Didymin Alleviates Hepatic Fibrosis Through Inhibiting ERK and PI3K/Akt Pathways via Regulation of Raf Kinase Inhibitor Protein

Liver fibrosis is reversible wound-healing response against a variety of acute and chronic stimulation in the liver, including ethanol, viral infection, cholestasis and metabolic disease [1,2]. Advanced liver fibrosis results in cirrhosis, liver failure, and portal hypertension [3]. It has been reported that the proliferation and activation of hepatic stellate cells (HSCs) play a key role in the development of liver fibrosis [4], and targeting the proliferation of HSCs can be considered as novel therapeutic strategy for treating liver fibrosis [5].

Natural products are rich source of compounds with enormous structural diversity and have been extensively studied in the field of drug discovery. Didymin is a flavone that is common in citrus fruits like oranges, lemons, mandarin and bergamot. It can cause cell death in non-small cell lung cancer in a p53 independent manner [6]. Moreover, it can inhibit neuroblastomas proliferation and induces apoptosis by inhibiting N-Myc and up-regulating Raf kinase inhibitor protein (RKIP) level making it a potential new approach for neuroblastoma therapy [7]. Although the anti-cancer potential of didymin is well known, its precise mechanisms has not been investigated. Especially, the effect of didymin on liver fibrosis remains unclear.

To understand the anti-fibrosis effect of didymin, the present study was designed to investigate the effects of didymin on liver injury and collage deposition in vivo and in vitro. Additionally, to investigate whether didymin alleviated hepatic fibrosis through up-regulating Raf kinase inhibitor protein (RKIP), locostatin, a specific inhibitor of RKIP [8], was used to interfere with RKIP expression, and the antagonism between both the drugs was observed.

Mentha spicata L. was purchased from Nanning Qianjinzi Chinese Pharmaceutical Co. Ltd (Nanning, China). Locostatin was obtained from Merck (Darmstadt, Germany). Fetal bovine serum (FBS) was purchased from Life Sciences (Grand Island, NY). MTT (3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide) and propidium iodide (PI) were purchased from Sigma Chemical Co. (St. Louis, MO, U.S.A).

M. spicata L. herb (30 kg) was extracted with 95% ethanol under reflux three times for 1.5 h each time; the combined solvent was evaporated in vacuo to remove the ethanol. The extracts were suspended in methanol and fractionated with petroleum ether. The methanol fraction was subjected to silica gel chromatographic column (200-300 mesh, Yantai, PR China; 10×300 cm) and eluted with trichloromethane /methanol (100:0→50:50, each 500 mL) to produce eight fractions (Fr. 1- Fr.8). Fr. 7 was applied to a silica gel column (trichloromethane/methanol, 95:5→50:50) to generate eight fractions (Fr.7.1- Fr. 7.8). Fr. 7.4 was successively separated by silica gel column chromatography (trichloromethane/methanol, 80:20) to generate a white compound (128.6 mg). Its spectroscopic data such as ESI-MS, ¹H-NMR and ¹³C-NMR were consistent with the data of didymin that was reported in the previous study [9], which indicates that the isolated compound is didymin.

Male SD rats, weighting 160 -180 g, were obtained from Guangxi Medical University Experimental Animal Center (Guangxi, China) and the research was conducted according to protocols approved by Guangxi Medical University Institutional Ethical Committee. Following one week acclimation period, rats were randomly divided into five groups (12 rats/group) including normal control group, didymin control group, CCl₄ model group, CCl₄+ didymin (CD) group and CCl₄ + didymin + locostatin (CDL) group. The rats in the last three groups received 2 ml/kg CCl₄ (mixed 1:1 in peanut oil) intragastrically twice a week for 12 weeks, while the animals in the normal and didymin control groups received equivalent peanut oil.

Additionally, during the period from weeks 8 to 12, the rats in the didymin control group and CD group were intraperitoneally administrated with 0.5 mg/kg didymin, and the rats in the CDL group received 0.5 mg/kg didymin plus 0.5 mg/kg locostatin once a day; while the animals in the normal control group and the CCl₄ model group received equivalent normal saline. At the end of treatment, all animals were sacrificed and blood and liver samples were obtained for further examination.

Liver tissues were fixed in 10% formalin, embedded in paraffin and sectioned at 5-µm thickness. The liver pathology was observed by hematoxylin-eosin (H&E) staining. Additionally, the collagen deposition was observed by Masson's trichrome staining.

Serum alanine transaminase (ALT) and aspartate transaminase (AST) activities were measured using commercial kits (Nanjing Jiancheng Institute of Biotechnology, Nanjing, China), and TNF-α in plasma was detected using an enzyme-linked immunosorbent assay kit (Beijing Yonghui Biological Technology, Beijing, China). In addition, hydroxyproline (Hyp), type III precollagen (PCIII), laminin (LN) and hyaluronic acid (HA) were determined by commercially available kits (Nanjing Jiacheng Bioengineering Institute, Nanjing, China).

HSC-T6 cells were seeded in 96-well plates at a density of 1×10⁵ cells/well. After 24 h incubation, cells were treated with 100, 50, 25, 12.5, 6.25 and 3.125 µM of didymin for 24, 48 and 72 h. Cell proliferation was then measured using MTT assay as previously described [9].

HSC-T6 cells were seeded into a 6-well plate at a density of 3× 10⁵ cells/well and allowed to adhere overnight. Then cells were divided into four groups including normal control group (cells were treated with equal amount of culture medium), PDGF-BB group (cells were treated with 10 nM PDGF-BB [10]), didymin + PDGF-BB (DP) group (cells were treated with 25 µM didymin [6,7] + 10 nM PDGF-BB), and didymin + locostatin + PDGF-BB (DLP) group (cells were treated with 25 µM didymin + 25 µM locostatin [11] + 10 nM PDGF-BB ). HSC-T6 cells were treated with drugs for 24 h, and cell apoptosis and cell cycle were analyzed according to our previous study [9].

HSC-T6 cells were seeded in a 96 well plate at a density of 1 × 10⁵ cells/well and administrated with drugs as described above for 24 h. Cells were added 5 mM rhodamine-123 dye (Rh-123, Sigma-Aldrich) for 30 min at room temperature in dark. MMP was then detected according to our previous study [9]. In addition, caspase-3 and -9 activities were detected using commercially available kits (BioVision Research Products, CA, USA) [9].

Whole cell proteins were prepared using RIPA buffer (Thermo Fischer Scientific, Inc., Waltham, MA) with 1% Halt protease inhibitor cocktail and 1% Halt phosphatase inhibitor cocktails (Thermo Fischer Scientific, Inc., Waltham, MA). Total hepatic proteins were extracted from liver tissues using radioimmunoprecipitation buffer containing a protease inhibitor cocktail (Sigma-Aldrich). The protein was separated on 10% sodium dodecyl sulfate (SDS)-polyacrylamide gels, then transferred onto nitrocellulose membranes. The expression of proteins were detected using the following antibodies: RKIP, ERK, p-ERK, MEK, p-MEK, p38, p-p38, JNK, p-JNK, Bcl-2, Bax, cyclin D1, cyclin B1, CDK4, PI3K, Akt and p-Akt (Santa Cruz Biotechnology), respectively. All the primary antibodies were incubated overnight at 4 °C. The bands were scanned and quantified by NIH IMAGEJ 1.38 Software (US National Institutes of Health, Bethesda). The detected proteins were normalized to GAPDH (Santa Cruz Biotechnology) or the respective total protein, as appropriate.

Statistical analysis was performed using SPSS 11.5 for Windows. Differences between the groups were assessed using a one-way analysis of variance (ANOVA) with a Tukey's test for post hoc multiple comparisons. The data are presented as the means ± SD. A p-value < 0.05 was considered statistically significant.

The effect of didymin on hepatic histological damage was assessed. The results showed that the livers in the normal and didymin control groups showed normal structure, and the hepatic cells were orderly arranged (Fig. 1 A1 and A2). In contrary, liver cells in the model group were edematous with fatty degeneration, and lobular architecture was destroyed or had disappeared (Fig. 1 A3). Didymin treatment significantly ameliorated histological damage, as evidenced by decrease in steatosis and hepatic lesions in the CCl₄ + didymin group (CD group) (Fig. 1 A4). In addition, Masson's trichrome staining showed that the liver tissues in the normal and didymin control groups showed traces of collagen only in the walls of major blood vessels (Fig. 1 B1 and B2). However, CCl₄ administration led to extensive accumulation of collagen in the liver tissues (Fig. 1 B3). Didymin treatment significantly decreased collagen deposition (Fig. 1 B4). Interestingly, the protective effect of didymin on liver injury (Fig. 1 A5) and collagen accumulation (Fig. 1 B5) were greatly abolished by the specific RKIP inhibitor locostatin.

As shown in Fig. 1C, serum AST and ALT activities were significantly increased in the CCl₄ model group compared to the normal control group. Treatment with didymin markedly decreased both enzymes activities, but its effect was partially abolished by locostatin. Similarly, didymin significantly decreased the content of TNF-α in the liver tissues, and the effect of didymin on TNF-α was reduced by locostatin (Fig. 1D).

The key collagen-related indicators including hepatic Hyp, PCIII, LN and HA were determined in this study. As shown in Fig. 1E, the contents of these indicators in the CCl₄ model group were significantly higher than those of the normal control group. These up-regulations of Hyp, PCIII, LN and HA were markedly inhibited by didymin treatment. It was not surprise that the effect of didymin on the collagen-related indicators was largely abolished by locostatin.

The effect of didymin on hepatic stellate cell proliferation was determined by MTT assay. As shown in Fig. 2A, didymin inhibited HSC-T6 cell proliferation in a distinct dose- and time-dependent manner. The 50% inhibitory concentrations (IC₅₀) were 45.9 ± 2.7, 26.2 ± 1.6 and 16.6 ±1.4 µM after incubation for 24, 48 and 72 h, respectively.

To understand the underlying basis for the cytotoxic effect of didymin on HSC-T6 cells, cells apoptosis was analyzed by flow cytometry. The result showed that didymin significantly promoted HSC-T6 cell apoptosis compared to the PDGF-BB group, but this effect was largely abolished by locostatin (Fig. 2B). We next analyzed the effect of didymin on cell cycle progression by FACS analysis. The result showed that treatment with didymin led to a significant accumulation of cells in the G2/M phase, suggesting that didymin exerts its anti-proliferative effects by disturbing cell cycle distribution (Fig. 2C).

Bcl-2 family proteins serve as critical regulators of pathways involved in apoptosis, acting to either inhibit or promote cell death [12]. As shown in Fig. 2D, didymin treatment significantly decreased the anti-apoptotic protein Bcl-2 expression, while increased the pro-apoptotic protein Bax expression. However, the effects of didymin on the apoptosis-related proteins were largely abolished by locostatin. In addition, cell cycle is controlled by cyclin-dependent kinases (CDKs), cyclin kinase inhibitors and cyclins [13]. Our result showed that didymin treatment significantly reduced the expressions of the cell cycle-related proteins including cyclin B1, cyclin D1 and CDK4 (Fig. 2E).

The change of MMP was evaluated by flow cytometry using rhodamin-123 (Rh-123). As shown in Fig. 3A, the intensity of fluorescence in the DP group was significantly decreased compared to the PDGF-BB group, indicating that didymin treatment induced apoptosis of HSC-T6 cells by disruption of MMP. Additionally, we further analyzed cytochrome c (Cyt c) level by Western blot assay. As shown in Fig. 3B, didymin treatment decreased Cyt c level in the mitochondria with a corresponding increase in cytosolic level, indicating that Cyt c was released from the mitochondria into the cytosol. These findings suggest that mitochondrial dysfunction is likely involved in didymin-induced apoptosis. Interestingly, locostatin did not abolish the effect of didymin on MMP or Cyt c. These results suggest that the underlying mechanism of didymin on apoptosis is partially RKIP-independent.

To fully understand the underlying mechanism of didymin on cell apoptosis, we also examined the protease activities of caspase-3 and -9. As shown in Fig. 3C, compared to the normal control group, PDGF-BB decreased both the protease activities. In contrast, didymin treatment significantly increased the activities of caspase-3 and -9.

RKIP has been shown to be involved in several cell signaling cascades [14]. In the present study, PDGF-BB-stimulated cells showed higher level of RKIP than that of normal control cells. Compared to the PDGF-BB group, didymin treatment led to significant increase in the RKIP expression in both the liver tissues (Fig. 4A) and HSC-T6 cells (Fig. 4B), but this effect was largely abolished by locostatin.

To investigate the underlying mechanism of didymin on HSC-T6 cell proliferation and apoptosis, we examined the effect of didymin on the key intracellular ERK/MAPK pathway. Compared to the normal control group, PDGF-BB administration significantly increased the phosphorylations of MEK and ERK (Fig. 4 C and D), while decreased the phosphorylations of p38 and JNK (Fig. 4 E and F). However, didymin treatment notably reversed these effects induced by PDGF-BB. Interestingly, the effect of didymin on ERK/MAPK pathway was greatly abolished by locostatin.

The PI3K signaling cascade regulates cell proliferation, survival, differentiation and apoptosis [15]. To fully understand the protective effect of didymin on HSC-T6 cell proliferation and apoptosis, we further investigated the effects of didymin on PI3K/Akt pathway. The results showed that the expression of PI3K and the phosphorylations of Akt in the PDGF-BB group were higher than those of the normal control group. Compared to the PDGF-BB group, didymin treatment significantly decreased the expressions of PI3K (Fig. 4G) and the phosphorylations of Akt (Fig. 4H). However, these effects were partially reversed by locostatin.

In the present study, hepatic fibrosis model was induced by CCl₄ in rats. We found that CCl₄ caused severe histological changes to the liver. But, these changes were significantly attenuated by didymin treatment. Additionally, our study revealed a significant increase in the activities of serum ALT and AST and the level of inflammatory cytokine TNF-α when exposed to CCl₄, indicating serious hepatocellular injury. Didymin treatment significantly decreased the activities of both the serum enzymes and the level of TNF-α. Furthermore, CCl₄ led to extensive accumulation of collagen in the liver tissues and significant increase in the contents of the collagen-related indicators including hepatic Hyp, PCIII, LN and HA. However, didymin treatment notably reversed these abnormal changes of collagen induced by CCl₄. These results indicate that didymin can alleviate liver injury and fibrosis induced by CCl₄.

HSCs play a critical role in the development of hepatic fibrosis. To fully understand the underlying mechanism of didymin against hepatic fibrosis, we further investigated the effect of didymin on HSCs biological behaviors including proliferation, apoptosis, cell cycle, mitochondrial function and the key intracellular signaling pathways such as ERK/MAPK and PI3K/Akt. As it is well known, apoptosis plays an important role in cells proliferation, differentiation, senescence and death. The potential to induce apoptosis has become an important topic in the study of anti-tumor drugs [16]. In the present study, didymin treatment decreased HSC-T6 cell proliferation in a dose-and time-dependent manner. Flow cytometric analysis showed that it significantly induced HSC-T6 cell apoptosis. These results suggest that didymin exerts anti-hepatic fibrosis mainly through inhibiting HSCs vitality and promoting cell apoptosis.

Cell cycle is regulated by a series of positive and negative mediators that act at sequential points throughout the cell cycle. We found that didymin treatment significantly increased the percentage of cells in G2 phase and inhibited the cell cycle-related regulators including cyclin B1, cyclin D1 and CDK4. Ii is well know that Cyclin D1 is a key G2/M phase regulator. The increase of cyclin D1 during G2 phase leads to progression to mitosis [17]. Thus, didymin induced cell cycle arrest in at G2 phase, likely through the inhibition of cyclin D1. In addition, we observed a decrease in the level of CDK4 commonly associated with the G1 transition and G2/M phase arrest. Cyclin B1 is a key regulatory protein controlling cell cycle progression in vertebrates. Depletion of cyclin B1 disrupts the spindle checkpoint functions and chromosomes alignment during mitosis [18]. In this context, the ability of didymin to down-regulate cyclin B1 in HSC-T6 cells represents a potentially significant mechanism of action for its anti-proliferative effects in HSC-T6 cells.

Mitochondria play an important role in regulating apoptosis. Mitochondrial dysfunction causes the collapse of mitochondria membrane potential (MMP), which results in mitochondrial permeability transition pore (MPTP) opening, enabling cytochrome c release from mitochondria into the cytosol [19]. Bcl-2 family proteins, including pro-apoptotic proteins (such as Bax) and anti-apoptotic proteins (such as Bcl-2), tightly regulate the mitochondria-dependent pathway [20]. In the present study, didymin treatment caused significant loss of MMP in HSC-T6 cells and obvious translocation of cytochrome c from the mitochondria to the cytosol. Furthermore, didymin treatment significantly decreased Bcl-2 expression and increased Bax level. These results suggest that the mitochondrial pathway of cell death, including Bcl-2 family and cytochrome c, may be involved in didymin-induced HSC-T6 cell apoptosis.

In general, cytochrome c releases into the cytosol and triggers caspase-9 activation. Subsequently, this initiator caspase activates caspase-3, which is a critical executioner of apoptosis. Following caspase activation, an increasing number of cellular substrates and the DNA repair protein poly (ADP-ribose) polymerase (PARP) are degraded or cleaved, resulting in cell death [21]. In the present study, didymin treatment significantly increased the levels of caspase-3 and −9, which suggests that the caspase-dependent mitochondrial signaling pathway is also involved in didymin-induced cell apoptosis.

The MAPK signaling pathway is highly conserved and is involved in cell growth, differentiation, survival, and invasion [22]. The MAPK family includes extracellular regulated kinase (ERK), c-Jun NH2-terminal kinase (JNK) and P38 kinase. ERK1/2 are key transducers of proliferation signals and are often activated by mitogens. In contrast, SAPK/JNK and p38 are poorly stimulated by mitogens but strongly activated by cellular stress. ERK and P38 have been recently reported to be involved in cancer cell death induced by anti-cancer reagents [23]. Once activated, ERK plays an important role in anti-apoptotic activity, while JNK and p38 are important for pro-apoptotic activity [24]. Our results showed that didymin treatment decreased the phosphorylations of MEK and ERK, but increased the phosphorylations of JNK and p38, suggesting that the MAPK signaling pathway is involved in didymin-induced apoptosis.

Additionally, PI3K/Akt inhibition by didymin may also contribute to its inhibitory effect on cell growth. PI3K/Akt is an important signaling pathway in regulating cell proliferation, growth, apoptosis, survival and metabolism by phosphorylating several substrates. The inhibition of Akt phosphorylation has been suggested as a novel target for therapeutic cancer agents [25]. Akt activation by PI3K reduces apoptosis and promotes tumor cell growth through phosphorylation to inactivate a variety of downstreamtargets [26]. Our results showed that didymin treatment significantly decreased PI3K expression and Akt phosphorylation, suggesting that didymin induced apoptosis, at least in part, by inhibiting the PI3K/Akt pathway.

Raf kinase inhibitor protein (RKIP) is a specific MAPK signaling pathway inhibitor. It directly interacts with both Raf-1 and MEK and disrupts the Raf-1/MEK interaction, thereby preventing MEK activation and its downstream targets [27]. Interestingly, didymin has been reported to partially exert its pro-apoptosis effects in neuroblastoma by up-regulating RKIP expression [7]. Therefore, to investigate whether the target of didymin against hepatic fibrosis was RKIP, locostatin was used to interfere with RKIP expression and the antagonism of didymin was analyzed in vivo and in vitro. Our results showed that didymin significantly increased RKIP expression both in the liver tissues and HSC-T6 cells. Importantly, the protective effects of didymin on liver injury and collagen deposition in rats with hepatic fibrosis were greatly abolished by the specific RKIP inhibitor locostatin. The cell experiments in vitro showed that locostatin revealed partial reversal of didymin-induced inhibitions of cell proliferation. The promotive effect of didymin on cell apoptosis was also significantly inhibited by locostatin. Moreover, locostatin caused a partial reversal in didymin-induced ERK/MAPK and PI3K/Akt signaling pathways inactivation. These contradictive findings suggest that RKIP up-regulation represents a potential mechanism of action of didymin, specifically for its anti-fibrosis effects.

In summary, our studies showed that didymin treatment significantly alleviated liver injury and collagen production induced by CCl₄ in rats. It notably inhibited HSC-T6 cell proliferation, predominantly by inducing apoptosis mediated by disrupting the cell cycle distribution, inducing mitochondrial dysfunction, regulating Bcl-2 family proteins and activating caspase-3/9. Moreover, didymin enhanced RKIP expression, thereby resulting in the inhibition of the ERK/MAPK and PI3K/Akt signaling pathways, which also contributed to didymin-induced apoptosis. These results indicate that didymin is likely to have potential as a new chemotherapeutic agent for the treatment of liver injury and fibrosis.

The authors gratefully acknowledge the financial support provided by the National Natural Science Foundation of China (No. 81260505; No.81473431) and the Guangxi Natural Science Foundation (2014GXNSFAA118155; 2014GXNSFAA118154).

The authors declare that there are no conflicts of interest.