The experimental protocol was approved by the Animal Care Committee of the Catholic University of Korea. Male ICR mice (Taconic Anmed, Rockville, MD, USA) initially weighing 25g, were housed in cages in a temperature- and light- controlled environment, and allowed free access to a low salt diet (0.01% sodium, Teklad Premier, Madison, WI, USA) and tap water. Mice were randomized into two groups and treated daily for 1 (n = 12) or 4 weeks (n = 12) as follows: 1) Vehicle group (VH, n = 6): subcutaneous injection with olive oil (Sigma Diagnostics, St. Louis, MO, USA, 1mL/kg); 2) CsA group (n = 6): subcutaneous injection with CsA (Novartis Pharma, Basel, Switzerland, 30mg/kg). This dosage and method of CsA administration were chosen according to a previous report.15,16

After the treatment was started, mice were pair-fed and daily body weight (BW) was monitored. Whole blood CsA levels were measured by a monoclonal radioimmunoassay (Incstar Co., Stillwater, MN, USA). Prior to sacrifice, animals were individually housed in metabolic cages (Tecniplast Gazzada S.A.R.L., Italy) for 24-h urine collection, and blood samples were obtained to evaluate serum creatinine (Scr) and blood urea nitrogen (BUN) levels. The creatinine clearance rate (ClCr) was calculated by the standard formula.

Mice were perfused through the left ventricle of the heart, and the kidney tissues were fixed with a periodate-lysine-paraformaldehyde (PLP) solution and then embedded in wax. After dewaxing, 4-µm sections were processed and stained with Masson's trichrome and hematoxylin.

After dewaxing, sections were incubated with a 0.5% Triton X100-PBS solution for 30 min and washed with phosphate buffered saline (PBS) three times. Non-specific binding sites were blocked for 1h with normal donkey serum diluted 1:10 in PBS, and then incubated overnight at 4℃ with antibodies diluted in 1:1000 in a humid environment. After rinsing in Tris-buffered saline (TBS), sections were incubated in peroxidase-conjugated donkey anti-mouse or rabbit IgG Fab fragment (Jackson ImmunoResearch Laboratories, West Grove, PA, USA) for 30 min. For staining, sections were incubated with a mixture of 0.05% 3,3'-diaminobenzidine containing 0.033% H₂O₂ at room temperature until a brown color was visible, washed with TBS, counterstained with hematoxylin and examined under light microscopy. The primary antibodies were as follows: C3 (Santa Cruz Biotechnology, Santa Cruz, CA, USA); C9, rabbit anti-rat polyclonal antibody (cross-reactive with mouse C9, kindly provided by Dr. B. P. Morgan, Cardiff University);17 C4d, a murine monoclonal antibody (Quidel Corporation, San Diego, CA, USA); CD55, a rabbit anti-rat polyclonal antibody (H-319, Santa Cruz Biotechnology, Santa Cruz, CA, USA).

For immunoblotting analysis, whole kidney tissues were homogenized in lysis buffer. Homogenates were centrifuged at 3000 rpm for 15 min at 4℃, and the protein concentration of the lysate was determined using the Bradford method (Bio-RAD, Hercules, CA, USA). Protein samples were resolved on a 15% SDS-polyacrylamide gel (SDS-PAGE) and then electroblotted onto Bio-Blot nitrocellulose membrane (Bio-RAD, Hercules, CA, USA). Ponceau S staining was performed to verify that equal amounts of protein (50µg) were loaded. The membrane was blocked for 1h in Tris-buffered saline with tween-20 (TBS-T, 10mM Tri-CL, 150mM NaCl [pH8.0], 0.05% tween-20) containing 5% nonfat powdered milk. C9 and CD46 were detected by incubating membranes for 1h with primary antibodies diluted 1:1000 (rabbit polyclonal anti-CD46 antibody, H-294, Santa Cruz Biotechnology, Santa Cruz, CA, USA). Primary antibody incubation was followed by six washes with TBS-T. The blot was then incubated with peroxidase-linked anti-mouse or rabbit IgG as a secondary antibody (Amersham Biosciences, Buckinghamshire, England) for 1h at room temperature. Antibody-reactive protein was detected using enhanced chemiluminescence (ECL, Amersham Life Science, UK). Optical densities were obtained using the VH group as a reference for 100%.

For double labeling with C4d or C3 and α-smooth muscle actin (α-SMA, Sigma Diagnostics), the tissue sections were prepared as described above. In brief, before incubation with primary antibodies, the tissue sections were permeabilized by incubation for 15 min in 0.5% Triton X-100 in PBS, and then blocked with normal donkey serum diluted 1:10 in PBS for 1h. Subsequently, the tissue sections were incubated overnight at 4℃ with monoclonal C4d or C3 antibody. The tissue sections were rinsed in PBS and incubated with Cy3-conjugated donkey anti-mouse antibody (diluted 1:500; Jackson ImmunoResearch Laboratories) for 2 h at room temperature. Tissues were rinsed with PBS, and then incubated with α-SMA antibody (diluted 1:300; Molecular probes, Inc. Eugene, OR, USA) for 2h at room temperature. The microscopic analysis was carried out using an MRC-1024 laser confocal microscope (Bio-Rad, Hercules, CA, USA) after the sections were mounted in Kaiser's glycerol gelatin (Merck, Dermstadt, Germany).

Data are expressed as means ± SEM. Multiple comparisons among groups were performed by one-way ANOVA with the posthoc Bonferroni test (SPSS software version 9.0, SPSS Inc, Chicago, IL, USA). Statistical significance was accepted when p < 0.05.

Mean body weights of both treatment groups increased during the experiment. Mean body weight (28 ± 0.8 vs. 35 ± 0.8, p < 0.05) and renal function (Scr: 0.70 ± 0.4 vs. 0.54 ± 0.04; BUN: 106 ± 24 vs. 42 ± 5; ClCr: 0.13 ± 0.03 vs. 0.24 ± 0.03, p < 0.05, respectively) was significantly lower in the CsA group than in the vehicle group after week 4 (Table 1). The mean concentration of CsA was 1305 ± 28ng/mL after week 1, and 1683 ± 83ng/mL after week 4. After week 4, the morphological features of the kidneys from CsA-treated mice (striped tubulointerstitial fibrosis, tubular atrophy, and inflammatory cell infiltration) were similar to those of kidneys from human cases of chronic CsA-induced nephrotoxicity (Fig. 1).

Immunoreactivity of C4d and C3 in the kidneys of mice treated with vehicle was slight. Expression of C4d and C3 in the renal tubular epithelium of the kidneys after week 1 was significantly increased by CsA treatment. A further increase was observed after week 4, when interstitial fibrosis and tubular atrophy appeared (Fig. 2 and Fig. 3). Glomeruli were negative for C3 and C4d. To clarify the localization of C4d-positive cells, double immunolabeling was performed using α-smooth muscle actin as a capillary marker. There were no merged cells within the tubulointerstitium (Fig. 4), which suggests that the C4d-positive cells are renal tubular cells.

We evaluated the expression of the terminal membrane attack complex (MAC) using an antibody against C9. CsA treatment enhanced MAC immunoreactivity in the interstitium and renal proximal tubules; this effect was time-dependent and restricted to subapical regions with an injured brush border (Fig. 5). Consistent with the immunohistochemical results, immunoblotting for MAC revealed that the expression of MAC protein was significantly increased by CsA in a time-dependent manner (1wk: 201 ± 36 vs. 100 ± 2; 4wk: 891 ± 37 vs. 100 ± 2, p < 0.05 vs. VH).

Expression of complement regulatory proteins (CD55 and CD46) was studied using immunohistochemistry or immunoblotting. Immunohistochemical analysis revealed that CD55 was expressed in the endothelial cells of the glomeruli and the vascular bundles of the outer medullas in the VH group and immunoreactivity of CD55 was increased by CsA treatment (Fig. 6). CD46 protein was also significantly increased by 4 weeks of CsA treatment (Fig. 7, 180 ± 11 (CsA) vs. 100 ± 20 (VH), p < 0.05).