HLA compatibility in organ transplantation

Nephrol Dial Transplant (2001) 16: 147–150

Nephrology Dialysis Transplantation

Brief Report

Altered antioxidant defence in a mouse adriamycin model of glomerulosclerosis An Deman1, Bart Ceyssens1, Marina Pauwels1, Jigang Zhang1, Katherina Vanden Houte2, Dierik Verbeelen3 and Christiane Van den Branden1 1Department of Human Anatomy, 2Division of Pathology and 3Department of Nephrology, Vrije Universiteit Brussel and Academic Hospital of the Vrije Universiteit Brussel, Brussels, Belgium

Abstract Background. Antioxidant enzyme status changes in experimental models of chronic renal disease with glomerulosclerosis. Most of the studies are performed in rats. We now investigate whether a mouse model with more rapid development of glomerulosclerosis is suitable for the study of radical-associated renal disease. Methods. Female BALB/c mice are injected intravenously with a single dose of adriamycin (10 mg/kg). The development of glomerular and interstitial injury is evaluated by means of renal function parameters and histology. Renal cortex activities of catalase, Cu/Zn and Mn superoxide dismutase and glutathione peroxidase are measured by enzymatic techniques, and their mRNA levels by Northern blot analysis. Results. The mice develop proteinuria and hypercholesterolaemia; glomerulosclerosis is present 20 days after adriamycin injection. Involvement of reactive oxygen intermediates in the disease process is supported by an increased cortex level of glutathione (1.77±0.13 vs 1.31±0.12 mmol/g kidney; P=0.021) and ferric iron deposition in the tubulointerstitial compartment. Glomerulosclerosis and tubulointerstitial lesions are accompanied by decreased cortex activities of catalase (0.19±0.01 vs 0.23±0.01 U/mg protein; P=0.024), glutathione peroxidase (0.28±0.01 vs 0.32±0.01 U/mg protein; P=0.049) and Mn superoxide dismutase (6.61±0.91 vs 9.25±0.99 U/mg protein, P=0.020). We find decreased cortex mRNA levels only for glutathione peroxidase. Conclusion. The fast development of glomerulosclerosis combined with an altered antioxidant status makes this mouse adriamycin model a suitable alternative for the slower rat models. Keywords: adriamycin; antioxidant enzymes; chronic renal failure; glomerulosclerosis; oxidative stress

Correspondence and offprint requests to: Dr Christiane Van den Branden, Human Anatomy, Vrije Universiteit Brussel, Laarbeeklaan 103, B-1090 Brussels, Belgium.

Introduction Progressive sclerosis of glomeruli and tubulointerstitial tissue are characteristics of evolutive renal diseases. These processes are accompanied by an increase of reactive oxygen intermediates (ROI ) and of changes in the antioxidant capacity of the cell [1]. The precise mechanism of these disturbances and the question of whether they play a major role in the pathogenesis of renal failure remain largely unknown. We previously reported changes of antioxidant enzyme (AOE) activities in chronic renal failure with glomerulosclerosis, using rat models. In both the remnant kidney as well as in adriamycin-induced renal disease, we documented a decrease of antioxidant enzymes and a beneficial effect of different treatments that influence AOE status [2–5]. A major disadvantage of rat models is that both the remnant kidney and the adriamycin model require several months before severe glomerulosclerosis develops. Recently, Chen et al. [6 ] showed that 18 days after the administration of adriamycin at a dose of 10 mg/kg body weight, glomerulosclerosis was present in BALB/c mice. In this study we investigate whether in their mouse adriamycin model with glomerulosclerosis, similar antioxidant defence system changes can be observed as in rat models.

Subjects and methods Experimental design Female BALB/c mice (Iffa Credo, Brussels, Belgium) with body weights of 17–19 g were injected intravenously with a single dose of adriamycin (doxorubicin (Pharmacia and Upjohn, Puurs, Belgium) diluted to 50% with 0.9% saline; 10 mg/kg; n=8). Control mice were injected with the same volume of saline (n=7). Intake of food was made easier by offering the animals a wet porridge of A04 meal ( UAR, Epinay, France) instead of pelleted chow. Mice were sacrificed on day 20. Blood samples were taken by transthoracic cardial puncture and the kidneys were rapidly removed and weighed. Transversal slices of renal tissue were fixed in 4% buffered formaldehyde at room temperature for 24 h and

© 2001 European Renal Association–European Dialysis and Transplant Association

148

embedded in paraffin for evaluation of sclerosis and ferric iron deposits. The remaining cortex of the same kidney was homogenized in appropriate buffers for determination of glutathione concentration and AOE activities. The second kidney cortex was snap frozen in liquid nitrogen and used for RNA extraction.

Clinical parameters Serum and urine urea, creatinine and total protein concentrations were analysed by the Kodak Ektachem method ( Kodak Eastman, Rochester, NY, USA) and serum cholesterol concentration was determined by enzymatic methods [7]; a semiquantitative measurement of proteinuria and a qualitative measurement of haematuria were performed daily using a Multistix strip (Bayer, Brussels, Belgium). Proteinuria was expressed as the protein:creatinine ratio.

Determination of renal cortex glutathione concentration The reduced form of glutathione (GSH ) was measured in renal cortex homogenates using the BIOTECH GSH-400 kit (OXIS International, Portland, OR, USA). Concentration was expressed as mmol GSH/g kidney cortex.

Determination of renal cortex AOE activities Cortex homogenates were prepared in 50 mM potassium phosphate buffer containing 0.1 mM EDTA and 1% Triton X-100, pH 7.8. Catalase (CAT ) activity was assayed by the method of Aebi [8]. Glutathione peroxidase (GPx) activity (selenium and non-selenium dependent) was determined by the method of Carmagnol et al. [9]. Total superoxide dismutase (SOD) and Mn SOD activities were measured by the method of Marklund and Marklund [10] and Cu/Zn SOD activity was calculated. The results are expressed in units/mg protein. Protein measurement was performed by the bicinchoninic acid method (PIERCE, Rockford, IL, USA).

A. Deman et al. Table 1. Clinical parameters in control and adriamycin-treated mice, 20 days after adriamycin treatment Parameter

Control

Adriamycin

serum urea (mg/100 ml ) serum creatinine (mg/100 ml ) serum total protein (g/100 ml ) serum cholesterol (mg/100 ml ) urine protein:creatinine ratio

22.86±1.60 0.215±0.006 5.34±0.10 60.14±2.20 1.50±0.50

21.00±1.51 0.236±0.016 5.51±0.07 133.44±15.70a 20.44±9.60a

Proteinuria is expressed as the ratio between urine total protein and creatinine concentration. Data are mean±SEM. aP