A total of 30 healthy male Sprague-Dawley rats (6-weeks old; 181.65±5.15 g) were purchased from the Animal Department at Xiangya Medical School, Central South University (Changsha, China). All rats had free access to food and water with a 12-h dark/light cycle in a climate-controlled room (temperature, 18–25°C; humidity, 65–80%; CO₂, 0.03%). Following feeding under adaptation conditions for 3 days, rats were randomly divided into normal control (n=6) and DKD model (n=24) groups. The normal control animals were fed with normal food and received once daily intraperitoneal injection of citrate buffer [0.1 mol/l, (pH 4.5)] at a dose of 10 ml/kg, while the rat model of DKD was performed as described previously (13,14). The rats were fed with a high-sugar and high-fat diet for 8 weeks followed by once daily intraperitoneal injection of streptozotocin (30 mg/kg; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) dissolved in citrate buffer. Blood and urine glucose was detected 72 h following streptozotocin injection. The diabetes model was considered to be successful when fasting blood glucose was >13.8 mmol/l, random blood glucose was >16.7 mmol/l, and urine glucose, detected by Benedict's test, was ++++. The rat model of DKD was considered to be successfully established if urine protein, determined by the Lowry procedure, was >30 mg for 24 h following the establishment of the diabetes model. During the experiment, blood glucose was measured every 3 days using a tail vein sample. A certain amount of long-acting insulins (0.4–3.2 units) was injected into the rats of which blood glucose was higher than 26.0 mmol/l, but random blood glucose was maintained above 16.7 mmol/l.

DKD model rats were subsequently randomly divided into the following three groups (n=8 for each group): DKD model group; low-dose tranilast group (200 mg/kg/day) and high-dose tranilast group (400 mg/kg/day) (11). The DKD model rats were only fed 1.5% sodium carboxymethyl cellulose solution, while the rats treated with tranilast had tranilast administrated twice daily, which was dissolved in 1.5% sodium carboxymethyl cellulose solution. By the 8th week following tranilast treatment, the 24 h urine samples of all rats were collected to determine the albumin concentrations, and rats were intraperitoneally anesthetized with ketamine/xylazine (80 mg/kg+10 mg/kg) and sacrificed. Blood samples for measuring creatinine were obtained from the abdominal aorta prior to the kidneys being harvested and weighed. All animals used in the present study were appropriately processed following protocols approved in advance by the Animal Care and Use Committee at Central South University. All procedures performed in experiments with animals were in accordance with the ethical standards of the Department of Nephrology, Second Xiangya Hospital at which the present study was conducted.

In the present study, the levels of blood glucose were estimated using an Accu-Chek glucometer (Roche Diagnostics, Basel, Switzerland). To assess the urinary albumin excretion rate (UAER), the albumin concentrations for the urine samples were detected using the Urine micro-albumin assay kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China). The serum for detecting creatinine were collected by centrifugation of whole blood (0.5 ml each of samples) at 3,000 × g for 15 min following blood agglutination at room temperature, and serum creatinine (Scr) was detected using a BioAssay Systems Creatinine assay kit (BioAssay Systems, Hayward, CA, USA).

The 4% paraformaldehyde-fixed (at room temperature for 24 h), paraffin-embedded tissue sections (3-µm) were dewaxed routinely in xylene for 10 min and then in graded alcohol (100, 95, 80 and 70%), 5 min each, and washed in dH₂O. Hematoxylin was used to stain nuclei for 3 min at room temperature. Following flushing with water for 30–60 sec, sections were stained in Ponceau acid fuchsin solution (ponceau 0.7 g, acid fuchsin 0.3 g, distilled water 99 ml and glacial acetic acid l ml) for 5–10 min, rinsed with tap water for 1 min, stained in 1% phosphomolybdic acid solution for 1 min and transferred into 2% aniline blue liquid to dye for 3 min. Following washing with water, sections were dehydrated in graded ethanol and made transparent in xylene for 1 min, and subsequently enclosed with neutral gum. All the procedures were done at room temperature. Collagen fibers were stained blue. Slides were observed under a light microscope (Olympus CX41; Olympus Corporation, Tokyo, Japan). Images were taken of 10 non-overlapping interstitial areas in each section. A high-definition color medical analysis system with image analysis software (HMIAS-2000; Qianping Image Technology Co., Ltd., Wuhan, China) was used for automatic measurement and analysis, and the percentage of tubulointerstitial area in the total area visualized in each section was calculated.

Modified toluidine blue staining was performed to detect mast cells (15,16). Paraffin sections (3-µm) were deparaffinized and rehydrated according to the above procedures, and immersed in toluidine blue for 3 min and differentiated with 0.5% glacial acetic acid for 10–20 sec until mast cell nuclei were visible at room temperature. Tissue sections were cleared with xylene and sealed with neutral gum, and subsequently photographed under a light microscope. A total of three images at high magnification (×200) were randomly selected for calculating the number of mast cells.

Paraffin sections (3-µm) were dewaxed and hydrated with the above method, and immersed in 0.01 mol/l citrate buffer (pH 6.0) at 90°C for 5 min for microwave antigen retrieval. Sections were then incubated with 3% H₂O₂ for 10 min at room temperature to eliminate endogenous peroxidase activity prior to blocking in 10% goat serum (Beyotime Institute of Biotechnology, Haimen, China) for 30 min at 37°C. Sections were subsequently incubated with anti-complement C3a receptor 1 (C3aR; 1:200; cat. no. sc-20138) and anti-proto-oncogene c-kit (c-kit; 1:400; cat. no. sc-5535) polyclonal antibodies (Santa Cruz Biotechnology Inc., Dallas, TX, USA), anti-fibronectin (FN; 1:400; cat. no. BA1772) and anti-stem cell factor (SCF; 1:400; cat. no. A01254) polyclonal antibodies (Boster Biological Technology Co., Ltd., Wuhan, China), and anti-collagen I polyclonal antibody (Col-I; 1:400; cat. no. T40227R; Meridian Life Science, Inc., Memphis, TN, USA) overnight at 4°C. Following rinsing, the sections were incubated with horseradish peroxidase-labeled goat anti-rabbit immunoglobulin (Ig)G secondary antibodies for 1 h at 37°C (1:500; cat. no. ab6721; Abcam, Cambridge, UK), stained with 3,3′-diaminobenzidine for 30–60 sec and counterstained with hematoxylin for 3 min at room temperature. Negative controls for specific labeling were performed in parallel by replacing the primary antibody with a normal rabbit serum (cat. no. A7016; Beyotime Institute of Biotechnology). The slides were examined and photographed with a light microscope (magnification, ×200). The Image-Pro P1us version 6.0 image analysis system (Media Cybernetics, Inc., Rockville, MD, USA) was used to measure the optical density of the positive area. The average optical density was used to represent the expression level of C3aR, FN, Col-I, SCF and c-kit.

A total of ~50–100 mg of kidney tissues were ground in liquid nitrogen and homogenized in 500 µl pre-cooled radioimmunoprecipitation lysis buffer (cat. no. P0013B; Beyotime Institute of Biotechnology) at 4°C. The lysate was cleared by centrifugation at 12,000 × g for 20 min at 4°C. Protein concentration was determined using a bicinchoninic assay kit (Beyotime Institute of Biotechnology). Subsequently, 20–30 mg soluble lysates were loaded in each lane and separated by 8 or 10% SDS-PAGE gel electrophoresis and electrophoretically transferred to polyvinylidene difluoride membranes. Membranes were blocked with 5% nonfat dry milk in TBS/0.5% Tween-20 for 1 h at room temperature, washed with TBS/Tween-20 and incubated with FN (cat. no. ab2413), Col-I (cat. no. ab34710), SCF (cat. no. ab101072) and c-kit (cat. no. ab46758) polyclonal antibodies (Abcam, Cambridge, UK) at 1:1,000 dilution overnight at 4°C. Blots were rinsed with TBS/Tween-20 and subsequently incubated with horseradish-peroxidase-conjugated secondary antibodies (at room temperature for 1 h), including goat anti-rabbit IgG (1:4,000; cat. no. ab6721) and rabbit anti-goat IgG (1:4,000; cat. no. ab97100) (Abcam, Cambridge, UK). Following washing with TBS/Tween-20, the blots were developed with enhanced chemiluminescence reagents (GE Healthcare, Chicago, IL, USA). The density of the identified bands was quantified by densitometry using ImageQuant software (version 5.2; Molecular Dynamics, Sunnyvale, CA, USA). Values were normalized with respect to β-actin (1:4,000; cat. no. ab8227; Abcam, Cambridge, UK) expression.

Tissue expression of FN, Col-I, SCF and c-kit mRNA was assessed by RT-qPCR. A TRIzol^® RNA extraction kit (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) was used to extract total RNA from the renal tissue. Total RNA was used for cDNA synthesis according to the instructions in the RevertAid First Strand cDNA Synthesis kit (Fermentas; Thermo Fisher Scientific, Inc.). Gene sequences were verified in the NCBI GenBank (https://www.ncbi.nlm.nih.gov/genbank/) and primers were designed according to primer design principles using Primer Premier version 5.0 (Premier Biosoft International, Palo Alto, CA, USA) and purchased from Sangon Biotech Co., Ltd. (Shanghai, China). The primer sequences are presented in Table I. qPCR amplification of the products was performed on an ABI 7300 Real Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.) using SYBR GreenER^® qPCR SuperMix Universal kit (Invitrogen; Thermo Fisher Scientific, Inc.) under the following reaction conditions: 50°C for 2 min and 95°C for 10 min; 40 cycles of 95°C for 10 sec, 60°C for 60 sec and 72°C for 60 sec; final step for 10 min at 72°C. The qPCR was repeated three times for each gene. Relative quantification of gene expression was performed using the 2^−ΔΔCq method with β-actin as an endogenous control (17).

SPSS software (version 17.0; SPSS, Inc., Chicago, IL, USA) was used for statistical analysis. The results are presented as mean ± standard deviation. Data were analyzed using one-way analysis of variance or the Kruskal-Wallis test. Pearson correlation analysis was performed to determine the correlation between the expression of SCF and c-kit with the infiltration of mast cells and indices of renal interstitial fibrosis. P<0.05 was considered to indicate a statistically significant difference.

The DKD rats demonstrated polydipsia, polyphagia, polyuria and other symptoms. Compared with the normal control group, the body weight of DKD model rats was significantly reduced, while the kidneys became bigger and heavier, and the kidney weight/body mass index was significantly increased (P<0.05; Table II). Blood glucose, UAER and Scr concentrations were significantly increased in the DKD model group compared with the normal control group (P<0.05; Table II). There was no significant difference in the body weight, the kidney weight/body mass index, UAER, Scr and blood glucose level among different tranilast groups and the DKD model group (P>0.05).

Toluidine blue staining and IHC of C3aR was performed to detect mast cells in the renal interstitium. Toluidine blue staining is a classic method for quickly identifying mast cells, which stains mast cells purple. IHC of C3aR is another method used for detecting mast cells as mast cells express C3aR. In normal rats, mast cells, which are stained purple by toluidine blue or stained brown by IHC, were primarily localized in perivascular areas and the tubulointerstitium, and only a few mast cells were detected in the renal interstitium, but not in glomeruli (Fig. 1). However, increased numbers of mast cells were demonstrated in the renal interstitium of DKD model rats (P<0.001; Fig. 1). Tranilast decreased the infiltration of mast cells dose-dependently (P<0.001; Fig. 1).

The tubulointerstitial fibrosis of the kidney was observed using Masson's trichrome staining, which demonstrated that there was no tubular expansion and fibrous proliferation in the normal control group (Fig. 2). In a DKD model, kidneys developed tubular vacuolar degeneration and atrophy with tubulointerstitial fibrosis (Fig. 2). The fibrotic areas detected by blue staining were significantly ameliorated in the kidneys treated with tranilast in a dose-dependent manner as compared with the DKD group (P<0.001; Fig. 2).

IHC staining demonstrated that FN and Col-I were primarily deposited in the glomeruli, renal tubular basement membrane and perivascular areas in the normal control rats (Fig. 3A and B). However, in the DKD rats, the expression of FN and Col-I was increased significantly in the renal interstitium, particularly in perivascular areas and inflammatory areas (P<0.001; Fig. 3A-C). Tranilast dose-dependently downregulated the expression of FN and Col-I (P<0.001; Fig. 3A-C). Western blot analysis and RT-qPCR demonstrated that the mRNA and protein levels of FN and Col-I were significantly upregulated in the kidneys of DKD rats, compared with those of normal control rats (P<0.001; Fig. 3D and E, respectively). Tranilast significantly downregulated the protein and mRNA expression of FN and Col-I in a dose-dependent manner compared with the DKD rats (P<0.05; Fig. 3D and E, respectively).

A small amount of SCF and c-kit protein was revealed by IHC staining in the renal tubules and interstitium of normal rat kidneys. In the DKD model rats, a significantly increased level of SCF and c-kit protein were observed in renal tubular epithelial cells and interstitial inflammatory cells, compared with control rats (P<0.001; Fig. 4A-C). Tranilast (400 mg/kg.d) downregulated the expression of SCF and c-kit (P<0.01; Fig. 4A-C). Compared with the normal control group, the protein and mRNA levels of SCF and c-kit in the DKD model group were significantly increased (P<0.001; Fig. 4D and E, respectively), and tranilast significantly decreased the protein and mRNA levels of SCF and c-kit in a dose-dependent manner (P<0.05; Fig. 4D and E, respectively), as determined by western blotting and RT-qPCR, respectively.

Pearson correlation analysis demonstrated a significant positive correlation between the level of SCF and the infiltration of C3aR-positive cells or the expression of FN and Col-I (r=0.941, P<0.01; r=0.896, P<0.01; and r=0.858, P<0.01, respectively; Table III). In addition, a strong positive correlation was demonstrated between the level of c-kit and the infiltration of C3aR-positive cells or the expression of FN and Col-I (r=0.951, P<0.01; r=0.976, P<0.01; and r=0.932, P<0.01, respectively; Table III). These results indicate that the expression of SCF and c-kit were positively correlated with the extent of mast cell infiltration and the expression of FN and Col-I.