Bruno Poirier1, Marilyne Lannaud-Bournoville1, Marc Conti2, Raymond Bazin3, .... kits respectively (Boerhinger, Mannheim, kidney samples homogenized in 3 ...
Nephrol Dial Transplant (2000) 15: 467–476
Nephrology Dialysis Transplantation
Original Article
Oxidative stress occurs in absence of hyperglycaemia and inflammation in the onset of kidney lesions in normotensive obese rats Bruno Poirier1, Marilyne Lannaud-Bournoville1, Marc Conti2, Raymond Bazin3, Odile Michel1, Jean Barie´ty1, Jacques Chevalier1 and Isaac Myara1,4 1INSERM U 430, Broussais Hospital, and Claude Bernard Association, Paris, 2Laboratory of Biochemistry, Biceˆtre Hospital, 3INSERM U 465, Institut des Cordeliers, Paris and 4Laboratory of Applied Biochemistry, Faculty of Pharmaceutical and Biological Sciences, Chaˆtenay-Malabry, France
Abstract Background. Several factors favour the development of kidney lesions. We examined the role of oxidative stress in the onset of renal alterations that occur in Zucker obese (ZO) fa/fa rats. Methods. Kidney structure, biological data, glycation parameters, advanced glycation end products (AGE), thiobarbituric acid-reactive substances ( TBARS), circulating antibodies anti-malondialdehyde (MDA)modified low-density lipoprotein (LDL), antioxidant defenses (Cu/Zn and Mn superoxide dismutase (SOD), catalase, glutathione peroxidase (GPx) activities, glutathione level ), were determined in plasma and/or kidney of young and old ZO rats and lean ( ZL) Fa/fa littermates. Results. Renal lesions and functional decline appeared at 3 months in hyperlipidaemic, hyperinsulinaemic, normotensive ZO rats, independently of any macrophage-ED +-cell infiltration. At 6 months and there1 after, kidney lesions and functional impairment worsened while numerous ED +-cells invaded the 1 interstitium. At 3 and 9 months, TBARS level in the LDL/very low-density lipoprotein fraction and in the kidney was higher in ZO than in ZL rats. Anti-MDALDL antibodies were increased in ZO rats. At 3 months, renal activity of Cu/Zn SOD was higher, and activities of catalase and GPx lower in ZO than in ZL rats, leading to an accumulation of hydrogen peroxide (H O ). At 9 months, a decrease in Cu/Zn SOD 2 2 activity and an increase in glutathione level were observed. Blood glucose and glycated proteins, as well as AGE in kidney, remained similar in both ZL and ZO rats, whatever their age. Conclusion. These data suggest that oxidative stress triggers, at an early age, the onset of kidney lesions and functional impairment in ZO rats, in absence of hyperglycaemia, hypertension and inflammation.
Correspondence and offprint requests to: Dr Jacques Chevalier, Immunopathologie Re´nale et Vasculaire, INSERM U 430, Hoˆpital Broussais, 96 Rue Didot, F-75674 Paris CEDEX 14, France.
Keywords: glomerulosclerosis; hyperinsulinaemia; lipid peroxidation; obesity; Zucker rat
Introduction Several mechanisms may contribute to the onset and/or the progression of kidney lesions that occur in patients or animals suffering from metabolic disorders such as diabetes mellitus, obesity and/or hyperlipidaemia. Among them, lipid peroxidation and oxidative stress, high glucose levels and glycated products, have been frequently proposed. Lipid peroxidation is initiated when polyunsaturated fatty acids (PUFA), principally located in cellular membranes and lipoproteins, interact with reactive oxygen species (ROS ) which include hydrogen peroxide (H O ), hypochlorous acid (HOCl ), nitric oxide 2 2 (NO) and free radicals such as superoxide anion (O Ω−), hydroxyl (OHΩ), alkoxyl (ROΩ) and peroxyl 2 (ROOΩ) radicals. In tissues, antioxidant defense, mainly made up of glutathione and antioxidant enzymes, opposes the toxic actions of ROS. Three major enzymes are involved in this defense: the Cu/Zndependent and Mn-dependent superoxide dismutases (Cu/Zn-SOD, Mn-SOD), catalase and glutathione peroxidase (GPx). Thus, oxidative stress is present in case of an imbalance between ROS production and antioxidant defense. Such oxidative stress has been demonstrated in obese [1], hyperlipidaemic [2] and diabetic patients and animals [3]. An abnormal glucose metabolism appears to be another key factor in the progression of kidney disease. A high glucose environment could intervene either directly, in stimulating kidney cells as suggested by in vitro experiments, or through the process of glycation, a non-enzymatic reaction of glucose with proteins, which produces Amadori compounds and, eventually, advanced glycation end products (AGE). Glycated proteins or AGE can directly activate glomerular cells. AGE receptors having been shown on rat mesangial
© 2000 European Renal Association–European Dialysis and Transplant Association
468
cells. Glycated proteins can also favour lipid peroxidation: glucose and fructose lysine, the first Amadori rearrangement products, generate ROS in the presence of trace amounts of iron or copper which act as catalysts. Signal transduction by the AGE receptor appears to involve generation of free radicals [4]. Although hyperglycaemia, non-enzymatic glycation and oxidative stress could intervene in the development of kidney lesions seen in obesity-related disorders such as diabetic nephropathy, the respective roles of these metabolic factors in the genesis of early renal injury is not clearly established, especially as they are most often accompanied by an inflammatory process. Indeed, most of the data on glomerular sclerosis suggested that a release of cytokines by infiltrating inflammatory cells is the key factor which launches the abnormal accumulation of extracellular matrix components. However, recent observations using the Zucker rat model suggested that initial steps of the extracellular matrix remodelling occurred independently of an inflammatory process [5,6 ]. Thus, in order to distinguish between the factors involved in the genesis of renal structural changes from those driving their progression, we examined in Zucker obese ( ZO) rats, the relationship between metabolic factors, oxidative stress, inflammation process and the time-course of renal morpho-functional changes leading to glomerulosclerosis and interstitial fibrosis. The Zucker rat is a useful model of genetic obesity presenting an autosomal recessive mutation of the fa gene encoding the leptin receptor. This strain shows abnormal glucose tolerance and peripheral insulin resistance similar to patients with type II non-insulin dependent diabetes mellitus (NIDDM ). ZO rats express, already at weaning, pronounced hyperlipidaemia which worsens with age. They rapidly develop glomerulosclerosis, focal and segmental glomerular hyalinososis (FSGH ) and interstitial fibrosis whose gravity increased with age [7]. These lesions occur in the absence of hypertension [7] or renal haemodynamic modification [8]. Male Zucker lean (Fa/fa) ( ZL) littermates have normal serum lipids, glucose and insulin and normal renal structure and function [7], and serve as a useful internal control.
Subjects and methods Animals Male ZL (Fa/fa) and ZO ( fa/fa) rats (Dr Raymond Bazin, INSERM U 465 husbandry, Paris, France) were identified and selected at 4 weeks of age by visual examination of inguinal fat deposits. They were raised in standard husbandry conditions, fed regular laboratory chow ad libitum (M25, Extralabo, Provins, France) and had free access to water. For determination of biological parameters, fasting animals were housed individually in metabolic cages with free access to water. Twenty-four-hour urine samples were collected and a blood sample was obtained by orbital sinus punction into tubes containing heparin. At the time of sacrifice at 1, 3, 6 and 9 months, animals were anaesthetized with pentobarbital (i.p., 0.1 ml/100 g body weight) and the kidneys were
B. Poirier et al.
removed and weighed. At 3 and 9 months blood was also collected from the aorta into ethylene-diamine-tetraaceticacid (EDTA) containing Vacutainers (Becton-Dickinson, Meylan, France) for lipid peroxidation assays. Animal care complied with the Principles of Laboratory Animal Care formulated by the National Society for Medical Research and the Guide for the Care and Use of Laboratory Animals (National Institutes of Health publication 86-23, revised 1989, authorization 00577, 1989, Paris, France).
Kidney structure and immunohistochemistry For histologic examination, kidneys were immediately placed on ice after removal and transverse sections were made at hilum and directly frozen in liquid nitrogen for immunohistochemistry study or fixed in alcoholic Bouin’s solution, embedded in paraffin, sectioned (4 mm thick) and stained with Masson’s trichrome, for routine histology. Overall tubulointerstitial injury was defined as tubular dilation, degenerating tubular cells and cellular debris in the lumen, proteinasceous tubular casts, interstitial fibrosis and interstitial inflammatory cell infiltrates. It was graded on Masson’s trichrome stained sections on a scale of 0–4 (0=normal; 0.5=small focal areas of damage; 1=involvement of less than 10% of the cortex; 2=involvement of 10–25% of the cortex; 3=involvement of 25–75% of the cortex; 4=extensive damage involving more than 75% of the cortex), to form a semi-quantitative index of cortical damage. Stained sections of at least 100 glomeruli were evaluated for the presence of FSGH lesions. The percentage of glomeruli with FSGH was determined for each tissue specimen. The differentiation and activation of interstitial mononuclear cells was assessed by incubating frozen kidney sections from rats aged 1, 3, 6 or 9 months with a mouse monoclonal antibody specific for a monocyte/macrophage cytoplasmic marker (ED antibody, 1 Serotec, Oxford, UK ), diluted 151000 in Tris-buffered saline pH 7.4, containing 0.1% bovine serum albumin (BSA, Sigma Chemical, St Louis, MO, USA) for 60 min at room temperature. The density of lymphocytes in the interstitium and in glomeruli was also estimated in frozen kidney sections of 3-month-old lean and obese rats, using a mouse monoclonal antibody directed against T cell receptors (15100, 30 min, clone RT3, Serotec). Then, the sections were washed in Trisbuffered saline and incubated with rabbit anti-mouse immunoglobulin antibody (Dako Corporation, Carpinteria, CA, USA) and alkaline anti-phosphatase alkaline complexes (diluted 1575) (Dako). The enzyme was revealed with freshly prepared Fast Red Substrate System (Dako) containing 0.33 mg/ml levamisole (Sigma) to reduce the staining background. Sections were counterstained with haematoxylin. The number of positive cells either in the glomerulus or in squares of 1.105 mm2 distributed over the interstitium area was counted, with a minimum of 50 glomeruli or 13 squares surveyed per kidney section as defined by convergent analysis. AGE deposits were detected by immunofluorescence on deparaffinized sections of 9-month-old ZL and ZO rats, using a rabbit anti-AGE antibody generously provided by Dr Bakala (Cell Biology Laboratory, University Paris 7, Paris, France), diluted 1520 in phosphate buffer saline supplemented with 2% fish gelatin (Sigma), and revealed using a FITC-labelled goat anti-rabbit IgG (Cappel, Cachranville, PA, USA). For positive control, kidney sections of 30-monthold Wistar rats were used. Immunolabelled sections were observed under a Leica TCS SP confocal microscope (Leica, Heidelberg, Germany).
Oxidative stress in normoglycaemic obese Zucker rat
Biological parameters Conscious systemic blood pressure was measured by a tailcuff system ( Ugo Basile Apelex, Varese, Italia). Orbital sinus blood samples were centrifuged at 2500 r.p.m. for 10 min at 4°C and aliquots of plasma were frozen and stored at −20°C. Triglycerides and total cholesterol were determined by the enzymatic and colorimetric GPO-PAP and CHOD-PAP detection kits respectively (Boerhinger, Mannheim, Germany). Glycaemia and plasma and urine creatinine concentrations were measured on a Synchron CX7 Beckman analyser (Beckman, Fullerton, CA, USA). Plasma insulin was measured by radioimmunoassay (CIS, Gif sur Yvette, France) with a rat insulin standard (Novo, Copenhagen, Denmark). Proteinuria was determined using the Coomassie Protein Assay Reagent (Pierce, Rockford, IL, USA) with bovine serum albumin as standard.
Sample preparation for lipid peroxidation parameters Blood obtained through aortic stick was centrifuged at 4000 r.p.m. for 10 min at 4°C and aliquots of plasma for lipoprotein separation were frozen and stored at −80°C. Plasma, rather than serum was used in order to reduce the release of lipoperoxides during clotting and sucrose (6 g/l, final concentration), was added to plasma aliquots before freezing. Previous experiments indicated that sucrose was a good cryopreservative for low-density lipoprotein (LDL) and did not modify lipid peroxidation parameters. All samples were stored for less than 8 weeks. For lipoprotein separation, we chose to investigate the combined LDL/very-low-density lipoprotein ( VLDL) fraction since rats have a relatively small amount of LDL. KBr (Merck, Darmstadt, Germany) solution (d=1.063 g/ml ) (3–5 ml ) containing 1 mM EDTA was carefully added to 5–7 ml of plasma previously adjusted to a density of 1.063 g/ml with solid KBr in 10 ml centrifuge tubes (Beckman, Gagny, France). Samples were centrifuged for 18 h at 4°C in a Beckman L8 ultracentrifuge using a 70.1 Ti rotor at 45 000 r.p.m. The top fraction (LDL+VLDL) was removed and the remaining plasma was adjusted to a density of 1.21 g/ml with solid KBr and, after having filled the centrifuge tube with KBr (d=1.21 g/ml )/EDTA as above, centrifuged again for 18 h to obtain high-density lipoproteins (HDL) in the upper phase and the remaining non-lipoprotein fraction in the pellet. Isolated lipoprotein and non-lipoprotein fractions were dialysed for 24 h at 4°C against four changes of 10 mM Tris–HCl buffer, pH 7.4, containing 1 mM EDTA, and stored at 4°C in the dark for less than 30 days. Tris buffer scavenges reactive chemical species such as hydroxyl radical and EDTA prevents the artifactual elevation of lipid peroxidation products. Protein was measured according to a Peterson assay with bovine serum albumin as standard. For kidney samples, small fragments (about 250 mg) were rinsed with 0.15 M NaCl containing 1 mM EDTA, frozen and stored in liquid nitrogen for less than 3 weeks. Tissues were homogenized in ice-bath for 30 s with 20 mM Tris–HCl buffer pH 7.4 containing 1 mM EDTA (1 ml/100 mg tissue) using an Ultra Turax homogenizer (Janken Kunkel IkaWerk, Staufen, Germany). The homogenate was centrifuged at 2300 g for 15 min at 4°C and the supernatant was used for TBARS and antioxidant enzyme activities. Tissue protein was determined using the Coomassie protein assay.
Lipid peroxidation parameters Among indicators used to explore the lipid peroxidation process, thiobarbituric acid-reactive substances (TBARS)
469
assay is the most popular owing to its sensitivity, although it is of relatively low specificity. TBARS content was assessed fluorometrically using 0.1 ml whole plasma, 0.1 mg plasma subfraction protein or 0.3 ml supernatant of tissue homogenate. In order to minimize within-run variations, each sample was performed in triplicate. Samples from each category were assayed at the same time. Glutathione concentration was determined from frozen kidney samples homogenized in 3 volumes of 5% trichloroacetic acid (Merck), according to the technique of Ellman [9] with 5,5∞-dithiobis (2-nitrobenzoic acid ) (DTNB) (Sigma) as reagent. The activity of Cu/Zn- and Mn-SOD, catalase and GPx was measured as described by The´rond et al. [10]. Determination of antibodies against MDA-LDL was performed by an ELISA method in 96-well plates coated with MDA-modified human LDL, as previously described [11].
Glycated protein Two methods based on boronate affinity chemistry were used for measuring the glycated haemoglobin: fully automated assay with the Abbott IMx analyser (Abbott, Laboratories, Abbott Park, IL, USA) and column chromatography (Glycogel II Boronate affinity gel, Pierce Inc). The percentage of glycated haemoglobin was determined according to the manufacturer. The Roche (Neuilly, France) reagent Unimate FRA was used for the colorimetric quantitative determination of plasma glycated protein (fructosamine). Total glycated proteins in plasma were also separated from nonglycated proteins by affinity chromatography on the Glycogel columns used above.
Statistical analysis Results were expressed as mean±SEM. Statistical analysis was carried out using a two-way ANOVA analysis of variance with age and group (genotype) as factors, followed by Bonferroni-Dunn tests (Statview 5.0 software, Abaccus Concept Inc, Berkeley, CA). The overall age effect, group effect and interaction reached statistical significance if P
