kidney may function as a potent antithrombotic principle, whereas this enzyme is highly sensitive for oxygen free radicals. To study whether O2- producing inflammatory cells are able to induce intraglomerular thrombosis via impairment of ADPase, we investigated influx of inflammatory cells in relation to glomerular ADPase activity and platelet aggregation in three models of GN. In two of these models (anti-Thy1 and anti-GBM GN) influx of neutrophils and thrombus formation occurs, whereas in anti-FX1A nephritis this aspect of the inflammatory phase is not present. The results show a relationship between influx of oxygen free radical-producing cells, reduction of glomerular ADPase activity and increased platelet aggregation. Moreover, it is shown that impairment of glomerular ADPase and increased platelet aggregation in anti-Thy1 and anti-GBM GN could be reduced by treatment with superoxide dismutase and catalase. The demonstration that activated neutrophils perfused ex vivo in the rat kidney can directly affect glomerular ADPase and antithrombotic potential in an O2- dependent manner, further supports the proposed sequence of events; oxygen free radicals produced by activated neutrophils reduce glomerular ADPase activity, leading to facilitation of thrombus formation.

Production of reactive oxygen species by tubular epithelial cells in culture

Reactive oxygen species (ROS) have been implicated in the pathogenesis of toxic, ischemic and immunologically-mediated renal injury. Although substantial evidence exists for the production of ROS by glomerular cells, little is known about production of these reactive oxygen metabolites by renal tubular cells. We examined the ability of cultured cells from different segments of the rabbit nephron to elaborate ROS. Under basal conditions, cells of the proximal tubule, cortical collecting duct, and papillary collecting duct produced superoxide anion and hydrogen peroxide. Exposure to opsonized zymosan or heat-aggregated gamma globulin significantly increased ROS production by all three tubular cell types. The production of superoxide anion and hydrogen peroxide was time dependent and increased with increasing concentrations of the stimulating factors. These experiments indicate that renal tubular cells have the potential to participate in renal injury via elaboration of highly-reactive oxygen metabolites.

Different types of segmental sclerosing glomerular lesions in six experimental models of proteinuria

From 133 to 615 glomeruli were examined in sections of kidneys from each of 60 animals, representing six rodent models of proteinuria. Particular attention was paid to the position of segmental lesions. Lewis rats given sheep anti-rat glomerular basement membrane antibodies had lesions almost exclusively at the glomerulo-tubular junction. Wistar rats on a diet of 24 per cent casein or with subtotal nephrectomy and a diet of 24 per cent soya had lesions mainly at the hilum. Wistar rats given bovine serum albumin had global lesions but virtually no segmental lesions. Wistar rats given puromycin aminonucleoside had lesions at the glomerulo-tubular junction and global mesangial abnormalities shortly after the treatment but later developed segmental lesions at all parts of the glomerulus. Untreated BUF/Mna rats had lesions at the glomerulo-tubular junction early in life but later had lesions at all parts of the glomerulus. Untreated NZB/NZW hybrid mice had various types of glomerulonephritis and also had lesions at the glomerulo-tubular junction. These findings showed that (1) segmental lesions at the glomerulo-tubular junction, or glomerular tip, occur in experimental animals, a fact not previously reported, and these tip changes are a common feature in several different models of proteinuria; (2) hilar segmental lesions are seen in conditions with hyperfiltration of protein; and (3) segmental lesions at various parts of the glomerulus are seen in some models of proteinuria and probably indicate late effects of random toxic damage to the glomerulus. Thus, there are at least three different types of segmental glomerular lesions in experimental animals--tip, hilar, and random--with different morphology and pathogenesis. It is likely that these findings can be extended to human renal diseases with segmental glomerular lesions. This will help to clarify the controversial and unsatisfactory term focal segmental glomerulosclerosis.