The current study was approved by the Animal Ethics Committee of Shandong University (Jinan, Shandong, China).

A total of 60 male Wistar rats (lot no. SCXL20030004; 6–7 weeks old; weight, 180–200 g) were obtained from the Shandong University Experimental Animal Center (Jinan, Shandong, China). These rats were housed in cages and allowed free access to food and tap water. They were maintained on a 12:12 h dark/light cycle, with a room temperature of 22±1.5°C and a humidity of 55±5%. A total of 24 rats were selected randomly and divided into 2 groups (n=12 in each group), and the control group (C group) was treated with a dosage of 250 mg·kg⁻¹·day⁻¹ GSPE (CT group) (lot no. G050412; Tianjin Jianfeng Natural Product R&D Co., Ltd., Tianjin, China) (14). The remaining 36 rats were injected with a dose of 55 mg·kg⁻¹ STZ (injected into the tail veins). After five days, rats with blood glucose levels higher than 16.7 mmol/l were considered as being diabetic, and six rats were excluded for failure to meet this criterion. One week after STZ injection, the diabetic rats were then divided randomly into 2 groups (n=15 in each group), a diabetic group without any treatment (diabetes mellitus; DM; group), and another diabetic group that was treated with a dosage of 250 mg·kg⁻¹·day⁻¹ GSPE (T group). The GSPE was intragastrically administrated in normal saline solution for 24 weeks. At the end of the experiments, the animals were fasted overnight for 18 h and then anesthetized with intraperitoneal injection of sodium pentobarbitone (60 mg/kg; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) and sacrificed. Kidneys were perfused with ice-cold PBS and then were stored at −80°C.

At the end of the experiments, the fasting plasma glucose (FPG), 24 h urinary albumin excretion, hemoglobin A1c (HbA_1C), urinary creatinine (Ucr), serum creatinine (Scr), Ccr = Ucr/ScrxV (V:ml/min, urine per minute), and the ratio of kidney weight to bodyweight and serum AGEs was measured for each rat. Scr and Ucr were measured using an automatic biochemical analysis instrument (DVI 1650; Bayer AG, Leverkusen, Germany). HbA_1C was measured by high-pressure liquid phase methods (ADAMS A1c HA-8180; Arkray, Inc., Kyoto, Japan). The level of AGEs in serum was measured by the enzyme-linked immunosorbent assay (ELISA) method (19) (Hitachi 850; Hitachi, Tokyo, Japan). To collect urine samples, rats were placed in individual metabolic cages for 24 h prior to sacrificing. ELISA (Nephrat II; Exocell, Philadelphia, PA, USA) was used to determine 24 h urinary albumin excretion levels.

For the light microscopic examination, the kidneys were immersion-fixed, embedded and two sections (4 µm thickness) were stained with periodic acid-Schiff (PAS) reagent. For electron microscope examination, the renal cortex was cut into small pieces and pre-fixed in glutaraldehyde, post-fixed in buffered sodium tetroxide for 1 h, and then embedded in EPON 812 embedding resin (Electron Microscopy Sciences, Fort Washington, PA, USA). The PAS-positive area present in the mesangial region excluding cellular elements indicated mesangial matrix expansion. By using Leica QWin version 3 image analysis software (Leica Microsystems GmbH, Wetzlar, Germany), the percentage of the PAS-positive area in the glomerulus was analyzed. A total of 10 glomeruli, randomly selected in the two slides from the rats (a minimum of 8 rats in each group), were evaluated by two investigators blinded of the origins of the slides.

For transmission electron microscopic examination, a JEOL JSM 1011 microscope (JEOL, Ltd., Tokyo, Japan), was used to photograph images covering one or two glomerular cross-sections. The images at ×15,000 magnification were used to measure the GBM length and the number of slit pores. To determine the FPW (foot process width), the slit pore length, and the GBM thickness, the images at ×50,000 or ×100,000 magnification were used according to published methods (20–22).

The total RNA was extracted from the renal cortex using TRIzol reagent (Thermo Fisher Scientific Inc., Waltham, MA, USA). Primers for RAGE, nephrin, podocin, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were designed and synthesized (Huada Gene Center, Beijing, China) based on the published sequence. The primers were as follows: RAGE forward, 5′-GCTCTGACCGAAGCGTGA-3′ and reverse, 5′-CCTTCAGGCTCAACCAACAG-3′, 240 bp product; nephrin forward, 5′-ACCAAGTCCAGTCGCCCAG-3′ and reverse, 5′-ATACCAGCCTCACCCGAGTCCC-3′, 306 bp product; podocin forward, 5′-GTGTCCAAAGCCATCCAGTT-3′ and reverse, 5′-GTCTTTGTGCCTCAGCTTCC-3′, 232 bp product; and GAP DH forward, 5′-GAGGGGCCATCCACAGTCTTCTG-3′ and reverse, 5′-CCCTTCATTGACCTCAACTACATGGT-3′. The Titan One Tube RT-PCR kit (Boehringer-Mannheim, Shanghai, China) was used to amplify a total of 0.5 µg RNA. Agarose gel electrophoresis was used to separate products and ethidium bromide staining was used for visualization. A Tanon-1000 Gel Image System (Tanon, Shanghai, China) was used to digitize bands and GAPDH was used as the control gene based on previous studies (5,23).

Western blot analysis was performed as previously described (24). Renal tissues were homogenized with ice-cold lysis buffer and were centrifuged at 10,000 × g for 15 min at 4°C. Equal amounts (100 µg) of protein were loaded and separated by 10% SDS-PAGE. Separated proteins were then transferred onto nitrocellulose paper (Bio-Rad Laboratories, Inc., Hercules, CA, USA). A sample of total protein from the lysate (20 µg) was separated and blotted. The membranes were incubated with the following primary antibodies: Anti-RAGE (1:250, cat no. MAB1179; R&D Systems, Inc., Indianapolis, IN, USA), anti-nephrin (1:1,000, cat no. ab58968; Abcam, Cambridge, UK), anti-podocin (1:1,000, cat no. sc-21009; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) and anti-β-actin (1:500, cat no. ab8226; Abcam) overnight at 4°C, and then reacted with anti-rabbit IgG horseradish peroxidase-conjugated secondary antibodies (1:5,000, cat no. sc-2004; Santa Cruz Biotechnology, Inc.) at room temperature for 1.5 h. Signal detection was performed via exposure of the blot to enhanced diaminobenzidenecolor reagents (OriGene Technologies, Inc., Rockville, MD, USA) for 5 min. Quantification of the luminosity of each identified protein band was performed using Adobe Photoshop software (AdobePhotoshop 7.0; Adobe Systems, Inc., San Jose, CA, USA).

Renal tissue sections (4 µm) were subjected to immunohistochemical staining based on the published method with specific antibodies anti-RAGE (1:100, ab37647; Abcam), anti-nephrin (1:500, ab58968; Abcam), and anti-podocin (1:400, H-130, sc-21009; Santa Cruz Biotechnology, Inc.) (25). Color was developed by incubating with diaminobenzidine and counterstaining with hematoxylin. Controls were obtained by replacing the primary antibody with PBS. Semi quantitative analysis of the percentage of positive staining area in the glomeruli and tubules was evaluated using a computer imaging analysis system (Image-Pro Plus software version 4·5; Media Cybernetics, Inc., Rockville, MD, USA). The brown areas were judged as positive.

Data were presented as the mean ± standard deviation. One way analysis of variance or the Mann-Whitney U test was performed to compare the differences between groups. P<0.05 was considered to indicate a statistically significant difference.

There were four treatment groups, the control group (C group) that received no treatment, rats treated with 250·mg·kg⁻¹·day⁻¹ GSPE (CT group), rats that received STZ injection to induce diabetes, however no additional treatment (DM group) and rats that were diabetic but received 250·mg·kg⁻¹·day⁻¹ GSPE (T group). Rats in each group were observed from 1 week to 24 weeks. At the start of the experiment, there were 12 rats in each of the C and CT groups and 15 rats in each of the T and DM groups. At the end of the experiment, there were 10, 10, 8, and 12 surviving rats in the C, CT, DM and T groups, respectively. As presented in Table I, the FPG, HbA_1C, 24 h urinary albumin excretion, Ccr and ratio of kidney weight to body weight in the DM group were significantly increased compared with the C group (P<0.01). The diabetic rats treated with GSPE (T group) had markedly reduced Ccr, urinary albumin excretion (P<0.01), and ratio of kidney weight to body weight than that of the DM group. However, no significant differences in FPG and HbA_1C were identified between the DM and the T groups.

As presented in Table II, animals that were injected with STZ exhibited a significantly higher level of AGEs in serum than that in the C group (P<0.01). Diabetic rats that were treated with GSPE (T group) had significantly lower AGEs than the level in the DM group (P<0.01).

The PAS-staining results in the glomeruli of each group are presented in Fig. 1A. Quantitative analysis of the percentage of the PAS-positive area in the glomeruli is summarized in Fig. 1B. The ECM accumulation was significantly higher in the glomeruli of the DM group than that of the C group (P<0.01), and diabetic rats treated with GSPE (T group) had a significantly lower ECM accumulation in the glomeruli than that of the DM group (P<0.01).

The foot processes of glomeruli were broader and flatter in the DM group than that in the other groups. It was also identified that the mean FPW was significantly greater in diabetic glomeruli as compared with control glomeruli (P<0.01). The mean slit pore length was significantly shorter in diabetic glomeruli than that of the control glomeruli (P<0.01). Additionally, there were significantly fewer slit pores (per 100 µm) of GBM in diabetic glomeruli than in the control glomeruli (P<0.01). The GBM was significantly thicker in diabetic glomeruli as compared with control glomeruli (P<0.01). However, these changes were significantly reversed by GSPE treatment (Fig. 2 and Table III).

In the kidney cortex of DM rats, the mRNA and protein expression of RAGE were markedly increased compared with the levels in the C group (P<0.01), and GSPE treatment significantly decreased the expression of RAGE (P<0.01) (Figs. 3 and 4). A significantly decreased expression of nephrin mRNA and protein was observed in the DM group as compared with the C group (P<0.05 and P<0.01, respectively), and GSPE treatment significantly increased the mRNA expression and protein level of nephrin (P<0.05 and P<0.01, respectively) (Figs. 3 and 4). However, no significant differences were identified in the mRNA or protein expression of podocin among the four groups (P>0.05) (Figs. 3 and 4).

Immunohistochemical staining indicated the localization of signals for RAGE, nephrin and podocin predominantly in the glomeruli. The intensity and area of RAGE staining were significantly increased in the glomeruli of diabetic rats as compared with the C group, and GSPE treatment significantly reversed this effect (Fig. 5A). The intensity and area of nephrin staining were significantly decreased in the glomeruli of diabetic rats compared with the C group, and GSPE treatment also significantly reversed these effects (Fig. 5B). No differences were observed in the intensity and area of podocin staining in the glomeruli for the four groups (Fig. 5C).