It is unknown as to how liver metastases are correlated with host immune status in colorectal cancer. In this study, we found that IL6, a proinflammatory cytokine produced in tumor-bearing states, promoted the metastatic colonization of colon cancer cells in association with dysfunctional antitumor immunity. In IL6-deficient mice, metastatic colonization of CT26 cells in the liver was reduced, and the antitumor effector function of CD8+ T cells, as well as IL12 production by CD11c+ dendritic cells, were augmented in vivo. IL6-deficient mice exhibited enhanced IFN-AR1–mediated type I interferon signaling, which upregulated PD-L1 and MHC class I expression on CT26 cells. In vivo injection of anti–PD-L1 effectively suppressed the metastatic colonization of CT26 cells in Il6−/− but not in Il6+/+ mice. Finally, we confirmed that colorectal cancer patients with low IL6 expression in their primary tumors showed prolonged disease-free survival. These findings suggest that IL6 may be a promising target for the treatment of metastasis in colorectal cancers by improving host immunity.

The suppression of antitumor effector cells in a tumor-bearing host is a critical problem for the immunotherapy of cancer patients. Immune-checkpoint inhibitors, such as anti–CTLA-4, anti–PD-1, and anti–PD-L1, have had success in some advanced cancers (1, 2). However, the clinical efficacy of immune-checkpoint blockade therapy for cancer varies according to the tumor type and between individual patients, suggesting that several different immunosuppressive mechanisms might operate in tumor-bearing hosts (3, 4). Therefore, alternative or additional targets to improve host immune status are required for the development of cancer immunotherapy with improved effectiveness.

Interleukin-6 (IL6), a multipotent cytokine, binds to IL6 receptor-alpha (IL6Rα) expressed on target cells. IL6/IL6Rα complexes with gp130, a signal transducer, to recruit Janus kinase, which phosphorylates signal transducer and activator of transcription 3 (STAT3). Phosphorylated STAT3 translocates to the nucleus and induces the transcription of various target genes to regulate cellular functions such as the proliferation, survival, and differentiation of various cell types, including epithelial cells, endothelial cells, fibroblasts, osteocytes, immune cells, and cancer cells (5–7).

We previously demonstrated that activation of the IL6/STAT3 signaling cascade in dendritic cells (DC) causes a reduction in antigen presentation ability and suppresses the subsequent antigen-specific helper and cytotoxic T-cell responses in vitro and in vivo (8, 9). We also revealed that blockade of IL6 signaling through the administration of an IL6Rα monoclonal antibody (mAb) to tumor-bearing mice significantly inhibits tumor growth by enhancing the effector function of CD8+ T cells (10, 11). We also previously demonstrated that IL6 inhibits the maturation and antigen presentation of human monocyte-derived DCs and suppresses antigen-specific T-cell responses in an IL12-dependent manner (12). In this study, IL6 gene expression was higher in CD11b+CD11c+ cells in the tumor tissues of colorectal cancer patients compared with peripheral blood mononuclear cells (PBMC). This confirmed that the TCR-mediated activation of both CD4+ T cells and CD8+ T cells was significantly reduced in the presence of CD11b+CD11c+ cells in tumor tissues compared with PBMCs.

Colorectal cancer, a leading cause of mortality, is the third most common cancer worldwide (13). It was reported that approximately 694,000 individuals died of colorectal cancer in 2012 (14, 15). Metastasis is one of the most severe risk factors for cancer death, and the liver is the most frequent site for the metastasis of colon cancer cells (16). However, the precise mechanism of liver metastasis with regard to the host's immune system has not been fully elucidated (17, 18). Identification of the causes of liver metastasis might, therefore, be a critical step toward controlling colorectal cancer.

This study aimed to elucidate the precise mechanisms involved in the regulation of liver metastasis and how they correlated to host immunity. We investigated whether IL6 was related to the metastatic colonization of colon cancer cells by inhibiting antitumor immunity in mouse models. We report that IL6 produced by tumor-bearing hosts contributed to the progress of metastatic colonization of colon cancer cells through the suppression of antitumor effector cells, including CD8+ T cells. These findings may help control the immunosuppressive colorectal cancer microenvironment.

Mice and cells

Wild-type BALB/c (Il6+/+) mice were obtained from Charles River Japan. BALB/c background IL6-deficient (Il6−/−) mice were obtained from the Center for Animal Disease Models, Research Institute for Biomedical Sciences, Tokyo University of Science. All mice were maintained in specific pathogen–free conditions in accordance with the guidelines of the Animal Department at Hokkaido University and were used at 6 to 8 weeks of age. All mouse experiments were approved by the Animal Ethics Committee of Hokkaido University (No. 14-0062) and conducted in accordance with the recommendations of the Guide for the Care and Use of Laboratory Animals of the University, an Institutional Animal Care and Use Committee.

The murine colon cancer cell line CT26 (CRL-2638) and murine breast cancer cell line 4T1 (CRL-2539) were obtained from the American Type Culture Collection in 2013 and frozen upon initial expansion. These cells were cultured, as described below. The cell lines used in experiments were cultured for a maximum of 20 passages before use. Cells were not reauthenticated and were tested as Mycoplasma pulmonis– and mouse hepatitis virus–negative. Murine IL6 CRISPR/Cas9 KO (sc-421114) and control CRISPR/Cas9 KO plasmids (sc-418922) were purchased from Santa Cruz Biotechnology. IL6, Il6 gene knockout, and control mock CT26 cells were established using the CRISPR/Cas9 KO system according to the manufacturer's protocol.

For ex vivo analysis, CT26 and 4T1 cells (2 × 105 cells/well in 12-well dishes) were transfected with pMX-IRES-GFP, obtained from Dr. T. Kitamura (The University of Tokyo), using Lipofectamine 3000 (Thermo Fisher Scientific). These GFP-transduced CT26 and 4T1 cells were used in the tumor-bearing mouse models, as described below.

Human subjects

Research protocols involving human subjects were approved by the institutional review board (IRB) of Hokkaido University Graduate School of Medicine (14-043) and the Institute for Genetic Medicine (14-005 and 14-0004). Written informed consent was obtained from each patient. In total, 108 colorectal cancer patients who underwent surgery at Hokkaido University Hospital (HUH; Sapporo, Japan) between 2003 and 2015 were included in this study. Inclusion criteria specified that the patients needed to have a diagnosis of an abdominal malignancy and be receiving their treatment at HUH. Excluded from this analysis were patients who required emergency surgery as a result of their underlying surgical condition and/or inadequate data concerning histology and follow-up. Samples (blood and tumor tissues) were collected at surgery. The material was provided as formalin-fixed, paraffin-embedded (FFPE) sections according to standard laboratory pathology practice and stored at room temperature until stained. The study was conducted in accordance with the Helsinki Declaration on ethical principles for medical research involving human subjects. Patients were followed up at 1- to 6-month intervals until death or until 31 March 2019. To monitor post-surgical tumor recurrence, blood tests for tumor markers and thoracoabdominal CT were done according to common protocols for surveillance. Patient clinical features are shown in Supplementary Table S1.

Antibodies and reagents

The following mAbs were obtained from BioLegend: fluorescent dye–conjugated anti-CD45 (30-F11), APC-conjugated anti-mouse CD11c (N418), APC-conjugated CD4 (RM4-5), APC-Cy7–conjugated anti-mouse CD8a (53-6.7), and APC-conjugated anti-mouse CD274 (10F.9G2); the following were obtained from BD Biosciences: FITC-conjugated CD44 (IM7), PE-conjugated anti-mouse CD62L (MEL-14), anti–H-2Kd (AF6-88.5), anti–I-Ad (AF6-120.1), and anti-CD16/32(2.4G2). PE-conjugated anti-mouse perforin (eBioOMAK-D) and PE-conjugated anti-mouse granzyme B (NGZB) were obtained from eBioscience. 7-AAD Viability Dye was purchased from Beckman Coulter. mAbs for IL6 neutralization (clone MP5-20F3), CD8 depletion (clone 53.6.7), IL12 neutralization (clone C17.8), IFN-AR1 signaling inhibition (clone MAR1-5A3), and an antagonist mAb against PD-L1 (clone 10F.9G2) were purchased from Bio X Cell. Clophosome-clodronate liposomes and control liposomes (Neutrl; F70101-N) were purchased from FormuMax. Recombinant murine IFNα (#BL752802) was purchased from BioLegend. Recombinant murine IFNβ (#8234-MB) was purchased from PeproTech EC Ltd. Phorbol 12-myristate 13-acetate (PMA) and A23187 calcium ionophore were purchased from Sigma-Aldrich. Phospho-Stat1 (Tyr701; 58D6) Rabbit mAb was purchased from Cell Signaling Technology.

Cell culture

CT26 cells were maintained in RPMI-1640 medium (Wako Pure Chemical Industries) supplemented with 10% fetal bovine serum (#172012, Nichirei Bioscience, Inc.), penicillin (200 U/mL), streptomycin (100 μg/mL; Meiji Seika Pharma Co., Ltd.), 10 mmol/L HEPES (Wako Pure Chemical Industries, Ltd.), and 2-mercaptoethanol (0.05 mmol/L; Nacalai Tesque, Inc.) at 37°C in a humidified atmosphere containing 5% CO2. For flow cytometry (protocol described below), CT26 cells (2.5 × 105 cells/well in 2 mL of RPMI-1640 media) were cultured in 12-well culture plates and treated with IFNα or IFNβ (50 ng/mL) in the presence or absence of STAT1 inhibitor (2 μg/mL; Selleckchem, #S1491) 2 hours prior to the cytokine stimulation for 24 hours and then harvested and analyzed for surface H-2Kd and PD-L1 on CT26 cells.

Metastatic colonization model

GFP-transfected CT26 cells (2 × 105) were inoculated intrasplenically (i.s.) or intravenously (i.v.) into Il6+/+ or Il6−/− BALB/c mice through 29-gauge needles (1 mL Terumo Myjector syringe) after mice were anesthetized with xylazine (3 mg/150 μL/mouse; Bayer) through 27-gauge needles (Terumo, #NN-2719S) with isoflurane (Baxter). The injected spleen was returned to the abdomen, and the wound was closed with surgical adhesive (Aron Alpha A; Daiichi Sankyo Pharmaceutical Co., Ltd.). Liver or lung tissues from Il6+/+ or Il6−/− BALB/c mice untreated or treated with all inhibitor and antibody were serially cut and stained by hematoxylin and eosin (H&E). Metastatic surface was calculated as the total surface occupied by metastasis divided by the total area of liver or lung sections, using ImageJ software. Metastatic colonization images of CT26-GFP+ tumors in liver or lung tissues were evaluated using epi-fluorescence on an IVIS Spectrum ex vivo imaging system (Xenogen) at day 14. Livers were removed immediately after mice were euthanized, and collected livers were submerged in PBS on ice. GFP-labeled CT26 cells were fluorescently imaged (540/465 em/ex filters). The conditions were as follows: exposure time = 10 seconds, Lamp level = high, binning = small, and F/Stop = 2. The signal intensity of the tumor burdens was expressed in total radiant efficiency (p/s/cm2/sr)/(μW/cm2). Images were analyzed using Living Image 4.0 software, and regions of interest were drawn around organs by manual contours. Anti-CD8, anti-CD4, anti-IL12, and anti–IFN-AR1 or control IgG (200 μg/mouse; Wako Pure Chemical Industries, Ltd., #143-09523), and clodronate liposomes (100 μL/mouse) or control liposomes (100 μL/mouse) were injected intraperitoneally into Il6+/+ and Il6−/− mice 1 day prior to tumor implantation and day 3 after tumor induction, and then every 4 days thereafter. Anti–PD-L1 (200 μg/mouse) or control IgG (200 μg/mouse) were i.p. injected into CT26 tumor-bearing Il6+/+ and Il6−/− mice at day 5 after tumor induction and then every 4 days thereafter. Anti-IL6 (200 μg/mouse) or control IgG (200 μg/mouse) were i.p. injected into CT26 tumor–bearing Il6+/+ mice 1 day prior to tumor implantation and day 3 after tumor induction, and then every 4 days thereafter. For the experiment reported in Fig. 1, blood samples were drawn from non–tumor-bearing or CT26-inoculated Il6+/+ and Il6−/− mice on day 14. The collected blood samples were centrifuged at 5,000 rpm for 10 minutes at 4°C three times, and the separated sera were stored at −80°C until analysis. Serum samples were also used for the biochemical analysis of alanine transaminase (ALT) and aspartate transaminase (AST). Measurements were performed using an autoanalyzer (JCA-BM6050, Jeol Ltd.) and AST/ALT assay kits: IATRO LQ AST/ALT (J) II (LSI Medience Corporation).

Figure 1.

In vivo reduction of metastatic colonization of colon cancer cells in the livers of Il6−/− mice. GFP-transfected CT26 murine colon cancer cells (2 × 105) were intrasplenically inoculated into wild-type (Il6+/+) and Il6−/− BALB/c mice (day 0). A, Experimental scheme is shown. B, Sera were collected from non–tumor- and CT26-inoculated wild-type and IL6-deficient mice. Serum IL6 concentrations are shown. n = 9–10 mice/group, cumulative results from three independent experiments. C, Serum aspartate transaminase (AST) and alanine transaminase (ALT) concentrations in liver metastases of mice (n = 7 mice/group, cumulative results from two independent experiments). D, Percentages of liver weight per total body weight were calculated (n = 4 mice/group, two independent experiments). E, H&E staining of liver tissues was performed. F, Ratios of tumor area relative to total liver tissue area were calculated by ImageJ software. G, Metastatic colonization of GFP-expressing CT26 cells in liver tissue was evaluated ex vivo using an imaging system at day 14. Representative images of non-tumor liver and GFP-expressing CT26 cell–bearing livers are shown. H, Photon flux ratios were determined from images of liver metastatic colonization of mice (n = 4 mice/group, three independent experiments). I, Survival of wild-type and IL6-deficient mice with liver metastasis monitored for 60 days (n = 11 mice/group, two independent experiments). J, CT26-Il6KO cells (2 × 105) were intrasplenically inoculated into Il6+/+ or Il6−/− BALB/c mice (day 0). H&E staining of liver tissue was performed 21 days after inoculation (two independent experiments). Representative micrographs are shown. K, Ratios of tumor area relative to total liver tissue area were calculated by ImageJ software. E and J, Bars in the images represent 500 μm. Dotted line, tumor areas. F,H, and K, Means and SDs of the data from 4 independent mice are shown. *, P < 0.05 by Dunnett test (B), Student t test (C,D,F,H, and K), and log-rank test (I).

Figure 1.

In vivo reduction of metastatic colonization of colon cancer cells in the livers of Il6−/− mice. GFP-transfected CT26 murine colon cancer cells (2 × 105) were intrasplenically inoculated into wild-type (Il6+/+) and Il6−/− BALB/c mice (day 0). A, Experimental scheme is shown. B, Sera were collected from non–tumor- and CT26-inoculated wild-type and IL6-deficient mice. Serum IL6 concentrations are shown. n = 9–10 mice/group, cumulative results from three independent experiments. C, Serum aspartate transaminase (AST) and alanine transaminase (ALT) concentrations in liver metastases of mice (n = 7 mice/group, cumulative results from two independent experiments). D, Percentages of liver weight per total body weight were calculated (n = 4 mice/group, two independent experiments). E, H&E staining of liver tissues was performed. F, Ratios of tumor area relative to total liver tissue area were calculated by ImageJ software. G, Metastatic colonization of GFP-expressing CT26 cells in liver tissue was evaluated ex vivo using an imaging system at day 14. Representative images of non-tumor liver and GFP-expressing CT26 cell–bearing livers are shown. H, Photon flux ratios were determined from images of liver metastatic colonization of mice (n = 4 mice/group, three independent experiments). I, Survival of wild-type and IL6-deficient mice with liver metastasis monitored for 60 days (n = 11 mice/group, two independent experiments). J, CT26-Il6KO cells (2 × 105) were intrasplenically inoculated into Il6+/+ or Il6−/− BALB/c mice (day 0). H&E staining of liver tissue was performed 21 days after inoculation (two independent experiments). Representative micrographs are shown. K, Ratios of tumor area relative to total liver tissue area were calculated by ImageJ software. E and J, Bars in the images represent 500 μm. Dotted line, tumor areas. F,H, and K, Means and SDs of the data from 4 independent mice are shown. *, P < 0.05 by Dunnett test (B), Student t test (C,D,F,H, and K), and log-rank test (I).

Close modal

After ex vivo IVIS imaging examination, excised liver tissues were finely chopped using scissors, dissociated in collagenase type IV (1 mg/mL; Sigma-Aldrich, Inc., #C5138) and then incubated at 37°C for 30 minutes. Samples were placed in gentleMACS C Tube (Miltenyi, #130-096-334). Tissues were dissociated using the gentleMACS Dissociator (Miltenyi, #130-093-235) for one minute. The dissociated liver samples were passed through a 100-μm cell strainer (Falcon, #352360) into a 50-mL conical. Lymphoid and myeloid cells were enriched by Percoll (GE Healthcare-Life Sciences, #17-0891-01) density centrifugation at 2,000 rpm for 20 minutes at room temperature and collected from the interface between the 33% and 70% Percoll layers. Erythrocytes were osmotically lysed using 1× RBC Lysis Buffer (eBioscience, #00-4333) and purified liver mononuclear cells (MNC) collected were resuspended in proper medium for further process. Population analysis, phosphorylated STAT1 (pSTAT1), and cytotoxicity assays were performed by flow cytometry, as described below.

For the experiment reported in Fig. 1, Supplementary Fig. S1, and Supplementary Fig. S2, we also assessed liver metastatic colonization by intrasplenic inoculation of GFP-transfected 4T1 cells (2 × 105) or CT26-Il6KO and control CT26-mock cells (2 × 105).

Cytotoxicity assay

Target cells CT26 (2.5 × 103 cells/well in 50 μL of RPMI-1640 media) were plated on 96-well V-bottom plates (NUNC, #249570), and the freshly isolated lymphocytes from metastatic livers suspended in 50 μL of RPMI (effectors) were added to a final volume of 250 μL at the ratios of 200:1, 100:1, 50:1, 25:1,12.5:1, 6.25:1, 3.1:1, and 1.5:1 [effector:target (E:T) ratio]. The plates were then incubated for 4 hours in a humidified CO2 chamber at 37°C and centrifuged at 5,000 rpm for 5 minutes at 4°C. The percentage cytotoxicity of antitumor effector cells was calculated by counting GFP+7AAD+ CT26 cells by flow cytometry.

Intracellular cytokine staining

To detect cytoplasmic perforin and granzyme B expression in CD8+ T cells, single-cell suspensions from liver tissue (1 × 106 cells in 250 μL/well in a 12-well culture plate) were stimulated with PMA (25 ng/mL) and A23187 calcium ionophore (1 μg/mL) for 4 hours in the presence of 5 μL per well of brefeldin A (10 μg/mL; Sigma-Aldrich, #B7651). Then, the cells were harvested and stained with anti-CD8 and 7-AAD, and fixed with 4% paraformaldehyde. After permeabilization using the Foxp3/Transcription Factor Staining Buffer set (eBioscience, #00-5523), the fixed cells were stained with anti-granzyme B or perforin. To detect cytoplasmic pSTAT1 in CT26 cells, single-cell suspensions from liver tissue (1 × 106 cells/well in a 12-well culture plate) were harvested, Fc-blocked for 15 minutes with anti-CD16/32, and stained with anti-CD45 and 7-AAD, and fixed with 4% paraformaldehyde. After permeabilization using the Foxp3/Transcription Factor Staining Buffer set (#00-5523), the fixed cells were stained with pSTAT1 mAb (#9167, at 1:200 dilution; Cell Signaling Technology). Samples were acquired by flow cytometry.

Flow cytometry

Population analysis: For surface staining, the cells were Fc-blocked for 15 minutes with anti-CD16/32 and stained with antibodies. The surface expression of CD11c, CD11b, CD8a, CD44, CD62L, I-Ad, H-2Kd, and CD274 were evaluated by FACSCanto II (BD Biosciences), and the results were analyzed with FlowJo software (Tree Star). The mean fluorescence intensity (MFI) ratio [sample ΔMFI (specific marker MFI − isotype control MFI)/control sample ΔMFI × 100] was calculated for samples. A FACSAriaII (BD Biosciences) was used for the isolation of CD11c+ cells from areas of metastatic tissue in the livers of Il6+/+ or Il6−/− mice.

Cytotoxicity analysis: As described above, GFP-labeled CT26 cells (target cells) were cocultured with liver MNCs (effector cells) at the ratios indicated above for 4 hours at 37°C. After the coincubation, dead cells were labeled with 7-AAD. The percentage of CT26 cell death was analyzed by flow cytometry.

pSTAT1 analysis: As described above, cytoplasmic pSTAT1 in CT26 cells was detected.

ELISAs

We determined IL6 in serum obtained as described above from CT26-inoculated Il6+/+ and Il6−/− mice at day 14 and non–tumor-bearing mice using an OptEIA Mouse IL6 ELISA Kit (BD Biosciences, #555240) according to the manufacturer's instructions. Flat-bottom 96-well ELISA plates were coated overnight at 4°C with 80 μL per well of capture antibody. The following morning, plates were washed three times with wash buffer (1× PBS with 0.05% Tween-20), and plates were blocked using 100 μL per well of assay diluent (PBS with 10% FBS) at room temperature for 1 hour and then washed. Standards, serially diluted, and 80 μL of samples were added to the plates. The plates were sealed and incubated for 90 minutes at room temperature. Following 4 washes, 80 μL of working detector (detection antibody and streptavidin–horseradish peroxidase) was added and incubated at room temperature for 45 minutes. Following 4 washing, 80 μL of substrate solution (tetramethylbenzidine and hydrogen peroxide) was added and left at room temperature in the dark. 50 μL of stop solution (2N sulfuric acid) was added, and the plates were read at 450 and 570 nm on a microplate reader (Iwaki Microplate Reader, EZS-ABS, # 58031, Asahi Techno Glass). EZScan-ABS ver1.20.00L for Mac program was used for analysis.

PCR analysis

As described above, 7AADCD45+CD11c+ cells, CT26, and total hepatic cell populations derived from liver metastatic tissues were isolated by sorting using a BD FACSAria. Total RNA was extracted from cells using ISOGEN (Nippon Gene; #311-07361) in accordance with the manufacturer's instructions. RNA concentration was measured by using NanoDrop Spectrophotometer (Thermo Fisher Scientific, #ND-1000). First-strand cDNA was synthesized using 1 μg of total RNA, 1 μL of oligo (dT; 0.5 μg/μL; Invitrogen), and 1 μL of Superscript III reverse transcriptase (Invitrogen) and then amplified by a thermal cycler (GeneAmp PCR System 9700, Applied Biosystems). 0.5 μL of template DNA was used for a final PCR reaction volume of 10 μL. Genes encoding murine Ifna, Ifnb, Il12a, Il12b, Arg1, Il10, Il6, and Actb were amplified and detected using a CFX connect Real-time PCR Detection System (Bio-Rad). The primer sequences and numbers of universal probes used in this study were as follows: Il6 (left: 5′-atcctctggaaccccacac-3′, right: 5′-gaactttcgtactgatcctcgtg-3′, universal probe: #53), Ifna (left: 5′-tcaagccatccttgtgctaa-3′, right: 5′-gtcttttgatgtgaagaggttcaa-3′, universal probe: #3); Ifnb (left: 5′-ctggcttccatcatgaacaa-3′, right: 5′-agagggctgtggtggagaa-3′, universal probe: #18); Il12a (p35) (left: 5′-tcagaatcacaaccatcagca-3′, right: 5′-cgccattatgattcagagactg-3′, universal probe: #49); Il12b (p40) (left: 5′-tgaactggcgttggaagc-3′, right: 5′-gcgggtctggtttgatga-3′, universal probe: #74); Arg1 (left: 5′-cctgaaggaactgaaaggaaag-3′, right: 5′-ttggcagatatgcagggagt-3′, universal probe: #2); Il10 (left: 5′-cagagccacatgctcctaga-3′, right: 5′-tgtccagctggtcctttgtt-3′, universal probe: #41); and Actb (left: 5′-aaggccaaccgtgaaaagat-3′, right: 5′-gtggtacgaccagaggcatac-3′, universal probe: #56). Sample signals were normalized to the reference gene Actb using the ΔΔCt method: ΔCt = ΔCtsampleΔCtreference. Percentages relative to the control sample were then calculated for each sample.

Immunohistochemistry

Metastatic livers or lungs obtained from CT26, 4T1, or CT26-Il6KO and control CT26-mock-inoculated Il6+/+ and Il6−/− mice at day 14 or 19 with and without treatment of all inhibitor and antibody were fixed in 4% paraformaldehyde phosphate buffer solution and then embedded in paraffin. After deparaffinization, antigen retrieval for CD3 and CD11c was performed with a reagent kit (Antigen Retrieval Solution pH9; #415211, Nichirei Bioscience, Inc.) at 95°C for 20 minutes or with 200 μL of proteinase K solution (0.4 mg/mL; #S3004, Dako) for each slide at room temperature for 5 minutes, respectively. Endogenous peroxidase activity was blocked by incubating samples with 0.3% hydrogen peroxide at room temperature for 10 minutes. After washing with Tris-buffered saline, sections were incubated with anti-CD3 (#ab134096, at 1:1,000 dilution, Abcam) or anti-CD11c (#GTX74940, at 1:100 dilution, GeneTex, Inc.) overnight at 4°C. Sections for CD3 and CD11c staining were incubated at room temperature for 30 minutes with Histofine Simple Stain MAX-PO (R; #424144, Nichirei Bioscience, Inc.) or with rabbit anti-hamster IgG (#6215-01, Southern Biotechnology Associates, Inc.), Histofine Simple Stain MAX-PO (R; #424144) at room temperature for 30 minutes, TSA PLUS Biotin Kit (#NEL749A001, PerkinElmer, Inc.) at room temperature for 5 minutes, and VECTASTAIN Elite ABC Reagent (#PK6100, Vector Laboratories, Inc.) at room temperature for 30 minutes. Protein expression was visualized using 3-3′-diaminobezidine-4HCL at room temperature for 5 minutes. The methods for pSTAT1 were performed as with CD3. Anti-phospho-STAT1 (#9167, at 1:300 dilution; Cell Signaling Technology) was used as a primary antibody against pSTAT1. All sections were counterstained with Mayer's hematoxylin.

Tumor specimens from 108 colorectal cancer patients were formalin-fixed and paraffin-embedded, and sections were stained with H&E. After deparaffinization, antigen retrieval for IL6 and PD-L1 was performed with a reagent kit (Citric Acid Buffer Solution pH 6.0; #RM102-C, LSI Medience Corporation) at 95°C for 20 minutes or with pH9 target retrieval solution (#K8004, Dako) at 95°C for 20 minutes, respectively. Endogenous peroxidase activity was blocked by incubating samples with 0.3% hydrogen peroxide at room temperature for 10 minutes. After washing with Tris-buffered saline, the slides were treated with anti-IL6 (#ab6672, at 1:500 dilution; Abcam) and anti–PD-L1 (clone SP142, #M4420, at 1:150 dilution, Spring, Bioscience, Inc.) overnight at 4°C. Sections for IL6 and PD-L1 staining were incubated with Histofine Simple Stain MAX-PO (R; #424144) at room temperature for 30 minutes or with rabbit LINKER (#K8019, Dako) at room temperature for 15 minutes and peroxidase (EnVision/HRP; #K5007, Dako) at room temperature for 20 minutes, respectively. After washing, IL6 and PD-L1 protein expression was visualized using 3-3′-diaminobezidine-4 HCL at room temperature for 5 minutes and 10 minutes, respectively. All assessments were made on viable tumor specimens at ×400 magnification. Each slide was evaluated by a pathologist who was blinded to the clinical outcomes.

Statistical analysis

In vitro experiments were repeated 3 to 5 times. In vivo experiments consisting of 4 to 10 mice per group were independently performed 2 to 3 times. For survival studies, we used 10 to 30 mice per experimental group. Single representative experiments are indicated in the figures. Mean values and SDs were calculated for each data set. Significant differences in the results were determined by a one-way analysis of variance (ANOVA) and Dunnett posttest. In some experiments, the two-tailed Student t test was used to evaluate differences between two groups. P values < 0.05 were considered statistically significant using the two-sided Student t test. The log-rank test was used to determine statistically significant differences in survival curves among CT26-inoculated mice and colorectal cancer patients. The prognostic implications of IL6 and PD-L1 expression and clinicopathologic parameters were analyzed by Cox univariate and multivariate proportional hazard models. Data were analyzed using JMP statistical software for Windows (version 13.1.0; SAS Institute Inc.). No statistical corrections were applied.

IL6 augments metastatic colonization of colon cancer cells

To investigate the effect of IL6 on the metastatic colonization of colon cancer cells in liver tissue, we intrasplenically inoculated GFP-transduced CT26 murine colon cancer cells into wild-type BALB/c (Il6+/+) and IL6-deficient (Il6−/−) mice (Fig. 1A). Serum IL6 was increased in Il6+/+ but not in Il6−/− mice at 14 days after inoculation (Fig. 1B), and serum ALT and AST were higher in Il6+/+ mice than in Il6−/− mice (Fig. 1C). The percentages of liver weight per total body weight of tumor-bearing Il6+/+ mice were significantly heavier than those of Il6−/− mice (Fig. 1D). H&E staining revealed that the metastatic foci of CT26 in Il6+/+ mice were thicker than in Il6−/− mice (Fig. 1E and F). Ex vivo imaging analysis of liver tissue showed that the tumorigenesis of CT26 cells was significantly reduced in Il6−/− mice compared with Il6+/+ mice (Fig. 1G and H). The survival of Il6+/+ mice was significantly shorter compared with Il6−/− mice (Fig. 1I). We also intravenously injected CT26 cells into Il6+/+ and Il6−/− mice to evaluate metastatic colonization of colon cancer cells in the lung. Metastatic colonization in the lungs of Il6−/− mice was significantly reduced compared with Il6+/+ mice (Supplementary Fig. S3A–S3D), and the survival of CT26-bearing Il6−/− mice was significantly prolonged compared with that of Il6+/+ mice (Supplementary Fig. S3E). These data suggested that IL6 promoted the metastatic colonization of colon cancer cells in the lung as well as in liver tissue.

We also intrasplenically inoculated GFP-transduced 4T1 breast cancer cells into Il6+/+ and Il6−/− mice to investigate the effect of IL6 on the metastatic colonization of breast cancer cells in the livers. Metastatic colonization of 4T1 cells in the livers of Il6−/− mice was significantly reduced compared with Il6+/+ mice (Supplementary Fig. S1A–S1D), and the survival of 4T1 intrasplenically injected Il6−/− mice was significantly prolonged compared with Il6+/+ mice (Supplementary Fig. S1E). These data confirmed that IL6 promoted the metastatic colonization in the liver in vivo.

To further confirm the effect of cancer cell–produced IL6 on metastatic colonization, we injected anti-IL6 into our Il6+/+ mouse model. We confirmed that anti-IL6 significantly suppressed the metastatic colonization of CT26 colon cancer cells in the liver (Supplementary Fig. S4A–S4D) and prolonged the survival of CT26-bearing Il6+/+ mice (Supplementary Fig. S4E).

Based on these data, we speculated that both host mouse- and CT26 cancer cell–derived IL6 might be involved in the metastatic colonization of the liver tissue in our experimental model. Therefore, we investigated Il6 gene expression in total hepatic cell population and CT26 cells isolated from the liver tissues of tumor-bearing Il6−/− and Il6+/+ mice. Il6 expression in total hepatic cell population was higher compared with CT26 cells inIl6+/+ mice, but Il6 expression in the total hepatic cell population from Il6−/− mice was not significantly different from that of CT26 cells isolated from Il6−/− mice (Supplementary Fig. S5). Il6 expression in CT26 cells isolated from Il6+/+ mice was higher compared with CT26 cells from Il6−/− mice. These data suggested that host-derived IL6 may augment Il6 expression in CT26 colon cancer cells in vivo.

In this study, we established Il6 gene-knockout CT26 (CT26-Il6KO) cells (Supplementary Fig. S2A), which had a lower proliferative ability in vitro compared with control CT26-mock cells (Supplementary Fig. S2B). When these cells were intrasplenically inoculated into Il6+/+ and Il6−/− BALB/c mice, in Il6+/+ mice, the metastatic colonization of CT26-Il6KO cells in the liver was significantly reduced compared with control cells (Supplementary Fig. S2C and S2D). We then confirmed that the metastatic colonization of CT26-Il6KO cells in the livers of Il6−/− mice was significantly reduced compared with those of Il6+/+ mice (Fig. 1J and K). These findings suggested that host-derived IL6, as well as CT26 cancer cells, facilitates the metastatic colonization of colon cancer cells in vivo.

Maturation of DCs and cytotoxicity of CD8+ T cells are enhanced in Il6−/− mice

Next, we examined the immune status of mice with CT26 metastatic colonization. CD11c+ and CD3+ cells were accumulated in the tumorous regions of liver tissue (Fig. 2A). I-AdhighCD11c+ mature DCs and CD44highCD62L effector memory CD8+ T cells infiltrated into the CT26-bearing liver tissues of Il6−/− mice to a greater degree than that in Il6+/+ mice (Fig. 2B). Il12a, Il12b, Ifna, and Ifnb expression in CD11c+ cells isolated from the liver metastases of Il6−/− mice were higher than those of Il6+/+ mice, whereas Arg1 and Il10 expression in Il6+/+ mice was higher than that in Il6−/− mice (Fig. 2C). The cytotoxicity of immune cells collected from the liver tissue of Il6−/− mice was higher than that of Il6+/+ mice (Fig. 2D). We then confirmed that both perforin and granzyme B in infiltrating CD8+ T cells from Il6−/− mice were significantly higher than those of Il6+/+ mice (Fig. 2E). These data suggested that the functions of antitumor effector cells, including mature CD11c+ DCs and CD8+ cytotoxic T cells, were augmented in the CT26-bearing liver tissue of Il6−/− mice.

Figure 2.

Accumulation of mature DCs and effector memory CD8+ T cells in liver metastases of Il6−/− mice. GFP-transfected CT26 murine colon cancer cells (2 × 105) were intrasplenically inoculated into Il6+/+ or Il6−/− mice (day 0). A, Immunohistochemistry staining of liver tissue was performed to evaluate CD11c and CD3 levels at day 14. Representative micrographs are shown (the zoomed areas are 9 × magnification; n = 2 mice/group). Bars, 200 μm. B, MNCs were collected from liver tissues of CT26 tumor–bearing Il6+/+ or Il6−/− mice at day 14. Surface expression of CD11c, I-Ad, CD44, CD62L, and CD8 on 7-AADCD45+ cells was evaluated by flow cytometry. Representative histograms for CD11chighI-Adhigh DCs and CD44+CD62LCD8+ T cells are shown. Percentages of CD11chighI-Adhigh cells and CD44+CD62LCD8+ cells relative to total CD45+ cells were calculated. Means and SDs are indicated (n = 3–4 mice/group, three independent experiments). *, P < 0.05 by Student t test. C, Relative expression of Il12a, Il12b, Ifna, Ifnb, Arg1, and Il10 relative to Actb in CD11c+ DCs from liver tissues at day 14 of CT26 tumor–bearing Il6+/+ or Il6−/− mice. Means and SDs are indicated (n = 3–10 mice/group, two–three independent experiments). *, P < 0.05 by Student t test. D, Percentage cytotoxicity of immune cells obtained from tumor-bearing liver tissues at day 14 against CT26 cells at the indicated E:T ratios was evaluated by flow cytometry. Means and SDs shown (n = 4 mice/group, three independent experiments). *, P < 0.05 versus Il6+/+ by Student t test. E, Perforin and granzyme B in CD8+ T cells from liver tissues were determined by flow cytometry at day 14. Change in (Δ)MFIs of perforin and granzyme B relative to each isotype control is shown. Means and SDs are shown (n = 5–7 mice/group, cumulative results from three independent experiments). *, P < 0.05 by Student t test.

Figure 2.

Accumulation of mature DCs and effector memory CD8+ T cells in liver metastases of Il6−/− mice. GFP-transfected CT26 murine colon cancer cells (2 × 105) were intrasplenically inoculated into Il6+/+ or Il6−/− mice (day 0). A, Immunohistochemistry staining of liver tissue was performed to evaluate CD11c and CD3 levels at day 14. Representative micrographs are shown (the zoomed areas are 9 × magnification; n = 2 mice/group). Bars, 200 μm. B, MNCs were collected from liver tissues of CT26 tumor–bearing Il6+/+ or Il6−/− mice at day 14. Surface expression of CD11c, I-Ad, CD44, CD62L, and CD8 on 7-AADCD45+ cells was evaluated by flow cytometry. Representative histograms for CD11chighI-Adhigh DCs and CD44+CD62LCD8+ T cells are shown. Percentages of CD11chighI-Adhigh cells and CD44+CD62LCD8+ cells relative to total CD45+ cells were calculated. Means and SDs are indicated (n = 3–4 mice/group, three independent experiments). *, P < 0.05 by Student t test. C, Relative expression of Il12a, Il12b, Ifna, Ifnb, Arg1, and Il10 relative to Actb in CD11c+ DCs from liver tissues at day 14 of CT26 tumor–bearing Il6+/+ or Il6−/− mice. Means and SDs are indicated (n = 3–10 mice/group, two–three independent experiments). *, P < 0.05 by Student t test. D, Percentage cytotoxicity of immune cells obtained from tumor-bearing liver tissues at day 14 against CT26 cells at the indicated E:T ratios was evaluated by flow cytometry. Means and SDs shown (n = 4 mice/group, three independent experiments). *, P < 0.05 versus Il6+/+ by Student t test. E, Perforin and granzyme B in CD8+ T cells from liver tissues were determined by flow cytometry at day 14. Change in (Δ)MFIs of perforin and granzyme B relative to each isotype control is shown. Means and SDs are shown (n = 5–7 mice/group, cumulative results from three independent experiments). *, P < 0.05 by Student t test.

Close modal

IL12-mediated activation of cytotoxic T cells reduces metastases in Il6−/− mice

We then investigated the antitumor effector cells in the CT26 model. In vivo injection of clodronate liposomes into Il6−/− mice to deplete phagocytes, including DCs and macrophages, significantly enhanced the metastatic colonization of CT26 tumor cells in the liver (Fig. 3A–C), including in the livers of Il6+/+ mice (Supplementary Fig. S6A–S6C). However, the difference of metastatic colonization between mice injected with control and clodronate liposomes was much more pronounced in the Il6−/− setting. Depletion of CD8+ T cells with anti-CD8 significantly increased the metastatic colonization of CT26 cells in the livers of Il6−/− mice (Fig. 3D and F), whereas the depletion did not significantly increase liver colonization of Il6+/+ mice (Supplementary Fig. S6D–S6F). In contrast, depletion of CD4+ T cells with anti-CD4 reduced CT26 liver metastasis in both Il6+/+ and Il6−/− mice (Supplementary Fig. S7A–S7F). In vivo injection of anti-CD4 into Il6+/+ mice also significantly enhanced the infiltration of I-AdhighCD11c+ mature DCs, CD44highCD62L effector memory CD8+ T cells, and perforin+ CD8+ T cells into liver metastases (Supplementary Fig. S7G–S7H). These results suggested that CD8+ T cells were required for the suppression of metastatic colonization of the livers of Il6−/− mice.

Figure 3.

IL12 and CD8+ effector T cells reduce metastatic colonization in the livers of Il6−/− mice. GFP-transfected CT26 cells (2 × 105) were intrasplenically inoculated into control liposome–, clodronate liposome–, control IgG–, anti-CD8–, or anti-IL12–treated Il6−/− mice (day 0). A,D, and G, H&E staining of liver tissue was performed 14 days after inoculation. Representative micrographs are shown (n = 2 mice/group). Bars, 500 μm. Dotted line, tumor areas. B,E, and H, Metastatic colonization of GFP-expressing CT26 cells in the liver were evaluated ex vivo by an imaging system at day 14. Representative images are shown. C,F, and I, Photon flux ratios were evaluated from the images (n = 4–6 mice/group, three independent experiments). Means and SDs are shown. *, P < 0.05 by Student t test. J, Immunohistochemistry was performed to evaluate CD11c and CD3 levels at day 14. Representative micrographs are shown (the zoomed areas are 6 × magnification; n = 2 mice/group). Bars, 200 μm. K, MNCs were collected from liver tissues of control IgG– or anti-IL12–treated CT26 tumor–bearing wild-type or Il6−/− mice at day 14. Surface expression of CD11c, I-Ad, CD44, CD62L, and CD8 on 7-AADCD45+ cells was evaluated by flow cytometry. Representative histograms for CD11chighI-Adhigh DCs and CD44+CD62LCD8+ T cells are shown. Percentages of CD11chighI-Adhigh cells and CD44+CD62LCD8+ cells relative to total CD45+ cells was calculated. Means and SDs are shown (n = 4 mice/group, two independent experiments). *, P < 0.05 by Student t test. L, Percentage cytotoxicity of immune cells obtained from liver tissues at day 14 against CT26 cells at the indicated E:T ratios was evaluated by flow cytometry. Means and SDs shown (n = 3 mice/group, two independent experiments). *, P < 0.05 versus corresponding control IgG group by Student t test. M, Perforin and granzyme B in CD8+ T cells from the liver were determined by flow cytometry at day 14. Change in (Δ)MFIs against each isotype control was calculated. Means and SDs shown (n = 3 mice/group, two independent experiments). *, P < 0.05 by Student t test.

Figure 3.

IL12 and CD8+ effector T cells reduce metastatic colonization in the livers of Il6−/− mice. GFP-transfected CT26 cells (2 × 105) were intrasplenically inoculated into control liposome–, clodronate liposome–, control IgG–, anti-CD8–, or anti-IL12–treated Il6−/− mice (day 0). A,D, and G, H&E staining of liver tissue was performed 14 days after inoculation. Representative micrographs are shown (n = 2 mice/group). Bars, 500 μm. Dotted line, tumor areas. B,E, and H, Metastatic colonization of GFP-expressing CT26 cells in the liver were evaluated ex vivo by an imaging system at day 14. Representative images are shown. C,F, and I, Photon flux ratios were evaluated from the images (n = 4–6 mice/group, three independent experiments). Means and SDs are shown. *, P < 0.05 by Student t test. J, Immunohistochemistry was performed to evaluate CD11c and CD3 levels at day 14. Representative micrographs are shown (the zoomed areas are 6 × magnification; n = 2 mice/group). Bars, 200 μm. K, MNCs were collected from liver tissues of control IgG– or anti-IL12–treated CT26 tumor–bearing wild-type or Il6−/− mice at day 14. Surface expression of CD11c, I-Ad, CD44, CD62L, and CD8 on 7-AADCD45+ cells was evaluated by flow cytometry. Representative histograms for CD11chighI-Adhigh DCs and CD44+CD62LCD8+ T cells are shown. Percentages of CD11chighI-Adhigh cells and CD44+CD62LCD8+ cells relative to total CD45+ cells was calculated. Means and SDs are shown (n = 4 mice/group, two independent experiments). *, P < 0.05 by Student t test. L, Percentage cytotoxicity of immune cells obtained from liver tissues at day 14 against CT26 cells at the indicated E:T ratios was evaluated by flow cytometry. Means and SDs shown (n = 3 mice/group, two independent experiments). *, P < 0.05 versus corresponding control IgG group by Student t test. M, Perforin and granzyme B in CD8+ T cells from the liver were determined by flow cytometry at day 14. Change in (Δ)MFIs against each isotype control was calculated. Means and SDs shown (n = 3 mice/group, two independent experiments). *, P < 0.05 by Student t test.

Close modal

We also found that IL6-deficient conditions augmented both Il12a and Il12b expression in CD11c+ cells in liver metastatic regions (Fig. 2C). Previous reports indicate that IL12 production by mature DCs induces antitumor effector cells through cell–cell interactions in tumor-bearing hosts (19, 20). To investigate the effect of IL12 production on metastatic colonization in the livers of Il6−/− mice, we administered anti-IL12 and observed significantly enhanced CT26 metastatic colonization in the liver (Fig. 3G–I), whereas this depletion did not enhance liver metastases in Il6+/+ mice (Supplementary Fig. S6D–S6F). We also noted the accumulation of CD11c+ and CD3+ cells in liver metastases (Fig. 3J) and that the infiltration of I-AdhighCD11c+ mature DCs and CD44highCD62L effector memory CD8+ T cells was significantly reduced by the administration of anti-IL12 to Il6−/− mice (Fig. 3K). Anti-IL12 significantly reduced the cytotoxicity of immune cells, as well as perforin and granzyme B in CD8+ T cells, in liver metastases compared with Il6−/− mice injected with control antibody (Fig. 3L and M). These results indicated that augmented IL12 production in the liver tissue of Il6−/− mice was related to the enhanced antitumor effector function of CD8+ T cells to prevent metastatic colonization in tumor-bearing hosts.

Augmented type I IFN signaling upregulates MHC class I and PD-L1 in Il6−/− mice

Ifna and Ifnb expression in CD11c+ DCs was enhanced in the metastasized livers of Il6−/− mice (Fig. 2C). To confirm the involvement of type I IFN production in the reduction of metastatic colonization, we intraperitoneally injected an IFN-AR1 antagonistic mAb into Il6−/− mice. The blockade of type I IFN signaling with anti–IFN-AR1 elevated CT26 metastatic colonization in the livers of Il6−/− mice (Fig. 4A–C). We found that the surface level of H-2Kd, an MHC class I molecule, on CT26 colon cancer cells was significantly increased after stimulation with IFNα and IFNβ in a STAT1-dependent manner in vitro (Fig. 4D). Previous studies report that stimulation with IFNγ upregulates surface PD-L1 and MHC class I on cancer cells (21–23). In this study, we found that stimulation with IFNα and IFNβ augmented surface expression of PD-L1 on CT26 colon cancer cells (Fig. 4E). The upregulation of PD-L1 expression was partially blocked by a STAT1 inhibitor, and immunohistochemistry (IHC) revealed that STAT1 was activated in the tumor sites of Il6−/− mice (Fig. 4F). Flow cytometry showed that pSTAT1 in CT26 liver metastases of Il6−/− mice were significantly higher than in Il6+/+ mice (Fig. 4G). We also observed that surface PD-L1 and MHC class I in CT26 liver metastases of Il6−/− mice were significantly higher compared with Il6+/+ mice (Fig. 4H). These findings suggested that type I IFN signaling was upregulated in the liver tissue of Il6−/− mice to activate antitumor immunity, which also resulted in the induction of PD-L1 and MHC class I expression on colon cancer cells in tumor-bearing mice.

Figure 4.

Upregulation of IFNα/β-STAT1 signaling and induction of PD-L1 expression on colon cancer cells in tumor-bearing Il6−/− mice. GFP-transfected CT26 cells were intrasplenically inoculated into control IgG– or anti–IFN-AR1–treated Il6−/− mice (day 0). A, H&E staining of the liver was performed at 14 days after inoculation. Representative micrographs are indicated (n = 2 mice/group). Bars, 500 μm. Dotted line, tumor areas. B, Metastatic colonization of GFP-expressing CT26 cells in the liver was evaluated ex vivo by an imaging system at day 14. Representative images are shown. C, Photon influx ratios were calculated from the images (n = 4 mice/group, three independent experiments). *, P < 0.05 by Student t test. D and E, CT26 cells were stimulated with IFNα and IFNβ in the presence or absence of a STAT1 inhibitor for 24 hours in vitro. Surface H-2Kd and PD-L1 on CT26 cells were evaluated by flow cytometry. Representative histograms are shown. Change in (Δ)MFIs against isotype controls was calculated (n = 3 mice/group, three independent experiments). *, P < 0.05 by Dunnett test. F, IHC staining of the liver from wild-type and Il6−/− mice was performed at day 14 to evaluate pSTAT1. Representative micrographs are shown (the zoomed areas are 9 × magnification). Bars, 200 μm. G, pSTAT1 expression on GFP+CD45 CT26 cells was determined by flow cytometry at day 14 for wild-type and Il6−/− mice. ΔMFIs of pSTAT1 expression against each isotype control were calculated (n = 3–5 mice/group, two independent experiments). *, P < 0.05 by Student t test. H, Surface H-2Kd and PD-L1 on GFP+CD45 CT26 cells obtained from livers of wild-type and Il6−/− mice at day 14 were evaluated by flow cytometry. Representative data are shown. ΔMFIs were calculated (n = 3–4 mice/group, three independent experiments). *, P < 0.05 by Student t test. C–E,G, and H, Means and SDs are shown.

Figure 4.

Upregulation of IFNα/β-STAT1 signaling and induction of PD-L1 expression on colon cancer cells in tumor-bearing Il6−/− mice. GFP-transfected CT26 cells were intrasplenically inoculated into control IgG– or anti–IFN-AR1–treated Il6−/− mice (day 0). A, H&E staining of the liver was performed at 14 days after inoculation. Representative micrographs are indicated (n = 2 mice/group). Bars, 500 μm. Dotted line, tumor areas. B, Metastatic colonization of GFP-expressing CT26 cells in the liver was evaluated ex vivo by an imaging system at day 14. Representative images are shown. C, Photon influx ratios were calculated from the images (n = 4 mice/group, three independent experiments). *, P < 0.05 by Student t test. D and E, CT26 cells were stimulated with IFNα and IFNβ in the presence or absence of a STAT1 inhibitor for 24 hours in vitro. Surface H-2Kd and PD-L1 on CT26 cells were evaluated by flow cytometry. Representative histograms are shown. Change in (Δ)MFIs against isotype controls was calculated (n = 3 mice/group, three independent experiments). *, P < 0.05 by Dunnett test. F, IHC staining of the liver from wild-type and Il6−/− mice was performed at day 14 to evaluate pSTAT1. Representative micrographs are shown (the zoomed areas are 9 × magnification). Bars, 200 μm. G, pSTAT1 expression on GFP+CD45 CT26 cells was determined by flow cytometry at day 14 for wild-type and Il6−/− mice. ΔMFIs of pSTAT1 expression against each isotype control were calculated (n = 3–5 mice/group, two independent experiments). *, P < 0.05 by Student t test. H, Surface H-2Kd and PD-L1 on GFP+CD45 CT26 cells obtained from livers of wild-type and Il6−/− mice at day 14 were evaluated by flow cytometry. Representative data are shown. ΔMFIs were calculated (n = 3–4 mice/group, three independent experiments). *, P < 0.05 by Student t test. C–E,G, and H, Means and SDs are shown.

Close modal

Anti–PD-L1 augments antitumor immunity in the livers of Il6−/− mice

To examine the effect of PD-L1 induction on liver metastatic CT26 cells in Il6−/− mice, we injected anti–PD-L1 into our mouse model. Anti–PD-L1 significantly reduced the metastatic colonization of CT26 cells in the liver of Il6−/− mice 19 days after inoculation (Fig. 5A–C). We further observed the accumulation of CD11c+ and CD3+ cells in metastatic regions (Fig. 5D), and the infiltration of I-AdhighCD11c+ mature DCs and CD44highCD62L effector memory CD8+ T cells was significantly increased with anti–PD-L1 in Il6−/− mice (Fig. 5E). Treatment with anti–PD-L1 also enhanced the cytotoxicity of immune cells, as well as the expression of perforin and granzyme B in CD8+ T cells infiltrating liver metastases compared with control antibody (Fig. 5F and G). The survival of CT26-bearing Il6−/− mice was significantly prolonged with anti–PD-L1 treatment compared with control IgG-treated mice (Fig. 5H). In contrast, we found that PD-L1 blockade did not reduce the metastatic colonization of CT26 cells in the livers of Il6+/+ mice (Supplementary Fig. S8A–S8F). These findings suggested that a lack of IL6 under tumor-bearing conditions augmented antitumor immunity and facilitated immune-checkpoint inhibition therapy using anti–PD-L1 to prevent the liver metastasis (Fig. 5I).

Figure 5.

Augmentation of the antitumor effect of PD-L1 blockade on metastatic colonization in the livers of Il6−/− mice. GFP-transfected CT26 cells were intrasplenically inoculated into Il6−/− mice (day 0). Control IgG or anti–PD-L1 was injected intraperitoneally into Il6−/− mice at day 5 and then every 4 days thereafter. A, H&E staining of liver tissue was performed 19 days after inoculation. Representative micrographs are indicated (n = 2 mice/group). Bars, 500 μm. Dotted line, tumor areas. B, Metastatic colonization of GFP-expressing CT26 cells in the liver at day 19 was evaluated ex vivo with an imaging system. Representative images are shown. C, Photon flux ratios were evaluated from the images (n = 3–4 mice/group, two independent experiments). *, P < 0.05 by Student t test. D, IHC staining of liver tissue was performed for CD11c and CD3 for anti–PD-L1–treated and control mice at day 19. Representative micrographs are shown (the zoomed areas are 6 × magnification). Bars, 200 μm. E, MNCs were collected from liver tissues of control IgG– or anti–PD-L1–treated CT26 tumor–bearing Il6−/− mice at day 19. Surface expression of CD11c, I-Ad, CD44, CD62L, and CD8 on 7-AADCD45+ cells was evaluated by flow cytometry. Representative histograms for CD11chighI-Adhigh DCs and CD44+CD62LCD8+ T cells are shown. Percentages of CD11chighI-Adhigh cells and CD44+CD62LCD8+ cells relative to total CD45+ cells were calculated. Means and SDs indicated (n = 4 mice/group, two independent experiments). *, P < 0.05 by Student t test. F, Percentage cytotoxicity of immune cells from the liver at day 19 of anti–PD-L1–treated and control mice against CT26 cells was evaluated by flow cytometry (n = 3 mice/group, two independent experiments). *, P < 0.05 versus control IgG group by Student t test. G, Perforin and granzyme B in CD8+ T cells from the liver of anti–PD-L1–treated and control mice were determined by flow cytometry at day 19. Change in (Δ)MFIs was calculated (n = 3 mice/group, two independent experiments). *, P < 0.05 by Student t test. H, Survival of anti–PD-L1–treated and control mice (n = 24 mice/group, two independent experiments) monitored for 40 days. *, P value determined by the log-rank test. I, Proposed model of the present study. Inhibitory and stimulatory actions are shown as blunted lines and arrows, respectively. CTLs, cytotoxic T lymphocytes. C and E–G, Mean and SDs are shown.

Figure 5.

Augmentation of the antitumor effect of PD-L1 blockade on metastatic colonization in the livers of Il6−/− mice. GFP-transfected CT26 cells were intrasplenically inoculated into Il6−/− mice (day 0). Control IgG or anti–PD-L1 was injected intraperitoneally into Il6−/− mice at day 5 and then every 4 days thereafter. A, H&E staining of liver tissue was performed 19 days after inoculation. Representative micrographs are indicated (n = 2 mice/group). Bars, 500 μm. Dotted line, tumor areas. B, Metastatic colonization of GFP-expressing CT26 cells in the liver at day 19 was evaluated ex vivo with an imaging system. Representative images are shown. C, Photon flux ratios were evaluated from the images (n = 3–4 mice/group, two independent experiments). *, P < 0.05 by Student t test. D, IHC staining of liver tissue was performed for CD11c and CD3 for anti–PD-L1–treated and control mice at day 19. Representative micrographs are shown (the zoomed areas are 6 × magnification). Bars, 200 μm. E, MNCs were collected from liver tissues of control IgG– or anti–PD-L1–treated CT26 tumor–bearing Il6−/− mice at day 19. Surface expression of CD11c, I-Ad, CD44, CD62L, and CD8 on 7-AADCD45+ cells was evaluated by flow cytometry. Representative histograms for CD11chighI-Adhigh DCs and CD44+CD62LCD8+ T cells are shown. Percentages of CD11chighI-Adhigh cells and CD44+CD62LCD8+ cells relative to total CD45+ cells were calculated. Means and SDs indicated (n = 4 mice/group, two independent experiments). *, P < 0.05 by Student t test. F, Percentage cytotoxicity of immune cells from the liver at day 19 of anti–PD-L1–treated and control mice against CT26 cells was evaluated by flow cytometry (n = 3 mice/group, two independent experiments). *, P < 0.05 versus control IgG group by Student t test. G, Perforin and granzyme B in CD8+ T cells from the liver of anti–PD-L1–treated and control mice were determined by flow cytometry at day 19. Change in (Δ)MFIs was calculated (n = 3 mice/group, two independent experiments). *, P < 0.05 by Student t test. H, Survival of anti–PD-L1–treated and control mice (n = 24 mice/group, two independent experiments) monitored for 40 days. *, P value determined by the log-rank test. I, Proposed model of the present study. Inhibitory and stimulatory actions are shown as blunted lines and arrows, respectively. CTLs, cytotoxic T lymphocytes. C and E–G, Mean and SDs are shown.

Close modal

IL6 expression in primary colorectal cancers correlates with disease-free survival

We performed IHC staining to assess IL6 and PD-L1 expression in the tumor tissues of colorectal cancer patients (Supplementary Table S1). As described in the Materials and Methods, 108 cases of colorectal cancer were stratified according to negative and positive expression of IL6 and PD-L1 (Fig. 6A). Statistical analysis revealed that disease-free survival (DFS) was significantly prolonged in IL6-negative patients (n = 55) compared with IL6-positive patients (n = 53; P = 0.006; Fig. 6B; Supplementary Table S2). However, PD-L1–positive patients (n = 71) had unfavorable DFS (P = 0.051) compared with PD-L1–negative patients (n = 37; Fig. 6B; Supplementary Table S2). IL6 positivity in both PD-L1–negative (n = 39) and PD-L1–positive patients (n = 14) was associated with significantly unfavorable DFS compared with IL6-negativity in both PD-L1–negative (n = 32, P = 0.041) and PD-L1–positive patients (n = 23, P = 0.009; Fig. 6C). To evaluate IL6 expression as an independent prognostic factor, we performed a univariate analysis of the 108 colorectal cancer patients using the Cox proportional hazards model. We found that pathologic T stage, N stage, lymphatic and venous invasion, CEA, CA19-9, and IL6 were significantly correlated with recurrence. Multivariate analysis confirmed IL6 was a significant predictor of DFS (relative risk 1.958; 95% confidence interval, 1.004–3.999, P = 0.049; Supplementary Table S2). IHC of IL6 and PD-L1 in liver metastases of colorectal cancer patients showed that 57 colorectal cancer patients could be stratified by the negative and positive expression of IL6 and PD-L1 (Supplementary Fig. S9A). However, DFS did not significantly differ between IL6-negative (n = 22) and IL6-positive (n = 35) patients (P = 0.929; Supplementary Fig. S9B). DFS was not significantly prolonged in PD-L1–negative patients (n = 31) compared with PD-L1–positive patients (n = 26; P = 0.446; Supplementary Fig. S9B). These data suggest that IL6 expression in the primary tumor sites of colorectal cancer patients was associated with the recurrence of colorectal cancer, including liver metastasis.

Figure 6.

Correlation between IL6 expression in primary tumor tissues and DFS of PD-L1–negative and PD-L1–positive colorectal cancer patients. IHC of primary tumor tissues of colorectal cancer patients (n = 108) was conducted to evaluate IL6 and PD-L1 expression using anti-IL6 and anti–PD-L1, respectively. A, Representative micrographs (the zoomed areas are 9 × magnification) of tumor tissues from IL6-positive and -negative and PD-L1–positive and –negative patients. Bars in the images represent 100 μm. B, Kaplan–Meier estimates of DFS for 108 colorectal cancer patients stratified into two groups: IL6-negative (n = 55) or IL6-positive (n = 53) patients and PD-L1–negative (n = 71) or PD-L1–positive (n = 37) patients. C, Kaplan–Meier estimates of DFS for 71 PD-L1–negative colorectal cancer patients stratified into two groups: IL6-negative (n = 32) or IL6-positive (n = 39) patients. Kaplan–Meier estimates of DFS for 37 PD-L1–positive colorectal cancer patients stratified into two groups: IL6-negative (n = 23) or IL6-positive (n = 14) patients. CI, confidence interval.

Figure 6.

Correlation between IL6 expression in primary tumor tissues and DFS of PD-L1–negative and PD-L1–positive colorectal cancer patients. IHC of primary tumor tissues of colorectal cancer patients (n = 108) was conducted to evaluate IL6 and PD-L1 expression using anti-IL6 and anti–PD-L1, respectively. A, Representative micrographs (the zoomed areas are 9 × magnification) of tumor tissues from IL6-positive and -negative and PD-L1–positive and –negative patients. Bars in the images represent 100 μm. B, Kaplan–Meier estimates of DFS for 108 colorectal cancer patients stratified into two groups: IL6-negative (n = 55) or IL6-positive (n = 53) patients and PD-L1–negative (n = 71) or PD-L1–positive (n = 37) patients. C, Kaplan–Meier estimates of DFS for 71 PD-L1–negative colorectal cancer patients stratified into two groups: IL6-negative (n = 32) or IL6-positive (n = 39) patients. Kaplan–Meier estimates of DFS for 37 PD-L1–positive colorectal cancer patients stratified into two groups: IL6-negative (n = 23) or IL6-positive (n = 14) patients. CI, confidence interval.

Close modal

Colorectal cancer, a common cancer worldwide, has a high recurrence rate after surgery that remains to be resolved (13–15). Because liver metastasis is a major cause of colorectal cancer–associated deaths (24), elucidation of the precise mechanisms involved is required for design of more effective treatments for patients with colorectal cancer. Our results indicated that IL6 modulates the immune status of the tumor microenvironment, which facilitated metastatic colonization of colon cancer cells in the liver by causing dysfunction of antitumor effector cells.

It is reported that chronic inflammation is closely related to tumorigenesis and cancer progression (25–27). Previous studies reveal that high serum IL6, a major proinflammatory cytokine, is associated with poor prognosis in metastatic colon cancer patients (28) and treatment refractory carcinomas (29, 30). IL6 can augment growth and epithelial–mesenchymal transition, metastatic spread (31–32), tumor angiogenesis, and the renewal and drug resistance of some cancer stem cells (33–35). In the current study, we revealed that IL6 produced in a tumor-bearing host promoted the metastatic colonization of colon cancer cells by suppressing antitumor immunity. From these findings, we speculate that IL6 has a critical role in antitumor immunity by controlling the metastasis of cancer cells in vivo.

In our metastatic colonization model, we found that depletion of CD4+ T cells enhanced antitumor effects in Il6+/+ mice. However, the efficacy was limited in Il6−/− mice. A previous paper reveals that CD4+ T cells in lymph nodes and tumor-infiltrating CD4+ and CD8+ T cells preferentially produce IFNγ in CT26 intradermally injected Il6−/− mice compared with Il6+/+ mice (23). It is well known that IFNγ suppresses the differentiation and function of regulatory T cells (Treg; ref. 36). Together with the data in this study, we speculate that immunosuppressive Tregs may not function in Il6−/− mice, whereas Tregs were present in Il6+/+ mice in our metastatic colonization model.

We found that Il12 expression in CD11c+ DCs, as well as the killer function of CD8+ T cells, was augmented in IL6-deficient mice in our liver metastatic colonization model. IL12 is a critical cytokine that mediates antitumor effects by activating both NK cells (37) and cytotoxic CD8+ T lymphocytes (38). Tumor gene therapy with Il12 is effective for colorectal liver metastasis treatment in an experimental mouse model (39). In our model using IL6-deficient mice, the neutralization of IL12 significantly enhanced the metastatic colonization of colon cancer cells and suppressed the killer function of tumor-infiltrating CD8+ T cells in the liver. Previous studies demonstrate that IL6 suppresses the maturation of DCs, resulting in a reduction in antigen presentation that is normally required for the induction of antigen-specific T cells (8, 12). Therefore, we speculate that IL6 production in tumor-bearing hosts suppresses IL12 production by DCs, which is essential for the activation of antitumor effector T cells, and that this promotes the subsequent metastatic colonization of colon cancer cells in the liver.

In this study, we found that Ifna and Ifnb expression in CD11c+ cells and the surface expression of MHC class I of colon cancer cells were enhanced in the liver tissues of our metastatic colonization model using Il6−/− mice. A previous study using mouse models indicates that colorectal cancer colonization in the liver is suppressed by the local delivery of Ifna (40). In our study, blockade of IFN-AR1–mediated type I IFN signaling led to an increase in metastatic colonization in the livers of Il6−/− mice. Therefore, we hypothesize that IL6 produced in tumor-bearing hosts suppresses the production of type I IFNs by host cells, including DCs, and reduces MHC class I expression on cancer cells. This, in turn, promotes the metastatic colonization of colon cancer cells by reducing the recognition of target cancer cells by antitumor effector T cells.

Previous studies report that stimulation with IFNγ upregulates surface PD-L1 and MHC class I on cancer cells (21–23). In our experiments, in vitro stimulation with IFNα and IFNβ increased surface PD-L1 expression, as well as MHC class I, on CT26 colon cancer cells. We confirmed that a lack of IL6 in tumor-bearing hosts enhanced IFN-AR1–mediated type I IFN signaling and augmented PD-L1 expression on CT26 colon cancer cells, whereas metastatic colonization was significantly reduced compared with IL6-sufficent conditions. Our previous study indicates that antitumor effector cells, including CD8+ T cells, accumulate at higher rates in the tumor tissue of IL6-deficient mice compared with wild-type mice (23). From these data, we speculated that administration of anti–PD-L1 might be more effective at inhibiting the metastatic colonization of colon cancer cells in the livers of Il6−/− mice, where we observed the upregulation of PD-L1 expression in an IFNα/β–IFN-AR1–dependent manner. As expected, metastatic colonization was significantly inhibited by anti–PD-L1 treatment in Il6−/− mice but not Il6+/+ mice, whereby a lack of IL6 led to the activation of antitumor effector T cells that prolonged the survival of mice.

When choosing the optimum therapy for cancer patients, it is important to evaluate their immune status (41). In our study using clinical specimens from colorectal cancer patients, low IL6 in primary tumor tissues (PD-L1–negative and –positive tumors) were significantly associated with improved DFS versus those with high IL6, suggesting that such cases have a beneficial antitumor immune status. A study has reported that high PD-L1 expression on tumor cells negatively affects the survival of colorectal cancer patients (42). Other reports indicate that high PD-L1 expression in colorectal carcinoma (43–45) and other cancer types (46–48) is associated with significantly improved survival. Therefore, the relationship between PD-L1 expression and the survival of colorectal cancer patients remains controversial (49). Based on previous reports and our findings in the present study, we speculate that IL6 present in primary tissues might be a potential marker to evaluate the immune status of both PD-L1–negative and –positive colorectal cancer patients, which may provide useful information for determining treatment strategies, such as using immune-checkpoint inhibitors targeting PD-1/PD-L1, and for the prognosis of recurrent tumors including liver metastasis.

In terms of therapeutic effects, a previous report demonstrates that the targeted inhibition of IL6 enhances the efficacy of anti–PD-L1 in murine models of pancreatic cancer (50). A clinical study using an anti-IL6R, tocilizumab, with carboplatin/doxorubicin and IFN-α2b was conducted in patients with recurrent epithelial ovarian cancer (51). They report that myeloid cells in IL6R mAb-treated patients produce more IL12 and that T cells are activated and secrete large amounts of effector cytokines, indicating a beneficial change in the antitumor immune status. The results of our metastatic colonization model with IL6-deficient mice are consistent with these findings. However, a limitation to our metastasis model is that it does not fully reflect the multistep process of metastasis because this is an experimental colonization assay that could skip the process of cancer cell detachment from the primary site and intravasation into vessels.

In conclusion, IL6 from tumor-bearing hosts suppressed the effective induction of antitumor immunity and promoted the colonization of metastasis in cancerous liver lesions. Lack of IL6 augments the effector functions of DCs and cytotoxic T cells that infiltrate tumor microenvironments. From these results, we speculate that the IL6 signaling cascade might be a target to regulate the liver metastasis of colon cancer cells in colorectal cancer patients through the augmentation of antitumor effector cells.

No potential conflicts of interest were disclosed.

Conception and design: H. Kitamura, Y. Ohno, A. Taketomi

Development of methodology: Y. Toyoshima, H. Kitamura, H. Xiang, Y. Ohno

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): Y. Toyoshima, H. Kitamura, H. Xiang, S. Homma, H. Kawamura, T. Kamiyama

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): Y. Toyoshima, H. Kitamura, S. Homma, M. Tanino

Writing, review, and/or revision of the manuscript: Y. Toyoshima, H. Kitamura, N. Takahashi

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): Y. Toyoshima, H. Kitamura, H. Xiang, Y. Ohno, S. Homma

Study supervision: H. Kitamura

The authors thank Dr. K. Sugiyama, S. Kii, Dr. N. Okada, Dr. N. Ichikawa, and Dr. T. Yoshida for their excellent technical assistance and thoughtful advice on this study. This work was partially supported by Grants-in-Aid for Scientific Research (C; 25460584 to H. Kitamura, and 16K10526, to N. Takahashi) from the Japan Society for the Promotion of Science (JSPS), the Platform Project for Supporting Drug Discovery and Life Science Research (Platform for Drug Discovery, Informatics, and Structural Life Science) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan (to H. Kitamura), the Japan Agency for Medical Research and Development (AMED; to A. Taketomi), and the Joint Research Program of the Institute for Genetic Medicine, Hokkaido University (to M. Tanino and A. Taketomi). The authors thank H. Nikki March, PhD, from Edanz Group (www.edanzediting.com/ac) for editing a draft of this manuscript.

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.

1.
Larkin
J
,
Chiarion-Sileni
V
,
Gonzalez
R
,
Grob
JJ
,
Cowey
CL
,
Lao
CD
, et al
Combined nivolumab and ipilimumab or monotherapy in untreated melanoma
.
N Engl J Med
2015
;
373
:
23
34
.
2.
Topalian
SL
,
Drake
CG
,
Pardoll
DM
. 
Immune checkpoint blockade: a common denominator approach to cancer therapy
.
Cancer Cell
2015
;
27
:
450
61
.
3.
Binnewies
M
,
Roberts
EW
,
Kersten
K
,
Chan
V
,
Fearon
DF
,
Merad
M
, et al
Understanding the tumor immune microenvironment (TIME) for effective therapy
.
Nat Med
2018
;
24
:
541
50
.
4.
Spranger
S
,
Gajewski
TF.
Impact of oncogenic pathways on evasion of antitumour immune responses
.
Nat Rev Cancer
2018
;
18
:
139
47
.
5.
Hirano
T
,
Yasukawa
K
,
Harada
H
,
Taga
T
,
Watanabe
Y
,
Matsuda
T
, et al
Complementary DNA for a novel human interleukin (BSF-2) that induces B lymphocytes to produce immunoglobulin
.
Nature
1986
;
324
:
73
6
.
6.
Tanaka
T
,
Kishimoto
T.
The biology and medical implications of interleukin-6
.
Cancer Immunol Res
2014
;
2
:
288
94
.
7.
Yu
H
,
Pardoll
D
,
Jove
R
. 
STATs in cancer inflammation and immunity: a leading role for STAT3
.
Nat Rev Cancer
2009
;
9
:
798
809
.
8.
Park
SJ
,
Nakagawa
T
,
Kitamura
H
,
Atsumi
T
,
Kamon
H
,
Sawa
S
, et al
IL-6 regulates in vivo dendritic cell differentiation through STAT3 activation
.
J Immunol
2004
;
173
:
3844
54
.
9.
Kitamura
H
,
Kamon
H
,
Sawa
S
,
Park
SJ
,
Katunuma
N
,
Ishihara
K
, et al
IL-6-STAT3 controls intracellular MHC class II alphabeta dimer level through cathepsin S activity in dendritic cells
.
Immunity
2005
;
23
:
491
502
.
10.
Sumida
K
,
Wakita
D
,
Narita
Y
,
Masuko
K
,
Terada
S
,
Watanabe
K
, et al
Anti-IL-6 receptor mAb eliminates myeloid-derived suppressor cells and inhibits tumor growth by enhancing T-cell responses
.
Eur J Immunol
2012
;
42
:
2060
72
.
11.
Narita
Y
,
Kitamura
H
,
Wakita
D
,
Sumida
K
,
Masuko
K
,
Terada
S
, et al
The key role of IL-6-arginase cascade for inducing dendritic cell-dependent CD4(+) T cell dysfunction in tumor-bearing mice
.
J Immunol
2013
;
190
:
812
20
.
12.
Ohno
Y
,
Kitamura
H
,
Takahashi
N
,
Ohtake
J
,
Kaneumi
S
,
Sumida
K
, et al
IL-6 down-regulates HLA class II expression and IL-12 production of human dendritic cells to impair activation of antigen-specific CD4(+) T cells
.
Cancer Immunol Immunother
2016
;
65
:
193
204
.
13.
Siegel
RL
,
Miller
KD
,
Jemal
A
. 
Cancer statistics, 2019
.
CA Cancer J Clin
2019
;
69
:
7
34
.
14.
Ferlay
J
,
Soerjomataram
I
,
Dikshit
R
,
Eser
S
,
Mathers
C
,
Rebelo
M
, et al
Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012
.
Int J Cancer
2015
;
136
:
E359
86
.
15.
Arnold
M
,
Sierra
MS
,
Laversanne
M
,
Soerjomataram
I
,
Jemal
A
,
Bray
F
. 
Global patterns and trends in colorectal cancer incidence and mortality
.
Gut
2017
;
66
:
683
91
.
16.
Hugen
N
,
van de Velde
CJ
,
de Wilt
JH
,
Nagtegaal
ID
. 
Metastatic pattern in colorectal cancer is strongly influenced by histological subtype
.
Ann Oncol
2014
;
25
:
651
7
.
17.
Lambert
AW
,
Pattabiraman
DR
,
Weinberg
RA
. 
Emerging biological principles of metastasis
.
Cell
2017
;
168
:
670
91
.
18.
Kitamura
T
,
Qian
BZ
,
Pollard
JW
. 
Immune cell promotion of metastasis
.
Nat Rev Immunol
2015
;
15
:
73
86
.
19.
Kitamura
H
,
Iwakabe
K
,
Yahata
T
,
Nishimura
S
,
Ohta
A
,
Ohmi
Y
, et al
The natural killer T (NKT) cell ligand alpha-galactosylceramide demonstrates its immunopotentiating effect by inducing interleukin (IL)-12 production by dendritic cells and IL-12 receptor expression on NKT cells
.
J Exp Med
1999
;
189
:
1121
8
.
20.
Fujii
S
,
Shimizu
K
,
Smith
C
,
Bonifaz
L
,
Steinman
RM
. 
Activation of natural killer T cells by alpha-galactosylceramide rapidly induces the full maturation of dendritic cells in vivo and thereby acts as an adjuvant for combined CD4 and CD8 T cell immunity to a coadministered protein
.
J Exp Med
2003
;
198
:
267
79
.
21.
Mandai
M
,
Hamanishi
J
,
Abiko
K
,
Matsumura
N
,
Baba
T
,
Konishi
I
. 
Dual Faces of IFNgamma in cancer progression: a role of PD-L1 induction in the determination of pro- and antitumor immunity
.
Clin Cancer Res
2016
;
22
:
2329
34
.
22.
Abiko
K
,
Matsumura
N
,
Hamanishi
J
,
Horikawa
N
,
Murakami
R
,
Yamaguchi
K
, et al
IFN-gamma from lymphocytes induces PD-L1 expression and promotes progression of ovarian cancer
.
Br J Cancer
2015
;
112
:
1501
9
.
23.
Ohno
Y
,
Toyoshima
Y
,
Yurino
H
,
Monma
N
,
Xiang
H
,
Sumida
K
, et al
Lack of interleukin-6 in the tumor microenvironment augments type-1 immunity and increases the efficacy of cancer immunotherapy
.
Cancer Sci
2017
;
108
:
1959
66
.
24.
Carpizo
DR
,
D'Angelica
M.
Liver resection for metastatic colorectal cancer in the presence of extrahepatic disease
.
Lancet Oncol
2009
;
10
:
801
9
.
25.
Mantovani
A
,
Allavena
P
,
Sica
A
,
Balkwill
F
. 
Cancer-related inflammation
.
Nature
2008
;
454
:
436
44
.
26.
Grivennikov
S
,
Karin
E
,
Terzic
J
,
Mucida
D
,
Yu
GY
,
Vallabhapurapu
S
, et al
IL-6 and Stat3 are required for survival of intestinal epithelial cells and development of colitis-associated cancer
.
Cancer Cell
2009
;
15
:
103
13
.
27.
Crusz
SM
,
Balkwill
FR
. 
Inflammation and cancer: advances and new agents
.
Nat Rev Clin Oncol
2015
;
12
:
584
96
.
28.
Thomsen
M
,
Kersten
C
,
Sorbye
H
,
Skovlund
E
,
Glimelius
B
,
Pfeiffer
P
, et al
Interleukin-6 and C-reactive protein as prognostic biomarkers in metastatic colorectal cancer
.
Oncotarget
2016
;
7
:
75013
22
.
29.
Mitsunaga
S
,
Ikeda
M
,
Shimizu
S
,
Ohno
I
,
Furuse
J
,
Inagaki
M
, et al
Serum levels of IL-6 and IL-1beta can predict the efficacy of gemcitabine in patients with advanced pancreatic cancer
.
Br J Cancer
2013
;
108
:
2063
9
.
30.
Makuuchi
Y
,
Honda
K
,
Osaka
Y
,
Kato
K
,
Kojima
T
,
Daiko
H
, et al
Soluble interleukin-6 receptor is a serum biomarker for the response of esophageal carcinoma to neoadjuvant chemoradiotherapy
.
Cancer Sci
2013
;
104
:
1045
51
.
31.
Lee
SO
,
Yang
X
,
Duan
S
,
Tsai
Y
,
Strojny
LR
,
Keng
P
, et al
IL-6 promotes growth and epithelial-mesenchymal transition of CD133+ cells of non-small cell lung cancer
.
Oncotarget
2016
;
7
:
6626
38
.
32.
Albino
D
,
Civenni
G
,
Rossi
S
,
Mitra
A
,
Catapano
CV
,
Carbone
GM
. 
The ETS factor ESE3/EHF represses IL-6 preventing STAT3 activation and expansion of the prostate cancer stem-like compartment
.
Oncotarget
2016
;
7
:
76756
68
.
33.
Gopinathan
G
,
Milagre
C
,
Pearce
OM
,
Reynolds
LE
,
Hodivala-Dilke
K
,
Leinster
DA
, et al
Interleukin-6 stimulates defective angiogenesis
.
Cancer Res
2015
;
75
:
3098
107
.
34.
Nagasaki
T
,
Hara
M
,
Nakanishi
H
,
Takahashi
H
,
Sato
M
,
Takeyama
H
. 
Interleukin-6 released by colon cancer-associated fibroblasts is critical for tumour angiogenesis: anti-interleukin-6 receptor antibody suppressed angiogenesis and inhibited tumour-stroma interaction
.
Br J Cancer
2014
;
110
:
469
78
.
35.
Gilbert
LA
,
Hemann
MT.
DNA damage-mediated induction of a chemoresistant niche
.
Cell
2010
;
143
:
355
66
.
36.
Overacre-Delgoffe
AE
,
Chikina
M
,
Dadey
RE
,
Yano
H
,
Brunazzi
EA
,
Shayan
G
, et al
Interferon-γ drives Treg fragility to promote anti-tumor immunity
.
Cell
2017
;
169
:
1130
41
.
37.
Uemura
A
,
Takehara
T
,
Miyagi
T
,
Suzuki
T
,
Tatsumi
T
,
Ohkawa
K
, et al
Natural killer cell is a major producer of interferon gamma that is critical for the IL-12-induced anti-tumor effect in mice
.
Cancer Immunol Immunother
2010
;
59
:
453
63
.
38.
Rubinstein
MP
,
Su
EW
,
Suriano
S
,
Cloud
CA
,
Andrijauskaite
K
,
Kesarwani
P
, et al
Interleukin-12 enhances the function and anti-tumor activity in murine and human CD8(+) T cells
.
Cancer Immunol Immunother
2015
;
64
:
539
49
.
39.
Martinet
O
,
Ermekova
V
,
Qiao
JQ
,
Sauter
B
,
Mandeli
J
,
Chen
L
, et al
Immunomodulatory gene therapy with interleukin 12 and 4–1BB ligand: long- term remission of liver metastases in a mouse model
.
J Natl Cancer Inst
2000
;
92
:
931
6
.
40.
Catarinella
M
,
Monestiroli
A
,
Escobar
G
,
Fiocchi
A
,
Tran
NL
,
Aiolfi
R
, et al
IFNalpha gene/cell therapy curbs colorectal cancer colonization of the liver by acting on the hepatic microenvironment
.
EMBO Mol Med
2016
;
8
:
155
70
.
41.
Chen
DS
,
Mellman
I.
Elements of cancer immunity and the cancer-immune set point
.
Nature
2017
;
541
:
321
30
.
42.
Koganemaru
S
,
Inoshita
N
,
Miura
Y
,
Miyama
Y
,
Fukui
Y
,
Ozaki
Y
, et al
Prognostic value of programmed death-ligand 1 expression in patients with stage III colorectal cancer
.
Cancer Sci
2017
;
108
:
853
8
.
43.
Droeser
RA
,
Hirt
C
,
Viehl
CT
,
Frey
DM
,
Nebiker
C
,
Huber
X
, et al
Clinical impact of programmed cell death ligand 1 expression in colorectal cancer
.
Eur J Cancer
2013
;
49
:
2233
42
.
44.
Dunne
PD
,
McArt
DG
,
O'Reilly
PG
,
Coleman
HG
,
Allen
WL
,
Loughrey
M
, et al
Immune-derived PD-L1 gene expression defines a subgroup of stage II/III colorectal cancer patients with favorable prognosis who may be harmed by adjuvant chemotherapy
.
Cancer Immunol Res
2016
;
4
:
582
91
.
45.
Li
Y
,
Liang
L
,
Dai
W
,
Cai
G
,
Xu
Y
,
Li
X
, et al
Prognostic impact of programed cell death-1 (PD-1) and PD-ligand 1 (PD-L1) expression in cancer cells and tumor infiltrating lymphocytes in colorectal cancer
.
Mol Cancer
2016
;
15
:
55
.
46.
Taube
JM
,
Anders
RA
,
Young
GD
,
Xu
H
,
Sharma
R
,
McMiller
TL
, et al
Colocalization of inflammatory response with B7-h1 expression in human melanocytic lesions supports an adaptive resistance mechanism of immune escape
.
Sci Transl Med
2012
;
4
:
127ra37
.
47.
Darb-Esfahani
S
,
Kunze
CA
,
Kulbe
H
,
Sehouli
J
,
Wienert
S
,
Lindner
J
, et al
Prognostic impact of programmed cell death-1 (PD-1) and PD-ligand 1 (PD-L1) expression in cancer cells and tumor-infiltrating lymphocytes in ovarian high grade serous carcinoma
.
Oncotarget
2016
;
7
:
1486
99
.
48.
Schalper
KA
,
Velcheti
V
,
Carvajal
D
,
Wimberly
H
,
Brown
J
,
Pusztai
L
, et al
In situ tumor PD-L1 mRNA expression is associated with increased TILs and better outcome in breast carcinomas
.
Clin Cancer Res
2014
;
20
:
2773
82
.
49.
Herbst
RS
,
Soria
JC
,
Kowanetz
M
,
Fine
GD
,
Hamid
O
,
Gordon
MS
, et al
Predictive correlates of response to the anti–PD-L1 antibody MPDL3280A in cancer patients
.
Nature
2014
;
515
:
563
7
.
50.
Mace
TA
,
Shakya
R
,
Pitarresi
JR
,
Swanson
B
,
McQuinn
CW
,
Loftus
S
, et al
IL-6 and PD-L1 antibody blockade combination therapy reduces tumour progression in murine models of pancreatic cancer
.
Gut
2018
;
67
:
320
32
.
51.
Dijkgraaf
EM
,
Santegoets
SJ
,
Reyners
AK
,
Goedemans
R
,
Wouters
MC
,
Kenter
GG
, et al
A phase I trial combining carboplatin/doxorubicin with tocilizumab, an anti-IL-6R monoclonal antibody, and interferon-alpha2b in patients with recurrent epithelial ovarian cancer
.
Ann Oncol
2015
;
26
:
2141
9
.