Caffeic acid attenuates the autocrine IL-6 in hepatocellular carcinoma via the epigenetic silencing of the NF-κB-IL-6-STAT-3 feedback loop

Abstract

Hepatocellular carcinoma (HCC) is the third leading cause of cancer-related mortality worldwide. The autocrine IL-6 signaling pathway plays a key role in HCC progression. Caffeic acid (CaA) is a novel anti-tumor agent; however, the functions of CaA in the regulation of autocrine IL-6 in HCC, and the molecular mechanisms it is involved in remain unclear. In our present study, we found that CaA blocked the expression/secretion of endogenous IL-6 by the microRNA-124 (miR-124)-mediated attenuation of the NF-κB-IL-6-STAT-3 feedback loop in HCC cells. Indeed, CaA elevated the expression of miR-124 by inducing DNA-demethylation. MiR-124, which targeted the 3′-UTR of NF-κB/p65 and STAT-3 mRNAs, attenuated the expressions/activations of these two proteins and led to a transcriptional suppression of the IL-6 gene. Knockdown of miR-124 reversed the CaA-induced inhibitions of NF-κB and STAT-3, as well as autocrine IL-6. By understanding a novel mechanism whereby CaA inhibits the autocrine IL-6 in HCC, our study would help in the design of future strategies of developing CaA as a potential HCC chemo-preventive agent when used alone or in combination with other current drugs.

---

1. Introduction

Hepatocellular carcinoma (HCC) is the third leading cause of cancer-related mortality worldwide.¹ Cancer-related inflammation has been proposed as a major hallmark of malignancy.² In HCC, tumors produce a cancer promoting inflammatory microenvironment via recruiting and/or auto-secreting chemotactic factors, such as interleukin-6 (IL-6), IL-8, tumor necrosis factor-alpha (TNFα), signal transducers and activators of transcription-3 (STAT-3), and eicosanoids, which promote cellular proliferation and promote the malignant properties.³ Of these, IL-6 is a potent inflammatory cytokine, which plays a key role in HCC progression.^1,4 In general, IL-6 activates its down-stream factors, Janus kinases/STAT-3.⁵ Previous studies reveal that the NF-κB-dependent production of IL-6 is critical for diethylnitrosamine (DEN)-induced HCC in mice.^1,2 Moreover, gluconeogenesis is severely compromised in HCC via IL-6/STAT-3-mediated signal transduction.⁶ Furthermore, autocrine IL-6 exhibits an effective potential for the maintenance of liver cancer stem cells, which facilitates chemotherapy resistance and post-surgical recurrences.⁷ Notably, recent studies have successfully identified the liver cancer progenitors whose malignant progression depends on autocrine IL-6 signaling pathway.⁴ Therefore, targeting autocrine IL-6 has become a new approach for the treatment of HCC.

Caffeic acid (3,4-dihydroxycinnamic acid, CaA), a naturally occurring hydroxycinnamic acid derivative, is an active component in phenolic propolis extracts and is also found in a wide variety of plants.⁸ Caffeic acid has several biological and pharmacological properties such as antiviral, antioxidant, anti-inflammatory, and immunomodulatory activities.⁹ In addition to these functions, CaA effectively exerts anticancer functions; moreover, it decreases the viability of HCC cells in a time-dependent manner.¹⁰ In addition, it has been identified as a matrix metalloproteinase-9 (MMP-9) inhibitor that inhibits tumor invasion and metastasis.¹¹ Furthermore, CaA and its derivative, caffeic acid phenethyl ester (CAPE), suppresses the angiogenesis of human renal carcinoma cells by blocking the secretion of vascular endothelial growth factors (VEGF).⁸ Recent studies suggest that CaA reduces the cutaneous TNFα, IL-6 and IL-1β levels in mice.¹² However, the roles of CaA in the autocrine IL-6 signaling in HCC, and the molecular mechanisms involved therein, remain largely uninvestigated. Herein, we treated the HCC cells with CaA to investigate the effects of CaA on endogenous IL-6 secretion and the underlying molecular mechanisms involved.

2. Experimental

2.1. Cell culture and reagents

HCC cell lines HepG2 and MHCC97H were obtained from the Shanghai Institute of Cell Biology, Chinese Academy of Sciences. Cells were maintained in 5% CO₂ at 37 °C in Dulbecco's modified Eagle's medium (Gibco, NY, USA) supplemented with 10% fetal bovine serum (Gibco), 100 U per ml penicillin, and 100 μg ml⁻¹ streptomycin (Gibco). Caffeic acid (CaA, purity ≥ 99.5%) and S-adenosylmethionine (SAM, a methyl donor) were purchased from Sigma Chemical Co. (St. Louis, MO, USA). All the other reagents used were of analytical grade or the highest grade available.

2.2. Determination of cell viability

A total of 2 × 10³ cells were seeded in 96-well plates. The next day, they were treated by 0, 10, 20, or 40 μM of CaA for 24 h. Then, the cells were incubated with 20.0 μl of WST-8 (CAS 193149-74-5) solution (Santa Cruz Biotechnology, TX, USA) for another 4 h. The absorbance at 450 nm was measured with a multimode microplate reader (Tecan, Männedorf, Switzerland). The values for medium control cells were used to determine the 100% level.

2.3. Enzyme-linked immunosorbent assay (ELISA)

To analyze IL-6 secretion, a total of 1 × 10⁵ cells were seeded in 6-well plates. The next day, they were treated by 0 or 20 μM of CaA for 24 h. Then, the mediums were collected, cleared by centrifugation and ELISA was performed using the human IL-6 Quantikine kit (R&D Systems, MN, USA) according to the manufacturer's protocol. Recombinant human IL-6 was used for calibration. The absorbance at 450 nm was measured with a multimode microplate reader (Tecan). The values for the 0 μM CaA-treated group was used to determine the 100% level.

2.4. Quantitative real-time polymerase chain reaction (qRT-PCR)

Total RNA was isolated using Trizol (Invitrogen, CA, USA) according to the manufacturer's recommendations. For the detection of mRNAs, total RNA (2 μg) was transcribed into cDNA using AMV Reverse Transcriptase (Promega, WI, USA). The primers used were NF-κB/p65, GGACTACGACCTGAATG (F), GGCACGATTGTCAAAGATA (R); STAT-3, GACCAGATGCGGAGAAGC (F), CGGTCTTGATGACGAGG (R); IL-6, CAGCCACTCACCTCTTCA (F), CACTGTCTTTGAGCCTGTC (R). For the detection of miR-124, total RNA, miR-124-specific stem-loop RT primers (RiboBio Co, Guangzhou, China), and MMLV reverse transcriptase (Promega) were used in reverse transcription as per the manufacturer's protocol. The qRT-PCR was performed using an ABI 7300 real-time PCR detection system (Applied Biosystems, NY, USA). Changes in the expression of each gene were calculated by a comparative threshold cycle (C_t) method using the formula 2^−(ΔΔC_t).

2.5. Western blots

Cells were scraped off in 0.2 ml of lysate buffer (Beyotime). Protein concentrations were measured with the BCA protein assay (Beyotime). Furthermore, the proteins (20 μg) were transferred to polyvinylidene fluoride membranes (Millipore, MA, USA) that were then probed with primary antibodies [RelA, p-RelA (ser536), STAT-3, and p-STAT-3 (Tyr 705), Cell Signaling Technology, Beverly, MA, USA, dilutions, 1 : 1000] at 4 °C overnight. Then, the membranes were incubated with horseradish peroxidase-conjugated secondary antibodies (Beyotime) for 1 h at room temperature. The antibody complexes were detected by enhanced chemiluminescence (Cell Signaling Technology). Glyceraldehyde phosphate dehydrogenase (GAPDH, Sigma) was used for normalization of protein loading.

2.6. MicroRNA transfection

Con-mimic, miR-124-mimic, anti-con, and anti-miR-124 were synthesized by RiBoBio Co. Cells were transiently transfected using the Lipofectamine 2000 reagent (Invitrogen) according to the manufacturer's protocol. Briefly, cells were seeded in 6-well plates at a density of 1 × 10⁶ per well. After 24 h, they were transfected with 50 nM anti-miR-124 or 20 nM miR-124-mimic for 12 h. After transfection, cells were conventionally cultured for another 24 h before being used for other experiments.

2.7. Luciferase reporter assay

The pGL3-STAT-3 (or NF-κB/p65)-3′-UTR (wild type, WT; or mutant, MT)-Luc constructs were synthesized by Shuntian Bio Co. (Shanghai, China). The plasmid phRL-tk, containing the Renilla luciferase gene, was purchased from Promega. Briefly, cells were plated in 24-well culture dishes. When cells proliferated from 60% to 80% confluence, con-mimics or miR-124-mimics were co-transfected with the respective reporter construct using the Lipofecamine 2000 reagent (Invitrogen) according to the manufacturer's protocol. The cells were lysed with passive lysis buffer (Promega), and the lysates were analyzed immediately with a 96-well plate luminometer (Berthold Detection System, Pforzheim, Germany). The amounts of luciferase and Renilla luciferase were measured with the Dual-Luciferase Reporter Assay System Kit (Promega) following the manufacturer's instructions.

2.8. DNA methylation analysis

The genomic DNA was isolated using DNA purification kits (Qiagen, Germantown, MD, USA) followed by the modification with sodium bisulfite using the EpiTect kit (Qiagen). DNA methylation was analyzed using a SYBR Green-based quantitative methylation-specific PCR (qMSP) as described previously.¹³ The primers used were: miR-124-methylated, CGAAATCGTTTTTTTGGACGCGG (F), GACGGAGGTCGCGATCGGTCGATT (R); miR-124-unmethylated, TGAAATTGTTTTTTTGGATGTGG (F), ATGGAGGTTGTGATTGGTTGAT (R). Briefly, 1 μl of bisulfite-treated DNA template was mixed with 10 μl of 2 × Power SYBR Green PCR Master Mix (Applied Biosystems) and a pair of primers in a final concentration of 400 nM. The PCR conditions included an initial incubation period at 50 °C for 2 min, denaturing at 95 °C for 10 min, and 40 cycles of denaturing at 95 °C for 15 s and annealing at 60 °C for 1 min.

2.9. Statistical analysis

Data were presented as the mean ± SD. A Student's t test and a one-way analysis of variance (ANOVA) followed by Dunnett's t test were used to assess significant differences between groups. In our present study, p values <0.05 were considered statistically significant.

3. Results

3.1. CaA attenuated the autocrine IL-6 in HCC cells

First, we determined the concentrations of CaA used in our study. MHCC97H and HepG2 cells were treated with different concentrations (10, 20, or 40 μM) of CaA for 24 h. As shown in Fig. 1A, there were viability decreases in the cells exposed to 40 μM CaA; however, no significant attenuations of viabilities were observed in the cells exposed to 10 and 20 μM CaA. Then, we determined the effects of CaA on IL-6 secretion. These two cells were treated by 10 or 20 μM of CaA for 24 h and the conditioned mediums were collected. Notably, CaA inhibited the secretion of IL-6 in a dose-dependent manner (Fig. 1B). As 20 μM of CaA effectively attenuated the secretion of autocrine IL-6 and had no significant cytotoxicity, we chose this concentration for further investigation.

Fig. 1 CaA attenuated the levels of autocrine IL-6 in HCC cells. (a) HepG2 and MHCC97H cells were treated by 0, 10, 20, or 40 μm CaA for 24 h. The cell viabilities were evaluated in triplicate by WST-8 hydrolysis using a cell counting Kit-8 assay. (b) Cells were treated by 0, 10 or 20 μm CaA for 24 h, and then the conditioned mediums were collected. The IL-6 secretions were determined in triplicate by ELISA. (c) Cells were treated by 0 or 20 μm CaA for 8, 16, or 24 h, qRT-PCR analyses in triplicate of IL-6 mRNA. *p < 0.05 and **p < 0.01 compared with cells not treated with CaA.

Next, we detected the effects of CaA on the expression of endogenous IL-6. MHCC97H and HepG2 cells were treated with 20 μM CaA for 8, 16, or 24 h. As shown in Fig. 1C, CaA blocked the expressions of IL-6 mRNA in a time-dependent manner. Collectively, these results suggest that CaA attenuated the secretion of autocrine IL-6 by the transcriptional inhibition of endogenous IL-6 in HCC cells.

3.2. Effects of CaA on the NF-κB and STAT-3 in HCC cells

For the past few years, two major factors that regulate the IL-6 expression in human cancers have been identified. These factors are NF-κB and STAT-3, both of which synergize in the regulation of the IL-6 gene at the transcriptional level.¹⁴ Herein, the knock down of either NF-κB or STAT-3 attenuated the expression of IL-6 mRNA in HepG2 and MHCC97H cells (Fig. 2A); the efficiency of the knock down of NF-κB or STAT-3 is exhibited in the ESI (Fig. S1†). Therefore, we next determined the effects of CaA on NF-κB and STAT-3 in HCC cells. As shown in Fig. 2B, CaA blocked the activations of NF-κB and STAT-3, as determined by the phosphorylations of NF-κB/p65 and STAT-3, in a time-dependent manner. Interestingly, with increased CaA exposure times, there were also greater decreased expressions of total NF-κB/p65 and STAT-3 proteins (Fig. 2B), as well as their mRNAs (Fig. 2C).

Fig. 2 Effects of CaA on the NF-κB and STAT-3 in HCC cells. (a) HepG2 and MHCC97H cells were transfected by si-NC, si-NF-κB/p65, or si-STAT-3 for 12 h. qRT-PCR analyses in triplicate of IL-6 mRNA. *p < 0.05 and **p < 0.01 compared with cells transfected by si-NC. (b and c) HepG2 and MHCC97H cells were treated by 0 or 20 μm CaA for 8, 16, or 24 h. (b) Western bolt analyses of p-NF-κB/p65, NF-κB/p65, p-STAT-3, and STAT-3 proteins. (c) qRT-PCR analyses in triplicate of NF-κB/p65 and STAT-3 mRNAs. *p < 0.05 and **p < 0.01 compared with cells not treated with CaA.

Collectively, these results indicated that, in HCC cells, the CaA-induced IL-6 inactivation was primarily attributed to the blockages of NF-κB and STAT-3. Furthermore, the repressive effects of CaA on these two transcription factors may not be via classical kinase signaling pathways because CaA decreased the expressions of total NF-κB/p65 and STAT-3 proteins/mRNAs. miRNAs are key regulators of gene expression through degrading mRNA or repressing translation in a post-transcriptional manner. Based on our current results, we hypothesized that miRNA might play a role in the CaA-induced attenuations of NF-κB and STAT-3 in HCC cells.

3.3. Identification of a functional miRNA in CaA-treated HCC cells

Previous studies revealed that miR-124 targets STAT-3, which attenuates the lipopolysaccharide (LPS)-induced IL-6 production.¹⁴ Herein, using TargetScan 6.2 (www.targetscan.org), we found a miR-124 binding site in the 3′-UTR of NF-κB/p65 mRNA. Therefore, we hypothesized that miR-124 might be involved in the CaA-induced inhibitions of NF-κB and STAT-3. Verifying this hypothesis, we first detected the effects of CaA on the expression of miR-124. As shown in Fig. 3, CaA enhanced the miR-124 level in MHCC97H and HepG2 cells in a time-dependent manner.

Fig. 3 CaA improved the miR-124 expression in HCC cells. HepG2 and MHCC97H cells were treated by 0 or 20 μm CaA for 8, 16, or 24 h, qRT-PCR analyses in triplicate of miR-124. **p < 0.01 compared with cells not treated with CaA.

Then, we determined the relationship between miR-124, NF-κB, and STAT-3 in HCC cells. The miR-124 binding sites are illustrated in the ESI (Fig. S2†). A luciferase reporter assay showed that the cells co-transfected with the wild type (WT) constructs and a miR-124-mimic led to significant decreases of the luciferase activities (Fig. 4A). Furthermore, we determined the functional relevance between the miR-124s′ binding effects and the expressions of NF-κB/p65 and/or STAT-3 mRNA. As shown in Fig. 4B, overexpression of miR-124 decreased the endogenous expressions of these two mRNAs in MHCC97H and HepG2 cells; the efficiency of overexpression of miR-124 is exhibited in the ESI (Fig. S3†).

Fig. 4 miR-124 targeted NF-κB/p65 and STAT-3 in HCC cells. (a) MHCC97H cells were co-transfected with the reporter constructs plus 20 nm mimic-NC or mimic-miR-124 for 12 h. luciferase reporter assay analyses in triplicate of the effects of miR-124 on the 3′-UTR of NF-κB/p65 (left) and STAT-3 (right). (b) HepG2 and MHCC97H cells were transfected by 20 nm mimic-NC or mimic-miR-124 for 12 h, qRT-PCR analyses in triplicate of the NF-κB/p65 (left) and STAT-3 (right) mRNAs. **p < 0.01 compared with cells transfected by mimic-NC.

Based on these results, we hypothesized that miR-124 might be involved in the CaA-induced attenuations of NF-κB and STAT-3, and in the attenuation of IL-6 expression/secretion in HCC cells. Therefore, we treated the wild type HCC cells or miR-124 knockdown HCC cells with CaA to compare the differences; the efficiency of the knockdown of miR-124 is exhibited in the ESI (Fig. S4†). In anti-NC-transfected cells, CaA decreased the expressions of NF-κB/p65 (Fig. 5A), STAT-3 (Fig. 5B), IL-6 (Fig. 5C) mRNAs, as well as IL-6 secretion (Fig. 5D); however, in anti-miR-124-transfected cells, these functions were attenuated (Fig. 5A–D). Collectively, these results suggest that the binding of miR-124 to the 3′-UTR of NF-κB/p65 and STAT-3 mRNAs facilitated the CaA-induced inhibitions of NF-κB and STAT-3, which in turn, decreased the autocrine IL-6 levels in HCC cells.

Fig. 5 CaA blocked NF-κB, STAT-3, and autocrine IL-6 by miR-124 in HCC cells. After the cells were pre-transfected by anti-NC or anti-miR-124 for 12 h, they were exposed to 20 μm CaA for 24 h. qRT-PCR analyses in triplicate of the (a) NF-κB/p65, (b) STAT-3, and (c) IL-6 mRNAs. (d) Analyses in triplicate of IL-6 secretion by ELISA (right). **p < 0.01 compared with anti-NC-transfected cells not treated with CaA, ##p < 0.01 compared with anti-NC-transfected cells treated with 20 μm CaA.

3.4. CaA improved the miR-124 expression by demethylation in HCC cells

Based on our findings mentioned above, we indicated that miR-124 played an important role in the attenuation of autocrine IL-6 secretion induced by CaA. Therefore, we further determined the molecular mechanisms whereby CaA enhanced miR-124. Studies indicate that there are large amounts of CpG-rich regions in the miR-124 promoter region, and that the silencing of miR-124 by DNA hypermethylation is a common event in HCC.¹⁵ Since the biotransformation of CaA results in a deficiency of methyl donors, which reduces DNA methylation (illustrated in ESI, Fig. S5†), and catechol-O-methyltransferase (COMT), which is involved in the biotransformation of CaA, is rich in human HCC cells,¹⁶ we hypothesized that CaA up-regulated the expression of miR-124 by inducing the DNA demethylation.

To confirm this hypothesis, DNA methylation analysis (qMSP) was conducted. As shown in Fig. 6A, the average methylation level of the miR-124 promoter was high in medium control MHCC97H cells, but low in CaA-treated cells. This data confirmed that CaA caused DNA demethylation in the miR-124 promoter region. We then tested the functional relevance of CaA-induced DNA demethylation and increased expression of miR-124. As shown in Fig. 6B, hypermethylation by SAM dramatically blocked the CaA-induced elevation of miR-124. In addition, the concentration of SAM used here had no detectable cytotoxicity (see the ESI, Fig. S6†). These results suggested that CaA improved the miR-124 expression by inducing the DNA demethylation in HCC cells.

Fig. 6 DNA demethylation was involved in CaA-improved miR-124 expression. (a) MHCC97H cells were treated by 0 or 20 μm CaA for 24 h, qMSP analyses (left) and average methylation level (right) of the miR-124 promoter. Annotation: m, methylated; u, unmethylated. (b) HepG2 and MHCC97H cells were treated by 0 or 20 μm CaA in the presence, or absence, of 100 μm SAM for 24 h, qRT-PCR analyses in triplicate of miR-124. *p < 0.05 and **p < 0.01 compared with cells not treated with no CaA; ##p < 0.01 compared with cells treated with 20 μm CaA alone.

4. Discussion

Hydroxycinnamic acid derivatives are reported to have anticancer, anti-inflammatory, and antioxidant properties. Their natural origin and ubiquitous occurrence have prompted strong interest in the use of them as anticancer agents.¹⁷ CaA is a major representative of hydroxycinnamic acids.¹⁷ Previous studies have indicated that the daily intake of coffee was associated with a reduced incidence of colon and rectal cancers.^18,19 In the human diet, daily intake of CaA in coffee drinkers is about 0.5–1 g (approximate 0.5–1 mM), and the absorption ratio of CaA is about 95%. However, a small part of CaA from foods will enter into the blood circulation, but most will reach the colon,²⁰ so we used a relatively lower concentration of CaA (20 μM) to investigate the functions of CaA on the progression of HCC. Herein, CaA did not appreciably affect the viability of MHCC97H and MHCC97L cells. However, CaA attenuates the autocrine IL-6 levels in these cells effectively, which may be associated with the transcriptional inhibition of the IL-6 gene.

For the transcriptional regulation of IL-6, two major factors in human cancers have been identified and they are NF-κB and STAT-3.¹⁴ In fact, IL-6, NF-κB, and STAT-3 constitute a positive feedback loop. On the one hand, the NF-κB, and/or STAT-3 elevate the IL-6 expression at the transcriptional level.¹⁴ On the other hand, IL-6 activates the NF-κB and TAT-3 via its specific manner of action. For example, IL-6 induces classical IKKβ-IκBα-NF-κB/p65 signal activation via the suppression of miR-200c.²¹ Moreover, by binding to its receptor complex, consisting of IL-6Rα and glycoprotein 130, IL-6 activates Janus kinases (JAK)/STAT-3 signaling.²² Furthermore, IL-6 mediates the cross-talk between NF-κB and STAT-3 in human cancer cells.²³ Therefore, the relationships among IL-6, NF-κB, and STAT-3 are very complex. In our present study, CaA decreased the expressions of total NF-κB/p65 and STAT-3, indicating an exhaustive abolishment of the NF-κB-IL-6-STAT-3 feedback loop. Furthermore, we identified that the repressive effects of CaA on NF-κB and STAT-3 were primarily attributed to the improvement of miR-124 expression.

MicroRNAs are small non-coding RNAs that regulate gene expression by binding to the 3′-UTR of the target mRNA thus inhibiting translation or targeting the mRNA for degradation. The abnormal expressions of miRNAs are associated with a variety of human cancers, including HCC.²⁴ miR-124 has been identified as an important tumor suppressor in HCC. For example, it suppresses the proliferation of HCC cells by targeting the phosphoinositide-3-kinase catalytic subunit alpha (PIK3CA) and/or STAT-3.^14,25 In addition, it also attenuates the aggressive properties in HCC cells by repressing ROCK2 and EZH2.²⁶ Moreover, miR-124 plays an important role in the hepatocyte nuclear factor 4α-mediated inhibition of NF-κB signal pathway.²⁷ Previous studies suggest that the epigenetic silencing of miR-124 by the hepatitis C virus core protein promotes the migration and invasion of intrahepatic cholangiocarcinoma cells.²⁸ In our present study, CaA elevated the expression of miR-124, which in turn, bound to the 3′-UTR regions of NF-κB/p65 and STAT-3 mRNAs, and attenuated the expressions of these two proteins. Furthermore, using qMSP, we found that the CaA-induced activation of miR-124 was dependent on DNA-demethylation.

5. Conclusions

In conclusion, CaA attenuated the expression/secretion of endogenous IL-6 by the miR-124-mediated abolishment of the NF-κB-IL-6-STAT-3 feedback loop in HCC cells. Indeed, CaA enhanced the expression of miR-124 by inducing DNA demethylation. miR-124, which targeted the 3′-UTR regions of NF-κB/p65 and STAT-3 mRNAs, decreased the expressions/activations of these two proteins. Knock down of miR-124 reversed the CaA-induced inhibitions of NF-κB and STAT-3, as well as the levels of autocrine IL-6.

Conflict of interest

The authors have no conflict of interest.

Acknowledgements

The authors are grateful to the National Natural Science Foundation of China (81402667), the Collegiate Natural Science Foundation of Jiangsu Province (14KJB330003), the project funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD-2014), the Research Fund for the Doctoral Program of Jiangsu Province (1302050C), and the Technology Development Fund of Nanjing Medical University (2013NJMU021).

References

1. J. Prieto, J. Hepatol., 2008, 48(2), 380–381.
2. W. E. Naugler, T. Sakurai, S. Kim, S. Maeda, K. Kim, A. M. Elsharkawy and M. Karin, Science, 2007, 317(5834), 121–124.
3. D. Capece, M. Fischietti, D. Verzella, A. Gaggiano, G. Cicciarelli, A. Tessitore, F. Zazzeroni and E. Alesse, BioMed Res. Int., 2013, 2013, 187204.
4. G. He, D. Dhar, H. Nakagawa, J. Font-Burgada, H. Ogata, Y. Jiang, S. Shalapour, E. Seki, S. E. Yost, K. Jepsen, K. A. Frazer, O. Harismendy, M. Hatziapostolou, D. Iliopoulos, A. Suetsugu, R. M. Hoffman, R. Tateishi, K. Koike and M. Karin, Cell, 2013, 155(2), 384–396.
5. J. C. Nault and J. Zucman Rossi, Int. J. Hepatol., 2013, 2013, 315947.
6. B. Wang, S. H. Hsu, W. Frankel, K. Ghoshal and S. T. Jacob, Hepatology, 2012, 56(1), 186–197.
7. D. Iliopoulos, H. A. Hirsch and K. Struhl, Cell, 2009, 139(4), 693–706.
8. J. E. Jung, H. S. Kim, C. S. Lee, D. H. Park, Y. N. Kim, M. J. Lee, J. W. Lee, J. W. Park, M. S. Kim, S. K. Ye and M. H. Chung, Carcinogenesis, 2007, 28(8), 1780–1787.
9. M. Touaibia, J. Jean-Francois and J. Doiron, Mini-Rev. Med. Chem., 2011, 11(8), 695–713.
10. E. Guerriero, A. Sorice, F. Capone, S. Costantini, P. Palladino, M. D'Ischia and G. Castello, Molecules, 2011, 16(8), 6365–6377.
11. W. H. Park, S. H. Kim and C. H. Kim, Toxicology, 2005, 207(3), 383–390.
12. M. Zhang, J. Zhou, L. Wang, B. Li, J. Guo, X. Guan, Q. Han and H. Zhang, Biol. Pharm. Bull., 2014, 37(3), 347–354.
13. L. Lu, D. Katsaros, I. A. de la Longrais, O. Sochirca and H. Yu, Cancer Res., 2007, 67(21), 10117–10122.
14. Y. Sun, Q. Li, H. Gui, D. P. Xu, Y. L. Yang, D. F. Su and X. Liu, Cell Res., 2013, 23(11), 1270–1283.
15. M. Furuta, K. I. Kozaki, S. Tanaka, S. Arii, I. Imoto and J. Inazawa, Carcinogenesis, 2010, 31(5), 766–776.
16. P. Wang, D. Heber and S. M. Henning, Food Funct., 2012, 3(6), 635–642.
17. Y. Yang, Y. Li, K. Wang, Y. Wang, W. Yin and L. Li, PloS one, 2013, 8(3), e58915.
18. K. B. Michels, W. C. Willett, C. S. Fuchs and E. Giovannucci, J. Natl. Cancer Inst., 2005, 97(4), 282–292.
19. A. Tavani, A. Pregnolato, C. La Vecchia, E. Negri, R. Talamini and S. Franceschi, Int. J. Cancer, 1997, 73(2), 193–197.
20. M. R. Olthof, P. C. Hollman and M. B. Katan, J. Nutr., 2001, 131(1), 66–71.
21. M. Rokavec, W. Wu and J. L. Luo, Mol. Cell, 2012, 45(6), 777–789.
22. J. V. Alvarez, P. G. Febbo, S. Ramaswamy, M. Loda, A. Richardson and D. A. Frank, Cancer Res., 2005, 65(12), 5054–5062.
23. C. H. Squarize, R. M. Castilho, V. Sriuranpong, D. S. Pinto Jr and J. S. Gutkind, Neoplasia, 2006, 8(9), 733–746.
24. Y. Zhou, Y. Li, J. Ye, R. Jiang, H. Yan, X. Yang, Q. Liu and J. Zhang, Liver Int., 2013, 33(8), 1271–1280.
25. Q. Lang and C. Ling, Biochem. Biophys. Res. Commun., 2012, 426(2), 247–252.
26. F. Zheng, Y. J. Liao, M. Y. Cai, Y. H. Liu, T. H. Liu, S. P. Chen, X. W. Bian, X. Y. Guan, M. C. Lin, Y. X. Zeng, H. F. Kung and D. Xie, Gut, 2012, 61(2), 278–289.
27. B. F. Ning, J. Ding, J. Liu, C. Yin, W. P. Xu, W. M. Cong, Q. Zhang, F. Chen, T. Han, X. Deng, P. Q. Wang, C. F. Jiang, J. P. Zhang, X. Zhang, H. Y. Wang and W. F. Xie, Hepatology, 2014, 60, 1607–1619.
28. B. Zeng, Z. Li, R. Chen, N. Guo, J. Zhou, Q. Zhou, Q. Lin, D. Cheng, Q. Liao, L. Zheng and Y. Gong, FEBS Lett., 2012, 586(19), 3271–3278.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra05878c

‡ These author contributed equally to this work.

---