Huh-7 cells, the human hepatoma cell, and Ava5 cell, the Huh-7 cells harboring HCV genotype 1b subgenomic RNA replicon (Blight et al., 2000), were used for antiviral studies. Both were maintained in Dulbecco’s modified Eagle’s medium with 10% heat-inactivated fetal bovine serum, 1% antibiotic-antimycotic, and 1% non-essential amino acids. All cells were incubated at 37°C with a 5% CO₂ supplement. The IH636 GSE was purchased from InterHealth Nutraceuticals (Benicia, CA, USA) (Shi et al., 2003). Interferon α-2a (Roferon-A) was purchased from Roche, Ltd. Telaprevir was purchased from Legend Stat International, Co., Ltd, and daclatasvir and sofosbuvir were purchased from Shanghai Haoyuan Chemexpress, Co., Ltd. The anti-HCV agents were prepared as stock solution at 100 mM in 100% DMSO. The final concentration of DMSO in the all experiments was constantly maintained at 0.1%. All chemicals were diluted by DMEM medium.

The standard procedure of Western blotting was performed as described previously (Lee et al., 2011). The membranes were probed with monoclonal antibodies specific for HCV NS5B (1:5000; Abcam, Cambridge, MA, USA), glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (1:10000; GeneTex, Irvine, CA, USA), anti-COX-2 antibody (1:1000; Cayman, MI, USA), anti-MAPK (phosphorylated and unphosphorylated forms of ERK1/2, p38, and JNK), anti-IKKα, anti-phospho-IKKα/β (Ser176/180), anti-NF-κB, anti-IκB-α, anti-phospho-IκB-α (Ser32) (1:1000; Cell Signaling Technology, Inc., Danvers, MA, USA), or anti-C-Myc antibody (1:1000; GeneTex, Irvine, CA, USA). The ECL detection kit was used for the signal detection (PerkinElmer, Shelton, CT, USA).

The total RNA of the cells were isolated with a total RNA miniprep purification kit (GMbiolab, Co., Ltd, Taiwan), according to the manufacturer’s instructions. The cDNA synthesis was performed by M-MLV Reverse Transcriptase (Promega, Madison, WI, USA), according to the manufacturer’s instructions. The levels of HCV NS5B, TNF-α, IL-1β, iNOS, and COX-2 RNA were detected by RT-qPCR with the following forward and reverse primer sets: NS5B (AJ238799), 5′-GGA AAC CAA GCT GCC CAT CA-3′ and 5′-CCT CCA CGG ATA GAA GTT TA-3′; TNF-α (NM_000594), 5′-CCT GTG AGG AGG ACG AAC-3′ and 5′-AAG TGG TGG TCT TGT TGC-3′; IL-1β (NM_000576), 5′-GGA GAA TGA CCT GAG CAC-3′ and 5′-GAC CAG ACA TCA CCA AGC-3′; iNOS (NM_000625), 5′-CTT TGG TGC TGT ATT TCC-3′ and 5′-TGT GAC CTC AGA TAA TGC-3′; and COX-2 (NM_000963), 5′-CCG AGG TGT ATG TAT GAG-3′ and 5′-TGG GTA AGT ATG TAG TGC-3′. The relative RNA level of these genes in each sample were normalized to cellular GAPDH mRNA with the forward primer: 5′-GTC TTC ACC ACC ATG GAG AA-3′ and reverse primer: 5′-ATG GCA TGG ACT GTG GTC AT-3′. The relative expression levels were analyzed by the ABI Step One Real-Time PCR-System in the standard procedure (ABI Warrington, UK).

The relative cell viabilities were determined by the CellTiter 96^® AQueous One Solution Cell Proliferation Assay (MTS) assay that depended on the measurement of mitochondria dehydrogenase enzyme activity from viable cells (Promega Corporation, Madison, WI, USA). Ava5 cells at the density of 5 × 10³ per well were treated with GSE at different concentrations. After a 3-day incubation, the supernatant was removed and added to the MTS mixture with 100 μl of phenol red-free medium and 20 μl of the MTS reagent at 37°C. After 4 h, the relative cell viabilities were analyzed by measuring the absorption at 490 nm on a microplate reader.

The cell culture-produced HCV (HCVcc) were generated by the transfection of in vitro transcribed full-length JFH-1 RNA into Huh-7.5 (Kato et al., 2006). The Huh-7 cells with a density of 5 × 10⁴ per well were infected with JFH-1 HCVcc at a multiplicity of infection (MOI) of 0.1 for 8 h. At the end of infection, the supernatant was removed and the cells were incubated with various concentrations of GSE for an additional 3 days. The total RNA of the cells was isolated and the relative HCV RNA levels were analyzed by a Step One Real-Time PCR-System.

Ava5 cells were incubated in a mixture containing GSE (0, 2.5, 5, 10, and 20 μg/ml) in combination with each of the anti-HCV agents, IFN-α (0, 7.5, 15, 30, and 60 U/ml), HCV NS3/4A protease inhibitor telaprevir (0, 0.075, 0.15, 0.3, and 0.6 μM), HCV NS5A inhibitor daclatasvir (0, 1, 2, 4, and 8 pM), and the RNA-dependent RNA polymerase nucleoside inhibitor sofosbuvir (0, 10, 20, 40, and 80 nM). After 3 days, the relative HCV RNA levels were analyzed by a Step One Real-Time PCR-System for calculating the drug dose effects. Based on the method of Chou and Talalay (1984), the combination index (CI) was calculated by CalcuSyn software (Biosoft, Cambridge, UK) with the presence of synergism (CI < 1), additivity (CI = 1), and antagonism (CI > 1).

The inactive COX-2 mutant (S516Q) expression vector (pCMV-COX-2^mut-Myc) without cyclooxygenase activity were generated from pCMV-COX-2-Myc with the primers: forward: 5′-GTTGGAGCACCATTCCAGTTGAAAGGACTTATG-3′, reverse: 5′-CATAAGTCCTTTCAACTGGAATGGTGCTCCAAC-3′ by QuikChange^® Site-Directed Mutagenesis Kit according to the manufacturer’s protocol (Stratagene, La Jolla, CA, USA) as previous described (Lecomte et al., 1994). All of the DNA fragments were confirmed by DNA sequencing.

All the transfection reactions were performed with a T-Pro^TM reagent (Ji-Feng Biotechnology, Co., Ltd, Taiwan), in accordance with the manufacturer’s instructions. The Ava5 cells with a density of 5 × 10⁴ per well were transfected with pCOX-2-Luc or pNF-κB-Luc (BD Biosciences Clontech, Palo Alto, CA, USA) for 8 h. After the transfection procedure, the medium was replaced with various concentrations of GSE for 3 days. The luciferase assay was performed using the Bright-Glo^TM Luciferase Assay System (Promega) according to the manufacturer’s protocol with cell extracts from each sample. To identify COX-2 regulation by GSE, Ava5 cells were transfected with indicated concentrations of the COX-2 expression vector (pCMV-COX-2-Myc) from 0.5 to 2 μg, and then treated with GSE for additional 3 days. To evaluate the effect of GSE on viral-induced inflammation response, the parental Huh-7 cells were infected with the JFH-1 virus for 8 h or transfected with HCV core or NS5A expression vector (pCMV-Core-Myc; pCMV-NS5A-Myc) for 12 h, followed by incubation with or without GSE (5 and 20 μg/ml) for another 3 days. To determine the effect of GSE on virus-induced inflammatory genes (TNF-α, IL-1β, iNOS, and COX-2), Huh7 cells were infected with JFH-1 or transfected with pCMV-Core-Myc or pCMV-NS5A-Myc in the presence of GSE for 3 days. Subsequently, the total RNA of the cells was isolated and the relative inflammation cytokine RNA levels were analyzed by a Step One Real-Time PCR-System.

Ava5 or Huh7 cells were treated with GSE at various concentrations for 3 days. Subsequently, the treated cells were washed thoroughly with cold phosphate-buffered saline (pH 7.4), and intracellular PGE₂ was extracted by a lysis reagent containing C₁₅H₃₄BrN to rupture cell membrane. The intracellular PGE₂ level was measured with the PGE₂ enzyme-linked immunosorbent assay system (Biotrak, Amersham Bioscience), according to the manufacturers protocol.

Nuclear extracts were prepared using hypotonic [10 mM HEPES, 1.5 mM MgCl₂, 10 mM KCl, 0.5 mM DTT, 10% Nonidet P-40 (pH 7.9)] and high-salt buffer [20 mM HEPES, 1.5 mM MgCl₂, 0.2 mM EDTA, 0.6 M KCl, 0.5 mM DTT (pH 7.9)] extraction as previously described (Abmayr et al., 2006). Briefly, Ava5 cells with a density of 4 × 10⁵ per well were treated with or without GSE at the indicated dose. After 3 days of incubation, the cells were lysed using the ice-cold hypotonic buffer. The cytoplasmic fraction were separated by centrifugation at 7,000 × g for 15 min. Then, the resulting nuclear pellets were extracted with high-salt nuclear extraction buffer at 4°C for 30 min and the nuclear proteins were collected after centrifugation at 20,000 × g for 15 min, which stored at -80°C until use. The protease inhibitors were added to hypotonic buffer and high-salt buffer immediately before use.

The results are presented as the mean ± SD for of least three independent experiments. Statistical data comparisons were analyzed with the Student’s t-test and ANOVA by using GraphPad software and ^∗P < 0.05 or ^∗∗P < 0.01 was considered statistically significant.

We first examined the anti-HCV activity of GSE using the HCV subgenomic RNA stably expressing cell line (Ava5). To this end, Ava5 cells were treated with various concentrations of GSE for 3 days or 20 μg/ml of GSE for 1–3 days, followed by Western blotting and qRT-PCR. The cytotoxicity of GSE was determined using an MTS assay under identical conditions. We found that GSE significantly inhibited HCV protein synthesis (Figures 1A,B) and RNA replication (Figure 1C) in a concentration- and time-dependent manner, and no significant cytotoxicity was observed. The inhibitory effect of GSE on HCV replication was also consistently observed in HCV-infected Huh-7 cells, with an IC₅₀ of 7.5 ± 0.3 μg/ml (Figure 1D).

Previous studies have reported that GSE exhibited a protective effect against cancer development and inflammation through the inhibition of the COX-2 signaling pathway (Derry et al., 2013). Besides, more reports demonstrated that suppression of HCV-elevated COX-2 expression could effectively block HCV replication (Trujillo-Murillo et al., 2007; Lee et al., 2011). To examine whether the anti-HCV activity of GSE correlates with suppression of the COX-2 expression, we first examined the effect of GSE on COX-2 expression using COX-2 promoter-driven reporter assay in GSE-treated Ava5 cells. The results revealed that GSE gradually decreased the luciferase activity compared to the parental Huh-7 cells and the GSE-untreated Ava5 cells (Figure 2A). The similar result were also observed in HCV infection assay (Supplementary Figure S1A). We next examined the COX-2 protein synthesis by Western blotting and measured the intracellular amount of COX-2 metabolite PGE₂ using the PGE₂ ELISA assay. Consistent with the COX-2 promoter-based activity assay, GSE decreased the HCV-elevated COX-2 protein (Figure 2B) and PGE₂ levels (Figure 2C) in a concentration-dependent manner. To investigate whether suppression of COX-2 expression contributes to the anti-HCV activity of GSE, we transfected increasing concentrations of the COX-2 expression vector pCMV-COX-2-Myc or inactive COX-2 mutant vector pCMV-COX-2^mut-Myc into GSE-treated Ava5 cells followed by a measurement of HCV protein synthesis and RNA replication after 3 days. As shown in Figure 2D, an increasing overexpression of extraneous COX-2-Myc gradually restored HCV protein levels (lanes 3–5) compared to GSE-untreated (lane 1), GSE-treated cells alone (lane 2), and GSE-treated cells with extraneous inactive COX-2-Myc mutant expression (lane 6), indicating that COX-2 function is essential to support HCV replication. Similar results were obtained from the measuring the HCV RNA level under identical conditions (Figure 2E). We also observed that the GSE-reduced-PGE₂ level was gradually restored by functional COX-2 expression, but not by inactive COX-2 mutant expression (Figure 2F). Taken together, these results revealed that GSE inhibited HCV replication by suppressing HCV-elevated COX-2 expression.

Both NF-κB and the mitogen-activated protein kinase (MAPK) signaling pathways are important regulators of COX-2 expression (Tsatsanis et al., 2006). To investigate whether GSE suppresses COX-2 expression by interfering with NF-κB activation, we first performed an NF-κB-mediated transcription reporter assay. The Ava5 cells were transfected a pNF-κB-Luc reporter plasmid and then incubated with increasing concentrations of GSE for 3 days. We found that the GSE concentration-dependently decreased luciferase activity compared to parental Huh-7 cells and GSE-untreated Ava5 cells (Figure 3A), indicating that GSE inhibited the HCV-stimulated NF-κB transcriptional activation. The similar results were also observed in HCV infection assay (Supplementary Figure S1B). A Western blot analysis further confirmed that the nuclear translocation of NF-κB p65 subunit was significantly interfered by GSE at a concentration of 20 μg/ml (Figure 3B). Phosphorylation of IκB by the active phosphorylated form of IKKα/β kinase is a key mechanism to control the nuclear translocation of the free form of NF-κB due to the ubiquitin-proteasome proteolysis of phospho-IκB (Chen, 2005). We further examined the effect of GSE on IKKα/β kinase and IκB phosphorylation in Ava5 cells. The results of Western blotting showed that GSE suppressed the amount of active phospho-IKKα/β kinase in a concentration-dependent manner (Figure 3C, upper panel). The total amount of IKKα or IKKβ did not change significantly. As expected, the phospho-IκB level was gradually decreased by GSE. In the meantime, the amount of unphospho-IκB accumulated (Figure 3C, middle panel). These results revealed that inactivation of the IKKα/β kinase was involved in the suppression of NF-κB-mediated COX-2 expression by a GSE treatment. In addition to the NF-κB signaling pathway, we next explored the effect of GSE on the MAPK pathways on the induction of COX-2 expression. Ava5 cells were treated with GSE at 20 μg/ml for 0–120 min and we then examined the effect of GSE on the phosphorylation status of the major regulators of the MAPK signaling pathway, including extracellular regulated protein kinases 1 and 2 (ERK1/2), p38 kinase, and c-Jun NH2-protein kinase (JNK). As shown in Figure 3D, GSE suppressed the amount of phosphor-ERK and phospho-JNK in a time-dependent manner (upper and middle panels). In contrast, there was no significant effect on the phospho-p38 protein level in GSE-treated Ava5 cells compared to the GSE-untreated control (lower panel). Consistent with the results of replicon system, GSE significantly suppressed the amount of phosphor-ERK and phospho-JNK instead of phospho-p38 in HCV infectious system (Supplementary Figure S2). Taken together, these results reveal that GSE inhibits HCV replication by blocking NF-κB and ERK/JNK MAPK signaling pathways to suppress COX-2 expression.

Acute or chronic inflammation is one of the hallmarks to hepatic injury, viral replication, and HCC development (Sukowati et al., 2016). HCV core and NS5A protein are known to be risk factors in the development of HCV-induced chronic inflammation (Nunez et al., 2004). To examine whether GSE exhibits a hepatoprotective activity against virus-induced inflammation, we measured the effect of GSE on the mRNA levels of several inflammatory mediators, including TNF-α, iNOS, COX-2 and IL-1 in HCV-infected, HCV core-expressing, or NS5A-expressing Huh-7 cells by a qRT-PCR analysis. As shown in Figure 4, the results demonstrated that HCV infection (A), core (B), or the NS5A (C) protein could transcriptionally activate those of the inflammatory mediators by approximately five to eightfold as compared to parental Huh-7 cells. With a GSE treatment, the mRNA levels of those stimulated pro-inflammatory mediators were concentration-dependently reduced by GSE compared to the levels observed in HCV-infected, core or NS5A-transfected Huh-7 cells with no GSE treatment, revealing that GSE can be used as a potential hepatoprotective agent against HCV-stimulated inflammation.

To examine the potency of GSE with supplementary usage, we examined the anti-HCV activity of GSE in combination with IFN-α or other USA Food Drug Administration (FDA) approved anti-HCV drugs, including NS3/4A protease inhibitor telaprevir, NS5A inhibitor daclatasvir, and NS5B polymerase inhibitor sofosbuvir. Ava5 cells were incubated with GSE and combined with either IFN-α or other HCV inhibitors at various fixed concentration ratios. The anti-HCV replication activities were analyzed by qRT-PCR. The CI values were calculated by CalcuSyn^TM software based on the principle described by Chou and Talalay (Chou and Talalay, 1984). The drug interaction in terms of synergism, CI < 1 for ED50, ED75, and ED90 (range: 0.86–0.34) and shown in Table 1, which indicated that a combination treatment exhibited a synergistic effect on anti-HCV activity. This means that GSE could be developed as a useful supplement or adjuvant to treat patients with chronic HCV infection under a combination regimen.