Renal fibrosis caused by deposition and accumulation of extracellular matrix (ECM) components is the common final pathway of chronic kidney disease. Multiple cellular events are involved in this process, including fibroblast activation, epithelial to mesenchymal transition (EMT), inflammatory cells infiltration, and tubular epithelial cell apoptosis (Zeisberg and Neilson, 2010; Liu, 2011). Currently, there are no generally effective treatments for preventing the progression of renal fibrosis (Nogueira et al., 2017).

Osthole (OST, 7-Methoxy-8-(3-methylbut-2-enyl)-2-chromenone) is a derivative of coumarin mainly found in plants of the Apiaceae family, Cnidium monnieri and Angelica pubescens, which are commonly used in traditional Chinese medicine (You et al., 2009; Zhang Z.R. et al., 2015). The pharmacological mechanism of its action is still under investigation, but several potentially therapeutic effects have been demonstrated. It has been found to reduce tumor progression (Lin et al., 2014; Feng et al., 2017), have antibiotic properties (Wang et al., 2009), reduce allergic reactions (Chiang et al., 2017) and ameliorate osteoporosis (Zhang et al., 2017). Previous studies have indicated that OST may reduce fibrosis in the lung, liver and heart (Chen et al., 2011; Liu et al., 2015; Hao and Liu, 2016), but in renal tissue it is less studied. Thus, the present investigation was to focus on the effect of OST on renal tissue.

Fibrogenesis is the deposition of pathological matrix. It is widely recognized that local fibroblast activation featured with proliferation and phenotypic appearance of myofibroblasts (α-SMA⁺ fibroblasts) mainly contributes to inappropriate matrix expansion and consequent renal structural deformations (Barnes and Gorin, 2011; Liu, 2011; LeBleu et al., 2013). Partial EMT is a feature of kidney fibrosis with tubular epithelial cells acquiring mesenchymal cell marker but the preserved integrity of the renal tubules (Grande et al., 2015). EMT program of renal epithelial cells and consequent G2 cell-cycle arrest promote kidney fibrosis through an altered secretome profile, leading to the accumulation of interstitial myofibroblasts and the compromised kidney parenchyma function (Lovisa et al., 2015; Djudjaj et al., 2017; Gewin, 2018). Transforming growth factor -beta1 (TGF-β1) is one of the most important profibrotic cytokines in chronic kidney disease (Meng et al., 2016), and exerts its effects mainly through the Smad pathway (Roberts et al., 1986; Desmouliere et al., 1993). Upon TGF-β1 stimulation, Smad2 and Smad3 are phosphorylated and activated, then form complex with Smad4 and translocate to the nucleus where they function as transcription factors for many fibrosis associated genes (Miyazono et al., 2000; Verrecchia et al., 2001). Activation of Smad3, especially in tubular epithelial cells or fibroblasts, is clearly involved in renal tubulointerstitial fibrosis through enhancing of EMT program, inflammatory cells influx and collagen accumulation (Sato et al., 2003; Meng et al., 2010). Smad7, an intracellular antagonist for TGF-β signaling, is degraded in progression of tubulointerstitial fibrosis (Fukasawa et al., 2004). Dysregulation of Smad7 promotes renal fibrosis (Chung et al., 2013) and conversely Smad7 treatment substantially attenuating progressive kidney diseases via inactivating TGF-β/Smad3 signaling (Zhou et al., 2016). Profibrotic cytokines secreted from inflammatory cells and injured tubular cells build up an inflammatory state in the kidney, arguably providing a directional or indirectional signal to trigger activation of resident fibroblasts and phenotypic transition of tubular cells (Liu, 2011). NF-kappaB (NF-κB), the major inflammatory response pathway (Blackwell and Christman, 1997), generates a loop that maintains the inflammatory signals is also the driving force for renal fibrosis (Miyajima et al., 2003; Tan et al., 2008). Indeed, increasing the levels of the endogenous inhibitor of NF-κB, I-κB, or inhibiting the pathway diminishes the expression of inflammatory genes, inhibits EMT phenotypes and decreases deposition of interstitial ECM (Inoue et al., 2010; Li et al., 2011).

The aim of this study was to assess the effect of OST on renal fibrosis in the mouse model of renal interstitial fibrosis induced by unilateral ureteral obstruction (UUO) and to examine if it could work through inhibiting renal fibroblast proliferation and EMT program. The main signaling pathways that involved in renal fibro-genesis such as TGF-β/Smad and the NFκB pathway were also investigated.

The findings of present study show that (1) treatment with the coumarin derivative osthole (7-Methoxy-8-(3-methylbut-2-enyl)-2-chromenone) reduces fibroblast activation and proliferation in vitro and in vivo; (2) OST treatment blocks EMT program in injured kidney from UUO mouse; (3) OST administration suppresses activation of TGF-β1/Smad signaling pathway in fibroblasts and renal tissue (4) as well as inflammatory response in obstructed kidneys after UUO, which contribute to renal fibrosis. Overall, these data reveal that in addition to the anti-inflammation property of OST in various renal diseases (Zheng et al., 2013; Song W. et al., 2014; Huang and Dong, 2017), OST mitigates UUO-induced renal fibrosis mainly through reduction of fibroblast activation and EMT, at least in part, by inhibiting the TGF-β1/Smad signaling pathway.

OST, an ingredient commonly used in traditional Chinese herb, has been known for their potent anti-inflammatory and anti-tumor actions (Liao et al., 2010; Shokoohinia et al., 2018). Recently, studies reported that OST could diminish extracellular matrix deposition in hepatic and myocardial tissues (Chen et al., 2011; Liu et al., 2015). Consistently, TGF-β1-induced cardiac fibroblast activation and production of ECM components were blocked by OST (Liu et al., 2017). Given the conception that activation and proliferation of renal resident interstitial fibroblasts are the main events in the origin of myofibroblasts, the primary matrix-producing cells (Strutz and Zeisberg, 2006), cultured renal interstitial fibroblasts were exposed to TGF-β1, the central mediator of renal fibrosis, or serum, a mixture of multiple growth factors, to investigate the anti-activation and anti-proliferation role of OST. Our results clearly showed that OST inhibited TGF-β1-induced NRK-49F activation with reduced expression of α-SMA and production of ECM, fibronectin and type I collagen. In addition, OST suppressed proliferation of NRK-49F, as verified by the decreased the cell number and DNA synthesis rate stimulated by FBS. Cell proliferation is typically regulated by multiple cell cycle proteins, and the effects of OST in modulating cell cycle- and proliferation-associated genes have been demonstrated in previous studies (Jarzab et al., 2014; Wang et al., 2015; Hu et al., 2016; Li et al., 2017). In this study, OST prevented FBS-induced up-regulation of cyclin D1 (a key regulator of G1 to S transition) and PCNA (cell proliferation marker) as well as down-regulation of p21 Waf1/Cip1 (the negative regulatory factor of cell cycle progression) in FBS-induced NRK-49F. Collectively, our data indicated that OST attenuated phenotypic transformation and proliferation of fibroblasts in vitro.

In vivo, the effect of OST on renal interstitial fibrosis was investigated in mice challenged with UUO, an animal model featured by chronic renal injury and fibrosis in the tubulointerstitial compartment of the kidney. Our data revealed that OST (40 and 80 mg/kg/day) ameliorated fibrotic lesions with less collagen deposited in the renal interstitium and lower presence of α-SMA⁺ myofibroblasts. As in cultured cells, in obstructed kidneys from UUO mice, myofibroblast proliferation, measured by co-staining for Ki-67 and α-SMA, was inhibited after OST treatment. Together, the findings in vitro and in vivo showed that regulating activation and proliferation of renal fibroblasts may be an important therapeutic role for OST treatment in progressive renal fibrosis.

Smad signaling, also known as the canonical TGF-β pathway, is critically involved in renal fibrosis by inducing activation of myofibroblasts, transformation of tubular epithelial cells to myofibroblasts and tubular apoptosis (Sato et al., 2003; Zhang et al., 2015; Guo et al., 2018). In this pathway, TGFβ1 mainly transduces its fibrotic signal through activation of TGFβreceptor 1 (TGFβR1) followed by phosphorylation and nuclear translocation of Smad3, thus driving the progress of renal fibrosis. During fibrogenesis, Smad3 was generally overactivated, while Smad7, an inhibitory Smad has been shown to be degraded (Fukasawa et al., 2004). In the present study, our experiments reproduced this phenomenon in TGF-β1-stimulated NRK-49F cells as wells as UUO injury-induced kidneys and OST treatment reversed them. The in vitro modulatory effect of OST on TGFβ/Smad signaling was in line with previous study that OST increased Smad7 expression and decrease Smad3 expression in TGF-β1-treated cardiac fibroblasts (Liu et al., 2017), and may explain the decreased phenotypic transformation in TGF-β1-stimulated NRK49F when pre-incubation with OST. However, it should be noted that the non-Smad signaling, such as MAPK, AKT, and Wnt pathway, could also be the mechanism for OST in inhibiting kidney fibroblast activation. In addition, strategy targeting rebalancing TGF-β/Smad3/Smad7 signaling in vivo has been demonstrated to be effective for reducing renal fibrosis in obstructive nephropathy (Chung et al., 2013; Zhou et al., 2016). Thus, our data provide evidence that OST may act as a regulator of TGF-β/Smad signaling, thereby attenuates interstitial fibrosis in UUO kidney.

Epithelial-mesenchymal transition, a process defined as epithelial cells acquiring of mesenchymal features and motile phenotype, normally occurs in embryonic development, tissue fibrosis, and cancer (Kalluri and Weinberg, 2009). A complete EMT normally concurred with the phenomenon that the epithelial cells lose their polarity and cell–cell adhesions through downregulation of E-cadherin and gain the ability to migrate and invasion with a spindle-shaped mesenchymal morphology. It was recently reported that the injured renal epithelial cells undergoing a partial EMT process, without directly generating interstitial myofibroblasts, result in prolonged proliferation, cell cycle arrest, secretion of profibrotic factors and abnormal metabolic rearrangements, thus promoting fibrosis and parenchymal damage (Grande et al., 2015; Lovisa et al., 2016). The functional significance of EMT program in renal fibrosis was evidenced by the improved tubular health and a lower degree of interstitial fibrosis in the mice with two master regulators of EMT, Snail 1 or twist 1, was genetically deleted in proximal tubular cells, when they were challenged with different models of renal fibrosis (Lovisa et al., 2015). Vimentin, a member of the intermediated filament family of proteins as well as the marker of mesenchymal cells, has been demonstrated to be important in the development of EMT in renal tubular cells (Wang et al., 2018). Normally vimentin is not present in epithelial cells but re-expresses in injured or dedifferentiation tubular cells (Witzgall et al., 1994; Rastaldi et al., 2002; Tan et al., 2009; Kusaba et al., 2014). Previous studies in tumor cell lines had revealed that OST was able to reverse EMT program with inhibiting EMT transcription factor Snail or Twist expression. As oncogenic EMT and organic fibrosis share many of the same important signaling pathways and transcription factors that implicated in the development of EMT, it is likely that OST attenuated renal fibrosis through suppression of EMT (Lin et al., 2014; Feng et al., 2017). In support of this hypothesis, our current study showed that OST rescued UUO injury-induced EMT with increased expression of E-cadherin and decreased expression of Vimentin in tubular cells of kidneys, and concomitantly the less expression of Snail 1, Twist and p-H3. Herein, the results regarding cell arrest at G2/M phase in our study were inconsistent with previous studies where OST had been shown to induce G2/M arrest in tumor cells (Xu et al., 2011; Lin et al., 2017). It was reported that G2/M arrest associated-cell cycle protein p21 was differently regulated by EMT transcription factors, snail or twist in epithelial and tumor cell lines (Lovisa et al., 2015; Srivastava et al., 2016; Yang et al., 2017). Different cell type as well as the context of the diseases may explain some of this discrepancy. Altogether, our data provides evidence that the EMT-like program in renal tubular cells provoked by UUO surgery was attenuated after OST treatment. However, in this experiment setting, we did not directly address how these tubular cells that undergoing partial EMT engaged in the progress of renal fibrosis, which still needs to be further warranted.

The chronic inflammatory microenvironment characterized by up-regulation of inflammatory cytokines and growth factors are required for the development and progression of organ fibrosis (Misseri et al., 2004; Meng et al., 2014). It has been demonstrated that OST largely inhibited inflammation response in multiple experimental animal models, including chronic kidney disease (Huang and Dong, 2017). Our study echoed with these studies that OST exhibited an anti-inflammatory property by reducing mRNA expressions of the cytokines inflammatory cytokines such as TNF-α, IL-6, IL-1β, and ICAM. Additionally, OST suppressed proinflammatory transcription factor, NF-κB activation in renal tissue of UUO mice, which was not surprised in our context as previous studies has showed that OST could function as an inhibitor of NF-κB (Hua et al., 2013; Yang et al., 2014; Xie et al., 2015). Furthermore, NF-κB could induce transcription and stabilization of the Snail1 protein (Wu et al., 2009; Berzal et al., 2015). Thus, NF-κB/Snail as well as TGF-β/Smad3/Snail1 axis maybe the underlying mechanisms for the anti-EMT action of OST we demonstrated in this study.

In summary, this study demonstrates that OST exhibites antifibrotic effects on obstructed nephropathy. The antifibrotic mechanisms of OST are associated with inhibition of fibroblast activation and proliferation, preservation of E-cadherin expression, and inactivation of the TGF-β/Smad and NF-κB signaling pathways. Despite the pleiotropic effects of OST on renal fibrosis indicated in our study, more detailed sets of investigation are warranted for the exact molecular basis and pharmacological action.

All authors have seen and approved the final version of the manuscript. SZ, JT, and EL conceived and designed the studies. SZ and QH performed most of the experiments with animal tissues and cultured cells, the statistical analysis, and contributed intellectually to the writing of the manuscript. XCai, SJ, NX, QZ, and XCao made the animal model and performed qPCR. SZ, MH, JT, and EL contributed to the conception of the article, the data interpretation, drafting of the manuscript, and revised the article for important intellectual content.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.