Background
Hepatocellular carcinoma (HCC) is the most common primary liver cancer and accounts for up to 70-85% of primary liver cancers [
1]. Advanced HCC carries a poor prognosis with a five-year survival rate <10% [
2]. It is well recognized that both genetic and environmental factors contribute to human HCC. Alcohol, an important environmental factor, was classified as a group I carcinogen and chronic alcohol consumption has been recognized as an important risk factor for liver cancers [
3,
4]. In addition to act as risk factor for carcinogenesis, epidemiological studies also indicated that alcohol consumption is often associated with advanced and aggressive tumors [
5]. Although alcohol is clearly recognized as a risk factor for liver cancers, its role in cancer progression and metastasis is currently unknown. We specifically hypothesized that chronic alcohol consumption might enhance HCC progression and metastasis.
Tumor angiogenesis, the formation of new blood vessels from endothelial precursors within tumors, is a prerequisite step for growth and progression of solid malignancies [
6]. Vascular endothelial growth factor (VEGF) is an essential player in tumor angiogenesis and mediates tumor aggressiveness [
7-
9]. Monocyte chemotactic protein-1 (MCP-1), a key CC chemokine responsible for trafficking and activation of monocytes/macrophages has been recognized as an important angiogenic chemokine [
10,
11]; it also plays a critical role in solid tumors [
12]. Previous studies indicated that alcohol exposure up-regulated the expression of VEGF and MCP-1 in other experimental models [
13]. It has also been reported that the expression of VEGF and MCP-1 was regulated by reactive oxygen species and NF-ĸB [
14,
15]. Based on these findings, we further hypothesized that chronic alcohol consumption might enhance HCC progression and metastasis through up-regulating the expression of VEGF/MCP-1, which, in turn, may be regulated by reactive oxygen species and the NF-ĸB signaling pathway. We tested this hypothesis by retrospectively analyzing clinical data, and functional analysis using experimental models.
Discussion
The association between alcohol consumption and the risk of HCC has been well-established. For example, a study with a cohort of 19,000 drinkers confirmed that alcohol enhances the risk of upper gastrointestinal tumors including liver cancer [
16]. Rehm et al. [
17] reported that RRs of liver cancer for low (0–39.99 g/d), hazardous (40–59.99 g/d) and harmful (60+ g/d) levels of alcohol consumption compared to never drinkers are 1.45, 3.03, and 3.60 respectively. Hennig et al. suggested the close association between alcohol consumption and HCC progression
in vitro [
18]. Huang et al. also confirmed that ethanol consumption induces hepatocellular carcinoma cell metastasis by changing the extracellular matrix
in vitro and this ability was reduced by food components such as pterostilbene and curcumin analogues [
19]. In a mouse model of hepatocaricinogenesis, Brandon-Warner et al. found that chronic ethanol feeding accelerated hepatocellular carcinoma progression in male mice [
20]. In our study, we demonstrate a positive correlation between alcohol consumption and enhanced HCC progression, even though the cohort of this study is relatively small (82 patients). Donato et al. showed that heavy alcohol consumption (>80 g of ethanol per day) is associated with the increased risk of HCC compared to non-drinkers [
21]. However, our data indicated that the moderate and heavy drinking (35–87.5 g/day and ≥ 87.5 g/day) was associated with enhanced HCC progression. There are many factors underlying this discrepancy. A larger cohort may be needed to provide more definite conclusion. In our cohort, the number of male patients is much larger than that of female; this is consistent with the epidemiological findings that males are about two to four times more likely to develop HCC than females [
22]. In future study, analysis of a larger cohort may be necessary to evaluate the interaction between alcohol consumption and other risk factors, such as Hepatitis B virus or hepatitis C virus which are known risk factors for the development of liver cancer due to chronic liver inflammation [
23]. Nevertheless, our clinical data provides the necessary foundation for our further experimental studies.
HCC is a vascular-dependent malignant tumor; an increase in tumor dimension above 0.5 mm will induce the proliferation of vascular endothelial cells and angiogenesis which is essential for tumor growth and metastatic dissemination [
24]. Our previous study and results from others showed that alcohol may promote tumor angiogenesis in animals [
13,
25,
26]. The current study for the first time demonstrates that alcohol consumption is closely associated with a higher micrcovessel density in HCC tissues, which implied that alcohol potentially enhances tumor angiogenesis in HCC patients. This interpretation is further supported by experimental data. For example, alcohol increases micrcovessel density in tumor tissues in a HCC xenograft model . Furthermore, we show that alcohol can stimulate tumor angiogenesis in a 3D model of tumor/endothelial cell co-culture system. Together, these results indicate that enhanced angiogenesis is a potential mechanism for alcohol-promoted progression of HCC.
VEGF and MCP-1 are key mediators for tumor angiogenesis and metastasis [
27,
28]. VEGF, as a pro-angiogenic factor, plays an essential role in the process of angiogenesis, including microvascular permeability, endothelial cell proliferation and tumor cell migration [
29,
30]. VEGF-positive tumors in HCC patients had much greater invasive potential and intrahepatic metastasis than VEGF-negative tumors [
31]. Tan et al. [
25] showed that alcohol increased the expression of VEGF in melanoma xenograft model. MCP-1 is also a potent pro-angiogenic chemokine which is associated with many malignant tumors progression by recruitment of macrophages and induction of angiogenesis [
32,
33]. In both ectopic and orthotropic xenograft models, the MVD was significantly increased in tumors over-expressing MCP-1 [
34]. We have previously demonstrated that alcohol can stimulate the expression of MCP-1 in breast cancer cells, which caused an enhanced angiogenesis [
13]. That a higher expression level of VEGF and MCP-1 is observed in HCC patients of alcohol-drinkers supports a role of VEGF and MCP-1 in alcohol promotion of HCC.
NF-κB plays a key role in tumorigenesis/progression and also regulates the expression of VEGF and MCP-1 [
35]. Consistently, we show that the expression of NF-κB is higher in alcohol-drinkers HCC patients than in that in non-drinkers. Furthermore, PDTC, an inhibitor of NF-κB and an antioxidant, attenuates alcohol-induced VEGF and MCP-1 expression in mice, supporting that ethanol-stimulated expression of VEGF/MCP-1 is mediated by NF-κB signaling.
NF-κB activity is regulated by reactive oxygen species (ROS) [
36]. We have previously shown that ethanol can induce intracellular ROS accumulation in breast cancer cells [
37,
38]. ROS plays an important role in carcinogenesis as a result of oxidative stress including oxidative injury, inflammation, and lipid peroxidation [
39]. C3G, a potent antioxidant, has been shown to effectively inhibit ROS and mitigate ethanol-induced oxidative stress [
38]. We confirm here that ethanol causes the accumulation of intracellular ROS in HepG2 cells and C3G scavenges ethanol-induced ROS. More importantly, C3G blocks ethanol-stimulated NF-κB promoter activities, suggesting that ethanol-induced ROS mediated NF-κB transcriptional activity. At cellular level, both PDTC and C3G effectively inhibit ethanol-induced migration/invasion of HepG2 cells and abolish the pro-angiogenic effect of ethanol in a 3-D model of tumor/endothelial cell co-culture. They also attenuate ethanol-induced tumor growth and metastasis in mice.
Materials and methods
Drugs and reagents
Ethanol, fibrinogen, aprotinin, thrombin and Pyrrolidine dithiocarbamate (PDTC) were purchased from Sigma Chemical Co. (St. Louis, MO). Cyanidin-3-glucoside (C3G) was prepared as previously described [
38]. Anti-MCP1 and anti-VEGF antibodies were obtained from Abcam (Cambridge, MA). Anti-NF-κB p65, IκBα, p-IκBα and anti-LMNB1 antibodies were purchased from Protein Tech Group (Chicago, IL, USA). Anti-β-actin was obtained from Cell signaling Technology (Danvers, MA). Anti-CD31 antibody was obtained from BD Pharmingen (San Diego, CA). Reactive oxygen species detection reagents were obtained from Invitrogen Molecular Probes (Eugene, OR). MTT assay kit was purchased from Roche Molecular Biochemicals (Indianapolis, IN). Matrigel Invasion Chambers were purchased from BD Biosciences (Bedford, MA). Cytodex 3 beads were purchased from Amersham Pharmacia Biotech (Piscataway, NJ).
Clinical patient data
The medical records of 52 HCC patients, admitted to first hospital of Anhui Medical University, were retrospectively analyzed between January and December 2009 (Table
1). Patients were asymptomatic at admission and diagnosed by examining the liver with computed tomography. All patients were performed anatomical liver resection. HCC and cirrhosis was confirmed by pathology. Cirrhosis developed in 76.9% (39 of 52) of patients. Forty-three patients (82.7%) were infected with hepatitis B virus. A series of demographic and clinical data were collected including alcohol consumption, tumor characteristics and pathologic stages. Survival and follow-up data were acquired by telephone interviews. With the approval of the first affiliated hospital of Anhui Medical University of Human Studies Committee, formalin-fixed paraffin-embedded tumor materials of 52 HCC patients were obtained. The histological sections of all cases were evaluated by oncopathologists.
Cell culture and ethanol exposure
Human hepatoma cell line HepG2 and mouse embryonic fibroblast cell line NIH3T3 were provided by the Institute of Cell and Biochemistry, Chinese Academy of Sciences (Shanghai, China) and maintained in Dulbecco’s modified Eagle’s medium(DMEM) containing 10% fetal bovine serum (FBS), penicillin (100 U/ml)/streptomycin (100 U/ml), at 37°C with 5% CO2. The original primary human umbilical vein endothelial cells (HUVECs) were isolated from human umbilical veins by using collagenase (Roche Diagnostics, Switzerland) as previously described [
40], and maintained on 0.2% gelatin coated in EGM-2 medium (Lonza, Walkersville, MD, USA) which consists of endothelial cell basal medium (EBM-2) with 10% FBS and other additives. Alcohol was added to the medium at concentrations of 0.2% v/v ethanol. Cell culture plates were placed on a rack inside a plastic container sealed with a tight-fitting lid. At the bottom of each container, there was a 200-ml water bath, which contained the same concentration of ethanol as in the cultural media. The containers were placed in a humidified environment and maintained at 37°C with 5% CO2.
Animals and ethanol administration
Male nude mice (4–5 weeks old, 15–20 g, n = 12/group) were purchased from the Animal Center, Chinese Academy of Sciences. The ethanol was administrated in drinking water as described previously [
13]. Briefly, mice were given 2% ethanol in drinking water for a 12 h period during the night and then replaced with water without ethanol at day time. The mice in the control group were provided with regular drinking water only. All procedures were carried out according to the guidelines of the Animal Welfare Act approved by the Institutional Animal Care and Use Committee of Anhui Medical University. The blood ethanol concentration (BEC) was determined at 6:00 a.m. using an Analox AM1 Alcohol Analyzer (Analox Instruments, Lunenburg, MA) as previously described [
41]. The BEC was 56.18 ± 11.6 mg/dl.
Mouse model of tumor xenograft
HepG2 human hepatoma cells were implanted subcutaneously in nude mice according to the previous study [
13]. Briefly, three days after ethanol exposure, HepG2 cells (1 × 10
6 in 100 μl PBS) were injected subcutaneously into mice using a 23-gauge needle. The mice were continually provided with 2% ethanol in drinking water or regular drinking water without ethanol. The size of the tumors was monitored every 2 days; two perpendicular dimensions of tumors were measured with a dial caliper. The volume was calculated based on the formula: V = 0.25 a × b
2; a is the longest and b is the shortest dimension. At the end of experiment, animals were killed and the tumors were removed. Some of the tumor tissues were fixed with 10% neutral formalin for and immunohistochemical studies. To evaluate the effect of PDTC on ethanol-mediated tumor growth, PDTC [100 mg/kg in 100 μl of Dimethyl sulfoxide (DMSO)] or DMSO alone was injected intraperitoneally 2 days following ethanol exposure. PDTC was administered every three days. The dosage of PDTC was selected based on previous studies [
42]. There were seven mice for each treatment group.
Histology and immunohistochemistry
Immunohistochemical (IHC) procedure was performed generally as described [
43]. Briefly, Five-micrometer-thick sections were cut and deparaffinaged in xylene prior to rehydration using gradient alcohol. For antigen retrieval, sections were treated with citrate buffer saline (pH 6.0) for 15 minutes at 95°C in a microwave oven. After blocking with 10% normal goat serum for 30 minutes at room temperature, the sections were incubated with primary antibodies: anti-MCP-1 (1:100), anti-VEGF (1:150), anti-NF-κBp65 (1:100), or anti-CD31 (1:100) overnight at 4°C. Following incubation, sections were washed with phosphate buffered solution and incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG for 1 h at room temperature. The avidin-biotin complex was added and the enzyme activity was visualized with 3, 3’-diaminobenzidine (DAB). Negative controls were prepared using PBS instead of the primary antibody. The average microvessel density (AMVD) of ten selected microscopic fields was calculated and expressed as the number of microvessels per mm
2 area, according to previous report [
44]. The integral optical density (IOD) of the stained sections was measured with a computer-assisted image-processing and analytical system. NF-κBp65 expression was analyzed by counting NF-κBp65-positive hepatocytes within 50 HPF with discrimination of the intracellular distribution serving as an indicator of NF-κB nuclear translocation. The nuclear-to-cytoplasmic ratio was calculated by dividing the number of cells with nuclear NF-κBp65 positivity by the number of cytoplasmic NF-κBp65 positive cells.
Analysis of experimental lung metastasis
After 5 weeks of ethanol exposure, mice in four groups (n = 22 for each group) were sacrificed. The fresh lung samples were harvested and fixed with 10% formalin for histopathology analysis. Tissues were paraffin-embedded and sectioned at a thickness of 5 μm. The sections were stained with hematoxylin–eosin (H&E) and then examined under the microscope to evaluate the morphological changes of metastasis tumors.
Detection of intracellular reactive oxygen species
Intracellular reactive oxygen species (ROS) levels were measured using the fluorescent dye CM-H2DCFDA (Invitrogen Corporation, Carlsbad, CA) as previously described [
45]. Briefly, HepG2 cells were trypsinized and seeded onto 6-well plates pretreated with/without C3G (20 μM) or PDTC (20 μM) for 0.5 hours before ethanol exposure. After the treatment, cells were washed with cold PBS and incubated with 5 μM CM-H2DCFDA for 30 min, followed by several additional washes with cold PBS. Cells were trypsinized and transferred into polystyrene round-bottom tubes; intracellular ROS levels were measured with a flow cytometer (FACScalibur, BD Biosciences, San Jose, CA) at an emission wavelength of 525 nm.
Cells (5 × 106) were harvested, resuspended in 50 μl of lysis buffer A (10 mM Hepes, 10 mM KCl, 0.1 mM MgCl2, 0.1 mM EDTA, 2 μg/ml leupeptin, 2 μg/ml pepstatin, 0.5 mM PMSF, pH 7.9), incubated on ice for 10 min and centrifuged for 10 min at 800 × g at 4°C. The supernatant was treated as the cytoplasmic fraction. The pellet (nuclei) was washed with buffer A and nuclear proteins were extracted in presence of 50 μl of buffer B (10 mM Hepes, 400 mM NaCl, 1.5 mM MgCl2, 0.1 mM EDTA, 2 μg/ml leupeptin, 2 μg/ml pepstatin, 0.5 mM PMSF, pH 7.9). Protein concentration was determined using a BCA protein assay kit (Pierce, Rockford, IL) according to the manufacturer's instructions.
Western blotting analysis
SDS–PAGE and Western blotting were performed according to standard procedures. Briefly, protein samples (10 μg) were subjected to electrophoresis on 4-12% SDS-polyacrylamide gradient gels and transferred to a PVDF membrane. After overnight blocking with 2% BSA in TBS-Tween 0.1%, the membranes were probed with anti-NF-κB p65 (1:1000), IκBα (1:500), p-IκBα (1:1000), β-actin (1:1000) or anti-LMNB1 (1:1000) antibodies overnight at 4°C. Anti-rabbit HPR conjugated antibody was used as a secondary antibody followed by enhanced chemiluminescence reaction. All Western blotting were repeated three times.
Luciferase reporter gene assay
NIH3T3 cells were trypsinized and seeded in 96-well plates at an initial density of 2 × 104 cells/well in DMEM containing 10% FBS at 37°C with 5% CO2 overnight. The cells at 50%–80% confluence were performed with 6 μg of pGL4.32 [luc2P/NF-κB-RE/Hygro] plasmid (Promega) transfected with Lipofectamine™ 2000 Transfection Reagent (Invitrogen) for normalizing transfection efficiency. The non-transfected cells were used as blank group. At 48 hours after transfection, cells were collected and resuspended in passive lysis buffer (Promega). Then the cells were exposed with/without ethanol in the presence or absence of PDTC or C3G. After incubation for 2 h, Luciferase activity was evaluated with a dual luciferase assay system (GloMax 96 Microplate Luminometer, Promega) according to manufacturers.
Cell proliferation assay
The MTT assay was employed to determine the number of viable cells in culture. Briefly, the cells were plated into 96-well plates and exposed to ethanol with/without C3G (20 μM) or PDTC (20 μM) for indicated times. After the treatment, 10 μl of MTT reagent was added into each well and the plates were incubated at 37°C for 4 h. After lightly vortexing the plate on an orbital shaker, the spectrophotometric absorbance was read on a microplate reader (Bio Rad Model 3550-UV) at 595 nm. Each individual experiment was performed at least three times.
Wound healing migration assay
Wound healing assay was carried out to determine the cell migration ability of tumor cells. The wound healing migration assay was performed as described previously [
46]. Briefly, HepG2 cells were plated on 6-well plates and grown to 80% - 90% confluence. Sterilized one-milliliter pipette tip was used to generate a wound across the cell monolayer, and the plates were washed with PBS. The cells were then treated with ethanol (0.2% v/v) in the presence or absence of PDTC (20 μM) for 24 h. Cells migrated into the wounded area or protruded from the border of the wound were visualized and photographed under the inverted microscope (Olympus, Japan).
Cell invasion assay
Cell invasion was performed in BD Bio Coat Matrigel Invasion chambers (BD Biosciences, Clontech; Bedford, MA, USA) with a porous polycarbonate membrane (8.0 micron pore size) as previously described [
47]. Briefly, HepG2 cells (5 × 10
5) were seeded onto upper chambers and treated with ethanol in the presence or absence of C3G or PDTC(20 μM). The lower compartment of the chamber was filled with medium containing 2%BSA served as chemoattractants for the cells. The chambers were cultured at 37°C in 5% CO2 for 48 h to allow tumor cell invasion through the matrix. Migrated cells were fixed with 100% methanol and stained with Giemsa for invasion assay. The filters were photographed and the cells were counted. The invasion assay was repeated in five separate experiments, with control and experimental groups performed in parallel.
Three-dimensional (3D) endothelial cell and tumor co-culture system
3-D model of endothelial cell and tumor cell co-culture system was performed to investigate the effect of ethanol on tumor angiogenesis as described previously [
13]. In this model, endothelial cells were induced to form a 3D capillary tube-like network on a fibrin gel bead system in the presence or absence of tumor cells. In brief, HUVEC or HepG2 cells (1 × 10
6) were trypsinized and mixed with Cytodex beads (3 × 10
3) in DMEM medium. After incubation, the mixtures of cells/Cytodex beads were re-suspended in medium containing 2.5 mg/ml fibrinogen and 0.15 U/ml aprotinin. A volume of 0.5 ml fibrinogen/bead solution was added to 24-well cell culture plates that were pre-coated with 0.625 U of thrombin. After coagulation, the resulting fibrin gels contained endothelial cells (HUVEC or HepG2) adhering to the beads. For co-culture of endothelial cells/tumor cells, the medium collected from HepG2 cells with/without ethanol in presence or absence of C3G or PDTC were added on top of the fibrin gels. The culture media was changed every two days. At different time points, cell morphogenesis was observed using an inverted phase-contrast microscope. Tube formation was quantified by counting branches from four representative fields per replicate. These experiments were performed in triplicate.
Statistical analysis
Data were presented as means ± standard deviations and analyzed by ANOVA, followed by Student-Newman-Keuls test. The correlations between alcohol consumption and clinicopathological features were analyzed by Pearson Chi-square test. Multivariable binary logistic regression was carried out to estimate odds ratio (OR) with its 95% confidence intervals (CI) in order to assess the risk factors for HCC aggressiveness. The multivariate Cox proportional hazards regression model was employed to estimate the prognostic factors associated with HCC recurrence. For the survival analysis after surgical resection, the data were processed using the Kaplan-Meier method and were compared using the log-rank test. Statistical analysis was conducted with SPSS 16.0 (IBM Corporation, Chicago, IL, United States). A value of P < 0.05 was considered statistically significant.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
FW, J-LY, K-KY, S-YW carried out the molecular studies, participated in the sequence alignment and drafted the manuscript; Y-ZX, MX, LC, F-FL, carried out the in vivo experiments; Y-ML, H-SF, X-YW and Z-QH participated in the in vitro experiments; JL, SYW participated in the design of the study and performed the statistical analysis; JL, J-LY, K-KY, Y-ZX and LK conceived of the study, and participated in its design and coordination and helped to draft the manuscript. All authors read and approved the final manuscript.