Methylated chrysin induces co-ordinated attenuation of the canonical Wnt and NF-kB signaling pathway and upregulates apoptotic gene expression in the early hepatocarcinogenesis rat model

Chemico-Biological Interactions 193 (2011) 12–21

Contents lists available at ScienceDirect

Chemico-Biological Interactions journal homepage: www.elsevier.com/locate/chembioint

Methylated chrysin induces co-ordinated attenuation of the canonical Wnt and NF-kB signaling pathway and upregulates apoptotic gene expression in the early hepatocarcinogenesis rat model Mahaboob S. Khan a,b, Devaraj Halagowder c, S. Niranjali Devaraj a,⇑ a b c

Department of Biochemistry, University of Madras, Guindy Campus, Chennai, India Department of Biochemistry, Govt. Home Science College, Panjab University, Chandigarh, India Unit of Biochemistry, Department of Zoology, University of Madras, Guindy, Chennai, India

a r t i c l e

i n f o

Article history: Received 9 February 2011 Received in revised form 19 April 2011 Accepted 19 April 2011 Available online 30 April 2011 Keywords: Preneoplastic nodules Wnt pathway NF-kB pathway Apoptosis Methylated flavone

a b s t r a c t Hepatocellular carcinoma (HCC), a highly aggressive form of solid tumor, has been increasing in South East Asia. The lack of effective therapy necessitates the introduction of novel chemopreventive strategies to counter the substantial morbidity and mortality associated with the disease. Recently, we reported that dimethoxy flavone (DMF), a methylated flavone derived from chrysin, significantly suppressed the development of preneoplastic lesions induced by N-nitrosodiethylamine (DEN) in rats, although the mechanism of action was not known. In the present study, we have investigated the effects of DMF administration on gene expression changes related to the inflammation-mediated NF-kB pathway, Wnt pathway and apoptotic mediators in DEN-induced preneoplastic nodules. There was a significant increase in inflammatory markers like cyclooxygenase-2 (COX-2) and inducible nitric oxide synthase (iNOS) and a decrease in apoptotic mediators like p53, caspase-3 and bax in DEN-treated rats when compared to the control group. Activation of NF-kB was noticed by an elevated expression of nuclear protein expression of NF-kB and cytoplasmic phospho-IkBaSer32/36 in the same animals. Likewise, upregulation of canonical Wnt pathway was noticed by elevated expression of nuclear protein levels of phospho-b-cateninThr393 and cytoplasmic casein kinase-2 (CK2), Dvl2 and cyclin D1 levels, along with a simultaneous decrease in expression of phospho-GSK3bSer9. Dietary DMF (100 mg/kg) administration inhibited liver nodule incidence and multiplicity by 82% and 78%, respectively. DMF also reversed the activation of NF-kB and Wnt pathway as shown by the decrease in protein expression of several proteins. Results of the present investigation provide evidence that attenuation of Wnt pathway and suppression of inflammatory response mediated by NF-kB could be implicated, in part, in the chemopreventive effects of methylated flavone. Therefore, the present findings hold great promise for the utilization of DMF as an effective chemotherapeutic agent in treating early stages of liver cancer. Ó 2011 Published by Elsevier Ireland Ltd.

1. Introduction Primary hepatocellular carcinoma (HCC) is one of the most frequently occurring forms of a solid tumor. It exhibits a high prevalence with 620,000 cases per year reported worldwide of which more than eighty percent of cases are reported from China, Africa and South East Asia [1]. It is highly aggressive, as shown by the mortality of 595,000 cases per year that nearly matches the incidence of this tumor type [2]. HCC presents with limited therapeutic

Abbreviations: DEN, diethylnitrosamine; qRT-PCR, quantitative real-time polymerase chain reaction; Wnt, wingless; DMF, 5,7-dimethoxy flavone. ⇑ Corresponding Author. Address: Dept. of Biochemistry, University of Madras, Guindy Campus, Chennai 600025, India. Tel.: +91 44 22202730; fax: +91 44 2223 5870. E-mail address: [email protected] (S.N. Devaraj). 0009-2797/$ - see front matter Ó 2011 Published by Elsevier Ireland Ltd. doi:10.1016/j.cbi.2011.04.007

options. Hence, a thorough understanding of the biological bases of this malignancy might suggest new strategies for effective treatment [3]. Hepatocarcinogenesis induced by DEN is an ideal animal model to investigate liver tumor formation because it proceeds in stages similar to that of human liver cancer i.e., formation of preneoplastic foci, neoplastic nodules and HCC nodules [4]. A recent transcriptome–genotype–phenotype correlation analysis in HCC revealed that 50% of the tumors were related to either Wnt or Akt pathway activation [5]. We previously reported that expression of genes involved in the Akt pathway is altered during liver cancer in rats [6]. Dietary flavonoids are promising chemopreventive agents since they influence pleiotropic intracellular functions. The core structure of the flavone group, 2-phenyl-4H-1-benzopyran-4-one (flavone) has already been shown to be a potent inducer of apopto-

M.S. Khan et al. / Chemico-Biological Interactions 193 (2011) 12–21

sis in vitro [7]. Chrysin can render a global hepato-protective effect irrespective of the cause of the underlying injury [8]. We recently reported that chrysin can abrogate early hepatocarcinogenesis and induce apoptosis in preneoplastic nodules induced by DEN [9]. 5,7-Dimethoxyflavone (DMF) is derived from methylation of chrysin and occurs naturally in pepper tree leaves [10]. Interestingly, DMF is postulated to be more effective than chrysin due to greater bioavailability and metabolic stability [11]. Recent evidences suggest a plausible mode of action of DMF. Treatment with DMF can influence mRNA levels of genes involved in apoptosis and cell cycle regulation favourably during DEN-induced preneoplastic lesions in rat liver [12]. To get insight into the mechanism of growth-inhibition, we assessed whether DMF affects two pathways, the canonical Wnt and NFkB pathways, recently shown to function cooperatively in breast and colon cancer cells [13]. NFkB has been reported to regulate the survival of tumor cells and to link inflammation with tumor progression [14–16]. As the Wnt/b-cat pathway is known to play a vital role in various aspects of liver biology [17] as well as regulate inflammation [18], we decided to investigate wnt signaling as a possible target for DMFmediated inhibition of preneoplastic lesion formation. Wnt’s are believed to exert their effect through two divergent intracellular signaling pathways: the ‘‘canonical pathway’’ that controls gene transcription through b-catenin, and the ‘‘non-canonical’’ pathway that leads to release of intracellular calcium. The canonical Wnt pathway influences the cytoplasmic turnover and accumulation of b-catenin, a primary event in the pathogenesis of hepatic tumors [19]. Aberrant activation of b-catenin has been implicated in the pathogenesis of hepatobiliary neoplasia ranging from benign lesions to liver cancer [20]. Significantly, the role of b-catenin in early experimental hepatocarcinogenesis is less well defined. Activation of Wnt pathway leads to accumulation of b-catenin in the cytoplasm that eventually translocates into the nucleus and initiates transcription from the TCF/LEF promoter leading to transcription of several target genes controlling cell cycle proliferation such as cyclin D1, c-Myc, PCNA, and c-Jun. Casein Kinase 2 (CK 2) has been reported to function as a positive regulator of Wnt signaling and was recently found in complex with b-catenin in Wnt-activated cells [21]. Notably, many naturally occurring flavonoids like quercetin, apigenin and emodin have been shown to inhibit CK2, an integral part of the Wnt pathway [22]. Recent studies also indicate that blocking the action of CK2 inhibits carcinogenesis [23]. Hence, CK2 is increasingly perceived as a druggable target in the treatment of many human diseases. However, the role of flavonoids in modulating expression of Wnt signaling pathway genes has not been studied, perhaps due to a paucity of an effective in vivo experimental model. Therefore, the present study was designed to closely examine changes in mRNA and protein expression levels of inflammatory mediators like COX-2, iNOS along with the NF-kB and Wnt pathway in the early hepatocarcinogenesis model. Briefly, alterations in mRNA levels were determined by semiquantitative RT-PCR for COX-2, NF-kB p65, b-catenin, CK2, cyclin D1, c-jun, p53 and Gapdh. Furthermore, to understand the pleiotropic nature of DMF action, protein expression related to the NF-kB and Wnt signaling pathway has been investigated by Western blotting in order to determine the possible targets of DMF-mediated abrogation of hepatocellular carcinogenesis.

(DMF), as shown in Fig. 1A, was prepared using the protocol described by Zheng et al. [24]. Primary antibodies for phospho-bcatenin Thr393, phospho-b-catenin Ser33/37, phospho-GSK3b Ser9, phospho-IkB Ser32/36,, phospho-NFkB p65 Ser529 and COX-2 were purchased from Santa cruz Biotechnology (Santa Cruz, CA, USA). Anti-p53 antibody was obtained from Calbiochem, anti-Bcl-xL antibody from BD Transduction Laboratories, anti-CK2a from Millipore, anti-NFkB from Abcam and anti-b-actin antibody was purchased from Chemicon. Primary antibodies for GST-Pi, Dvl2, Bax, Caspase 3, and Cyclin D1 were purchased from Cell signaling Technology (Beverly, MA, USA). Secondary antibodies of anti-mouse and anti-rabbit origin conjugated with horseradish peroxidase (HRP) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA) and anti-goat HRP conjugate was purchased from Bangalore Genei, India Ltd. Protein assay kit was purchased from Bio-Rad Laboratories.

2.2. Animal treatment All animal experiments were conducted in accordance with the guidelines of committee for the purpose of control and supervision of experiments on animals (CPCSEA), Government of India, and prior permission was sought from the institutional animal ethics committee (IAEC No: 01/089/09) for conducting the study. Briefly, 64 male Wistar rats (Kings Institute, Chennai), weighing 140–160 g at the beginning of the study, were randomised into four different groups of 16 animals each as described in Khan et al. [9]. They were kept in plastic cages on rice husk as bedding and provided with feed and water ad libitum, at room temperature [12 h light (7 a.m to 7 p.m, 12 h dark); 26 ± 2 °C]. Group A served as normal control rats and were fed with a standard basal diet. Dosage fixation of DMF was done as described earlier by Khan et al. [12]. Group B rats were fed normal diet along with DMF (100 mg/kg), three times weekly by oral gavage, beginning at experimental week 8 until week 11 i.e., the end of experimental period. Group C rats were induced by intraperitoneal DEN (200 mg/kg) injections, three times in total, given at experimental week 2, 4 and 6 to elicit preneoplastic lesion formation as described by Khan et al. [9]. Group D rats were induced with DEN, and subsequently received DMF (100 mg/kg bodyweight), three times weekly through oral gavage, with the first dose beginning at experimental week 8 as in group B rats. All animals were examined daily for signs of toxicity such as ill appearance, circling rashes, tremors, roughened coat, and rhinitis. At the end of the study (11 weeks), the livers from all animals were perfused and subsequently excised under anesthesia. Morphological analysis for presence of visible hepatocyte nodules was performed. Data on incidence, size and distribution of nodules were also collected. A portion of non-nodular liver tissue from different study groups was collected and frozen in liquid nitrogen and stored in a 80 °C freezer. Serial sections (10 lm) of liver tissue were prepared from different groups for immunohistochemical analysis to identify the preneoplastic lesions. Six to eight livers per group were analysed for various parameters.

2. Materials and methods 2.1. Chemicals DEN and chrysin (97% purity) were purchased from Sigma Chemical Company, St. Louis, MO, USA. Dimethoxy flavone

13

Fig. 1. (A) The chemical structure of 5,7-dimethoxy flavone (DMF).

14

M.S. Khan et al. / Chemico-Biological Interactions 193 (2011) 12–21

2.3. RNA isolation and relative quantification by real-time PCR To isolate total RNA, approximately 50 mg from each liver (centrilobular region) in the region containing the nodules was placed into the TRIzol reagent (Sigma Chemical Company, USA) and immediately homogenized using a Potter homogenizer. The samples were further homogenized with a 1 ml sterile syringe and 18 gauge needle. Finally, total RNA was extracted using the chloroform–phenol method. Complementary DNA (cDNA) was obtained from total RNA (2 ug/ml) using the Quantitect Reverse Transcription Kit (Qiagen) according to the manufacturer’s instructions. qPCR amplifications were performed in triplicate using the SYBR Green I assay in Stratagene M3000 (Stratagene, USA). The reactions were carried out in a 96-well plate in 20-ll reactions containing 2X SYBR Green Master Mix (Invitrogen, USA), 2 pmol each of forward and reverse primer, and a cDNA template corresponding to 10 ng total RNA. Glyceraldehyde-3-phosphate dehydrogenase (Gapdh) was used as the housekeeping gene. SYBR Green PCR conditions were maintained as previously described [9]. Melting curve analysis of amplification products was performed at the end of each PCR reaction to confirm that a single PCR product was detected. Each sample was run in triplicate. PCR reactions without the addition of the template were used as blank controls. Finally, relative quantitative values of each sample was determined with 1/25 diluted cDNA and were normalised with those of the Gapdh gene. Gene expression was quantified using a modification of the 2-DDct method as previously described [12]. Primer sequences were designed using the NCBI-Primer BLAST online tool and synthesised commercially. Specific set of primers (MWG Biotech, Ebesberg, Germany) were used: COX-2 (forward 50 CTG AGG GGT TAC CAC TTC CA 30 and reverse 50 TGA GCA AGT CCG TGT TCA AG 30 ), NF-kB p65 (forward 50 ACG ATC TGT TTC CCC TCA TCT 30 and reverse 50 TGC TTC TCT CCC CAG GAA TA 30 ), b-catenin (forward 50 TGA CCT CAT GGA GTT GGA CA 30 and reverse 50 CAG CTA CTT GCT CTT GCG TG 30 ), CK-2 (forward 50 CGT CAG TTT CTC TCT CCT GAC GGG A 30 and reverse 50 ACG GAG CCC ACA GAA CCA GGA AAT 30 ), cyclin D1 (forward 50 GTC GAG AAG AGA AAG CTC TG 30 and reverse 50 TTA AAA GCC TCC TGT GTG AA 30 ), p53 (forward 50 TTG GAC CCT GGC ACC TAC AAT G 30 and reverse 50 GCA GAC AGG CTT TGC AGA ATG G 30 ) and Gapdh (forward 50 TCA AGA AGG TGG TGA AGC AG 30 and reverse 50 AGG TGG AAG AAT GGG AGT TG 30 ). 2.4. Histology of liver and Immunohistochemistry Hematoxylin and eosin staining and immunohistochemical detection of GST-Pi in 10 lm-thick liver sections corresponding to surface as well as deep-seated nodules was done by standard immunohistochemical techniques. Briefly, the sections were incubated for 10 min at 65 °C in 10 mmol/L sodium citrate buffer (pH 6.5) for antigen retrieval. Following a 5-min was with PBS, the endogenous peroxidases were blocked by 1% H2O2 in PBS for 5 min. The sections were again washed and blocked for 1 h in PBS containing 5% normal goat serum. The slides were washed and then incubated overnight with the GST-Pi primary antibody (1:200) at 4 °C in a humidified chamber. After washing with PBS, the sections were incubated with HRP-conjugated secondary antibody (goat anti-rabbit, 1:200 dilution) for 30 min at 37 °C. The chromogenic reaction was developed with 3,30 -diaminobenzidine tetrahydrochloride solution. Negative control sections were processed simultaneously using the same procedure, except the primary antibody. All sections were viewed under a light microscope. 2.5. Western blot analysis Livers harbouring the preneoplastic nodules were homogenized in ice-cold lysis buffer [100 mM Tris buffer (pH 7.4) containing a

protease inhibitor cocktail containing 100 mM PMSF, 1 mM Bestatin, 1.5 mM Pepstatin A, 1.4 mM E-64, 0.08 mM Aprotinin and 1 mM Leupeptin (Sigma Aldrich, USA)]. Lysates were centrifuged at 7500g for 10 min at 4 °C, and aliquots of supernatant containing 50 lg protein were boiled in sodium dodecylsulphate (SDS) sample buffer for 5 min before electrophoresis on 12% SDS–polyacrylamide gel. After electrophoresis, proteins were transferred to PVDF membrane, and the blots were blocked with 5% fat-free dry milkPBST (phosphate-buffered saline containing 0.1% Tween 20) for 1 h at room temperature and then washed in PBST buffer. The membranes were incubated for 2 h at room temperature with 1:1000 dilution of primary antibodies for b-actin, COX2, iNOS, p53, caspase-3, Bcl-xL, Bax, phospho-b-catenin Ser33/37, phosphoNF-kB p65 Ser529, phospho-IkB Ser32/36, Wnt 3, Dvl2, phosphoGSK3b Ser9, Cyclin D1 and CK2. Blots were washed three times with PBST at 10 min intervals followed by incubation with 1:5000 dilution of horseradish peroxidise-conjugated secondary antibodies for 1 h and again washed with PBST three times. The transferred proteins were visualised by enhanced chemiluminescent detection system (SuperSignalÒ West Femto Maximum Sensitivity Substrate, Thermo Scientific Pierce, USA). b-actin served as loading control of protein. Intensity of digitised bands was analysed using Quantity One software. 2.6. Preparation of nuclear extracts from rat liver The nuclear extract from rat liver was prepared as described earlier by Chun et al. [25]. Briefly, liver samples were homogenized in 1 ml of hypotonic buffer A [10 mM HEPES (pH 7.8), 10 mM KCl, 2 mM MgCl2, 1 mM DTT, 0.1 mM EDTA, 0.1 mM phenylmethylsulfonyl-fluoride (PMSF)]. To the homogenates was added 80 ll of 10% Nonidet P-40 (NP-40) solution, and the mixture was then centrifuged for 2 min at 14,000g. The supernatant was collected as a cytosolic fraction. The precipitated nuclei were washed once with 500 ll of buffer A and 40 ll of 10% NP-40, centrifuged, resuspended in 200 ll of buffer C [50 mM HEPES (pH7.8), 50 mM KCl, 300 mM NaCl, 0.1 mM EDTA, 1 mM DTT, 0.1 mM PMSF, 20% glycerol] and centrifuged for 5 min at 14,800g. The supernatant containing nuclear proteins was collected, and after determination of protein concentration, used for blotting with primary antibodies for NF-kB and phospho-b-catenin Thr393. 2.7. Statistical analysis Tumor incidences (number of rats with tumors) and tumor multiplicity (total number of tumors per rat) among the different study groups were analysed by ANOVA, with post hoc comparisons made using Tukey’s test. All other data is represented as mean ± standard deviation (SD) A probability level, as shown by P values less than 0.05 and less than 0.001 were considered statistically significant. 3. Results 3.1. Effect of DMF on hepatic nodule formation Administration of DMF (100 mg/kg) had no significant effect on body weight and toxicity in control rats. No major difference in food and water intake was apparent among the various groups during the entire study period. Consistent with our previous studies based on this tumor model, DEN-exposure induced significant liver nodule formation (Table 1) with a 68.8% incidence and 3.6 ± 0.23 (mean ± SD) multiplicity in group C animals. In contrast, none of the control animals developed any nodules or any detectable lesions. Treatment with DMF significantly inhibited liver nodule incidence by 82% (P < 0.05) and multiplicity by 78% (P < 0.05).

M.S. Khan et al. / Chemico-Biological Interactions 193 (2011) 12–21 Table 1 Effect of DMF administration on nodule incidence and multiplicity in DEN-induced male Wistar rats.

** ***

Experimental group

No. of rats

Hepatic nodule incidence (%)

Nodule multiplicity

Group Group Group Group

16 16 16 16

0 0 11/16 (68.8)** 3/16 (18.7)***

0 0 3.6 ± 0.23** 0.78 ± 0.14***

A (Control) B (DMF Alone) C (DEN-induced) D (DEN+DMF)

P < 0.001, significantly different from control group (Tukey’s test). P < 0.05, significantly different from DEN-alone group (Tukey’s test).

3.2. Effect of DMF on liver histology, and GST-Pi positive preneoplastic nodules The general morphology of liver was examined by hematoxylin and eosin staining, and that of preneoplastic foci by their characteristic GST-Pi expression pattern. Liver sections from control rats (Fig. 2A and E)and drug control rats (Fig. 2B and F) treated with DMF alone exhibited normal morphology whereas DEN-exposed rats (Fig. 2C and G) showed significant loss of hepatocyte architecture, presence of binucleated, oval-shaped and irregular cells. Besides, Fig. 2G corresponding to group C rats showed distinct GSTPi positive foci following exposure to DEN, when compared with control animals. Administration of DMF significantly suppressed formation of GST-positive foci (Fig. 2H), when compared to the DEN-induced rats and also led to a marked improvement in the gross microscopic appearance of liver tissue (Fig. 2D). 3.3. Treatment with DMF decreases the expression of inflammatory mediators As inflammation plays a key role in hepatocarcinogenesis, the ability of dimethoxy flavone to suppress the induction of inflammatory markers iNOS and COX-2 was observed. Normal animals showed a minimal hepatic expression of these two markers whereas, a significant (P < 0.05) upregulation in mRNA and protein levels of COX-2 was noticed in hepatic preneoplastic nodules from DEN-challenged animals (Figs. 3 and 4). Further, DEN-induced overexpression of iNOS was suppressed at the protein level from 1-fold to 6-fold (P < 0.001) in rats fed DMF-supplemented diet (100 mg/kg). Similarly, DEN-induced overexpression of COX-2 was suppressed at the mRNA level from 1-fold to 4-fold, and at the protein level from 1.5 fold to 6-fold in the same group of animals. Hence, these results indicate that DMF supplementation was able to reverse the high levels of inflammatory markers initiated by exposure to DEN. 3.4. DMF administration increases expression of apoptotic mediators in preneoplastic lesions We recently showed that DEN-exposure alters the expression of apoptosis mediator p53 in preneoplastic nodules. We also reported that a small, but significant increase in p53 mRNA was noticed in DEN-induced animals supplemented with DMF (12). Accordingly, in this study, we have investigated the protein expression levels of more apoptosis mediators like p53, Bcl-xL, Bax and caspase-3 in the livers of rats treated with DEN and fed with DMFsupplemented diet. As shown in Fig. 4C, DEN markedly increased the protein expression levels of anti-apoptotic member Bcl-xL (P < 0.05). A very small, but significant increase in caspase-3 as well as non-significant increases in p53 and Bax was also seen following exposure to the genotoxic carcinogen DEN. In contrast, a 3.6-fold decrease in Bcl-xL protein and 2–3 fold increase in protein levels of p53, Bax and caspase-3 were noticed following

15

DMF treatment at a dose of 100 mg/kg. Taken together, these data suggest that there was a significant (P < 0.001) induction of apoptosis in rats fed with DMF compared with the DEN alone group. 3.5. DMF administration abrogates NF-kB signaling in preneoplastic lesions To confirm the requirement of NF-kB translocation to the nucleus following DEN exposure, the blots were performed using the nuclear fraction with anti-NF-kB antibody. This was further confirmed by using antibodies that specifically recognises the phosphorylated forms of phospho-NF-kB p65Ser529 and phosphoIkB Ser32/36. It was noticed that, during DEN-induced preneoplastic lesions the protein levels of p65 and phosphorylated IkB in the cytoplasm were significantly increased (P < 0.05) leading to a concomitant increase in NF-kB levels in the nucleus indicating impaired NF-kB signaling (Fig. 4A). As shown in Fig. 4A, there was a significant (P < 0.001) downregulation of cytoplasmic levels of phospho-IkB Ser32/36 and nuclear levels of NF-kB proteins in DEN-induced, DMF administered rats compared with the DEN alone induced rats. Taken together, all these results indicate that DMF was able to suppress NF-kB signaling during the DEN-induced preneoplastic lesion formation. 3.6. Impaired NF-kB signaling concurrently activates canonical Wnt signaling in DEN-induced preneoplastic lesions Because NF-kB has been shown to synergistically activate Wnt signaling through GSK3b, we next tested whether expression of proteins in Wnt pathway are altered in the present study. Protein lysates from different study groups were examined for levels of two different forms of phosphorylated b-catenin, inactive form (phospho-b-catenin Ser33/37) in cytoplasm and active form (phospho-b-catenin Thr393) in nucleus. As shown in Figs. 3 and 4B, DEN exposure induced Wnt signaling in preneoplastic lesions by significantly accumulating b-catenin mRNA and active b-catenin (phospho-b-catenin Thr393) protein in the nucleus. Treatment with DMF led to a significant decrease in nuclear levels of active b-catenin (phospho-b-catenin Thr393). Likewise, levels of CK2, a kinase responsible for activation of b-catenin by phosphorylating the Thr 393, were decreased following administration of DMF. Interestingly, levels of phosphorylated p65 Ser529, another cellular target of CK2, were also decreased (as shown in Fig. 4A) following DMF treatment confirming a decrease in activity of CK2. To confirm that DMF administration led to turning-off of Wnt/ b-catenin signaling, we used antibody that specifically recognised the phosphorylated, inactive form of GSK3b (phospho-GSK3bSer9). It was seen that levels of phospho-GSK3bSer9 were significantly elevated in the DMF-treated group. Finally, levels of Wnt-target protein, cyclin D1 (Fig. 4B) along with Dvl2 and inactive phospho-b-catenin Ser33/37 (data not shown) were checked in the protein lysates from different animals. It was noticed that DMF exposure led to a significant decrease in Dvl2 and cyclin D1, with concomitant increase in cytoplasmic levels of phospho-b-catenin Ser33/37 confirming that Wnt signaling was mitigated (Fig. 5B). Altogether, these data collectively demonstrate that the administration of the methylated flavonoid DMF downregulates inflammatory mediators like COX-2 and iNOS in DEN-induced preneoplastic lesions and may inhibit key mediators of IkB degradation thereby abrogating NF-kB signaling. As shown in Fig. 5A, this may induce synergistic upregulation of key pro-apoptotic mediators like p53, caspase-3 and Bax thereby leading to apoptosis. Our data also demonstrate a plausible mechanism for the pleiotropic effects of DMF inhibition based on inhibition of the Wnt pathway in this model system. Treatment with DMF abrogated

16

M.S. Khan et al. / Chemico-Biological Interactions 193 (2011) 12–21

Fig. 2. Effect of DMF on liver morphology and GST-Pi foci formation. Representative immunohistochemical staining of neoplastic nodules dissected and stained with hematoxylin and eosin (A–D) and antibodies specific for GST-Pi (E–H). Immunohistochemistry was performed as described in Section 2. (A and E) Normal liver (Group A) showing cells with uniform size and absence of GST-Pi positive foci. (B and F) HE and GST-Pi staining in rats treated with DMF (100 mg/kg, Group B) showing pattern similar to control. (C and G) sections from DEN-induced (200 mg/kg, Group C) rats. Arrows indicate binucleate, irregular shaped cells shown by hematoxylin staining and the presence of GST-Pi positive foci. (D and H) liver sections from DMF + DEN (Group D) rats showing significant amelioration in liver morphology and less common GST-Pi foci. Magnification 200.

Wnt signaling as demonstrated by an increase in cytoplasmic levels of GSK3b, with a concomitant decrease in CK2, leading to accumulation of p-b-catenin Ser33/37 that is degraded by ubiquitination in the cytoplasm. This was confirmed by a decrease in expression of Wnt-target gene cyclin D1. These results further establish a novel molecular link among inflammation, NF-kB and Wnt signaling in the chemopreventive effect of methylated flavones.

4. Discussion In our previous study using a preneoplastic model of liver carcinogenesis initiated with diethylnitrosamine, we showed that the expression levels of several genes involved in apoptosis and cell cycle regulation were favourable altered [12]. Nevertheless, the precise mechanism by which DMF exerts a chemopreventive action in

M.S. Khan et al. / Chemico-Biological Interactions 193 (2011) 12–21

17

Fig. 3. Quantitative changes in gene transcripts following DMF administration in DEN-induced preneoplastic lesions. Rats were sacrificed 11 week following commencement of the study, preneoplastic lesions were removed and total RNA was extracted. Complementary DNA obtained from the RNA was amplified by real-time PCR using genespecific primers and quantified. Lane 1 – Control, Lane 2 – DMF 100 mg/kg alone, Lane 3 – DEN 200 mg/kg, and Lane 4 – DEN 200 mg/kg along with DMF 100 mg/kg. All expression values were normalized to the value of Gapdh gene used as an internal control. Each column represents the mean ± SD (n = 6–8 livers). Data values obtained from analysis in triplicate. Relative amount was calibrated based on the transcript amount of the corresponding gene in untreated controls. DEN exposure (group C) led to significant decrease in p53 mRNA levels while increasing levels of COX-2, NF-kB p65, b-catenin, CK2, as well as Wnt target gene cyclin D1, when compared with the control (P < 0.05). Treatment with DMF (group D) significantly increased p53 mRNA levels while suppressing the levels of COX-2, NF-kB p65, b-catenin, CK2 and cyclin D1, when compared to DEN-induced rats (P < 0.001, Group C). ⁄P < 0.05 compared with normal group; ⁄⁄P < 0.001 compared with DEN-induced group.

rats has not been completely elucidated. In recent years, an overwhelming number of studies provide convincing evidence that persistent neoplastic nodules may be employed as stable endpoints for carcinogenecity testing [26–28]. Recent studies have shown that naturally occurring bioflavonoids possess considerable chemopreventive potential [11,29]. However, our present results provide the first evidence that attenuation of subcellular signaling pathways by the flavone plays a valuable role in inhibition of the early stages of hepatocellular carcinogenesis in rats. Mounting evidence underlines the importance of inflammation during various stages of hepatocarcinogenesis [30]. As hallmarks of inflammatory processes, COX-2 and iNOS expression were measured in preneoplastic liver sections of DEN-treated rats. Cyclooxygenase-2 (COX-2), induced by several stimuli associated with inflammation, is involved in carcinogenesis, including liver tumorigenesis of human [31] and rodents [6]. It has been established that exposure to DEN elevates the expression of inflammatory markers like COX-2 and NF-kB [9] in preneoplastic lesions.Furthermore, COX-2 specific inhibitors such as celecoxib [32] and omega-3 fatty acids [33] can ameliorate liver injury or inhibit growth of hepatocellular carcinoma. Chronic inflammatory reactions can be triggered by nitric oxide production in hepatocytes [34] by the

inducible nitric oxide synthase (iNOS) enzyme. Recent evidences suggest that iNOS plays a crucial role in development and progression of HCC and is frequently overexpressed during liver cancer in rats [35]. In addition, selective iNOS inhibitor like aminoguanidine, can suppress human HCC growth, suggesting that iNOS signaling could serve as an important target for prevention and treatment of HCC [36]. In agreement with previous observations cited above, we have observed an elevated hepatic expression level of iNOS and COX-2 in DEN-treated animals. Additionally, we have shown that DMF is able to reverse the elevated expression of COX-2 and iNOS in the preneoplastic lesions. Our results substantiate a clear antiinflammatory effect and highlight the possibility that iNOS could be a potential target of DMF-mediated chemoprevention in the current experimental conditions. This is in line with a recent observation that flavonoids like resveratrol can exhibit antitumor responses through inhibition of iNOS activity [37]. The transcription factor nuclear factor-kappaB (NFkB), belongs to a family of different proteins that can form homodimers and heterodimers with each other [38]. NF-kB is a key regulator of inflammation and has been reported to be activated constitutively in human HCC [39]. The activated form of NF-kB is a heterodimer, consisting of a p50 (NF-kB1) or p52 (NF-kB2) subunit and the

18

M.S. Khan et al. / Chemico-Biological Interactions 193 (2011) 12–21

Fig. 4. Effect of DMF on hepatic protein expression related to Wnt, NF-kB pathway and apoptosis during DEN-induced preneoplastic nodules. Rats were sacrificed 11 wk following commencement of the study, preneoplastic lesions were removed, total cellular protein was separated, transferred to PVDF membrane, and probed with corresponding antibodies. The same blots were stripped and reprobed with b-actin to verify equal protein loading. Representative blots of three independent experiments. Blots correspond to proteins in the cytoplasmic fraction, unless specifically indicated. Lane 1 – Control, Lane 2 – DMF 100 mg/kg alone, Lane 3 – DEN 200 mg/kg, and Lane 4 – DEN 200 mg/kg along with DMF 100 mg /kg. (A) Western blot and densitometric analysis of proteins associated with inflammation and NF-kB pathway. (B) Representative Western blot and densitometric analysis of Wnt pathway proteins. (C) Changes in expression of apoptotic mediators. Increased expression of inflammatory mediators (iNOS, COX-2), NF-kB and Wnt pathway proteins, and decreased expression of apoptotic markers were noticed in the DEN-induced neoplastic nodules (P < 0.05). DMF administration reversed the expression of the proteins to a marked extent (P < 0.001). Each column represents the mean ± SD (n = 6–8 livers). Y-axis represents fold induction (relative to bactin). ⁄P < 0.05 compared with normal group; ⁄⁄P < 0.001 compared with DEN-induced group.

transactivating subunit p65 (RelA). Genes encoding c-Rel, p50 and p52 are located at sites of recurrent genomic arrangements in cancer patients [40]. Inactive NF-kB resides in the cytoplasm associated with eight regulatory proteins called inhibitors of kB (IkB), of which IkBa, IkBb and IkBe may be the most common. Inflammatory insult and most other related stimuli result in degradation of IkBa that is essential for release and activation of NF-kB and eventual nuclear translocation of the classical NF-kB dimer p50-RelA which activates the expression of target genes. Various stimuli can induce rapid phosphorylation of IkBa by Inhibitor of IkB Kinase (IKK) on Ser 32 and Ser 36, leading to ubiquitination and degradation of IkBa [41]. The importance of IKK can be confirmed from the fact that conditional deficiency in IKKc leads to heaptic steatohepatitis and hepatocellular carcinoma [42]. In the present study, we have observed for the first time a decrease in phospho-IkBa Ser32/36 protein in the cytoplasm following DMF treatment in DEN-induced animals. This has been further supported by a parallel decrease in nuclear NF-kB protein levels in the same animals. Recent studies have shown that flavonoids directly inhibit IKK activity with a consequent decrease in phosphorylation of IkB [43]. Non-phosphorylated form of IkB exhibits more

stability and lesser degradation thereby sequestering NFkB in the cytoplasm, resulting in reduced levels of NF-kB relocating to the nucleus. In view of these observations, we postulate that the inhibitory effect of DMF could be achieved by a direct inhibition of IKK’s, as shown by a decrease in cytoplasmic phosphorylated IkB levels and the concomitant nuclear NF-kB expression. All these studies indicate that NF-kB signaling pathway is a key target in chemoprevention of HCC by flavonoids. It is believed that liver injury occurs due to extent and/or nature of NF-kB pathway inhibition as well as the presence of concurrent subcellular changes in apoptosis and other pathways [44]. Apoptosis provides an innate cellular defence against tumorigenesis characterised by removal of cells with genetic instability that have developed during oncogenesis [45] and by deletion of cells suffering DNA insult from genotoxic agents [46]. A recent study in human HCC cells indicates that constitutive NF-kB activity plays an anti-apoptotic role [47]. Mice lacking RelA subunit of Rel/NF-kB transcription factors in embryonic development induces massive apoptosis of fetal liver cells [48]. NF-kB family proteins have been shown to directly activate expression of apoptosis inhibitor Bcl-xL [49]. Additionally, IKKb inhibitors, apart from directly killing

M.S. Khan et al. / Chemico-Biological Interactions 193 (2011) 12–21

19

Fig. 5. Schematic representation of the mechanism of action of dimethoxy flavone. (A) DEN-induced inflammation triggers concerted activation of the NF-kB and Wnt signaling cascades. IKK, a key component in the NF-kB pathway, increasingly phosphorylates IkB to form pIkB Ser32/36 that is rapidly degraded in the cytoplasm. This leads to accumulation of the active form of NF-kB which eventually translocates to the nucleus and activates gene expression. NF-kB can also directly activate the anti-apoptotic factor BcL-xL, thereby inhibiting apoptosis. DEN exposure also induces levels of CK2, a key Wnt pathway component that phosphorylates b-catenin at Thr393 leading to its accumulation in the cytoplasm followed by translocation to the nucleus, activating transcription of target genes such as COX-2 and cyclin D1. (B) Dimethoxy flavone administration leads to down-regulation of nuclear NF-kB, perhaps through inhibition of IKK’s, as shown by a decrease in levels of cytoplasmic pIkB Ser32/36. Simultaneously, down-regulation of Bcl-xL and up-regulation of p53, Bax and caspase 3 levels establish apoptosis. Finally, DMF treatment also resulted in up-regulation of GSK3b and downregulation of CK2, thereby phosphorylating b-catenin at Ser33/37 leading to its degradation in the cytoplasm and attenuation of the Wnt signaling pathway.

20

M.S. Khan et al. / Chemico-Biological Interactions 193 (2011) 12–21

NF-kB-dependent cancer cells also potentiate the apoptotic response to cytotoxic chemotherapy [50]. In our study several apoptotic markers showed altered expression in the liver preneoplastic lesions of rats administered DMF 2 weeks after the last DEN injection. Analysis of the present data showed that increase in expression levels of pro-apoptotic factors like p53, Bax and caspase-3 with a concomitant decrease in levels of anti-apoptotic factor Bcl-xL correlated well with the induction of apoptotis in liver preneoplastic lesions. This is consistent with in vitro data that shows apoptosis is induced by the flavonoid chrysin, in Hep G2 cells [51]. Few in vivo studies have investigated the apoptotic response by chrysin or DMF administration at the early time point of preneoplastic lesion formation. Previous studies from our laboratory have shown that flavonoids can increase the acute apoptotic response to the genotoxic carcinogen DEN, furthermore this increase correlates with DNA fragmentation within preneoplastic lesions [9,12]. Interestingly, DMF administration alone did not elicit induction of apoptosis nor toxicity in the liver of control rats. Increased apoptosis during initiation events might enhance the elimination of mutated cells that might otherwise progress to malignancy. Therefore, alterations in the homeostatic maintenance of the liver epithelium via reduction in cell proliferation and flavonoid-induced inhibition of chymotrypsin-like and trypsin-like proteasome activity [52] may account for the synergistic changes observed in p53 as well as Bax, observed in group D rats alone, leading to simultaneous regulation of cell cycle and apoptosis, respectively. Flavonoids may also deduce synergy through broader mechanisms associated with apoptosis and the NFkB pathway. Recently Li et al., [51] showed that the flavonoid chrysin sensitizes several human cancer lines, including Hep G2, to apoptosis via suppression of nuclear factor-kappa B. Indeed, it was shown that chrysin inhibited degradation of IkB protein and subsequent nuclear translocation of p65. As a result, the expression of a NF-kB-targeted antiapoptotic gene, c-FLIP-L was suppressed, thereby triggering apoptosis. Our results also showed a decrease in cytoplasmic pIkB Ser32/36 alongwith a concomitant decrease in nuclear p65 level, as well as induction of p53, Bax and caspase-3 in the DEN-induced rats fed with DMF. This undoubtedly confirms the synergistic link between induction of apoptosis and inhibition of the NF-kB pathway in the DEN-induced preneoplastic lesions. Wnt’s are a family of highly conserved secreted proteins that act as mediators of key cell–cell signaling events. Wnt signals are pleiotropic, with effects that include mitogenic stimulation, cell fate specification, and differentiation. Augmentation of canonical Wnt signaling has been shown to be coupled to the progression of colon cancer from microadenoma to macroscopic tumors [53]. Recently, Kaler et al. [54] established a previously unknown link among inflammation, IL-1b, Wnt signaling and growth of colon cancer cells. It was shown that inflammation triggered by proinflammatory cytokines like interleukin 1b (IL-1b) can enhance TCF4/b-catenin transcriptional activity in tumor cells through inactivation of GSK3b. Activation of Wnt pathway proceeds via phosphorylation of GSK3b, stabilization of b-catenin, enhancement of TCF4-dependent gene activation and the expression of Wnt target genes like cyclin D1 in tumor cells. The pool of GSK3b that participates in Wnt signaling exists as a multi-protein complex that includes axin, b-catenin and APC [55]. The type of kinase, and consequently, the aminoacids that are phophorylated in b-catenin is thought to be crucial in deciding the subcellular outcome. In the absence of Wnt signaling, phosphorylation of b-catenin on Ser33 and Ser37 by the Casein kinase-1 (CK1)-GSK3b-Axin complex in naive cells targets b-catenin for ubiquitination and eventual degradation [56,57]. Whereas, in Wnt-activated cells, casein kinase 2 (CK2) phosphorylates b-catenin on Thr393 within the armadillo region, which prevents binding with axin as well as ubiquitination, thereby increasing stability and accumulation in the cytoplasm [58].

Stabilized phospho-b-cateninThr393 translocates to the nucleus, where it binds to members of the TCF family of transcription factors, leading to expression of target genes such as c-myc, c-jun, cyclin D1 and cox-2 [59]. Importantly, b-catenin translocation is often detected at the invasive front composed of transformed cells (tumor), akin to preneoplastic lesions and surrounding tissue [60], suggesting that the tumor microenvironment may provide soluble inflammatory mediators that promote nuclear translocation of b-catenin that promote tumor progression. A decrease in levels of nuclear phospho-b-cateninThr393 level as well as crucial Wnt signaling pathway components like CK2, Dvl2, and target proteins like cyclin D1 may suggest a possible deactivation of the canonical Wnt pathway during chemoprevention by DMF. This is further confirmed by the increase in cytoplasmic levels of phospho-b-catenin Ser33/37, seen in parallel with an increase in GSK3b supporting the assumption of increased b-catenin degradation in the cytoplasm, that eventually disrupts the canonical wnt signaling. The effects of DMF on mediators as well as upstream elements of Wnt signaling need to be explained by further investigation of molecular mechanisms in this model. We are currently in the pursuit of unravelling the mechanism behind the pleiotropic effects observed during chemoprevention by DMF using the in vitro model that permits the use of pathway-specific inhibitors. A number of studies have established that inflammation and deranged apoptosis contributes to initiation, promotion and progression of hepatocarcinogenesis. In this report, we present data which demonstrate that DMF abrogates NF-kB signaling in DEN-induced preneoplastic nodules, perhaps through inhibition of IKK activity, as shown by a decrease in cytoplasmic levels of phospho-IkBa Ser32/36. We also present data that shows that NF-kB activation is coupled to Wnt signaling in liver preneoplastic lesions exposed to DEN (Fig. 5A and B). Similar to NF-kB, inhibition of Wnt signaling by DMF administration is apparently achieved by the twin mechanism of decrease in level of a key Wnt pathway activator kinase CK2, along with a concomitant increase in levels of GSK3b, leading to an increased formation of phospho-b-catenin Ser33/37, which is ubiquitinated and eventually degraded. Our demonstration that DMF, a dietary flavone, can simultaneously inhibit NF-kB and Wnt signaling in preneoplastic lesions holds great promise for use as an effective chemotherapeutic agent in treating early stages of liver cancer. Funding information This study was supported by a minor grant from University Grants Commission [F.No.37-600/2009] to S.M.K. Conflict of interest statement None declared. Acknowledgement We thank Dept. of Endocrinology, IBMS for technical assistance during various stages of the project. References [1] J. Ribes, R. Cleries, L. Esteban, V. Moreno, F.X. Bosch, The influence of alcohol consumption and hepatitis B and C infections on the risk of liver cancer in Europe, J. Hepatol. 49 (2008) 233–242. [2] D.M. Parkin, F. Bray, J. Ferlay, P. Pisani, Estimating the world cancer burden: Globoscan 2000, Int. J. Cancer 94 (2000) 153–156. [3] N. D’Alessandro, P. Poma, G. Montalto, Multifactorial nature of hepatocellular carcinoma drug resistance: could plant polyphenols be helpful?, World J Gastroenterol. 13 (2007) 2037–2043.

M.S. Khan et al. / Chemico-Biological Interactions 193 (2011) 12–21 [4] R. Peto, R. Gary, P. Brantom, P. Grasso, Dose and time relationships for tumor induction in the liver and esophagus of 4080 inbred rats by chronic ingestion of N-nitrosodiethylamine or N-nitrosomethylamine, Cancer Res. 51 (1991) 6452–6469. [5] R. Boyault, S. De reynie, Transcriptome classification of HCC is related to gene alterations and to new therapeutic targets, Hepatology 45 (2007) 42–52. [6] V. Sivaramakrishnan, S.N. Devaraj, Morin fosters apoptosis in experimental hepatocellular carcinogenesis model, Chem. Biol. Interact. 183 (2010) 284–292. [7] U. Wenzel, S. Kuntz, M.D. Brendel, H. Daniel, Dietary flavone is a potent apoptosis inducer in human colon carcinoma cells, Cancer Res. 60 (2000) 3823–3831. [8] G. Pushpavalli, C. Veeramani, K.V. Pugalendi, Infleunce of chrysin on hepatic marker enzymes and lipid profile against D-galactosamine-induced hepatotoxicity rats, Food Chem. Toxicol. 48 (2010) 1654–1659. [9] S.M. Khan, H. Devaraj, S.N. Devaraj, Chrysin abrogates early hepatocarcinogenesis and induces apoptosis in N-nitrosodiethylamineinduced preneoplastic nodules in rats, Toxicol. Appl. Pharmacol. 251 (2011) 85–94. [10] F. Ahmad, S.A. Bakar, A.Z. Ibrahim, R.W. Read, Constituents of the leaves of Piper caninum, Planta Med. 63 (1997) 193–194. [11] T. Walle, Methoxylated flavones, a superior cancer chemopreventive flavonoid subclass?, Semin Cancer Biol. 17 (2007) 354–362. [12] S.M. Khan, H. Devaraj, S.N. Devaraj, Methylated chrysin, a dimethoxy flavone, partially suppresses the development of liver preneoplastic lesions induced by N-Nitrosodiethylamine in rats, Food Chem. Toxicol. 49 (2011) 173–178. [13] M. Moreau, S. Mourah, C. Dosquet, b-catenin and NF-kB cooperate to regulate the uPA/uPAR system in cancer cells, Int. J. Cancer 128 (2011) 1280–1292. [14] M. Karin, Nuclear factor-kB in cancer development and progression, Nature 441 (2006) 431–436. [15] F.R. Greten, L. Eckmann, T.F. Greten, J.M. Park, Z.W. Li, L.J. Egan, M.F. Kagnoff, M. Karin, IKK beta links inflammation and tumorigenesis in a mouse model of colitis-associated cancer, Cell 118 (2004) 285–296. [16] J.L. Luo, S. Maeda, L.C. Hsu, H. Yagita, M. Karin, IKKb couples hepatocyte death to cytokine-driven compensatory proliferation that promotes chemical hepatocarcinogenesis, Cancer cell 6 (2004) 297–305. [17] M.D. Thompson, S.P. Monga, Wnt/b-catenin signaling in liver health and disease, Hepatology 45 (2007) 1298–1305. [18] Q. Du, X. Zhang, J. Cardinal, Z. Cao, Z. Guo, L. Shao, D.A. Geller, Wnt/betacatenin signaling regulates cytokine-induced human inducible nitric oxide synthase expression by inhibiting nuclear factor-kappaB activation in cancer cells, Cancer Res. 69 (2009) 3764–3771. [19] J. Hu, A. Dong, V. Fernandez-Ruiz, M. Shan, M. Kawa, E. Martinez-Anso, J. Prieto, C. Qian, Blockade of Wnt signaling inhibits angiogenesis and tumor growth in hepatocellular carcinoma, Cancer Res. 69 (2009) 6951–6959. [20] J. Behari, The Wnt/ b-catenin signaling pathway in liver biology and disease, Exp. Rev. Gastoenterol. Hepatol. 6 (2010) 745–756. [21] S. Wang, K.A. Jones, CK2 controls the recruitment of Wnt regulators to target genes in vivo, Curr. Biol. 16 (2006) 2239–2244. [22] S.P. Davies, H. Reddy, M. Caivano, P. Cohen, Specificity and mechanism of action of some commonly used protein kinase inhibitors, Biochem. J. 351 (2000) 95–105. [23] D. Zhu, J. Hensel, R. Hilgraf, M. Abbasian, O. Pornillos, G.D. Yazdi, Inhibition of protein kinase CK2 expression and activity blocks tumor cell growth, Mol. Cell Biochem. 333 (2010) 159–167. [24] X. Zheng, W.D. Meng, Y.Y. Xu, J.G. Cao, F.L. Qing, Synthesis and anticancer effect of chrysin derivatives, Bioorg. Med. Chem. Lett. 13 (2003) 881–884. [25] K.S. Chun, Y.S. Keum, S.S. Han, Y.S. Song, S.H. Kim, Y.J. Surh, Curcumin inhibits phorbol ester-induced expression of cyclooxygenase-2 in mouse skin through suppression of extracellular signal regulated kinase activity and NF-kB activation, Carcinogenesis 24 (2003) 1515–1524. [26] G.M. Williams, Phenotypic properties of preneoplastic rat liver lesions and applications to detection of carcinogens and tumor promoters, Toxicol. Pathol. 10 (1982) 3–10. [27] K. Doi, M. Wei, M. Kitano, N. Uematsu, M. Inoue, H. Wanibuchi, Enhancement of preneoplastic lesion yield by chios mastic gum in a rat liver medium-term carcinogenesis bioassay, Toxicol. Appl. Pharmacol. 234 (2009) 135–142. [28] R. Puatanachokchai, K. Morimura, H. Wanibuchi, M. Oka, A. Kinoshita, F. Mitsuru, S. Yamaguchi, Y. Funae, S. Fukushima, Alpha-benzene hexachloride exerts hormesis in preneoplastic lesion formation of rat hepatocarcinogenesis with the possible role for hepatic detoxifying enzymes, Cancer Lett. 240 (2006) 102–113. [29] A. Pick, H. Muller, I.K. Pajeva, M. Weight, R. Mayer, B. Haenisch, H. Bonisch, C.E. Muller, M. Wiese, Structure-activity relationships of flavonoids as inhibitors of breast cancer resistance protein (BCRP), Bioorg. Med. Chem. doi:10.1016/ j.bmc.2010.12.043. [30] J. Prieto, Inflammation, HCC and sex: IL-6 in the centre of the triangle, J. Hepatol. 48 (2008) 380–381. [31] H. Akkiz, S. Bayram, A. Bekar, E. Akgollu, Y. Ulger, Functional polymorphisms of cyclooxygenase-2 gene risk for hepatocellular carcinoma, Mol. Cell Biochem. 347 (2011) 201–208. [32] Y. Washino, E. Koga, Y. Kitamura, C. Kamikawa, K. Kobayashi, T. Nakagawa, C. Nakazaki, I. Ichi, S. Kojo, Effect of celecoxib, a selective cyclooxygenase-2 inhibitor on carbon tetrachloride intoxication in rats, Biol. Pharm. Bull. 33 (2010) 707–709.

21

[33] K. Lim, C. Han, Y. Dan, M. Shen, T. Wu, Omega-3 polyunsaturated fatty acids inhibit hepatocellular carcinoma cell growth through blocking beta-catenin and cyclooxygenase-2, Mol. Cancer Ther. 8 (2009) 3046–3055. [34] P.C. Kuo, K.Y. Abe, R.A. Schroeder, Oxidative stress increases hepatocyte iNOS gene transcription and promoter activity, Biochem. Biophys. Res. Commun. 234 (1997) 289–292. [35] B. Ahn, B.S. Han, D.J. Kim, H. Ohshima, Immunohistochemical localization of inducible nitric oxide synthase and 3-nitrotyrosine in rat liver tumors induced by N-nitrosodiethylamine, Carcinogenesis 20 (1999) 1337–1344. [36] D.F. Calvisi, F. Pinna, S. Ladu, R. Pellegrino, M.R. Muroni, M.M. Simile, M. Frau, M.L. Tomasi, M.R. De Miglio, M.A. Seddaiu, L. Daino, V. Sanna, F. Feo, R.M. Pascale, Aberrant iNOS signaling is under genetic control in rodent liver cancer and potentially prognostic for the human disease, Carcinogenesis 29 (2008) 1639–1647. [37] Z. Yang, S. Yang, B.J. Misner, R. Chiu, F. Liu, F.L. Meyskens, Nitric oxide initiates progression of human melanoma via a feedback loop mediated by apurinic/ apyrimidinic endonulease-1/redox factor-1, which is inhibited by resveratrol, Mol. Cancer Ther. 7 (2008) 3751–3760. [38] M.S. Hayden, S. Ghosh, Shared principles in NF-kappaB signaling, Cell 132 (2008) 344–362. [39] D.I. Tsai, S.L. Tsai, Y.H. Chang, S.N. Huang, T.C. Chen, K.S. Chang, Y.F. Liaw, Constitutive activation of nuclear factor kappaB in hepatocellular carcinoma, Cancer 89 (2000) 2274–2281. [40] W. Wang, J.L. Abbruzzese, D.B. Evans, L. Larry, K.R. Cleary, P.J. Chiao, The nuclear factor-kappa B RelA transcription factor is constitutively activated in human pancreatic adenocarcinoma cells, Clin. Cancer Res. 5 (1999) 119–127. [41] Q.K. Huynh, N. Kishore, S. Mathialagan, A.M. Donnelly, C.S. Tripp, Kinetic mechanisms of IkappaB-related kinases (IKK) inducible IKK and TBK-1 differ from IKK-1/IKK-2 heterodimer, J. Biol. Chem. 277 (2002) 12550–12558. [42] T. Luedde, N. Beraza, V. Kotsikoris, G.V. Loo, A. Nenci, R. De Vos, T. Roskams, C. Trautwein, M. Pasparakis, Deletion of NEMO/IKKgamma in liver parenchymal cells causes steatohepatitis and hepatocellular carcinoma, Cancer Cell 11 (2007) 119–132. [43] C.C. Chen, M.P. Chow, W.C. Huang, Y.C. Lin, Y.J. Chang, Flavonoids inhibit tumor necrosis factor-alpha-induced upregulation of intercellular adhesion molecule-1 (ICAM-1) in respiratory epithelial cells through activator protein-1 and nuclear factor-kappaB: structure–activity relationships, Mol. Pharmacol. 66 (2004) 683–693. [44] L.M. Staudt, Oncogenic activation of NF-kappaB, Cold Spring Harb. Perspect. Biol. doi:10.1101/cshperspect.2010.a000109 [45] C.B. Thompson, Apoptosis in the pathogenesis and treatment of disease, Science 267 (1995) 1456–1462. [46] C.S. Potten, H.K. Grant, The relationship between ionizing radiation-induced apoptosis and stem cells in the small and large intestine, Brit. J. Cancer 78 (1998) 993–1003. [47] P.J. Chiao, R. Na, J. Niu, G.M. Sclabas, Q. Dong, S.A. Curley, Role of Rel/NFkappaB transcription factors in apoptosis of human hepatocellular carcinoma cells, Cancer 95 (2002) 1696–1705. [48] A.A. Beg, W.C. Sha, R.T. Bronson, S. Ghosh, D. Baltimore, Embryonic lethality and liver degeneration in mice lacking the RelA component of NF-kappaB, Nature 376 (1996) 167–170. [49] C. Chen, L.C. Edelstein, C. Gelinas, The Rel/NF-kappaB family directly activates expression of the apoptosis inhibitor Bcl-x(L), Mol. Cell Biol. 20 (2000) 2687– 2695. [50] A.S. Baldwin, Control of oncogenesis and cancer therapy resistance by the transcription factor NF-kappaB, J. Clin. Invest. 107 (2001) 241–246. [51] X. Li, Q. Huang, C.N. Ong, X.F. Yang, H.M. Shen, Chrysin sensitizes tumor necrosis factor-a-induced apoptosis in human tumor cells via suppression of nuclear factor-kappa B, Cancer Lett. 293 (2010) 109–116. [52] Y.X. Wu, X. Fang, Apigenin, chrysin and luteolin selectively inhibit chymotrypsin-like and trypsin-like proteasome catalytic activities in tumor cells, Planta Med. 76 (2010) 128–132. [53] P. Kaler, B.N. Godasi, L. Augenlicht, L. Klampfer, The NF-kB/AKT-dependent induction of Wnt signaling in colon cancer cells by macrophages and IL-1b, Cancer Microenviron. 2 (2009) 69–80. [54] P. Kaler, L. Augenlicht, L. Klampfer, Macrophage-derived IL-1 beta stimulates Wnt signaling and growth of colon cancer cells: a crosstalk interrupted by vitamin D3, Oncogene 28 (2009) 3892–3902. [55] S. Ikeda, S. Kishida, H. Yamamoto, H. Murai, S. Koyama, A. Kikuchi, Axin, a negative regulator of the Wnt signaling pathway, forms a complex with GSK3beta and beta-catenin and promotes GSK-3beta-dependent phosphorylation of beta-catenin, EMBO J. 17 (1998) 1371–1384. [56] H. Clevers, Wnt/beta-catenin signaling in development and disease, Cell 127 (2006) 469–480. [57] R. Nusse, Wnt signaling in disease and development, Cell Res. 15 (2005) 28–32. [58] D.H. Song, I. Dominguez, J. Mizuno, M. Kaut, S.C. Mohr, D.C. Seldin, CK2 phosphorylation of the armadillo repeat region of beta-catenin potentiates Wnt signaling, J. Biol. Chem. 278 (2003) 24018–24025. [59] L.G. Van der Flier, L. Sabates-Bellver, I. Oving, A. Haegebarth, M. De Palo, M. Anti, M.E. Van Gijn, S. Suijkerbuijk, M. Van de Wetering, G. Marra, H. Clevers, The intestinal Wnt/TCF signature, Gastroenterology 132 (2007) 628–632. [60] T. Brabletz, A. Jung, K. Hermann, K. Gunther, W. Hohenberger, T. Kirchner, Nuclear overexpression of the oncoprotein beta-catenin in colorectal cancer is localized predominantly at the invasion front, Pathol. Res. Pract. 194 (2001) 701–704.