Inhibitory Effect of Tanshinone IIA on Rat Hepatic Stellate Cells

Background Anti-inflammation via inhibition of NF-κB pathways in hepatic stellate cells (HSCs) is one therapeutic approach to hepatic fibrosis. Tanshinone IIA (C19H18O3, Tan IIA) is a lipophilic diterpene isolated from Salvia miltiorrhiza Bunge, with reported anti-inflammatory activity. We tested whether Tan IIA could inhibit HSC activation. Materials and Methods The cell line of rat hepatic stellate cells (HSC-T6) was stimulated with lipopolysaccharide (LPS) (100 ng/ml). Cytotoxicity was assessed by MTT assay. HSC-T6 cells were pretreated with Tan IIA (1, 3 and 10 µM), then induced by LPS (100 ng/ml). NF-κB activity was evaluated by the luciferase reporter gene assay. Western blotting analysis was performed to measure NF-κB-p65, and phosphorylations of MAPKs (ERK, JNK, p38). Cell chemotaxis was assessed by both wound-healing assay and trans-well invasion assay. Quantitative real-time PCR was used to detect gene expression in HSC-T6 cells. Results All concentrations of drugs showed no cytotoxicity against HSC-T6 cells. LPS stimulated NF-κB luciferase activities, nuclear translocation of NF-κB-p65, and phosphorylations of ERK, JNK and p38, all of which were suppressed by Tan IIA. In addition, Tan IIA significantly inhibited LPS-induced HSCs chemotaxis, in both wound-healing and trans-well invasion assays. Moreover, Tan IIA attenuated LPS-induced mRNA expressions of CCL2, CCL3, CCL5, IL-1β, TNF-α, IL-6, ICAM-1, iNOS, and α-SMA in HSC-T6 cells. Conclusion Our results demonstrated that Tan IIA decreased LPS-induced HSC activation.


Introduction
Liver fibrosis is a wound-healing, scarring response to such chronic injuries as hepatitis B and C, alcoholic liver disease, nonalcoholic steatohepatitis, and autoimmune hepatitis [1,2]. Hepatic fibrosis has been recognized as a dynamic and at some stages, reversible process [3]. In the normal liver, hepatic stellate cells (HSCs) are myofibroblasts located in the space of Disse. Quiescent HSCs are primarily responsible for storing huge amounts of vitamin A in lipid droplets [4][5][6]. Activation of HSCs is the contributing cause of liver fibrogenesis and characterized by phenotypic transformation with diverse functional changes, including proliferation, contractility, cytokine secretion, chemotaxis, fibrogenesis, and matrix degradation [6]. Namely, HSCs are induced to trans-differentiate from quiescent into activated form by inflammatory mediators or growth factors [1,5,7]. Lipopolysaccharide (LPS, or endotoxin) is a major component of the outer membrane of Gram-negative bacteria. It has been reported that LPS levels in patients with cirrhosis are increased in both systemic and portal venous blood [8,9]. LPS is the ligand of the pattern recognition receptor, Toll-like receptor 4 (TLR4), and powerfully stimulates host innate immune responses [8,9]. LPS instigates the consequent signals downstream and leads to the activation of the transcription factors NF-kB and AP-1, resulting in the induction of potent innate immune responses [10,11]. Scientific reports mentioned that myofibroblasts are also critical innate immune sensors induced by LPS and can themselves generate various inflammatory effectors including chemokines, cytokines, and oxygen radicals [12]. LPS triggers high levels of TLR4 expression in HSCs [13,14]. In both in vitro and in vivo studies, gut derived LPS is an important activator for Kupffer cells and HSCs in liver injury [14][15][16].
Salvia miltiorrhiza Bae. (S. miltiorrhiza), the root of Labiatae plant, has been used to treat heart, metabolic diseases and hepatitis for a long time in Chinese medicine [17]. Several reports including ours have shown that S. miltiorrhiza extracts exert both in vitro and in vivo inhibitory effects against various models of hepatic fibrosis [18][19][20]. Salvianolic acid B (C 36 H 30 O 16 , Sal B) is a major component in aqueous hydrophilic extract of S. miltiorrhiza. Sal B has been demonstrated to inhibit the activation of ERK and NF-kB in TNF-a-stimulated endothelial cells [21]. Previous studies indicated that S. miltiorrhiza extracts inhibited carbon tetrachloride (CCl 4 )-induced P450 2E1 and iNOS protein in rats. Furthermore, Sal B suppressed CCl 4 -stimulated accumulation of reactive oxygen species (ROS) in hepatocytes and HSCs, and also HSC activation [22,23]. Tanshinone IIA (C 19 H 18 O 3 , Tan IIA) is one of the main lipophilic compounds of S. miltiorrhiza. In an in vitro study, Tan IIA has been revealed to suppress production of NO, expressions of iNOS and IL-1b in activated RAW 264.7 cells [24]. An in vivo study showed that Tan IIA induces activation of the cytochrome P450 mixed-function oxidase system in C57B/6J mice [25,26].
There is still no clinically efficacious drug specifically against liver fibrosis [2]. The potential of Tan IIA as an agent against liver fibrosis is not clarified. The present study was undertaken to explore the potential effect of Tan IIA on LPS-induced HSCs, in comparison with Sal B.

Preparation of drugs
Tan IIA was obtained from Sigma-Aldrich, and Sal B was isolated from the roots of S. miltiorrhiza, as previously reported [23]. For in vitro experiments, Tan IIA and Sal B were dissolved in dimethyl sulfoxide (DMSO, Sigma-Aldrich) and diluted with medium to give a DMSO concentration below 0.1%. LPS (100 ng/ml, Sigma-Aldrich) was used in all experiments for stimulation.

Cell viability assay
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay was performed to exclude the cytotoxicity of drugs to HSCs. Cells (7.5610 4 cells/well) were seeded in 24-well plates with FBS-free medium for 24 hr, and with Tan IIA (1, 3 and 10 mM) and Sal B (200 mM) for another 24 hr at 37uC. After incubation with drugs, MTT assay was used to evaluate the cell viability of HSC-T6 cells. The cells were incubated with minimum essential medium containing 0.1 mg/ml MTT (Sigma-Aldrich) for 30 min. The formazan particle was dissolved with DMSO and measured using an enzyme-linked immunosorbent assay reader, according to the method of our previous report [27].
NF-kB responsive luciferase assay Cells (7.5610 4 cells/well) were seeded in 24-well plates with FBS-free medium at 37uC the day before transfection. Transfection was performed with transfection reagent Fugene-6 (Roche, Basel, Switzerland) according to manufacturer's specifications. NFkB-Luc reporter construct (1 mg/well) (Stratagene, La Jolla, CA, USA) was added to cells along with plasmid CMV-bgalactosidase (CMV-b-gal, 0.2 mg/well; Promega, Madison, WI, USA) and transfection reagent (15 ml/24-well plate). CMV-b-gal served as an internal control to normalize the transfection efficiency. After 24 hr incubation, cell cultures were treated with Tan IIA (1, 3 and 10 mM) or Sal B (200 mM) for 30 min. LPS was then added to stimulate NF-kB activity for 6 hr. Cell lysates were harvested with diluted reporter lysis 5X buffer (Promega). 24-well plates were added reporter lysis buffer (100 ml/well), and 20 ml cell lysate mixed with 100 ml luciferin in white 96-well plates. Luminescence was detected by luminometer-VICTOR2 Multi-label Counter (Perkin Elmer Inc., Waltham, MA, USA), as previously reported by us [27,28].

Western blotting analysis
HSC-T6 cells were seeded on 6 well-plates (1610 6 cells/well). After culturing for 24 hr, serum-starved cells were pretreated with Tan IIA (10 mM) or Sal B (200 mM) for 1 hr, then exposed to LPS for 15 min for harvesting NF-kB-p65 and MAPKs (ERK, JNK and p38), and exposed to LPS for 24 hr for harvesting a-SMA. Nuclear extracts of HSCs for NFkB-p65 translocation were prepared according to our previous study [29]. Whole cell protein extracts for measurement of ERK, JNK, p38 and a-SMA proteins were harvested with RIPA buffer. Cell lysates were separated with

Immunofluorescent staining
HSC-T6 cells were seeded on glass cover slips in 6 well-plates (1610 6 cells/well) with medium at 37uC for 1 day incubation. Serum-starved cells were pretreated with Tan IIA (10 mM) or Sal B (200 mM) for 1 hr, then exposed to LPS for 15 min. The cells were fixed, permeabilized, and blocked. Samples were incubated with an antibody against NFkB-p65 (1:500, Santa Cruz Biochnology, Santa Cruz, CA, USA) incubated in 4uC overnight, and then mounted with 49,6-diamidino-2-phenylindole dihydrochloride (DAPI, Sigma-Aldrich). The cells were visualized on LSM780 confocal microscope (Zeiss, Oberkochen, Germany) using a digital imaging system. The scale bar was added using Metamorph software.
Wound-healing assay HSC-T6 cells were seeded on collagen coated 24-well plates [28,30,31] at a density of 9610 4 cells/well. After 24 hr serum starvation, a monolayer of cells was scratched by a 200 ml tip in serum-free medium. Cells were pre-treated with Tan IIA (10 mM) and Sal B (200 mM) for 1 hr, then incubated in the presence of LPS for 24 hr. HSC-T6 cell migration was quantified by the area of migrated cells to the scratched cell-free zone after 24 hr, and measured by Image-Prop PLUS software.

Trans-well invasion assay
Serum-starved HSC-T6 cells (1610 5 cells/trans-well) were cultured on 24-well chemotaxis chamber Millicell (Merck Millipore, Darmstadt, Germany) with 8 mM pores, which were precoated Matrigel Matrix (100 ml/trans-well; BD Biosciences, San Jose, CA, USA) for 1 hr at room temperature, and then washed with serum-free medium. The bottom wells were filled with serumfree medium co-treating with LPS plus Tan IIA (10 mM) or Sal B (200 mM) for 24 hr, and the entire chamber was incubated at 37uC for 24 hr. Cells migrated to the lower surface of the membrane were stained with hematoxylin (Sigma-Aldrich) to define the cell nuclei, as previously reported by us [32].

Quantitative real-time polymerase chain reaction analysis
HSCs (1610 6 cells/well) were seeded on 6 well-plates. After culturing for 24 hr, serum-starved cells were pretreated with Tan IIA (10 mM) or Sal B (200 mM) for 1 hr, then exposed to LPS for 12 hr. Total RNA was extracted from HSC-T6 cells using the TRIzol Reagent (Life technologies, Carlsbad, CA, USA). RNA was prepared by reverse transcription using oligo-dT and dNTP, and each sample was added with RT kit, Life technologies). Quantitative real-time PCR was performed by the StepOnePlus Quantitative Real-Time PCR System (Life technologies), according to the manufacturer's instructions. The sequences of primers for quantitative real-time PCR are listed in Table S1.

Statistical analysis
Data were expressed as mean 6 S.E.M. Statistical analysis was obtained using one-way ANOVA. A p value less than 0.05 was considered as significantly different.

Results
Tanshinone IIA has no cytotoxicity in HSC-T6 cells Both Sal B and Tan IIA ( Figure 1A and B) are key constituents of S. miltiorrhiza, the former coming from aqueous extract and the latter being in lipophilic part. According to our previous studies, Sal B has already been found to protect hepatic function, such as inhibiting CCl 4 induced ROS accumulation in hepatocytes and HSC activation [21,23]. Therefore, we further studied bioactivities of Tan IIA against HSC activation. Using MTT assays, neither Tan IIA (1, 3 and 10 mM) nor Sal B (200 mM) showed cytotoxicity to HSC-T6 cells for 24 hr ( Figure 1C). To confirm whether there was proliferative effect of Tan IIA and Sal B in HSC-T6 cells, we used immunofluorescent staining of Ki67. The results show that there was no increase in proliferation by Tan IIA or Sal B ( Figure  S1).
Tanshinone IIA attenuated LPS-induced NF-kB activation in HSC-T6 cells NF-kB is a transcription factor crucially involved in inflammatory responses. We used a reporter gene assay to detect whether LPS-induced NF-kB activities were downregulated by Tan IIA. Cells were pretreated with Tan IIA (1, 3 or 10 mM) or Sal B (200 mM) for 30 min, and stimulated with LPS (100 ng/ml) for 6 hr. LPS-induced NF-kB luciferase activities were significantly suppressed by both Tan IIA (10 mM) and Sal B (200 mM, Figure 2A). We used NF-kB inhibitor-pyrrolidine dithiocarbamate (PDTC) for comparison in the Luciferase assay. Furthermore, PDTC also effectively downregulated LPS-stimulated mRNA expressions of IL-1b, TNF-a, iNOS, ICAM-1 and IL-6 genes ( Figure S2). The NF-kB heterodimer (p50/p65) is sequestered in the cytoplasm by IkB-a in the un-stimulated normal cells. If cells are stimulated with diverse agents, p50/p65 is released from IkB-a. NF-kB activity is induced following p65 translocation into nucleus from cytoplasm. In Western blotting analysis, we found that nuclei contents of NF-kB-p65 subcomponent was increased by LPS (100 ng/ml), which was attenuated by both Tan IIA and Sal B significantly ( Figure 2B). In addition, with immunofluorescent imaging, we also observed that LPS-triggered NF-kB-p65 nuclei translocation from the cytoplasm was ameliorated by Tan IIA and Sal B ( Figure 2C). NF-kB-p65 is green and DAPI represents the nuclei part which was pseudo-colored from blue to red by Metamorph software. When NF-kB-p65 translo-cation was induced by LPS, merged images indicated colocalization of NF-kB-p65 with nuclei showing orange color (bar, 100 mm).
We used gene transcription assays to test our hypothesis that inflammatory signaling is affected by Tan IIA and Sal B in vitro. Quantitative real-time PCR analysis of whole cell RNA demonstrated that Tan IIA suppressed mRNA expressions of IL-1b, TNF-a, iNOS, ICAM-1, and IL-6 genes in LPS-stimulated HSC-T6 cells, however, there was no significance in ICAM-1 and IL-6 genes in LPS-stimulated HSC-T6 cells with Sal B treatment ( Figure 2D, gene subsequences of forward and reverse primers are described in Table S1). LPS-induced IL-1b reaches to 7.8 folds compared with control, and iNOS is dramatically to 49.2 folds. Results suggested that LPS had great ability to trigger iNOS production in HSCs. Of note, Tan IIA exerted better and significant inhibition of LPS-induced mRNA expressions of ICAM-1 and IL-6 in HSC-T6 cells.
Taken together, our results suggest that both Tan IIA and Sal B might have inhibitory effect in LPS-mediated NF-kB signaling by decreasing NF-kB luciferase activities and NF-kB-p65 nuclear translocation. Otherwise, there was distinct inhibition of Tan IIA and Sal B in related genes expression of inflammatory signaling.

LPS-stimulated phosphorylations of mitogen-activated protein kinases (MAPKs) were inhibited by Tan IIA in HSC-T6 cells
Mitogen-activated protein kinases (MAPKs) are involved in key cellular functions, such as proliferation, inflammatory responses, differentiation and migration. MAPKs include ERK, JNK and p38 [33]. LPS-induced phophorylations of ERK, JNK and p38 were mostly attenuated by Tan IIA (10 mM) and Sal B (200 mM), except that Sal B had a slight but non-significant tendency of inhibiting p38 ( Figure 3A and B). To sum up, Tan IIA was superior to Sal B in inhibiting LPS-stimulated phosphorylations of MAPKs in HSC-T6 cells. Reports showed that several chemokines are crucially associated with MAPKs pathway, such as chemokine (C-C motif) ligands-CCL2, CCL3 and CCL5 [34][35][36]. Therefore, we used quantitative real-time PCR to assess whether Tan IIA and Sal B showed the ability in downregulating LPS-induced CCL2, CCL3, and CCL5. It has been observed that Tan IIA and Sal B abolished LPS-induced mRNA expressions of CCL2, CCL3 and CCL5 ( Figure 3C). Remarkably, LPS triggered gene expression of CCL3 to 9.3 times (9.30 61.38), and Tan IIA and Sal B inhibited CCL3 efficiently.

Tan IIA suppressed LPS-stimulated chemotaxis of HSC-T6 cells
Chemokines secretion is required to cell migration [35,37], and elevated HSC motility is crucially involved in hepatic fibrogenesis. On the basis of Tan IIA and Sal B against LPS-induced CCL2, CCL3 and CCL5 mRNA, which are key chemokines associated with chemotaxis. We determined to use wound-healing assay and trans-well invasion assay to exam Tan IIA and Sal B on LPSinduced HSC motility. Wound-healing assay revealed that with 24 hr exposure to LPS, HSC-T6 cells showed marked migration to the cell-free zone, which was attenuated by both Tan IIA and Sal B (Figure 4A and B). The trans-well invasion assay also showed that both Tan IIA and Sal B reduce 24 hr LPS-stimulated migration of HSC-T6 cells ( Figure 4C and D). Both assays demonstrated Tan IIA and Sal B had great ability to suppress LPS-induced HSC migration. HSC activation is the major change of liver fibrosis. a-SMA production and accumulation are main characteristics in HSC activation. We observed clear chemotaxis of LPS-induced HSC, and results showed that Tan IIA and Sal B both inhibited cell mobility caused by LPS. We determined to use Western blotting analysis to assess the change of a-SMA in LPS-stimulated HSC. Results revealed that Tan IIA and Sal B downregulated LPSinduced a-SMA protein production ( Figure 5A), and this is supported by gene transcription assays that showed Tan IIA and Sal B reduced mRNA of a-SMA ( Figure 5B). Our previous reports also indicated Sal B could inhibited TGF-b and CCl 4 -induced HSC activation and our present data give more information about ability of Sal B in inhibiting LPS-induced HSC activation. Moreover, our results demonstrated that the lipophilic compound Tan IIA showed more powerful inhibition in a-SMA production than the hydrophilic compound Sal B.

Discussion
In the present study, we have observed that Tan IIA and Sal B, two different compounds of S. miltiorrhiza could decrease LPSstimulated HSC activation by reducing NF-kB luciferase activity, NF-kB-p65 translocation, phosphorylations of MAPKs, chemotaxis of HSCs, and mRNA expressions of several pro-inflammatory and pro-fibrotic genes. There was no cytotoxicity to HSC-T6 cells at the working concentrations of both Tan IIA (1, 3 and 10 mM) and Sal B (200 mM).
Mammalian cells recognize the presence of pathogen associated molecular patterns (PAMPs) through a group of receptor complexes [38]. Originally, TLRs were identified as mammalian homologues of Drosophila Toll, and the function is pattern recognizing [39,40]. TLRs crucially initiate innate immunity and activate downstream intracellular signalling pathways resulting in an induction of innate immune response. LPS is the most important ligand of TLRs, and TLRs are expressed on various kinds of cells in the liver, including Kupffer cells, hepatocytes and HSCs [38]. Ligands binding to TLRs activate MyD88-dependent and -independent signalling pathways. The MyD88-dependent pathway is related to NF-kB translocation into nuclei, and activation of MAPKs and proinflammatory cytokines [10,41]. Activation of TLRs in HSCs by LPS has been reported to be involved in various acute and chronic liver diseases. Our present results demonstrated that LPS efficiently induced the NF-kB signaling pathway, phosphorylations of MAPKs, and several gene transcriptions related to inflammation and fibrogenesis. In addition, LPS stimulated HSC chemotaxis notably.   S. miltiorrhiza has been prescribed for a variety of maladies in Chinese medicine [42]. Several studies including ours have demonstrated inhibitory effects of S. miltiorrhiza extracts against hepatic fibrosis [17,18,43]. Tan IIA and Sal B are lipophilic and hydrophilic compounds of S. miltiorrhiza, respectively. In the literature, Tan IIA has been shown to inhibit LPS-induced NF-kB mobilization and ERK phosphorylation in rat macrophages [44]. Tan IIA has also been reported to ameliorate ischemia/ reperfusion injury, with inhibition of NF-kB and MAPKs signaling pathways [45]. Moreover, there had been evidence that Tan IIA protected mice from immune-mediated liver injury via NF-kB and IFN-c/STAT1 signaling pathways [46]. Additionally, Tan IIA has also been reported to suppress LPS-and ethanol-induced Kupffer cell sensitization, and inhibit formation of reactive oxygen/ nitrogen species [47]. Sal B has been shown to diminish LPSinducible TNF-a production in macrophages [48]. Our present study illustrated that LPS-induced NF-kB activation and phophorylations of MAPKs were reduced by Tan IIA and Sal B in another cell type, HSCs. Cell migration is a characteristic of HSC activation. A previous report demonstrated that Tan IIA inhibits macrophage migration [45]. Consistently, our present study showed that Tan IIA attenuated LPS-induced HSC migration in both wound-healing and trans-well invasion assays.
HSCs express a multitude of chemokines, including the CC chemokines CCL2, CCL3, and CCL5, which are involved in cell chemotaxis and liver fibrosis [49,50]. Chemokines and their receptors are up-modulated in the injured liver [51]. In line with this, we observed that Tan IIA and Sal B suppressed LPS-induced mRNA expressions of CCL2, CCL3, and CCL5 genes. Notably, LPS triggered gene expression of CCL3 to 9.3 times, in addition, Tan IIA and Sal B inhibited CCL3 efficiently. Seki et al. and Brenner et al. both indicated that HSC is directly stimulated by TLR4 to trigger proinflammatory features, such as upregulation of CCL2, CCL3 and ICAM-1 [15,16]. Furthermore, researchers found that in human HSC, all three MAPKs induce CCL2 secretion [34] and CCL3 stimulates cell migration [35]. There is report mentioned JNK-2 and p38 are required for LPS related RANTES (CCL5) mRNA expression [36].
A previous study showed that Tan IIA exerts anti-inflammatory effects against LPS by downregulating iNOS gene expression, NO production, and expressions of inflammatory cytokines (IL-1b, IL-6, and TNF-a) in macrophages [52]. TNF-a and IL-1 are reported as acute-response cytokines [53,54], and mediate activation of NF-kB pathway in HSCs [55]. On the other hand, IL-6 directly promotes HSC survival and proliferation during enhanced liver fibrosis [56]. Our present study showed that Tan IIA inhibited LPS-stimulated inflammatory response, reducing mRNA expressions of IL-1b, TNF-a, IL-6, and iNOS genes, which are NF-kB dependent. In our previous paper, Sal B reduced ROS accumulation in primary rat hepatocytes and HSCs, and diminished a-SMA formation in HSCs [23]. In the present study, we found that both Tan IIA and Sal B inhibited mRNA expression of a-SMA gene, with Tan IIA exerting more powerful effects. Tan IIA exerted inhibited LPS-induced NF-kB activation and translocation, downregulated MAPKs phosphorylations as well, together with attenuating chemotaxis and mRNA expressions of CCL2, CCL3, and CCL5 genes in HSCs. Tan IIA also suppressed mRNA expressions of a-SMA and IL-1b, TNF-a, IL-6, and iNOS genes. In conclusion, our results suggest that Tan IIA inhibited LPSinduced HSC activation, which might hold potential for treating hepatic fibrosis. Figure S1 There was no proliferation in HSC-T6 cells with Tan IIA and Sal B treatment. HSC-T6 cells were seeded on glass cover slips in 6 well-plates (1610 6 cells/well) and pretreated with Tan IIA (10 mM) or Sal B (200 mM) for 1 hr, then exposed to LPS for 24 hr. The protocols were followed according to the method described previously. The primary antibody was against Ki67