Soluble fibrinogen-like protein 2 promotes the growth of hepatocellular carcinoma via attenuating dendritic cell-mediated cytotoxic T cell activity
Soluble fibrinogen-like protein 2 (sFGL2), a secretory protein expressed by regulatory T cells (Tregs) with immunosuppressive activity, is highly expressed in both the peripheral blood and tumor tissue of patients with hepatocellular carcinoma (HCC); however, sFGL2 function in HCC remains largely unknown. Here, we elucidated the potential role of sFGL2 in HCC progression.
T cells, dendritic cells (DCs), and related cytokines in the tumor microenvironment were comparatively analyzed in BALB/c and C57BL/6 mice bearing transplanted hepatomas harboring Fgl2-knockout or receiving sFGL2-antibody treatment. Additionally, the effects of sFGL2 on DCs and T cells were evaluated in vivo and ex vivo.
The growth of both subcutaneously and orthotopically transplanted hepatomas was inhibited in Fgl2-knockout mice and those treated with the sFGL2 antibody, respectively, as compared with controls. Moreover, sFGL2 depletion enhanced the proportion and cytotoxicity of cytotoxic T cells, promoted DC maturation, and improved DC activity to proliferate T cells in the tumor microenvironment. Furthermore, we detected lower levels of interleukin (IL)-35 in both types of transplanted hepatomas and higher level of IL-6 in orthotopically transplanted hepatomas following sFGL2 depletion. Mechanistically, we found that sFGL2 impaired bone-marrow-derived DC (BMDCs) function by inhibiting phosphorylation of Akt, nuclear factor-kappaB, cAMP response element binding protein, and p38 and downregulating the expression of major histocompatibility complex II, CD40, CD80, CD86, and CD83 on BMDCs in vitro.
Our data suggest that sFGL2 promotes hepatoma growth by attenuating DC activity and subsequent CD8+ T cell cytotoxicity, suggesting sFGL2 as a novel potential therapeutic target for HCC treatment.
KeywordsSoluble fibrinogen-like protein 2 Hepatocellular carcinoma Tumor microenvironment Dendritic cells Immunosuppression
Bone marrow-derived dendritic cells
Cytometric beads array
Carboxyfluorescein succinimidyl amino ester
CAMP response element binding protein
Cytotoxic T-lymphocyte-associated antigen 4
Cytotoxic T lymphocytes
Draining lymph nodes
Enzyme linked immunosorbent assay
Extracellular signal-regulated kinase
Fibrinogen-like protein 2
Forkhead box P3
Granulocyte macrophage-colony stimulating factor
Integrated optical density
Magnetic cell isolation and cell separation
Mitogen-activated protein kinase
Myeloid-derived suppressor cells
Major histocompatibility complex II
Mechanistic target of rapamycin
Programmed cell death 1
Programmed cell death-ligand 1
Sodium dodecyl sulfate polyacrylamide gel electrophoresis
Soluble fibrinogen-like protein 2
Tris-buffered saline with Tween-20
Transforming growth factor
Type 1 T helper cells
Type 2 T helper cells
Tumor necrosis factor
Regulatory T cells
Hepatocellular carcinoma (HCC), the second leading cause of cancerous mortality, is associated with hepatitis B or C virus infection, which otherwise induces immune tolerance [1, 2]. Resistance to HCC treatment is widely attributed to immune-regulatory cells, such as regulatory T cells (Tregs), myeloid-derived suppressor cells (MDSCs), and tumor-associated macrophages (TAMs) in the tumor microenvironment . Interleukin (IL)-10 and transforming growth factor (TGF)-β, immunosuppressive cytokines secreted mainly by Tregs, are commonly detected in the serum of HCC patients in many cases, with both cytokines capable of inhibiting immune surveillance and protecting tumor growth by attenuating T cell activation [4, 5]. Additionally, immune checkpoint proteins, including cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4), programmed cell death 1 (PD-1), and programmed cell death-ligand 1 (PD-L1), are expressed on Tregs, TAMs, MDSCs, and hepatoma cells and inhibit activation of effector immune cells, such as cytotoxic T cells (CTLs) and natural killer (NK) cells, in tumor tissue . Although attempts have been made to treat HCC with immune-checkpoint inhibitors, efficiency varies in individuals, which limits their use [7, 8]. Therefore, it is important to elucidate an expanded spectrum of immune regulators in the tumor microenvironment in order to identify potential therapeutic targets.
Fibrinogen-like protein 2 (FGL2)/fibroleukin is a member of the fibrinogen-related protein superfamily and comprises both membrane and soluble subtypes . Soluble FGL2 (sFGL2) is an immunosuppressive factor that inhibits dendritic cells (DCs)  by binding to the FcγRIIB receptor . As an immune regulator, sFGL2 plays a critical role in the immune balance in autoimmune diseases and can restrict the progression of autoimmune glomerulonephritis  and T cell-induced colitis . However, in viral hepatitis, sFGL2 attenuates antiviral immunity, leading to poor prognosis, with impaired Treg function and prolonged survival time observed following FGL2 blockage in murine models of viral fulminant hepatitis . Another study showed that sFGL2 depletion inhibits glioma growth and decreases the numbers of MDSCs, alternatively activated macrophages (M2 macrophages), and CD39+ Tregs . Additionally, sFGL2 promotes the accumulation of MDSCs via C-X-C motif chemokine ligand 12 and increases the number of activated cancer-associated fibroblasts in a murine model of lung cancer . In an HCC study, levels of serum sFGL2 were reportedly higher in HCC patients , and hepatic stellate cells were found to secrete sFGL2 and inhibit the proliferation of CD8+ T cells, thereby hindering antitumor immunity . However, the function of sFGL2 in HCC remains largely unknown.
In this study, we established transplanted hepatoma models and investigated the role of sFGL2 in HCC growth. Our data showed that sFGL2 blockage by antibody interference or genetic deletion decreased hepatoma burden by enhancing DC activation and increasing CTL number and cytotoxicity in tumor tissue while having minimal effect on MDSC and M2 macrophage numbers. Mechanistically, we found that sFGL2 hindered the expression of major histocompatibility complex II (MHCII), CD40, CD80, CD86, and CD83 and attenuated the phosphorylation of Akt, nuclear factor-kappaB (NF-κB), cAMP response element binding protein (CREB), and p38 on bone marrow-derived DCs (BMDCs) in vitro, which influenced DC activation.
Female wide-type (WT) BALB/c and C57BL/6 mice (aged 6–8 weeks) were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China). Syngeneic Fgl2−/− mice were generated by the Beijing Genomics Institute (Beijing, China). All mice were kept in micro-isolator cages, and the experimental protocols were approved by the Animal Ethics Committee.
BNL 1ME A.7R.1 cells (BNL cells; 8 × 106) and Hepa1–6 cells (8 × 106) were respectively transplanted into the left flank of BALB/c and C57BL/6 mice to create subcutaneous HCC models. BNL cells are a liver epithelial cell line from BALB/c mice that exhibit malignant properties. Hepa1–6 cells are derived from hepatomas from BW7756 mice and arise in C57BL/6 mice. Tumor volumes were measured every 2 days, and 100 μg of the FGL2 antibody or isotype were injected intratumorally twice weekly once the tumor volume reached > 100 mm3. In a separate experiment, the same number of liver cancer cells was inoculated into the left flank of WT and syngeneic Fgl2−/− mice. Additionally, to explore the effect of IL-35 on the hepatoma environment, 100 μg IL-35 antibody or isotype (Abcam, Cambridge, UK) were injected intra-tumorally once weekly when the tumor volume reached > 100 mm3. To establish an orthotopically transplanted HCC model, 1 × 106 BNL cells were implanted into the left lateral liver lobes of WT and syngeneic Fgl2−/− BALB/c mice.
Anti-FGL2 polyclonal antibody
For treatment of hepatoma-burdened mice, we used a rabbit polyclonal antibody against a partial form of FGL2 (amino acids 338–356) that included the fibrinogen-related domain critical for immunosuppressive function . The preparation, purification, and identification of the antibody were completed by the Proteintech Group Inc. (Rosemont, IL, USA), which also provided the isotype IgG antibody.
BNL and Hepa1–6 cells were cultured in Dulbecco’s minimum essential medium containing 10% (v/v) fetal bovine serum (Gibco, Gaithersburg, MD, USA) and 100 μg/mL penicillin/streptomycin (Invitrogen, Carlsbad, CA, USA). In the logarithmic growth phase, cells were harvested and implanted into mice subcutaneously or orthotopically.
Magnetic cell isolation and cell separation (MACS; Miltenyi Biotec, Bergisch Gladbach, Germany) was used to isolate CD4+CD25− and CD8+ T cells from tumor tissue of untreated WT mice. Subsequently, these cells were mixed with DCs from tumors derived from different groups at various proportions. T cells were dyed with 5 μM carboxyfluorescein succinimidyl amino ester (CFSE) prior to 3-day culture in the presence of 200 IU/mL murine IL-2 (PeproTech, Rocky Hill, NJ, USA) in 96-well round-bottomed plates for 72 h. The proliferated T cells were detected according to the percentage of CFSE dilution. For T helper (Th) cell-differentiation analysis, CD4+CD25− T cells were mixed with DCs from tumors at a 5:1 ratio and cultured in the presence of 200 IU/mL murine IL-2 in 96-well round-bottomed plates for 6 days. Th1 and Th2 cells and Tregs were measured and respectively characterized as interferon (IFN)-γ+, IL-4+, and CD25+forkhead box P3+ (Foxp3+) cells among CD4+ T cells.
To analyze CD8+ T cell cytotoxicity in tumors, 104 ultraviolet inactivated BNL cells were mixed with 105 CD8+ T cells for 72 h. Subsequently, the CD8+ T cells were mixed with 104 BNL cells in the logarithmic growth phase for 12 h in order to detect BNL cell apoptosis by 7-aminoactinomycin D (7-AAD) staining.
To obtain BMDCs, bone-marrow cells were separated from mouse tibias, and immature BMDCs were harvested after 6 days of culture with 10 ng/mL IL-4 and 10 ng/mL granulocyte macrophage-colony stimulating factor (GM-CSF; PeproTech).
IL-12p70, IL-35, and TGF-β in the homogenate were detected using enzyme-linked immunosorbent assay (ELISA) kits (Biolegend, San Diego, CA, USA) at an absorbance of 540 nm. Cytometric bead array (CBA; BD Biosciences, San Diego, CA, USA) was used to measure IL-4, IL-6, IL-10, TNF-α, and IFN-γ levels.
Flow cytometric analysis
Cell phenotype was assessed by flow cytometry (BD LSRFortessa; BD Biosciences) after incubation with the following fluorescein-labeled antibodies: CD45-allophycocyanin (APC)-eFluor780, CD3e-fluorescein isothiocyanate (FITC), CD4-APC, CD8a-phycoerythrin (PE)-cy7, CD25-PE-cy7, Foxp3-PE, IFN-γ-AlexaFluor488, IL-4-PE-cy7, CD107a-PE, granzymeB-peridinin chlorophyll (PerCP)-eFluor710, perforin-FITC, CD11c-PE-cy7, CD80-PerCP-eFluor710, CD83-FITC, B7-H4-PE, and CD31-PE. All antibodies were purchased from eBioscience (San Diego, CA, USA), except CD4-APC, CD8a-PE-cy7, and Foxp3-PE (BD Biosciences). Intracellular antigens were determined after incubation with ionomycin (500 ng/mL; Abcam, Cambridge, UK) and phorbol-12-myristate13-acetate (10 ng/mL; Sigma-Aldrich, St. Louis, MO, USA) for 1 h and monensin (2 μM; eBioscience) for an additional 4 h. Fixation and permeabilization were performed prior to antibody incubation, and data were analyzed using FlowJo software (TreeStar, Ashland, OR, USA).
Analysis of DC surface markers and cell-signaling pathways in vitro
BMDCs were incubated with 4 μg/mL recombinant murine FGL2 (R&D Systems, Minneapolis, MN, USA) for 16 h and stimulated with 500 ng/mL lipopolysaccharide (LPS; Sigma-Aldrich) for 12 h. BMDCs were then harvested, and the expression of MHCII, CD40, CD80, CD86, and CD83 was assessed by flow cytometry. To analyze cell-signaling pathways, BMDCs were treated with recombinant murine FGL2 (0, 0.5, 1, or 2 μg/mL) for 16 h, followed by 500 ng/mL LPS stimulation for 30 min. Relevant proteins in cells were extracted and their concentrations measured using a standard bicinchoninic acid assay (Pierce Biotechnology, Rockford, IL, USA). After sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transfer to polyvinylidene fluoride membranes for western blot, 5% not-fat milk was used for blocking for 1 h at room temperature. Primary antibodies were incubated with membranes at 4 °C overnight. Following several wash steps with Tris-buffered saline with Tween-20 (TBST), secondary antibodies were incubated for 1 h at room temperature. Proteins were detected using enhanced chemiluminescence reagent (Servicebio, Wuhan, China), and the integrated optical density (IOD) of the proteins was calculated using a Gel-pro analyzer (Media Cybernetics, Rockville, MD, USA). Antibodies against Akt, phosphorylated (p)-Akt, mechanistic target of rapamycin (mTOR), p-mTOR, NF-κB-p65, p-NF-κB-p65, CREB, p-CREB, p38, p-p38, Erk1/2, and p-Erk1/2 were obtained from Cell Signaling Technology (Danvers, MA, USA), and antibodies for glyceraldehyde 3-phosphate dehydrogenase and histone-3 were purchased from Servicebio.
Data were expressed as mean ± SEM unless otherwise specified. Significance between groups of absolute values was determined by Student’s t test, and the Mann–Whitney U test was used to analyze percentages. The log-rank test was used to compare survival rates between groups. A P < 0.05 was considered significant.
FGL2-antibody treatment or Fgl2 knockout inhibits tumor growth in subcutaneously transplanted HCC models
sFGL2 blockage reduces IL-35 levels, elevates the number of tumor-infiltrated CD8+ T cells with enhanced cytotoxicity, and promotes DC maturation in subcutaneously transplanted HCC models
Previous studies demonstrated that sFGL2 hinders DC maturation and function in vitro . To explore whether sFGL2 hampers DCs during HCC progression, we analyzed DCs from the tumors and DLNs of mice subcutaneously transplanted with HCC (Fig. 2f). The proportion of CD31+ DCs decreased significantly in the tumor, whereas the number of MHCII+ DCs increased in DLNs from Fgl2−/− mice. Moreover, analysis of the number of MDSCs and M2 macrophages in tumors and DLNs revealed similar levels between Fgl2−/− and WT mice (Additional file 1: Figure S2a).
We then evaluated the numbers of T cells and DCs in Hepa1–6-derived tumors in C57BL/6 mice, finding a larger number of CD3+ T cells in tumors (Additional file 1: Figure S3a) and Th1 cells in tumors and DLNs (Additional file 1: Figure S3b) in Fgl2−/− mice. Additionally, analysis of CD8+ T cells revealed that Fgl2 knockout elevated the number of tumor-infiltrated CD8+ T cells and their production of granzyme B in DLNs (Additional file 1: Figure S3c). Moreover, we observed elevated expression of CD83 on DCs in Fgl2−/− mice. Furthermore, analysis of the immunological features of T cells and DCs in tumors from BALB/c (Additional file 1: Figure S4) and C57BL/6 (Additional file 1: Figure S5) mice following anti-FGL2 treatment showed similar results. These data suggested that sFGL2 impaired antitumor immunocytes in the hepatoma microenvironment.
Our results showed IL-35 levels was much lower in tumor from Fgl2−/− mice. To further explore the influence of diminished IL-35, 8 × 106 BNL cells were s.c. injected into BALB/c mice, then 100 μg IL-35 antibody were injected intra-tumorally once weekly when the tumor volume reached > 100 mm3. Immunocytes were examined 14 days after tumor inoculation and we found significantly more Th1 cells and fewer Tregs in tumor from anti-IL-35 treatment group compared with control (Fig. 2g).
sFGL2 blockage promotes DC stimulation of T cells in the tumors of subcutaneously transplanted HCC models
Fgl2 knockout suppresses the growth of orthotopically transplanted hepatoma and activates tumor-infiltrated CD8 lymphocytes and DCs
sFGL2 inhibits the expression of MHCII, CD40, CD80, CD86, and CD83 and phosphorylation of Akt and p38 in BMDCs in vitro
Blockage of immunosuppressive factors is currently regarded as a primary strategy for cancer immune therapy. Tregs, MDSCs, and TAMs are immunosuppressive cells that abate antitumor immunity via the release of immune-regulatory cytokines and presentation of surface molecules, particularly through immune-checkpoint proteins, such as PD-1 and CTLA-4 . Antibodies targeting immune-checkpoint proteins have been demonstrated as effective at controlling the progression of tumor growth in many Phase III clinical trials ; however, clinical efficacy is influenced by tumor type, individual response, targets, and drug administration. Moreover, PD-1 and PD-L1 antibodies exhibit limited clinical efficacy in HCC therapy , making it necessary to identify new targets.
The immunosuppressive activity of sFGL2 might represent a potentially important mediator of tumor growth. Previous studies report reductions in tumor growth following Fgl2 knockout in glioma and lung cancer models [15, 16]. In the present study, we reported for the first time that sFGL2 promoted tumor-microenvironment remodeling and HCC progression, and that hepatoma growth was inhibited following blockage of sFGL2 in both subcutaneously and orthotopically transplanted murine models of hepatoma. In glioma and lung tumor models, levels of MDSCs and M2 macrophages decreased in tumors in Fgl2-knockout groups, suggesting an sFGL2-mediated mechanism promoting tumor growth through increases in these two subsets [15, 16]. However, in the present study, we found that sFGL2 blockage did not alter the numbers of MDSC and M2 macrophages in HCC models, indicating that sFGL2 might accelerate HCC progression through other pathways.
sFGL2 hinders DC maturation of T cell proliferation in vitro . Therefore, we hypothesized that sFGL2 might promote hepatoma growth by attenuating the number of DCs and T cells, as the main effectors of antitumor immunity. Our results revealed changes in the number, phenotype, and function of DCs and T cells according to sFGL2 status, and that sFGL2 hindered the phosphorylation of mTOR, CREB, Akt, and p38 in BMDCs in vitro. Furthermore, sFGL2 reduced phosphorylated levels of the NF-κB p65 subunit, which subsequently influenced the expression of inflammatory cytokines and surface molecules in DCs. The effect is mediated by sFGL2 binding to FcγRIIB  which downregulates the activation of the factors . However, phosphorylation of these factors was unaltered in T cells (Additional file 1: Figure S7), indicating that sFGL2 might not directly modulate T cells. Because changes in Akt and p38 signaling can influence DC maturation and activation [30, 31], we analyzed related surface markers on BMDCs following sFGL2 treatment in vitro, finding that BMDC phenotype was altered by sFGL2 according to decreased levels of MHCII, CD40, CD80, CD86, and CD83 expression. Moreover, in hepatoma-transplanted murine models, sFGL2 blockage promoted MHCII and CD83 expression and downregulated the expression of CD31, which is reportedly an immune-inhibitory marker on the DC surface . These results indicated that sFGL2 blockade promoted DC maturation and inhibited tolerogenic molecule expression on DCs. As the upstream event of adaptive immunity, DC activation can lead to enhanced T cell activation. Our results showed that DCs from the tumor microenvironment in models undergoing sFGL2 blockage upregulated T cell proliferation to a greater degree than that observed in controls. Additionally, DCs from mice treated with the sFGL2 antibody induced higher levels of Th1 differentiation from CD4+CD25− T cells. These data confirmed a role for sFGL2 in suppressing DC maturation, resulting in inhibited T cell function.
An increased number of infiltrated CD8+ T cells in the HCC microenvironment produced higher levels of perforin and granzyme B following sFGL2 blockade, with this effect augmented by an elevated number of Th1 cells. Here, we observed fewer Tregs in the tumors of sFGL2-depleted mice, suggesting that a deficiency in tolerogenic DCs might have decreased Treg number. However, our data showed that DCs from sFGL2-blocked tumors did not inhibit Treg induction from CD4+CD25− cells, indicating that a lower percentage of Tregs in these tumors might not be the results of decreases in the number of DC-induced Tregs. Additionally, we demonstrated that sFGL2 blockage diminished IL-35 levels, which are produced by Tregs. IL-35 plays a vital role of self-tolerance, where IL-35 deficiency promotes autoimmunity in the form of multiple sclerosis, aplastic anemia, allergic rhinitis, and allergic diseases . Moreover, IL-35 promotes tumor growth, with tumor cells overexpressing IL-35 growing faster than controls , whereas anti-IL-35 treatment enhances antitumor immunity in vivo . Furthermore, IL-35 in tumor tissue promotes HCC progression and is associated with HCC recurrence . In the present study, we treated BALB/c mice harboring BNL tumors with intratumoral injection of anti-IL-35 once weekly, with the results showing that IL-35 blockage upregulated the number of Th1 cells and diminished the number of Tregs (Fig. 2g). This suggested that IL-35 might be an indicator capable of assessing the tumor microenvironment based on its promotion of tumor growth; therefore, downregulation of IL-35 levels in the hepatoma microenvironment might promote elevations in the number of Th1 cells and decreased Treg infiltration.
Conceptualization: QN, DX, XL. Methodology: MY, DX, ZZ. Software: MY, MW, WL. Validation: ZZ, JC, MX. Formal analysis: MY, ZZ, JC. Resources, Investigation and Data Curation: MY, ZZ, MX. Writing – original draft preparation: MY. Writing – review and editing: QN, DX, M-FY, XW. Visualization and Supervision: JH, DX. Project administration and Funding acquisition: QN. All authors read and approved the final manuscript.
This work was supported by National Natural Science Foundation of China (NSFC81571989) and the National Major Science and Technology Special Project on Major New Drug Innovation (2018ZX09733001–002-006).
Ethics approval and consent to participate
All the protocols of animal experiments were in accordance with the Guide for the Care and Use of Laboratory Animals approved by Huazhong University of Science and Technology.
Consent for publication
All authors consent for publication.
The authors declare that they have no competing interests.
- 15.Yan J, Kong LY, Hu J, et al. FGL2 as a multimodality regulator of tumor-mediated immune suppression and therapeutic target in gliomas. J Natl Cancer Inst. 2015;107. https://doi.org/10.1093/jnci/djv137.
- 24.Kim EK, Choi EJ. Pathological roles of MAPK signaling pathways in human diseases. Biochim Biophys Acta. 1802;2010:396–405.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.