Nodakenin alleviated obstructive nephropathy through blunting Snail1 induced fibrosis

Abstract Tubulointerstitial fibrosis plays an important role in end‐stage renal failure, and there are only limited therapeutic options available to preserve organ function. In the present study, we identified that nodakenin, a coumarin isolated from the roots of Angelicae gigas, functions effectively against unilateral ureteral obstruction (UUO)‐induced fibrosis via down‐regulating Snail1 expression. We established UUO‐induced renal fibrosis in mice and then administered with nodakenin orally ata a dose of 1 and 10 mg/kg. The in‐vivo results indicated that nodakenin protected obstructive nephropathy through its anti‐inflammatory and anti‐fibrotic properties. Nodakenin prevented the infiltration of inflammatory cells, alleviated the levels of pro‐inflammatory cytokines, reduced the polarization of macrophages and down‐regulating the aberrant deposition of extracellular matrix at the site of injury. Of note, nodakenin dramatically impeded Smad3, NF‐κB p65 phosphorylation and Snail1 expression. In line with in vivo studies, nodakenin suppressed the expression of Snail1, Smad3 phosphorylation and fibrogenesis in TGF‐β1‐treated renal epithelial cells in‐vitro. Furthermore, we found that the effect of nodaknin against fibrosis was reversed in Snail1 overexpressing cells, whereas nodakenin could not further reduce expression of fibrogenesis in Snail1 silenced cells, suggesting that nodaknein may function through a Snail1‐dependent manner. Collectively, this study reveal a critical role of nodakenin in the cure of renal fibrosis.

It is a widely held view that TGF-β/Smad signalling plays a critical role in renal fibrosis. 3 Transforming growth factor (TGF-β)1 functions in various cell types including renal tubular epithelial cells (TECs) and initiates action via direct (Smad) and indirect (non-Smad) pathways to cause renal fibrosis. 4,5 Direct blockade of TGF-β1 signal transduction with anti-TGF-β antibody seems to be an effective strategy.
However, researchers recently have proved that administration of TGF-β1 neutralizing monoclonal antibody failed to alleviate the progression of diabetic nephropathy. 6 In view of this, treatment of renal fibrosis should focus specifically on the downstream molecules related to fibrosis, rather than blocking the general effect of TGF-β1.
As the critical cellular mediator of TGF-β1, SMAD family member 3 (Smad3) plays a vital role in renal fibrosis by regulating the transcription of fibrogenic genes within α-smooth muscle actin (α-SMA). 7 Snail1 (encoding snail family zinc finger 1, known as Snail1), activated by Smad3/4 and cooperation of Smad3/4, has been reported to work as a transcription factor to control the expression of junction components like E-cadherin, claudins and so on. Collectively, inhibition of Smad3 or Snail1 may be a potential therapeutic target for renal fibrosis.
In recent years, drugs isolated from the medicinal herbal have drawn increasing attractions from the scientific researchers as an alternative therapy for the prevention and treatment of various renal disease. [8][9][10][11] Nodakenin, a coumarin derived from the roots of Angelicae gigas, has been reported to treat against multiple diseases including alleviating inflammation to treat against lethal endotoxin shock and allergic inflammation, enhancing cognitive function and adult hippocampal neurogenesis and so on. [12][13][14] However, little reports have been suggested the effect of nodakenin against renal fibrosis.
Considering the opposite roles on renal fibrosis between nodakenin and Snail1, we hypothesized that nodakenin protect against renal fibrosis by manipulating Snail1 expression. Hence, the goals of this study are to reveal the renoprotective effect of nodakenin on UUO and whether by targeting Snail1. In the current study, we have disclosed that the effects of nodakenin on the inflammation and fibrosis in vitro and in vivo. Furthermore, we also examined the underlying mechanisms by lentivirus-mediated up-regulation or siR-NA-induced down-regulation of Snail1.

| UUO-induced renal fibrosis model and nodakenin treatment
The UUO-induced renal fibrosis model was constructed in the male BALB/C mice (6 to 8-week-old, 20-25 g) as described previously. 15 After anaesthetizing with pentobarbital, the left ureter was obstructed by two-point ligations and snipping the midline of the ureter. The mice were randomly divided into the following groups (n = 6 per group): (a) sham group; (b) UUO with nodakenin at a dose of 1 mg/kg, 10 mg/kg or vehicle solvent (10% tween-80) (in which mice received 7 consecutive days of nodakenin/vehicle solvent). The next day after surgery, the mice were received with drug treatment by daily gavage and sacrificed at day 7 after treatment.
For western blotting and Quantitative-PCR, mice kidneys were excised and stored in liquid nitrogen before usage. All experimental manipulations were approved by the Ethics Committee for Animal Experiments of Southwest Medical University.

| Cell culture, lentivirus infection, siRNA transfection and MTT assay
Rat renal tubular epithelial cell line (NRK-52E) was obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China).
NRK-52E cells were cultured in DMEM (high glucose) supplemented with 5% (v/v) FBS (Gibco, USA), 100 U/ml penicillin/streptomycin (Life Technologies, USA). For in vitro cellular assay, DMSO was used as a solvent to dissolve nodakenin, and the final concentration of DMSO in the medium was under 0.1%.
The gene of the rat snail1 coding sequence region was synthesized by Sangon (Shanghai, China) and cloned into pLVX-IRES-EGFP plasmid. To make lentiviruses, DNA vectors were transfected into human 293T cells. Next, medium supernatant was harvested, centrifuged at 72 000 g for 2 hours (Optima XPN-100, Beckman, USA) and the sediments were resuspended in fresh medium, stored at −80 °C before usage. Finally, lentiviruses were used to infect NRK-52E cells and 3 days later, 1.5 μg/ml puromycin (Solarbio, China) was added to the medium for 7 days to obtain a stable snail1-overexpression cell lines. Cells within 3-10 passages were used for further research. MTT assay was conducted as described previously. 16 In brief, 2 × 10 3 cells were seed in 96-well plates and incubated overnight before treatment. Thereafter, cells were received with various concentrations of nodakenin (1-80 μmol/L) for 48 hours.
Then, the supernatant was discarded and 100 μL of MTT solvent (0.5 mg/mL) was added for incubation. After indicated time (4 hours), the supernatant was again discarded and 150 μL of DMSO was added to dissolve the deposits for 15 minutes. Finally, The absorbance was obtained by a microplate reader (VT, Biotek, USA).

| Histological analysis and immunofluorescence staining
The mice kidneys were fixed with paraformaldehyde, embedded in paraffin and sectioned at 4 μm for further analysis. HE staining and Periodic acid-Schiff staining (PAS) was firstly conducted to evaluate the tubular injury level. In brief, the percentage of cortical tubular injury with 0 to 4 grading scale: 0, normal; 1, less than 25%; 2, 25% to 50%; 3, 50% to 75%; and 4, >75%. 17 Next, we performed Masson's trichrome staining and Sirius staining to assess the degree of fibrosis according to the manufacturer's instructions (Nanjing Jiancheng, China). Immunohistochemistry was performed with a microwave-based antigen retrieval technique. 18 After incubating with the antibody against α-SMA (1: 100, Boster, China), the slides were exposed with Biotin-Streptavidin HRP-based SPlink Detection Kits (ZSGB-Bio, China) and counterstained with haematoxylin. The sections were then photographed with Virtual Slide Microscope (VS120, Olympus, Japan).
For immunofluorescence staining, the kidneys were firstly fixed in 4% paraformaldehyde at 4 °C overnight, followed by dehydrating in 10% to 30% sucrose and embedding in OCT. 5 μm sections were prepared by using a Leica cryostat. The sections were then blocked with 3% goat serum, incubated with anti-F4/80 antibodies (Santa Cruz, USA), followed by incubation with FITC-conjugated secondary antibodies. After rinsing with PBS for 3 times, the sections were immediately photographed with confocal microscopy (A1R-PLUS, Nikon, Japan).

| Western blot
Total proteins were extracted from cells or mice kidneys using RIPA lysis buffer (Beyotime, China) and protein concentrations were quanti- anti-Fibronectin/Collagen I antibody, Abcam, USA), followed by HRPconjugated secondary antibodies for 1 hour at room temperature.
Signals were detected with ChemiDoc TM (Bio-Rad, USA) and quantified using Gel-Pro analyser (Media Cybernetics, USA).

| Flow cytometry for detection of macrophage polarization and the expression levels of E-cadherin
For the detection of macrophage polarization, mice kidneys were harvested, digested with Blendzyme 4 (Roche, USA) and grinded into cell suspension as previously report. 19 In order to detect the expression levels of E-cadherin, NRK-52E cells were digested with 0.25% trypsin.

| Statistical analysis
All the data were expressed as the mean ± standard deviation (SD).
Statistical analyses were performed with student's t-test or one-way ANOVA with SPSS 20.0 software. P < 0.05 was considered to be statistically different.

| Nodakenin treatment alleviates renal fibrosis
To explore whether nodakenin could affect the process of pathologic changes, a UUO-induced mouse renal fibrosis model was constructed. As shown in Figure 1A, the obstructed kidney in solvent-treated mice displayed a pale colour compared with the sham group, suggesting a weaker blood supply. Whereas nodakenin treatment significantly reversed the impact. Meanwhile, we also investigated the expression of KIM-1, a biomarker of proximal tubular injury by western blot. As shown in Figure 1B Figure 1D,E). Moreover, we also investigated the blood biochemical indexes in each group. In line with the outcomes of the kidney pathologic changes, the levels of creatinine and bun after nodakenin treatment restored the aberrant alterations in a dose-dependent manner ( Figure 1F,G). Collectively, these results suggested that nodakenin treatment could effectively alleviate UUO-induced the pathologic changes.
Next, we determined whether nodakenin treatment could inhibit kidney fibrosis. Firstly, Masson trichrome staining and Sirius red staining was performed to investigate the degree of fibrosis in various groups. As presented in Figure 2A, the results revealed that abundant extracellular matrix deposition in UUO kidneys.
Overall, these results indicated that nodakenin obviously improved renal fibrosis at least partially ascribed to suppress the activation of myofibroblast cells.

| Administration of nodakenin attenuates UUOinduced inflammatory responses in mice
Inflammation responses have the critical effect on renal fibrosis as a priming factor, and the macrophage infiltration in renal fibrosis has a dominant role in the production of chemokines such as TNF-α, IL-1β, iNOS. 21,22 To investigate whether nodakenin displayed an antiinflammation effect, the macrophage infiltration, polarization and pro-inflammatory cytokine expression in the mice kidneys were detected by advantage of flow cytometry, immunofluorescence assay, quantitative PCR and western blot. As seen in Figure 3A, the results of flow cytometry indicated that nodakenin treatment dramatically decreased the infiltration of the macrophage. Besides, flow cytometry analysis suggested that nodakenin treatment also reduced M1 macrophage polarization, which is characterized by the production of a range of pro-inflammatory molecules such as iNOS. In line with flow cytometry, immunofluorescence assay also revealed that nodakenin treatment significantly decreased the macrophage infiltration compared with UUO group by in situ immunofluorescence (data presented in Figure 3B). Next, we investigated pro-inflammatory cytokine expression by quantitative PCR and western blot. As expected, nodakenin treatment reduced the levels of pro-inflammatory cytokines in a dose-dependent manner ( Figure 3C-G). Notably, nodakenin treatment reversed the aberrant increase of NF-κB p65, which lays an upstream of chemokines ( Figure 3F,G). Taken together, these results demonstrated that nodakenin could alleviate UUOinduced inflammatory responses in mice.

| Nodakenin inhibits fibrosis in vitro
Tubular epithelial cells (TECs) play a crucial role during the development of renal fibrosis progression. 23 The molecular structure of nodakenin was presented in Figure 4A. Considering the potential cytotoxicity, we first evaluated the cytotoxicity of nodakenin by MTT assay. As shown in Figure 4B, no obvious injury was observed under 40 µM nodakenin for 48 hours. We then investigated the effects of nodakenin in TGF-β1 induced fibrosis in rat tubular epithelial cells.
As shown in Figure 4C, the results indicated that after stimulation

| Nodakenin inhibits Smad3 phosphorylation and Snail1 expression in vitro and in vivo
Based on the above results, we have preliminarily concluded that effect of nodakenin on renal fibrosis, but the specific mechanism remains to be clarified. Hence, we investigated the potential critical pathway in vitro and in vivo. We first evaluated the expression of phosphorylated smad3 and snail1 by immunohistochemical staining. As shown in Figure 5A, a dramatic increase of Snail1 and phosphorylated smad3 was observed in renal tubules of UUO model.
Meanwhile, we also observed an obvious translocation of phosphorylated smad3 from the cytoplasm to nuclei. Whereas, nodakenin treatment reversed the above phenomenon. We also evaluated the indicated protein levels in vitro and in vivo by western blot.
Consistent with immunohistochemical staining, the results revealed that snail1, TGF-β1 and phosphorylated smad3 were significantly blunted both in cell-culture samples and tissue samples (details listed in Figure 5B-E).

| Nodakenin alleviates the fibrosis via a Snail1dependent mechanism in renal tubules
To further elucidate whether nodakenin functions against renal fibrosis by suppressing the Snail1 expression, we overexpressed and silenced the Snail1 respectively. We first evaluated the fluorescence intensity of cells after transduction by converted microscopy. As shown in Figure 6A, the cells after lentivirus transduction induced a marked increase in fluorescence, suggesting successful transduction. Next, we measured the levels of Snail1 by western blot after puromycin treatment. The results indicated that Snail1 was obviously increased compared with the control group (details listed in Figure 6B,C). On the above basis, we determined the ef-

| D ISCUSS I ON
Nodakenin, a small-molecule coumarin isolated from the roots of  The transcription factor Snail1 is a crucial protein in EMT and is mediated by multiple pathways. 26 Of note, NF-κB can induce the expression and stabilize Snail1 protein. In addition, Snail1 can also be activated by Smad3/4 complex. After that, Snail1 provoked the progression of fibrogenesis. Given the previous observations, inhibition of Snail1 protein should be an effective way to inhibit fibrosis. In this study, we found that nodakenin impacted the phosphorylated smad3.
Meanwhile, findings from our research indicated that nodakenin is a potent inhibitor of Snail1, which suggested that nodakenin relieve the fibrosis at least partly by the suppression of Smad3/Snail1.
In conclusion, this study has shown that nodakenin can significantly inhibit UUO-induced renal fibrosis in mice and TGF-β1treated renal epithelial cells by regulating NF-κB and Smad3 induced Snail1 expression, subsequently improving the inflammation and fibrogenesis in the obstructive kidney. The present study lays the groundwork for future research into a natural therapeutic agent for the treatment of renal fibrosis.

CO N FLI C T O F I NTE R E S T
The authors declare that there are no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.