Icariin was obtained from Yangtze River Pharm Co., Ltd. (Taizhou, China). Irbesartan was obtained from Sanofi San de la Fort Group Co., Ltd. (Paris, France). Streptozocin (STZ) was purchased from Sigma-Aldrich (St. Louis, MO, USA). Urine protein test kit, total cholesterol assay kit, low-density lipoprotein cholesterol assay kit, high-density lipoprotein cholesterol assay kit, triglyceride assay kit, creatinine assay kit, Heme oxygenase 1 (HO1), Quinone oxidoreductase 1 (NQO1) enzyme activity detection kits and malondialdehyde (MDA) were gained from Nanjing JianCheng Bioengineering Institute (Nanjing, China). Rat insulin ELISA kit was obtained from Jiangsu Yutong Biotechnology Co., Ltd (Nanjing, China). Blood glucose meter and Blood glucose test strips were obtained from Roche (Basel, Switzerland). Nuclear and cytoplasmic protein extraction kit and Phenylmethanesulfonyl fluoride (PMSF) were obtained from KeyGen Biotech (Nanjing, China). TRIzol reagent was purchased from Invitrogen (California, USA). DAPI, SOD, GPx, CAT enzyme activity detection kits and the goat anti-rabbit secondary antibodies were obtained from Beyotime Biotechnology (Shanghai, China). PI3K, Akt and mTOR antibodies were obtained from Cell Signaling Technology, Inc (Massachusetts, USA). ML385 was obtained from MCE (New Jersey, USA). Heme oxygenase 1 (HO1), superoxide dismutase 2 (SOD2) /MnSOD, and thioredoxin 1 (Trx1) antibodies were purchased from ProteinTech (Chicago, USA). Fibronectin, NQO1 and Nrf2 antibodies were purchased from Abcam (Chicago, USA). p62 antibody was obtained from Wanleibio (Shenyang, China). G15 was purchased from ApexBio (Houston, USA). Fetal bovine serum (FBS) was purchased from Biological Industries (BI, Israel). GAPDH and β-actin antibodies were purchased from Abways Technology (beijing, China). Dihydroethidium (DHE) was purchased from Yeasen (Shanghai, China). TRIzol reagent, qPCR SYBR® Green Master Mix and Enhanced chemiluminescent (ECL) plus reagent kit were purchased from Vazyme (Nanjing, China).

Female Spraque–Dawley rats (weighing between 180 to 200 g, n = 104) were obtained from Zhejiang Experimental Animal Center. Throughout experimental process, rats were housed in a controlled environment condition. The experimental protocol was reviewed and approved by Institutional Animal Care and Use committee of China Pharmaceutical University. The approved number of animal experiments was 2019-05-16.

For intraperitoneal injection, STZ was dissolved in citrate buffer (PH 4.5) and administered in a dose of 55 mg/kg. After 1 week of STZ administration, the rats with fasting blood glucose level over 11.1 mmol/L and persistent hyperglycemia over 16.7 mmol/L were considered as diabetic. Four weeks later, the serum urea nitrogen, urine protein and creatinine levels of rats were measured, and the rats (n = 98) meeting the requirements were selected for experimentation (Figure 1B). Rats were randomly divided into six groups: normal group, model group, icariin (20, 40, 80 mg/kg, i.g.) group and irbesartan (13.5 mg/kg, i.g.) group. After 9 weeks of administration, the blood and urine samples of all the rats were gathered. Finally, the rats were sacrificed, and their kidney tissues were obtained and weighed. For hematoxylin-eosin (H&E), Masson and PAS staining, part of kidney tissues were fixed in 4% paraformaldehyde. The rest of kidney tissues were preserved at −80°C for western blot analysis. The serum of rats was collected and stored for the detection of certain biochemical indexes.

For immunohistochemistry assays, kidney tissues were fixed in 4% paraformaldehyde and embedded in paraffin. Then, the renal sections were daparaffinized, subjected to antigen retrieval and treated with 0.5% H₂O₂ to eliminate endogenous peroxidase activity. Next, the sections were incubated in blocking buffer (2% BSA) with primary antibody overnight at 4°C. Nrf2 and Fibronectin used at a dilution of 1:200 and 1:100, respectively.

To detect the levels of urea nitrogen and serum creatinine, ELISA assays were performed according to the manufacturer's instructions. All ELISA kits were purchased from Nanjing JianCheng Bioengineering Institute (Nanjing, China).

Human mesangial cells (HMC) was obtained from ScienCell Research Laboratories, Inc. (San Diego, California, USA). Cells were cultured in Dulbecco's Modified Eagle's medium (DMEM) containing 10% FBS (fetal bovine serum) and Penicillin-Streptomycin Solution (1X) in an atmosphere of 5% CO₂ at 37°C. For the experimental study, HMCs were maintained in DMEM with normal glucose (5.6 mM glucose, NG), high glucose (30 mM glucose, HG), icariin of different concentrations and irbesartan of a certain concentration in the presence of high glucose for 48 h.

The whole and nucleus proteins were lysed by certain lysis buffer containing 1% PMSF on the ice, and then clarified by centrifugation at 4°C. The samples were then separated by 10 % SDS-PAGE and electroblotted onto PVDF membranes. The membranes were blocked with 5% non-fat milk in TBST for 1.5 h, and incubated with Nrf2 (1:1,000), SOD2 (1:1,000), NQO1 (1:1,000), HO1 (1:1,000), Trx1 (1:1,000), Fibronectin (1:1,000), β-actin (1:1,000), and GAPDH (1:1,000) at 4°C overnight. Next, the membranes were washed and incubated with goat anti-rabbit IgG (1:5,000) for 1.5 h at room temperature. Finally, using ECL reagent (vazyme) to make the membrane visible and analytical, and all these images were quantified by ImageJ Software (Media Cybernetics, Silver Spring, MD, USA). Data were expressed as the ratio of integrated optical density to area.

HMC (1 × 10⁶ cells per well) were plated on glass coverslips. The cells were fixed in 4% paraformaldehyde, permeabilized with 0.3% Triton X-100 in PBS for 20 min at room temperature and incubated in blocking buffer (1% BSA) for 1 h. Then, Nrf2 antibody (1:200) was added in the glass coverslips at 4°C overnight. After washing cells with PBS for 3 times, goat anti-rabbit IgG H&L (FITC) was added and incubated for 1 h at room temperature. For nuclear staining, cells were incubated with DAPI (1:100,000) for 10 min. Finally, the images of cells were gathered by fluorescence microscope (Olympus, Japan) in the China Pharmaceutical University.

HMC were treated with indicated drugs for 48 h, washed with cold phosphate-buffered saline (PBS) three times, and harvested by centrifugation. Then, the nuclear and cytoplasmic fractions were prepared by commercial nuclear and cytoplasmic extraction reagents. Proteins isolated from HMC were incubated with Nrf2 or Keap1 antibody at 4°C overnight. Next, the proteins were precipitated using protein A/G-agarose beads for 4 h at room temperature. After washing the precipitated proteins with PBS for 3 times, SDS-PAGE was used to separate proteins as illustrated in the western blot assay.

Before transfection, 1 × 10⁶ HMC cells were seeded in 6-well. 24 h later, the cells were transfected with the Control siRNA, siNrf2 or siGPER by lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA, USA) and culture in opti-MEM reduced serum medium for 4 h. Next, the opti-MEM reduced serum transfection medium was removed from plates and DMEM with 10% FBS was added for 36 h incubation. Then, cells were treated by indicated drugs and harvested. Lastly, the samples were subjected to Q-PCR assay or western blot assay according to manufacturer's instruction with some modification.

Kidneys cortexes from control and diabetic mice were collected or HMC cells were isolated to extract RNA using Trizol followed by the manufacturer's instructions. Total RNA was dissolved in the DEPC water. The RNA concentration was determined with a spectrophotometer (Nano 1000, Thermo Fisher Scientific, Waltham, MA, USA) by absorbance at 260 nm. Then, the reverse transcription of purified total RNA (1 μg) was conducted by HiScript™ reverse transcriptase kit and then reverse transcriptase–polymerase chain reaction was conducted on ABI 7500 (Applied Biosystems, Foster City, CA, USA) with Hieff™ qPCR SYBR® Green Master Mix. GAPDH was used as a control for target gene. The sequences of primers were listed in Tables 1, 2.

To detect the superoxide, HMC were treated with irbesartan (1 μM) and icariin (1, 3, 10 μM) for 48 h with or without high glucose. The final concentration of dihydroethidium (DHE) was 10 μM (Yeasen, Shanghai, China) with serum-free medium. Remove the cell culture medium, add appropriate volume of DHE, and incubate in 37°C cell incubator for 30 min. After washing cells with serum-free cell culture solution for 3 times, cells were collected and treated according to the manufacturer's instructions. Lastly, cells were detected by flow cytometry (Macsqant, Germany) in the China Pharmaceutical University.

To detect the superoxide level in the kidney tissues of rats, frozen slides were incubated with 10 μM DHE for 30 min at 37°C in dark. After washing slices with PBS for 3 times, they were photographed by fluorescence microscope.

HMC cells (1 × 10⁶ cells per well) were cultured in a 6-well plate at the logarithmic growth stage. The cells were completely adherent to the wall. Added low sugar culture medium to the normal group, high sugar culture medium (30 mM) to the model group, and treat the administration group with corresponding concentration of drugs and high sugar culture medium, respectively. After 48 h of administration, discarded the culture medium, washed the residual complete culture medium twice with PBS, and discarded all the residual PBS as much as possible. The activities of SOD, HO-1, NQO1, GPx and CAT antioxidant enzymes were detected according to the product manual of commodity manufacturer.

EMSA was performed to detect the DNA binding activity of Nrf2 by a commercial kit (Pierce, Rockford, IL, USA). Briefly, HMC (1 × 10⁶ cells per well) were cultured in 6-well plates and treated with indicated drugs for 48 h and the nuclear proteins were extracted. Then, the proteins were incubated with biotin-labeled Nrf2-specific oligonucleotides, poly (dI-dC) and binding buffer for 20 min at room temperature. Next, the protein-DNA complexes were separated on a 5% non-denatured polyacrylamide gels and transferred onto PVDF membranes. Lastly, the biotin end-labeled DNA was detected by chemiluminescence.

For Luciferase reporter assay, HMC were co-transfected with pGL3-ARE reporter gene plasmid by Lipofectamine 2000 for 8 h. Then, fresh medium with indicated drugs were added for another 48 h. Lastly, the relative luciferase activity was detected by the Luciferase assay kit.

All experiments were performed for at least 3 times and the data was shown as mean ± SD. Data among groups were analyzed by T-test (two groups) or one-way ANOVA (multiple groups) by using SPSS statistical software (SPSS, Chicago, IL, USA). p < 0.05 was considered as statistically significant.

Four weeks after STZ injection, diabetic rats showed a remarkably increase in food intake (Figure 2B), water intake (Figure 2C), blood glucose (Figure 2D), blood pressure (Figure 2E), 24 h urine volume (Figure 2F), 24 h proteinuria (Figure 2G), blood urea nitrogen (BUN) (Figure 2H), serum creatinine (Src) (Figure 2I), and creatinine clearance rate (Figure 2K) while reduction in urine creatinine (Ucr) (Figure 2J) and weight (Figure 2A) when compared with the normal rats. The result showed that diabetic rats were developed with diabetes nephropathy. Subsequently, healthy rats with the injection of 0.1 M citrate buffer and successful established DN rats were divided into six groups.

Body weight, food intake and water intake were recorded every day during the 9-weeks period. Diabetes is often associated with weight loss and increased water and food intake. Our results showed that diabetic rats took more water and food, along with the hyperglycemia and high blood pressure, indicating that a DN animal model was successfully established. Interestingly, icariin dose-dependently attenuated the progression of DN, as exhibited by decreased weight loss (Figure 3A) and inhibited the increase in water (Figure 3C), food intake (Figure 3B) and blood pressure (Figure 3G), but no significant changes in blood glucose (Figure 3D), fasting insulin level (Figure 3E) and insulin sensitivity index (Figure 3F) were observed in the different groups, which suggested that the improvement of diabetes symptoms by oral icariin was probably independent of glycemic control.

Nine weeks after being grouped, kidneys in diabetes mellitus (DM) rats in model group were heavier than those in other groups (Figure 4H). However, these indicators of kidney function were dose-dependently increased in icariin treatment groups of diabetic rats. The 24 h urinary volume (Figure 4A), 24 h proteinuria (Figure 4B), blood urea nitrogen (BUN) (Figure 4C), serum creatinine (Src) (Figure 4D), and 24 h urinary albumin excretion (Figure 4G) were all significantly higher in five groups of diabetic rats than those in normal group. These values were decreased dose-dependently in icariin treatment groups. Compared with the irbesartan treatment group, the ratio of kidney weight to body weight, serum creatinine, and 24 h urinary albumin excretion were all further decreased, whereas urinary creatinine (Figure 4E) and creatinine clearance (Figure 4F) were increased in icariin (40, 80 mg/kg) groups but these values did not differ between the two groups. Dyslipidemia often occurs in diabetic nephropathy, including increased triglyceride (TG) (Figure 4I), total cholesterol (Figure 4J), non-esterified fatty acid (Figure 4K), low density lipoprotein cholesterol (LDL-C) (Figure 4L) and decreased high density lipoprotein cholesterol (HDL-C) (Figure 4M), these values were dose-dependently improved in icariin (40, 80 mg/kg) groups.

Glomerular lesions are characterized by extracellular matrix deposition in the mesangium and glomerular hypertrophy. In order to observe renal cortical morphological changes in diabetic rats, the representative photomicrographs of rat kidneys in HE, PAS and Masson's trichrome-stained from the six groups were performed. As shown in Figures 5A–C, in model group, the glomerular exhibited alterations in several aspects including extracellular matrix deposition, mesangial region enlargement and basement membrane thickening compared with normal group. Icariin treatment obviously attenuated glomerular hypertrophy and mesangial matrix expansion in diabetic rats. Similarly, the HE staining exhibited increased ECM deposition in the glomeruli of model group, after icariin treatment, the renal injury dose-dependently improved (Figure 5E). Furthermore, the Masson stain also manifested that icariin-decreased collagen deposition in the glomeruli of diabetic rat. Fibronectin (FN), the major component of glomerular ECM, play a key role in promoting glomerular fibrotic lesions. In this study, icariin significantly dampened the increase in FN protein expression in the diabetic rat's kidney by immunohistochemical stain (Figure 5D) and western blot (Figure 5F) assay. Overall, these results showed that icariin alleviated renal damage by reducing the ECM deposition.

Oxidative stress has been considered to be one of the main causes of excessive generation of ECM and consequent leading to damage to kidney function and structure in DN. To explore the effects of icariin on oxidative stress in DN, the malondialdehyde (MDA) content (Figure 6A), superoxide dismutase (SOD) activity (Figure 6B) and superoxide anion (Figure 6C) as determined by DHE staining in kidney were detected. Compared with normal group, MDA content and superoxide anion were higher, while SOD activity were significantly lower in model group. In contrast, the level of MDA and superoxide anion were markedly lowered, whereas the activity of SOD increased dose-dependently in icariin groups.

The level of ROS in HMC-induced by high glucose, as determined by DHE staining, was markedly increased. The content of superoxide anion (Figure 6D) was dose-dependently attenuated by icariin (1, 3, 10 μM) detected by flow cytometry. During the progression of DN, the antioxidant scavenging system does not work, accompanying by decreased activity and expression of antioxidase, including catalase (CAT), heme oxygenase (HO-1), NADPH: quinone acceptor oxidoreductase 1 (NQO1), superoxide dismutase (SOD) and glutathione peroxidase (GPx). As illustrated in Figure 6E, the stimulation with high glucose significantly decreased the activity of the scavenging system against reactive species in HMC, while icariin promoted the activity of antioxidant enzymes. Moreover, Q-PCR and western blot assays indicated that icariin (1, 3, 10 μM) significantly increased the mRNA and protein expressions of HO-1, NQO1, SOD and Trx1 (Figure 6F) in HMC.

To illuminate the mechanism of icariin on anti-oxidative stress with its effect on the expression and activation of Nrf2, a key regulator of antioxidative stress system was firstly investigated. As shown in the Figures 7A–C, icariin increased the Nrf2 protein level in vivo and in vitro. Subsequently, we investigated whether icariin played a role in the regulation of Nrf2, and the protein levels of Nrf2 in the nucleus and cytoplasm were detected by western blot assay. The results showed that icariin exerted promotion effects on Nrf2 nuclear translocation compared with the HG group and also in icariin treatment groups of diabetic rats (Figure 7D). To further confirm that icariin promoted HG-induced distribution of Nrf2 in HMC, immunofluorescence staining was performed by laser scanning confocal microscopy to investigate the distribution of Nrf2 (Figure 7E). The images indicated that icariin could significantly increase Nrf2 staining (green) in the nucleus, suggesting that icariin promote the nuclear translocation of Nrf2, and the results were consistent with the western blot. In addition, the DNA binding activity and the transcriptional activity of Nrf2 were detected by EMSA and luciferase reporter gene assay, respectively. As shown in Figures 7F,G, HG stimulation had no effect on the binding activity and luciferase activity of Nrf2 in HMC, while icariin significantly increased the DNA-binding activity of Nrf2 and Nrf2 luciferase activity. The massive deposition of fibronectin (FN) plays an important role in the diabetic glomerular pathology, therefore, inhibition of ECM accumulation in kidneys would repress the glomerulopathy and delay the progression of DN. Figure 7H illustrated that Nrf2 siRNA and ML385 (Nrf2 Inhibitor) also abolished the inhibitory effect of icariin on ECM deposition. Therefore, icariin might alleviate the oxidative stress and ECM generation in HG-induced HMC by activating the Nrf2 signaling pathway.

Under normal conditions, the degradation of Keap1 mainly through the proteasomal degradation and autophagy-lysosome pathway (Tkachev et al., 2011). In this process, p62 compete with the Keap1-binding site on Nrf2, subsequently the Keap1-p62 complex is recruited to the autophagosome by LC3, and degraded by autophagy (Fan et al., 2010; Taguchi et al., 2012). To explore the mechanisms by which icariin stimulates Nrf2 activation under high glucose conditions, the protein level of Keap1 was detected by western blot assay (Figures 8A,B). The result showed that Keap1 was significantly increased in model group and HG group as well as it significantly decreased in icariin treatment group. To further confirm that Keap1 is degraded through p62-dependent pathway, the level of co-immunoprecipitated p62 significantly increased after treat with icariin in HMC (Figure 8C), suggesting that p62 was involved in the enhancement effect of icariin on Nrf2 level in HMC under high glucose conditions. In addition, the protein levels of total and phosphorylated PI3K, Akt, and mTOR1 (Figures 8D,E) were examined by western blot assay. Icariin treatment decreased the phosphorylation levels of PI3K, Akt, and mTOR in diabetic rats, these findings revealed that PI3K/Akt/mTOR signal pathway mediated autophagy inhibition is activated by icariin.

We have previously demonstrated that icariin attenuated high glucose-induced TGF-β production through G protein-coupled estrogen receptor 1 (GPER) (Li et al., 2013). Here, we additionally demonstrated significant increases in ROS accumulation as indices of oxidative stress during the progression of DN, and explored associated mechanisms. To confirm the key role of GPER played in icariin mediated protection of DN, the level of ROS (Figure 9A) was detected by flow cytometry. The results showed that icariin markedly reduced the superoxide anion in HMC, while G15 (a selective antagonist of GPER) prevented icariin decreasing ROS. Then, the protein levels of Nrf2 in the nucleus and cytoplasm (Figures 9B,C), the interaction between Nrf2 and Keap1 (Figure 9D), the antioxidant enzymes (Figure 9E), Keap1 (Figure 9F), PI3K/Akt pathway (Figure 9G), the interaction between p62 and Keap1 (Figure 9H), and FN (Figure 9I) were detected. siGPER and G15 significantly reduced the inhibitory effects of icariin on Keap1, PI3K/Akt pathway, p62 and FN expression. In addition, icariin also significantly up-regulated the Nrf2 and the antioxidant enzymes protein level, increased the dissociation of Nrf2 and Keap1, association of Nrf2 and p62 in HMC, while these effects could be abolished by siGPER and G15. The main mechanisms about anti-CKD effect of icariin has been showed in (Figure 10).