Lobohedleolide suppresses hepatitis C virus replication via JNK/c-Jun-C/EBP-mediated down-regulation of cyclooxygenase-2 expression

Hepatitis C virus (HCV) chronically infects 2–3% people of the global population, which leads to liver cirrhosis and hepatocellular carcinoma. Drug resistance remains a serious problem that limits the effectiveness of US Food and Drug Administration (FDA)-approved direct-acting antiviral (DAA) drugs against HCV proteins. The objective of our study was to discover new antivirals from natural products to supplement current therapeutics. We demonstrated that lobohedleolide, isolated from the Formosan soft coral Lobophytum crassum, significantly reduced HCV replication in replicon cells and JFH-1 infection system, with EC50 values of 10 ± 0.56 and 22 ± 0.75 μM, respectively, at non-toxic concentrations. We further observed that the inhibitory effect of lobohedleolide on HCV replication is due to suppression of HCV-induced cyclooxygenase-2 (COX-2) expression. Based on deletion-mutant analysis of the COX-2 promoter, we identified CCAAT/enhancer-binding protein (C/EBP) as a key transcription factor for the down-regulation of COX-2 by lobohedleolide, through which lobohedleolide decreased the phosphorylation of c-Jun NH2-terminal protein kinase and c-Jun to suppress HCV-induced C/EBP expression. The combination treatment of lobohedleolide with clinically used HCV drugs synergistically reduced HCV RNA replication, indicating that lobohedleolide exhibited a high biomedical potential to be used as a supplementary therapeutic agent to control HCV infection.

Hepatitis C virus (HCV) is a pathogen of high risk causing chronic liver diseases, including hepatic fibrosis, liver cirrhosis and hepatocellular carcinoma (HCC) 1 . The current clinical standard therapy for chronic hepatitis C (CHC) infection is a combination treatment with pegylated interferon-α (peg-IFN-α) and ribavirin (RBV), but the sustained virologic response (SVR) with only about 40% for HCV genotype 1 patients. As of 2011, two US Food and Drug Administration (FDA)-approved direct-acting antiviral (DAA) agents, boceprevir and telaprevir, against the viral protease activity were shown to significantly improve the antiviral response rate in the treatment of CHC infection 2 . However, the lack of pangenotypic activity and the rapid selection of drug-resistant viral mutants have limited the antiviral response in patients infected with HCV genotypes 2-6 3 . More recently, the Lobohedleolide suppresses HCV replication by inhibiting HCV-induced COX-2 expression and its activity. Inhibition of COX-2 expression has been reported to interfere with HCV replication 21 . In the LPS-induced inflammatory model, lobohedleolide has been shown to exert an inhibitory effect on COX-2 expression 20 . To investigate whether the suppression of COX-2 plays an important role in the activity of lobohedleolide against HCV replication, we first detected the amount of COX-2 protein in the HCV self-replicating cells, Ava5, and HCV JFH-1-infected Huh-7 cells after lobohedleolide treatment. The results of Western blotting indicated that lobohedleolide markedly suppressed the HCV-induced COX-2 protein levels in a dose-dependent manner compared with that in the parental Huh-7 cells, lobohedleolide-untreated Ava5 cells, and HCV-infected Huh-7 cells in the absence of lobohedleolide treatment (Fig. 2a,b). Next, we used a COX-2 promoter-based reporter assay to evaluate the inhibitory effect of lobohedleolide on COX-2 at the transcriptional level. The pCOX-2-Luc, a plasmid encoding firefly luciferase under the control of the COX-2 promoter, was transfected into Ava5 cells or HCV-infected Huh-7 cells, and then, these plasmid-transfected cells were incubated with lobohedleolide at increasing concentrations for 3 days. As shown in Fig. 2c,d, lobohedleolide dose-dependently decreased the HCV-elevated COX-2 promoter activity in Ava5 and HCV-infected cells. We further investigated the effect of lobohedleolide on COX-2 catalytic activity by monitoring the levels of PGE 2 in Huh-7 and Ava5 cells. The levels of HCV-elevated PGE 2 were significantly reduced with lobohedleolide treatment in a dose-dependent manner compared with that in the untreated cells (Fig. 2e).
To characterize whether the lobohedleolide -mediated down-regulation of COX-2 expression and its catalytic activity is involved in the suppression of HCV replication, we overexpressed exogenous COX-2 to evaluate the inhibitory activity of lobohedleolide on HCV replication. Ava5 cells were transfected with vehicle or various concentrations of pCMV-COX-2-Myc vector encoding the cox-2 gene, and the plasmid-transfected cells were incubated with 40 μM of lobohedleolide or 0.1% of DMSO, as a negative control. As shown in Fig. 2f, the lobohedleolide-reduced viral RNA level (top panel, lane 2) was gradually rescued following the increasing expression of exogenous COX-2-Myc (top panel, lanes 3-5) compared with that in the vehicle-transfected Ava5 cells incubated with DMSO (top panel, lane 1). Consistently, the exogenous expression of COX-2 attenuated the inhibitory effect of lobohedleolide on the HCV protein levels, as analyzed by Western blotting. These results clearly indicated that lobohedleolide inhibited HCV replication by down-regulating the virus-induced COX-2 expression.
SCIeNTIfIC REPoRTS | (2018) 8:8676 | DOI:10.1038/s41598-018-26999-w Lobohedleolide inhibits C/EBP transcription factor activity. COX-2 induction is mediated by multiple transcription factor pathways, such as NF-κB, C/EBP, and AP-1 22 . To identify which binding elements of transcription factors on the COX-2 promoter are responsible for modulating the lobohedleolide -mediated inhibition of COX-2 expression, we created a series of COX-2 promoter reporter constructs carrying the deleted promoter fragments, including wild-type (WT), ΔNF-κB, and ΔNF-κB/C/EBP, linked to firefly luciferase gene to elucidate the effect of lobohedleolide on COX-2 transcription, in which the WT promoter region contained NF-κB, C/EBP, and AP-1 responsive elements (Fig. 3a). Huh-7 and Ava5 cells were transiently transfected with pCOX-2-Luc, pCOX-2(ΔNF-κB)-Luc, or pCOX-2(ΔNF-κB/C/EBP)-Luc, and the plasmid-transfected cells were respectively treated with lobohedleolide at 0, 20 and 40 μM. The relative COX-2 promoter activities were analyzed using luciferase activity assay. As shown in Fig. 3b, lobohedleolide dose-dependently reduced the HCV-induced luciferase activity driven by the COX-2/WT promoter and the COX-2/ΔNF-κB promoter, but exhibited no significant effect on that of luciferase activity driven by the COX-2/ΔNF-κB/C/EBP promoter, suggesting that the  To further verify which transcription factors are directly involved in the reduction effect of lobohedleolide on COX-2 promoter activity, Ava5 cells were transfected with individual luciferase reporter vector carrying each transcription element, including NF-κB, C/EBP, and AP-1. The results of the reporter assay showed that lobohedleolide significantly reduced the HCV-induced luciferase activity mediated by C/EBP (Fig. 4b), but exhibited no significant effect on that of luciferase activity driven by NF-κB or AP-1 (Fig. 4a,c). To confirm the inhibitory effect of lobohedleolide on the transcription factor activity of C/EBP in Ava5 cells, we created a COX-2 promoter reporter construct with a point mutation on C/EBP site alone, named as pCOX-2-(mC/EBP)-Luc (Fig. 4d). Ava5 cells were transfected with pCOX-2-Luc or pCOX-2-(mC/EBP)-Luc, and then the plasmid-transfected cells were incubated with lobohedleolide at concentrations ranging from 5 to 40 μM. As shown in Fig. 4e, the COX-2/mC/ EBP promoter activity was not affected by lobohedleolide compared with COX-2/WT promoter activity, confirming that the C/EBP transcription factor contributed to the suppression of HCV-induced COX-2 expression by lobohedleolide.

Lobohedleolide reduces HCV-induced C/EBPβ binding and expression, c-Jun phosphorylation, and the activation of JNK.
To further examined whether lobohedleolide could reduce the binding of AP-1 and C/EBPβ to the endogenous COX-2 promoter, we performed Chomatin Immunoprecipitation assay. The result showed that lobohedleolide decreased the C/EBP binding to COX-2 promoter (Fig. 5a), but lobohedleolide did not reduce the binding of AP-1 (Fig. 5b). A previous study has shown that the activated c-Jun could control the expression of C/EBP 23 . Therefore, we further examined the levels of C/EBPβ protein and phosphorylated c-Jun in lobohedleolide-treated Ava5 cells, and the results of Western blotting showed that lobohedleolide dose-dependently reduced the C/EBPβ protein level and the phosphorylation level of c-Jun (Fig. 5c). The activated JNK is the major upstream activator of c-Jun 24 . To investigate whether the phosphorylation of JNK is involved in the inhibitory effect of lobohedleolide on COX-2 transcription, Ava5 cells were treated with lobohedleolide at 40 μM for the indicated duration ranging from 0.5 to 6 h. The results of Western blotting showed that lobohedleolide significantly reduced the phosphorylation level of JNK in a time-dependent manner (Fig. 5d). In addition to the JNK/c-Jun-C/EBPβ signaling pathway, COX-2 expression is also controlled by other transcriptional factors, including NF-κB and MAPK 25 . We found that lobohedleolide had no significant impact on the phosphorylation level of NF-κB, ERK and p38 induced by HCV (Fig. 5e,f), suggesting that the JNK/c-Jun-C/EBPβ signaling pathway is the most possible mechanism of lobohedleolide-mediated inhibition of COX-2 expression.
Combination treatment of lobohedleolide with clinically used anti-HCV agents synergistically reduces HCV replication. To examine the potential of lobohedleolide as a supplementary therapeutic agent, we investigated the antiviral effect using combination treatments of lobohedleolide with clinically used HCV Cell lysates were subjected to luciferase activity assay. Data are presented as mean ± SD of at least three independent experiments, with each measurement carried out in triplicate. Asterisks indicate significant difference between lobohedleolide-and DMSO-treated Ava5 cells. *P < 0.05; **P < 0.01. drugs, including IFN-α, the NS3/4 A protease inhibitor telaprevir, the NS5B polymerase inhibitor sofosbuvir, and the NS5A inhibitor daclatasvir. The antiviral effects of the combination treatments were compared with the inhibitory effects of single treatments of the corresponding inhibitor in Ava5 cells, resulting in three CI values (ED 50 , ED 75 , and ED 90 ), through calculations as described in the Methods section. As shown in Table 1, the CI values are <1 for all ED 50 , ED 75 , and ED 90 values, ranging from 0.12 to 0.59, showing that lobohedleolide exhibited a synergistic effect with various inhibitors on HCV replication. There was no apparent cytotoxicity in all combination treatments (data not shown).

Discussion
Several studies have indicated that the inhibition of COX-2 by selective inhibitors or naturally available compounds could possibly be used as a strategy for inflammation treatment, cancer prevention, and suppression of virus replication 26-28 . In the present study, we demonstrated that lobohedleolide significantly reduced the HCV-induced COX-2 transcription, protein synthesis, and its metabolite PGE 2 production, resulting in the suppression of HCV replication (Figs 1 and 2). These observations indicate that lobohedleolide could possibly be used for the treatment of HCV-related liver diseases with chronic inflammation in addition to a potential agent to conquer HCV infection.
Resistance to DAA agents is an incoming threat in treating chronic HCV infection with long-term treatment due to high mutation frequencies of RNA viruses. Therefore, host-targeting antiviral agents can be considered as an alternative strategy for overcoming the viral resistance due to the extremely low mutation rate of host genome in eukaryotic cells and the increasing drug susceptibility in all genotypes and serotypes 29 . In addition, cocktail therapy has become a promising strategy to increase SVR and reduce drug resistance in HCV-infected patients 30,31 . In the present study, we suggest lobohedleolide as a suitable anti-HCV agent based on its potential to target host COX-2 expression. Furthermore, lobohedleolide exhibited a synergistic anti-HCV effect with either IFN-α or other FDA-approved DAA agents (Table 1). Accordingly, lobohedleolide might be considered as a An earlier study had demonstrated that lobohedleolide exhibited a reducing effect on LPS-induced iNOS and COX-2 expression in RAW264.7 macrophage cells 20 , but the precise anti-inflammatory mechanism of lobohedleolide was not clearly investigated. In this study, we clearly demonstrated that lobohedleolide reduced the HCV-induced COX-2 transcription level by mediating the C/EBP regulatory element (Figs 3 and 4). The C/EBP family is a homodimeric DNA-binding basic-leucine zipper (bZIP) transcription factor and consists of six proteins divided into two subgroups, namely, C/EBP-α, -β, and -δ and C/EBP-γ, -ε, and -ζ. 32 . C/EBP-β and -δ are considered as the major mediators for regulating COX-2 expression 33,34 . The effect of lobohedleolide on C/EBP isoform(s) involved in COX-2 expression in Ava5 cells remains to be investigated. In addition to C/EBP as an up-regulator for COX-2 expression during HCV infection, HCV-induced production of C/EBP mRNA causes an increase in the expression of inflammatory cytokines and chemokines, such as IL-1α, TNF-α, and CXCL1 35 , which contributes to chronic fibrogenesis and hepatocarcinogenesis 36 . Clinically, a high expression level of IL-1α, TNF-α, and IL-2 was observed in HCV-infected patients 37 . In addition, the expression of a wide range of genes and microRNAs, such as miR-181c and miR-122, are also regulated by C/EBP, which participate in cell growth and proliferation in HCV-related liver diseases or HCC 38,39 . Further investigation of the inhibitory effect of lobohedleolide on C/ EBP-regulated inflammation in different chronic injuries could be performed, which would provide another potential therapeutic possibility of using lobohedleolide against HCV-related diseases. In contrast to the upregulator of COX-2 expression, the WT p53 has been considered as a suppressor of COX-2 transcription as it competes with the TATA-binding protein 40 . Previous studies have shown that p53 expression is downregulated by the far upstream element (FUSE)-binding protein, leading to persistent HCV replication in most HCC tumors 41 . However, the relationship between p53 and COX-2 during HCV replication is not yet clearly verified. It will be useful to investigate the effect of lobohedleolide on p53-depedent inhibition on COX-2 expression during HCV infection.
In conclusion, our study results have revealed that lobohedleolide suppressed HCV replication by suppressing JNK phosphorylation, leading to the downregulation of c-Jun phosphorylation and C/EBP expression. Reduced COX-2 expression by lobohedleolide contributed to the suppression of PGE 2 , finally causing a reduction on HCV RNA replication (Fig. 6). The synergistic anti-HCV effect of lobohedleolide with IFN, telaprevir, or sofosbuvir showed that lobohedleolide may serve as a therapeutic supplementary agent for increasing the treatment efficacy or decreasing the possibility of drug resistance in HCV-infected patients.

Materials and Methods
Cell culture and reagents. Ava5, an engineered HCV subgenomic replicon cell line, Huh-7 and Huh7.5 cells were cultured in Dulbecco's modified Eagle's medium (DMEM) containing 10% heat-inactivated fetal bovine serum, 1% antibiotic-antimycotic, and 1% nonessential amino acids at 37 °C with a 5% CO 2 supplement. Ava5 cells have to be maintained in DMEM with 1 mg/ml G418 to maintain the stable expression of the replicon.
Preparation of lobohedleolide. Lobohedleolide was isolated from the Formosan soft coral L. crassum following a method that has been described previously 20 .
Western blotting. Ava5 cells were seeded into a 24-well plate at a density of 5 × 10 4 cells/well and into a 6-well plate at a density of 5 × 10 5 cells/well. After 12-16 h of incubation, cells were treated with the reagents for appropriate duration at the indicated concentrations. Then, the cells were washed with ice-cold PBS and lysed using RIPA lysis buffer. The insoluble protein was removed by centrifugation at 12,000 rpm for 30 min at 4 °C, and the protein concentration of the soluble lysate was measured by Bio-Rad protein assay kit (Hercules, CA, USA). Immunoblotting analysis was performed as previously described. Briefly, an equal amount of protein was loaded onto 10% SDS-polyacrylamide gel electrophoresis and transferred to PVDF membranes. The levels of the protein of interest were measured using specific antibodies against HCV NS5B (

HCV particle preparation and infection assay.
Full-length and linearized JFH-1 RNA was transfected into Huh7.5 cells for producing infectious HCV genomic type 2a JFH-1 virus particles 42 . After 3 days, the supernatant was collected and filtered through a 0.45-μm filter and stored at −70 °C until use. The infectivity rate of HCV JFH-1 was determined through serial diluting and infecting the Huh7.5 cells for 3 days. Immunostaining was performed with anti-core antibodies. Huh7.5 cells were infected with HCV JFH-1 particles at a multiplicity of infection of 0.1 for 6 h, and the infected cells were washed with PBS and replaced with fresh medium. Lobohedleolide was applied at various concentrations for an additional 3 days. Total RNA was extracted and subjected to qRT-PCR as described above.
Transfection and luciferase activity assay. The Ava5 cells were seeded into a 24-well plate at a density (Merck, Darmstadt, Germany). Briefly, Ava5 cells were treated with lobohedleolide for 3 days. Cells were then cross-linked with 1% formadehyde for 10 minutes. After washing with PBS, cells were lysed with Cell Lysis Buffer containing protease inhibitor cocktail. The chromatin was sheared by sonication. 100 ng of DNA were used as input. The remaining chromatin fraction were immunoprecipitated with anti-C/EBPβ (4 μg; GeneTex, CA, USA) or anti-AP-1 antibodies (10 μg; Sigma, MO, USA) overnight at 4 °C. The C/EBPβ-bound COX-2 promoter DNA was analyzed by PCR with specific primers: 5ʹ-cggagggtagttccatgaaa-3ʹ (forward), and 5ʹ-caggcttttacccacgcaaa-3ʹ (reverse Analysis of the drug synergism. Ava5 cells were seeded into a 24-well plate at a density of 5 × 10 4 cells/well and treated with serially diluted lobohedleolide at 2.5, 5, 10 and 20 μM in combination with diluted IFN-α (7.5, 15, 30, and 60 U/mL), the HCV protease inhibitor telaprevir (0.075, 0.15, 0.3, and 0.6 μM), or the RNA-dependent RNA polymerase nucleoside inhibitor sofosbuvir (10,20,40, and 80 nM). Each of the combination treatment was performed by adding lobohedleolide horizontally with various HCV inhibitors vertically in a checkerboard cross in the 24-well plate. After 3 days of treatment, total cellular RNA was extracted, and the RNA levels were quantified by qRT-PCR with specific primers as described above. The combination index (CI) values of each combination achieving 50%, 75%, or 95% reduction in the HCV RNA level were calculated using the CalcuSyn2TM computer program (Biosoft, Cambridge, UK), which was based on the Chou and Talalay analysis method 43,44 . Primarily, CI values of 1, <1, and >1 indicate additive, synergistic, and antagonistic effects, respectively.
Statistical analysis. All data were obtained from at least three independent experiments and are presented as mean ± SD. Statistical significance was determined using Student's t test for differences between two data groups (lobohedleolide-treated and -untreated cells). *P < 0.05 or **P < 0.01 was considered to be statistically significant.