A Preclinical Model of Chronic Alcohol Consumption Reveals Increased Metastatic Seeding of Colon Cancer Cells in the Liver

Metastasis is a crucial factor for the death of cancer patients. The liver is the most frequent and critical site of metastasis of colorectal cancer, which is the third most prevalent cancer worldwide (1). Approximately 50% of colorectal cancer patients develop liver metastasis, and it is responsible for at least two-thirds of the deaths from colorectal cancer (2). Recent studies on metastasis have focused on the “seed-and-soil” hypothesis, in which metastasis depends on the interactions between cancer cells and specific organ microenvironments (3).

Alcohol is among the most widely consumed beverages around the world, and alcohol consumption has a great impact on the microenvironments of various organs, especially the liver (4). The relationship between alcohol and cancer has been extensively studied with a primary focus on carcinogenesis (4). Several epidemiologic studies presented controversial effects of alcohol consumption on cancer-associated outcomes, including survival rate and metastasis (5). An animal study showed chronic alcohol consumption accelerated melanoma metastasis into the lung by suppression of natural killer (NK) cells (6), whereas another study presented suppression of melanoma by alcohol intake via activating IFNγ-producing immune cells (7).

The liver is the central organ of alcohol metabolism. A certain level of alcohol consumption contributes to cancer development and may affect the metastasis-related microenvironment, which includes inflammatory cytokines, adhesion molecules, and angiogenesis factors (4, 8). In addition, alcohol has been well known as one of the highest risk factors for the development of colorectal cancer (9). One epidemiologic study showed a positive association between alcohol consumption and liver metastasis in colorectal cancer patients (10). However, no experimental evidence has been established implicating alcohol as a causative factor for liver metastasis of colorectal cancer.

Herein, we adapted a liver metastasis model via splenic injection with the MC38 colon cancer cell line under three different conditions of alcohol consumption, to investigate the influence on the steps of seeding and soiling as well as the microenvironment.

Syngeneic MC38 murine colon cancer cells were kindly provided by Prof. Eui-Cheol Shin (KAIST, Daejeon, Korea) in 2014. The cells were cultured in DMEM (Welgene) supplemented with 10% heat-inactivated FBS (Welgene) and 1% penicillin-streptomycin-amphotericin antibiotics (Welgene) at 37°C in 5% CO₂ humidified atmosphere. The cell line was tested and validated to be mycoplasma free. The cell line was authenticated by its morphology and ability to generate tumor in syngeneic mice throughout the experiment.

Cell proliferation was evaluated using the EZ-Cytox kit (DoGen). Briefly, MC38 cells were cultured in a 24-well microplate (5 × 10³ cells per well). The cells were incubated for 24, 48, or 72 hours after treatment with various concentrations of ethanol (0%, 0.001%, 0.01%, 0.1%, 1%, or 2%). After incubation with 10% WST-1 reagent for 60 minutes, absorbance of the supernatants (150 mL) was measured at 450 to 600 nm using VERSAmax and SoftMax 5.1 (Molecular Devices).

Specific pathogen-free C57BL/6N (male, 8-weeks-old, 22–25 g) mice were obtained from Koatech Co. After a 1-week acclimation period, the mice were given free access to either water or 20% volume/volume (v/v) alcohol as the sole drinking fluid. When first administered, the alcohol concentration was gradually increased from 10% to 20% (v/v) over the first week. Liquid consumption, food consumption, and the body weight of the mice were recorded twice per week.

All of the experiments were approved by the Institutional Animal Care and Use Committee of Daejeon University (DJUARB2014-029) and were conducted in accordance with the Guide for the Care and Use of Laboratory Animals, published by the NIH (Bethesda, MD).

C57BL/6 mice were divided randomly into four groups (n = 7–8 per group): control (water), prealcohol (ethanol for 4 weeks before tumor injection), postalcohol (ethanol for 3 weeks after tumor injection), and alcohol group (ethanol for 7 weeks, 4 weeks before and 3 weeks after tumor injection), respectively. Tumor injection was performed after 4 weeks of water (control and postalcohol group) or alcohol (alcohol and prealcohol group) intake. After an additional 3 weeks of consumption of water (control and prealcohol group) or alcohol (alcohol and post-alcohol group), the mice were sacrificed and liver metastases were evaluated (Sacrifice 1, Supplementary Fig. S1A).

To observe the altered liver status by ethanol consumption, two groups of mice were fed either water or 20% alcohol for 4 weeks. The mice were sacrificed immediately for microenvironment-related assays (Sacrifice 2), or sacrificed 24 hours after splenic injection with fluorescence-labeled MC38 cells for the examination of cancer cell arrest (Sacrifice 3, Supplementary Fig. S1B).

Liver metastasis was generated using a previously described splenic injection model (11). Briefly, the mice were anesthetized using a ketamine/xylazine mixture, and the spleen was exteriorized via a 5 mm incision in the left upper abdomen. MC38 cells (1 × 10⁵) in 100 μL of PBS were slowly injected into the spleen and allowed to flush to the liver for 1 minute. Then, the spleen was removed and homoeostasis was assured by ligation with a suture. The mice were sacrificed at 21 days postinjection. All of the mice were sacrificed under ether anesthesia, and the livers were removed and weighed. Lastly, surface liver metastases were counted under blinded conditions.

For histologic evaluation, the liver tissues were fixed in 10% formalin, embedded in paraffin blocks, and cut into 5 μm sections. The sections were stained with hematoxylin and eosin. Immunostaining was performed with an anti-ICAM-1 antibody (Abcam), a biotinylated secondary antibody (Vector), and a Vectastain ABC kit (Vector). The sections were visualized using 3,3′-diaminobenzidine (Sigma), and counterstained with Mayer Hematoxylin (Dako). The slides were imaged using a Leica EZ4 HD microscope (Leica Microsystems). For immunofluorescence, 10-μm cryosections were fixed in −20°C methanol and then treated with antibodies against NIMP-R14 (Santa Cruz Biotechnology) for neutrophils or F4/80 (Abcam) for macrophages followed by an Alexa-conjugated secondary antibody (Invitrogen), CD8a-PE for CD8⁺ T cells (BD Biosciences), or CD49b-FITC (eBioscience) for NK cells. The sections were observed using a confocal microscope (Olympus FV1200).

MC38 cells were labeled with 1 μmol/L Cell Tracker Green (Molecular Probes) and 2 × 10⁵ cells were injected intrasplenically. The mice were sacrificed at 24 hours postinjection (Sacrifice 3, Supplementary Fig. S1B). The livers were harvested and the left lateral lobes were observed using an inverted fluorescent microscope (Olympus IX71) at ×100 magnification. Eight pictures were randomly captured from each sample (n = 4 per group) using EOS Utility (Canon). Fluorescent (green) tumor cells were counted under blinded condition. The data were presented as the mean total number of cells per mouse.

The serum levels of aspartate transaminase (AST), alanine transaminase, alkaline phosphatase (ALP), triglycerides, and total cholesterol were determined using an autoanalyzer (Chiron Diagnostics Co.). Hepatic lipid profiles were measured using commercial kits (Triglyceride and total cholesterol kit, ASAN Pharmacy) according to the previous method (12).

Hepatic levels of inflammatory cytokines were measured using commercial ELISA kits: TNFα, IFNγ, IL1β, IL6 (TNFα, and IFNγ from BD Biosciences; IL1β, and IL6 from R&D Systems).

Total RNA was extracted from liver tissues using TRIzol reagent (Molecular Research Center). cDNA was synthesized from total RNA (2 μg) in a 20-μL reaction using the High-Capacity cDNA Reverse Transcription kit (Ambion). RT-PCR was performed using SYBRGreen PCR Master Mix (Applied Biosystems) with PCR amplification performed in accordance with a standard protocol using the IQ5 PCR Thermal Cycler (Bio-Rad). The 2 (−ΔΔC_t) method was used to express relative mRNA levels after correcting for β-actin mRNA expression. The following primers (5′→3′, forward and reverse) were used: Sele (NM_011345), Icam1 (NM_010493), Vcam1 (NM_011693), Tlr4 (NM_021297), and Actb (NM_007393).

Lymphocyte subclasses in the peripheral blood were analyzed using a FACS Caliber and Cell Quest Pro software (BD Biosciences) after staining with fluorescent dye–conjugated antibody mixtures containing CD335 (NKp46)-FITC for NK cells, CD3e-PE for T cells, CD8-PerCP for CD8⁺ T cells, and CD19-APC for B cells (BD Biosciences). Complete blood counts were obtained using HEMAVET (CDC Technologies).

The data are expressed as the means ± SD or fold changes in means. Statistical significance was determined by Student t test or Mann–Whitney U test. In all analyses, P < 0.05 was taken to indicate statistical significance.

The liver is the first organ encountered by most metastatic colorectal cancer cells due to the hemodynamics between the liver and intestine; the liver also provides a relatively favorable microenvironment for these cells (13). Because liver metastasis is the primary determinant of colorectal cancer patient survival, identifying the associated risk factors is necessary in the clinical field. Previous studies have presented the liver metastasis-enhancing effect of a high-fat diet and endotoxins on colorectal cancer (11, 14); however, regarding alcohol consumption and liver metastasis in colorectal cancer, no experimental study has been conducted, and their relationship is only based on clinical observation (10)

To investigate the effect of alcohol intake on the two main stages of cancer metastasis to the liver (survival and adhesion vs. metastatic colonization), we established a splenic injection model with MC38 cells under four different conditions of alcohol intake (control, prealcohol, postalcohol, or thorough alcohol consumption). Alcohol consumption significantly increased liver metastasis of colorectal cancer as evidenced by metastatic nodule numbers under naked view, histologic findings, as well as liver weights (Fig. 1A–1E). The number of tumor nodules was significantly increased by all three conditions of alcohol intake comparing water group (6.0 ± 2.7 as mean ± SD): prealcohol (14.4 ± 4.0, 2.4-fold, P < 0.001), postalcohol (12.3 ± 4.0, 2.0-fold, P < 0.01), and thorough alcohol consumption (13.4 ± 3.4, 2.2-fold, P < 0.001, Fig. 1C). The increased liver metastasis in prealcohol group may indicate that alcohol promotes survival and adhesion of cancer cells whereas the postalcohol data suggest the promotion of colonization by alcohol consumption. There was no significant difference among the three alcohol-consuming groups.

In the current study, C57BL/6 mice were fed with 20% ethanol in the drinking water as described by others. This alcohol dosage mimics chronic alcohol drinking in human (15). The food intake was reduced by 0.8-fold during alcohol consumption; however, the bodyweight gain was not affected. The above alcohol intake elevated the blood alcohol concentration by 0.5% in mice as previously reported (16). We considered whether ethanol might directly affect cancer cells in the blood. Our results demonstrated that MC38 cell proliferation was notably inhibited by various concentrations of ethanol (from 0.001% to 2% over 48 hours, Fig. 2F). These results may suggest that the alcohol-associated promotion of liver metastasis of MC38 cells was affected by an alteration of the host microenvironment rather than directly by alcohol.

The organ microenvironment plays a key role in cancer development as well as metastasis (13). Chronic alcohol intake increases the production of hepatic inflammatory cytokines and reactive oxygen stress (17, 18). Inflammation regulates cancer progression by acceleration of cell proliferation, inhibition of apoptosis, and stimulation of angiogenesis (8). We found significantly increased levels of inflammatory cytokines (TNFα, 1.3-fold; IL1β, 1.5-fold; IL6, 1.5-fold; and IFNγ, 1.2-fold increase) in hepatic tissue by 4-week alcohol intake (Fig. 3A), which is in agreement with the elevated AST and ALP levels in the serum (Supplementary Table S1), and the relatively increased infiltration of immune cells including neutrophils, macrophages, and NK cells into the liver (Fig. 3D). This notion suggested that the inflammatory microenvironment would be involved in the mechanisms underlying the accelerated hepatic metastasis of colon cancers in alcohol consumption conditions. This alcohol intake increased hepatic lipid contents by approximately 1.4-fold comparing with the basal level of water group (Supplementary Table S1), thus alcohol-induced fatty change also would affect liver microenvironment.

Metastasis consists of multiple steps: (i) invasion (ii) survival and adhesion to the distant site, and (iii) metastatic colonization (13). The numbers of cancer cell colonies 3 weeks after MC38 cell injection, before and after different conditions of alcohol consumption, were determined. The results showed that alcohol consumption promoted the seeding and soiling events of MC38 cells in liver tissue. The seeding of cancer cells to a distant site is a critical step for the development of cancer metastasis. A high-fat diet was reported to alter the liver environment to be favorable for hepatic seeding of colorectal cancer in an animal model (19). We further confirmed that chronic alcohol intake facilitates the seeding of cancer cells to the liver (2.5-fold increase, P < 0.001) using a fluorescence-based metastasis tracking assay (Fig. 2A, d and E).

The inflammation-associated microenvironment is also closely linked to increased adhesion molecules in various tissues (8). In particular, adhesion molecules such as E-selectin, ICAM-1, and VCAM-1 play key roles in cancer cell arrest in the liver (20). In the current study, the hepatic expression of Icam1 was significantly elevated by 1.4-fold in alcohol (4 weeks) group, whereas no significant change was observed in E-selectin (Sele) and Vcam1 (Fig. 3B). The increased hepatic level of ICAM-1 is further demonstrated by immunohistochemical analysis (Fig. 3C). Previous studies have reported the potential role of ICAM-1 in cancer metastasis. Clinical research on colorectal cancer has revealed a positive relation between liver metastasis and the sinusoidal expression of ICAM-1 (21); however, a recent study showed that ICAM-1 deficiency in mice accelerated liver metastasis of SL4 colon cancer cells (22). Further studies are needed to determine the role of ICAM-1 in liver metastasis under condition of alcohol consumption.

The activation of toll-like receptor 4 (TLR4) by gut-derived endotoxin is one mechanism that can explain alcohol-induced liver injury (23). Furthermore, the stimulation of TLR4 is previously reported to promote liver metastasis of colorectal cancer (14); however, we observed no difference in the expression of Tlr4 in the alcohol (4 weeks) group compared with the water group (Fig. 3B). This result indicated that 4 weeks of 20% ethanol intake induced only mild liver inflammation without the hepatic activation of TLR4, suggesting that the alcohol-derived acceleration of hepatic metastasis was not caused by endotoxin.

Cancer cells are under immune surveillance for tumor development, growth, and metastasis (13). Alcohol is well known as an immunosuppressant; however, several studies have also reported that chronic alcohol intake can activate immune system (5). In the current study, 4 weeks of alcohol intake significantly decreased the numbers of NK cells and CD8⁺ T cells and their ratios by 0.6- and 0.7-fold, respectively, in the peripheral blood (Fig. 4), which is consistent with previous studies (6, 16). This result somewhat conflicts with above observation of NK cells' infiltration in liver, whereas the hepatic NK cells are also known as contributor in alcohol-induced liver injury (24). NK cells and CD8⁺ T cells are the key immune cells responsible for cytotoxic antitumor activity (25). Thus, the reduction of NK cells and CD8⁺ T cells likely contributed to facilitate liver metastasis in our mouse model.

Taken together, the current study establishes the first experimental evidence that alcohol consumption facilitates liver metastasis of colorectal cancer by augmenting both the seeding and soiling steps of metastasis via an alteration of the liver microenvironment and immune surveillance.

No potential conflicts of interest were disclosed.

Conception and Design: H.-J. Im, H.-G. Kim, C.-G. Son

Development of methodology: H.-J. Im, H.-G. Kim, S.-J. Park, C.-G. Son

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): H.-J. Im, H.-G. Kim, J.-S. Lee, H.-S. Kim, I.-J. Jo, S.-J. Park, C.-G. Son

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): H.-J. Im, H.-G. Kim, J.-S. Lee, H.-S. Kim, I.-J. Jo, C.-G. Son

Writing, review and/or revision of the manuscript: H.-J. Im, H.-G. Kim, J.-H. Cho, I.-J. Jo, C.-G. Son

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): C.-G. Son

Study supervision: C.-G. Son

The authors thank Prof. Eui-Cheol Shin (KAIST, Daejeon, Korea) for providing the MC38 cell lines.

This study was supported by a grant (HI12C1920 to C.-G. Son) from the Oriental Medicine R&D Project.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.