Abstract
Background/Aim: The adenovirus vector– carrying reduced expression in immortalized cell (REIC) gene (Ad-REIC) increases endoplasmic reticulum stress chaperone GRP78/BiP expression and induces the JNK-mediated apoptotic pathway. We aimed to determine whether Ad-REIC–induced apoptotic cell death can trigger immunogenic cell death (ICD). Materials and Methods: We examined the emission of damage-associated molecular patterns in vitro and the vaccination effect in vivo. We determined the immunological changes in the tumour microenvironment by putative ICD inducers and the combined effects of immune checkpoint blockade therapies. Results: Ad-REIC induced the release of high-mobility group box 1 and adenosine triphosphate and the translocation of calreticulin in murine mesothelioma AB12 cells. The vaccination effect was elicited by Ad-REIC treatment in vivo. The effect of Ad-REIC was potentiated by anti-cytotoxic T-lymphocyte–associated protein 4 antibody treatment in a murine mesothelioma AB1-HA cell model. Conclusion: Ad-REIC induces ICD in malignant mesothelioma.
- REIC
- immunogenic cell death
- anti-cancer immunity
- gene therapy
- adenovirus vector
- immune checkpoint inhibitors
The reduced expression in immortalized cell (REIC) gene was first identified at the Okayama University (Okayama, Japan), and was described as a gene whose expression is down-regulated in immortalized human fibroblasts OUMS-24 (1). The REIC gene is identical to the human Dkk (dickkopf)-3 gene (2), a member of the Dkk family of genes (hDkk-1, hDkk-2, hDkk-3, and hDkk-4), which encodes secreted regulatory factors of the Wnt/β-catenin signalling pathway and plays critical roles in many essential biological processes and a variety of diseases (3). REIC expression is epigenetically silenced in a variety of cancers (4), and REIC re-expression suppresses the proliferation of different types of cancer cells (3). Among these physiological functions, REIC has been recently described as a physiological endoplasmic reticulum (ER) stressor (5). Indeed, gene therapy using an adenovirus-carrying REIC increases ER stress chaperone BiP (GRP78) expression and thereafter induces JNK-mediated apoptotic cell death in different types of cancer cells (6) and does not cause non-cancer cell death but the production of interleukin (IL) 7 to reinforce the survival of natural killer cells (7). In addition, REIC protein is capable of inducing the differentiation of mononuclear leukocytes into dendritic cell (DC)–like cells (8), resulting in systemic augmentation of anti-tumour immune responses. These characteristics are fundamental properties of adenovirus-mediated REIC gene therapy. Based on these findings, phase I–IIa studies of Ad-REIC gene therapy in several cancer patients are ongoing (9, 10).
Immune checkpoint inhibitors, a type of immunotherapy, block immune checkpoint proteins from binding to co-inhibitory molecules (11). Anti-programmed death 1 (anti-PD-1) antibody and anti-cytotoxic-T-lymphocyte-associated protein 4 (CTLA-4) antibody therapy has demonstrated to have a durable response across different cancer types, and these initial clinical successes have unveiled a new area in cancer immunotherapy (12). However, durable clinical responses were only seen in 20-30% of cancer patients, implying that pre-existing anti-tumour immunity is crucial for the reactivity to immune checkpoint blockade (ICB) therapies (12). To date, several positive prognostic factors, such as the number of tumour-infiltrating lymphocytes (TILs), tumour PD-L1 expression, high tumour mutational burden, the existence of tumour neoantigens, and an interferon γ (IFN-γ) gene expression signature, have been identified (13). Tumours, which intrinsically lose the expression of antigen-presenting molecules or are deficient in T cells, are less likely to respond to immune checkpoint inhibitors (14). The tumour microenvironment comprises cancer cells, immune cells, fibroblasts, adipocytes, endothelial cells, mesenchymal cells, and extracellular matrix. Metabolic reprogramming of the tumour microenvironment is induced by tumour-derived exosomes (15), and metabolic competition between immune cells and tumour cells promotes local immune escape and immune exhaustion (16). Therefore, a novel therapeutic strategy is required to generate an immunogenic tumour microenvironment and expand the number of patients who could benefit from immune checkpoint inhibitor therapy. One strategy to induce this conversion involves the use of pharmaceutical interventions that elicit immunogenic cell death (ICD) within the tumour (17). However, a limited number of anti-cancer drugs, including anthracyclines (18), oxaliplatin (19), radiation therapy (20), photodynamic therapy (21), and oncolytic viruses (22), have been reported to cause ICD.
ICD is defined as a type of cancer cell death that occurs with the release of highly immunostimulatory damage-associated molecular patterns (DAMPs), such as the extracellular emission of high-mobility group box 1 (HMGB-1) (23), adenosine triphosphate (ATP) (24), and the translocation of calreticulin (CRT) (25). Induction of ICD in the tumour microenvironment leads to efficient antigen presentation by DCs, and dying tumour cells function as tumour-specific neoantigen vaccines. Consequently, tumour neoantigen-specific CD8+ T cells are generated and moved out of the tumour-draining lymph nodes into the tumour microenvironment, resulting in effector functions. However, until now, only a limited number of anticancer modalities have been reported as ICD inducers. Furthermore, a particular drug has usage limitations due to the genetic background, tissue origin, and local tumour microenvironment; therefore, it is highly desirable to explore other anti-cancer therapies that can augment tumour immunogenicity and thereafter determine the advantage of immune checkpoint inhibitors.
In this study, we determined whether Ad-REIC–induced apoptotic cell death can trigger ICD by examining the emission of DAMPs in vitro and the vaccination effect in vivo. For confirmation testing, we compared the anti-tumour effects with and without the host immune system. Finally, we determined immunological changes in the tumour microenvironment by putative ICD inducers and the combined effects of ICB therapies.
Materials and Methods
Cell culture. Murine mesothelioma AB12 cells were kindly provided by Suzawa et al. (26) and cultured in high-glucose Dulbecco’s modified Eagle’s medium (Thermo Fisher Scientific, Waltham, MA, USA; #11965) supplemented with 10% foetal bovine serum (Thermo Fisher Scientific) and 1% penicillin and streptomycin (Thermo Fisher Scientific; #15070063) and maintained on a 100-mm plastic dish (BD Biosciences, Franklin Lakes, NJ, USA; #353003) at 37°C in an atmosphere containing 5% CO2. Murine mesothelioma AB1-HA (European Collection of Animal Cell Cultures, Porton Down, UK; #12070506) and AB22 (European Collection of Animal Cell Cultures; #10092307) were cultured in RPMI-1640 (Thermo Fisher Scientific, Gibco; #11875) supplemented with 10% foetal bovine serum (Biowest, Nuaillé, France) and 1% penicillin and streptomycin. In the case of AB1-HA cell culture, the aminoglycoside antibiotic G-418 (Sigma-Aldrich, St Louis, MO, USA; #4727878001) solution was added at a final concentration of 400 μg/ml to the RPMI-1640 complete medium.
Antibodies and reagents. InVivoPlus polyclonal Syrian hamster IgG (Bio X Cell, West Lebanon, NH, USA; clone: N/A; #BP0087), InVivoPlus anti-mouse CTLA-4 (CD152) (Bio X Cell; clone: 9H10; #BP0131), InVivoPlus rat IgG2a isotype control, anti-trinitrophenol (Bio X Cell; clone: 2A3; #BE0290), and InVivoPlus anti-mouse PD-1 (CD279) (Bio X Cell; clone: RMP1-14; #BP0146) were purchased and used. These antibodies were diluted with endotoxin-free Dulbecco’s phosphate-buffered saline (Merck Millipore, Billerica, MA, USA; #TMS-012-A) for intraperitoneal (i.p.) administration. Rat anti-mouse CD16/CD32 mouse BD Fc block (BD Biosciences, San Jose, CA, USA; clone: 2.4G2; #553142), PE-Cy7 rat anti-mouse CD44 (BD Biosciences; clone: IM7; #560569), FITC-Rat Anti-Mouse CD62L (BD Biosciences; clone: MEL-14; #561917), APC-Cy7 rat anti-mouse CD8a (BD Biosciences; clone: 53-6.7; #561967), and PerCP-Cy5.5 Mouse Anti-Mouse CD45.2 (BD Biosciences; clone: 104; #552950) were purchased and used. When these antibodies were used for immunostaining, all were diluted 100-fold before use. Mitoxantrone (Sigma-Aldrich; #M6545) and cisplatin (Sigma-Aldrich; #479306) were purchased and used.
Adenovirus infection. The full-length complementary DNA of the human REIC/DKK-3 gene was integrated into a cosmid vector pAxCAwt and transferred into an E1/E2-deleted replication-deficient adenovirus type 5 vector with the CAG (cytomegalovirus early enhancer/chicken β-actin) promoter using the cosmid cassettes and Ad DNA-terminal protein complex (COS/TPC) method (Takara Bio, Shiga, Japan) (27). The super gene expression (SGE) system was constructed by inserting the triple translational enhancer sequences of human telomerase reverse transcriptase, simian virus 40, and cytomegalovirus downstream of the BGH polyA sequence (28). An adenoviral vector carrying the LacZ gene with a CAG (Figure 1 and Figure 2) or SGE promoter (Figure 3 and Figure 4) was used as the control vector. These adenoviral vectors were generated using five replication-incompetent serotype adenoviruses. AB12 cells were plated at a concentration of 1.0×106 cells per well in 6-well plate culture plates 24 h before infection (Greiner; Bio-One GmbH, Frickenhausen, Germany). On the day of infection, the cells were counted, and AB12 cells were infected at a multiplicity of infection (MOI) of 0, 100, or 300. After adenovirus incubation at 37°C for 1 h, the complete medium was added to each well and incubated thereafter for 24 or 48 h.
Quantification of cell death by annexin V and propidium iodide (PI) staining. Cell death was quantified using the Annexin V Apoptosis Detection Kit APC (Thermo Fisher Scientific; #88-8007-72) according to the manufacturer’s instructions. The cells were detached using 0.05% trypsin/ethylenediaminetetraacetic acid (EDTA) solution and prepared at concentrations from 1.0 to 5.0×106 cells/ml, 24 or 48 h after adenovirus infection. One aliquot (100 μl) of the cell suspension was washed once with annexin V binding buffer and stained with 5.0 μl of APC-annexin V and 2.5 μl of PI for 15 min at room temperature. Stained cells were filtered using a 5-ml round-bottom tube with a 35-μm pore-size cell strainer cap to remove cell aggregates, and the annexin V+ PI− cells were measured using BD FACSAria I (BD Biosciences).
Quantification of ER stress using quantitative polymerase chain reaction (PCR). Quantification of ER stress (unfolded protein response) was performed using real-time quantitative PCR. Infected cells were harvested 24 h after infection, and total RNA was purified using the RNeasy Plus Mini Kit (Qiagen, Valencia, CA, USA; #74134) according to the manufacturer’s protocol. The concentration of the obtained RNA was measured using a Nanodrop 1000 (Thermo Fisher Scientific), and 2,000 ng of RNA was used along with SuperScript VILO Master Mix (Thermo Fisher Scientific; #11754050). Purified RNA samples were mixed with SuperScript IV VILO Master Mix according to the manufacturer’s instructions, and the mixture was incubated at 25°C for 10 min, 50°C for 10 min, and 85°C for 5 min. Reverse transcripts were diluted (1:10) with Ambion Nuclease-Free Water (Thermo Fisher Scientific) and used as a template for real-time PCR. Real-time PCR was performed using the StepOne PCR (Applied Biosystems, Foster City, CA, USA) in a MicroAmp Fast Optical 96-Well Reaction Plate (Thermo Fisher Scientific; #4346907). The PCR reaction was performed in a total volume of 20 μl using TaqMan Gene Expression Assays (20×) (listed in Table I) and TaqMan Fast Advanced Master Mix (2×) (Applied Biosystems; #4444557). The reaction conditions for amplification of DNA were 50°C for 2 min, 95°C for 20 s, 40 cycles of 95°C for 1 s, and 60°C for 20 s. The quantification of gene expression was performed by the ddCT method (29) using StepOne Software version 2.2 (Applied Biosystems). 18S CT was used as an endogenous control.
Quantification of extracellular ATP and HMGB-1 release. Conditioned medium derived from AB12 mesothelioma cells was harvested, centrifuged at 300×g for 5 min, and stored at −80°C. Extracellular ATP levels in the supernatant were quantified using an ATP Bioluminescence Assay Kit HS II (Roche Diagnostics, Mannheim, Germany; #11699709001) according to the manufacturer’s protocol. The supernatant extracellular HMGB-1 level was determined using the HMGB-1 ELISA Kit II (Shino-Test Corporation, Tokyo, Japan; #326078738) according to the manufacturer’s protocol. Absorbance and luminescence were measured using a FlexStation III (Molecular Device, San Jose, CA, USA).
Determination of cell surface CRT exposure. Cell surface CRT exposure was determined by staining with an antibody without permeabilization. The infected cells were harvested and washed twice with cell-staining buffer (BioLegend, San Diego, CA, USA; #420201). Aliquots (100 μl) of the cell suspension were blocked with 2 μl of purified anti-mouse CD16/32 (BioLegend; #101302) at 4°C for 10 min. Thereafter, 1.0 μl of anti-CRT antibody (Abcam, Cambridge, MA, USA; #ab2907) or 10 μl of rabbit IgG, polyclonal Isotype Control (ChIP Grade) (Abcam; #ab171870) was added and incubated for 30 min at 4°C. Further, these cells were washed once with cell-staining buffer and stained with goat anti-rabbit IgG (H+L) cross-adsorbed secondary antibody, Alexa Fluor 647 (Thermo Fisher Scientific; #A-21244, 1,000-fold dilution) for 30 min at 4°C. After washing once with cell staining buffer, aliquots (200 μl) of the cell suspension were stained with 5.0 μl DAPI (4¢,6-diamidino-2-phenylindole, dilactate) (BioLegend; #422801). Stained cells were filtered using a 5-ml round bottom tube with a 35-μm pore-size cell strainer cap to remove cell aggregates and analysed using BD FACSAria I. The median fluorescence intensity (MFI) was assessed for non-apoptotic–gated cells (DAPI−). The MFI from each sample was calculated by subtracting the anti-isotype-antibody-stained MFI from the anti-CRT antibody-stained MFI.
Co-culture of dying tumour cells with bone marrow-derived DCs. Bone marrow cells were harvested from BALB/c mouse bone marrow and cultured in 20 ng/ml GM-CSF (Peprotech, Princeton, NJ, USA; #315-03) containing complete RPMI-1640 with 2-mercaptoethanol in a 100-mm Petri dish (BD Falcon; #351029) for 7 days to generate DCs, as previously described by Lutz et al. (30). Forty-eight hours after Ad-CAG-REIC infection (at an MOI of 100), infected AB12 mesothelioma cells, including both dying and living cells with or without condition medium, were harvested and co-cultured with DCs at 1.0×106 cells/well. Twenty-four hours after co-culture of DCs and AB12 mesothelial cells, cells were harvested and then washed with cell-staining buffer (BioLegend; #420201), blocked with FcR blocking (Miltenyi Biotec; #130-092-575) and then stained with PE anti-mouse CD11c (BioLegend; #117307, 100-fold dilution) and APC anti-mouse CD86 (BioLegend; #104713, 100-fold dilution). Stained cells were filtered using a 5-ml round bottom tube with a 35-μm pore-size cell strainer cap to remove cell aggregates and were analysed using BD FACSAria I (BD Biosciences). The mean fluorescence intensity of CD86 gated on CD11c+ DCs was analysed.
Evaluation of ICD (in vivo vaccination assay). All animal protocols were in compliance and approved by the Animal Use Committees of Okayama University Medical School (approval no. OKU-2015173) or approved in agreement with the Guidelines of the Institutional Animal Care or Use Committee of Watarase Research Center of Kyorin Pharmaceutical (approval no. 17-117). Female BALB/cAJcl mice (6 weeks old), purchased from Japan Clea Inc. (Tokyo, Japan), were used when 7 weeks old after domestication in a novel environment. ICD in vivo by vaccination assays was performed as reported previously (31). Murine mesothelioma AB12 cells were treated in vitro with a presumptive inducer of ICD, 10 μM mitoxantrone (25), or a putative non-ICD inducer, 300 μM cisplatin (19), for 24 h or adenovirus infection at an MOI of 100 or 300 for 48 h. Cells were then washed and resuspended in phosphate-buffered saline (PBS), and 3.0×106 AB12 cells were finally injected subcutaneously into the left flank of mice as vaccination under isoflurane anaesthesia (3%-4% for induction and 1%-3% for maintenance). Mice were divided into PBS control (n=18), mitoxantrone (n=9), Ad-CAG-REIC at an MOI of 100 (n=15), Ad-CAG-REIC at an MOI of 300 (n=4), Ad-SGE-REIC at an MOI of 100 (n=9), and cisplatin (n=5). The experiment was conducted as four independent examinations. Seven days after the vaccination, mice were inoculated with intact 1.0×106 AB12 cells in the right flank as a challenge site under isoflurane anaesthesia. The tumour incidence rate was constantly observed at both vaccination and challenge sites over 41 days after AB12 challenge.
Evaluation of Ad-SGE-REIC on the immunocompetent mouse and immunodeficient mice models of mesothelioma AB12 and AB1-HA. As a confirmatory test for immunogenic cell death, specific pathogen-free immunodeficient BALB/cAJcl-nu/nu mice (aged 6 weeks) were purchased from Japan Clea Inc. and acclimatized for 1 week. When the tumours (AB12 and AB1-HA) reached a volume of 50-100 mm3, the mice were divided into three groups, PBS treatment (n=5), Ad-LacZ (n=5), and Ad-SGE-REIC (n=5). Tumour volume was monitored at least twice a week by measurement using a digital calliper (Mitutoyo Corp, Tokyo, Japan) and approximated according to the formula V=½×a×b2 (where a is the long diameter and b is the short diameter of the tumour). Tumour diameters were measured at least twice a week until 11 and 19 days after inoculation with AB12 and AB1-HA, respectively. The humane endpoint for euthanasia of mice was single tumour size greater than 2,000 mm3 or total tumour (ipsilateral and contralateral) size greater than 3,000 mm3.
Pathogen-free immunocompetent female BALB/cAJcl mice (aged 6 weeks) were purchased from Japan Clea Inc. and acclimatized for 1 week. One million cells of AB12 and AB1-HA cells were suspended in 100 μl of endotoxin-free PBS, and each cell line was inoculated intradermally into the bilateral flanks of mice (n=15 per each type of mesothelioma) under isoflurane anaesthesia. When the tumours reached a volume of 50-100 mm3, the mice were divided into three groups according to the tumour volume, PBS-treatment (n=5), Ad-LacZ (n=5), and Ad-SGE-REIC (n=5) in the case of AB12 and AB1-HA. Tumour diameters were measured until 12 and 21 days after inoculation of tumours with AB12 and AB1-HA, respectively.
Evaluation of anti-PD-1 antibodies on syngeneic mouse models of mesothelioma (AB12, AB1-HA, and AB22). Pathogen-free immunocompetent female BALB/cAJcl mice (aged 6 weeks) were purchased and acclimatized for 1 week. One million cells were suspended in 100 μl of endotoxin-free PBS. Each cell line was inoculated intradermally into the bilateral flanks of mice (n=15 per type of mesothelioma). When the tumours reached a volume of 50-100 mm3, the mice were divided into two groups, non-treatment (n=10) and anti-PD-1 treatment group (n=5), according to the tumour volume, and then anti-PD-1 antibody (100 mg) was administered by i.p. injection every 3 or 4 days for a total of three injections. Tumour diameters were measured at least twice a week. Tumour diameters were measured at least twice a week until 31, 19, and 40 days after tumour inoculation with AB12, AB1-HA, and AB22, respectively.
Evaluation of combinational therapy on syngeneic mouse models of mesothelioma (AB1-HA). Pathogen-free immunocompetent female BALB/cAJcl mice (aged 6 weeks) were purchased from Japan Clea Inc. and acclimatized for 1 week. When the tumours reached a volume of 50-100 mm3, the mice were divided into four groups, isotype antibody control (n=5), anti-PD-1 treatment group (n=5), Ad-SGE-REIC (n=5), and Ad-SGE-REIC with anti-PD-1 treatment (n=5). Tumour diameters were measured until 24 days after tumour inoculation. To assess the effect of combination therapy with CTLA-4 inhibitors, the mice were divided into isotype antibody control (n=5), anti-CTLA-4 treatment group (n=5), Ad-SGE-REIC (n=5), and Ad-SGE-REIC with anti-CTLA-4 treatment (n=5). Tumour diameters were measured until 35 days after tumour inoculation.
Analysis of CD8+ T-cell memory phenotypes in secondary lymphoid tissue. Pathogen-free immunocompetent female BALB/cAJcl mice aged 6 weeks were purchased from Japan Clea Inc. and used after 1 week of acclimatization. AB12 cells (1×106) were suspended in 100 μl of endotoxin-free PBS and inoculated intradermally into the bilateral flanks of mice (n=15) under isoflurane anaesthesia. When the tumours reached a volume of 50-100 mm3, the mice were divided into groups according to the tumour volume and then administered by intratumoural injection of Ad-SGE-REIC (n=5), Ad-LacZ (n=5), or PBS (n=5) at a dose of 1.0×109 infection-forming units (IFUs) into the left flank tumour under isoflurane anaesthesia. At the indicated time, mice were euthanized by cervical dislocation, and both the spleen and lymph nodes were harvested, minced with a razor, and transferred into gentleMACS C Tube (Miltenyi Biotec) filled with FACS buffer (PBS containing 5 mM EDTA and 2% FBS) and homogenized with gentleMACS Octo Dissociator with Heaters using gentleMACS m_Spleen01 program. Cell suspensions were passed through a 70-μm nylon cell strainer (Falcon; #352350) and used for immunostaining. In the case of splenocytes, red blood cells were removed using ACK lysis buffer (Thermo Fisher Scientific, Gibco; #A1049201). Cells were washed twice with PBS, and dead cells were stained with LIVE/DEAD Fixable Aqua Dead Cell Stain Kit, for 405-nm excitation (Invitrogen; #L34957), according to the manufacturer’s protocol. The cells were washed twice and suspended in 100 μL of FACS buffer. Cells were blocked with Purified Rat Anti-Mouse CD16/CD32 and stained with each antibody for 30 min on ice. Thereafter, the cells were washed with 2 ml of FACS buffer and fixed with 1% paraformaldehyde (Wako Pure Chemical Industries, BD; #553141) for 20 min at room temperature. Further, the cells were washed with 2 ml of FACS buffer and analysed using BD FACS Verse (BD Bioscience).
Analysis of gene expression profile of the tumour microenvironment. At the indicated time, AB1-HA-bearing mice were killed by cervical dislocation, and tumour tissues were harvested, minced using a razor, and immersed in RNAlater Stabilization Solution (Invitrogen; #AM7021) overnight at 4°C, and subsequently stored at −80°C. To isolate total RNA, frozen tissues were transferred into gentleMACS M Tube (Miltenyi Biotec) filled with Buffer RLT Plus, which was part of the RNeasy Plus Mini Kit and homogenized with gentleMACS Octo Dissociator with Heaters using gentleMACS Program RNA_02. Thereafter, total RNA was purified using the RNeasy Plus Mini Kit according to the manufacturer’s protocol. After measuring the concentration of RNA by Nanodrop 2000 (Thermo Fisher Scientific), 2,000 ng of RNA was reverse transcribed using SuperScript VILO Master Mix SuperScript (Life Technologies, Foster City, CA; #11754050) following the manufacturer’s instructions, and the mixture was incubated at 25°C for 10 min, 50°C for 10 min, and 85°C for 5 min. Reverse transcripts were diluted (1:10) with Ambion Nuclease-Free Water (Thermo Fisher Scientific) and used as a template for real-time PCR. Real-time PCR was performed in combination with TaqMan Gene Expression Assays and TaqMan Fast Advanced Master Mix (Applied Biosystems; #4444557) on the ViiA 7 Real-time PCR System (Life Technologies). The TaqMan Gene Expression Assay IDs are listed in Table I.
Statistical analyses. Statistical analyses were performed using GraphPad Prism version 8.4.3 (GraphPad Software, Inc., San Diego, CA, USA). Data are expressed as the mean±standard error of the mean. A two-tailed unpaired Student’s t-test was used to compare the differences between two groups, whereas one-way analysis of variance followed by Dunnett’s post hoc test was used to compare the difference between multiple groups. In the case of vaccine experiments, log-rank tests were performed to compare treated cells to the control (PBS). For all comparisons, p<0.05 was considered statistically significant.
Results
Ad-CAG-REIC induces the emission of ATP and HMGB-1 and exposure of cellular surface CRT in murine mesothelioma AB12 cells. First, AB12 mesothelioma cells were selected as a validation model since Suzawa et al. (26) showed potent systemic anti-tumour immunity by intratumoural injection of Ad-SGE-REIC. We investigated whether apoptotic cell death induced by Ad-REIC shows the typical features of ICD, such as CRT surface exposure, HMGB-1, ATP release, and ER stress in AB12 mesothelioma cells (Figure 1A). We confirmed an increase in the population of annexin V+ and PI− cells compared with the mock-infected control 24 h after infection (Figure 1B). Furthermore, cell death induced by Ad-CAG-REIC infection was approximately equal to that of the control virus 48 h after infection (data not shown). To confirm that apoptotic cell death was associated with ER stress, we analysed the induction of unfolded protein response (UPR) gene expression (gadd34, XBP1, DNAjb9, and Grp78 were up-regulated by PERK, ATF6, IRE1, and the common UPR pathway, respectively). Among these genes, gadd34 expression was selectively up-regulated by Ad-CAG-REIC infection (Figure 1C). Next, we investigated whether Ad-REIC treatment stimulates the extracellular release of HMGB-1 and ATP from AB12 mesothelial cells. As a result of the analysis, we observed an increase in HMGB-1 and ATP release into the conditioned medium 48 h after the infection compared with the mock-infected control (Figure 1D and E). Furthermore, we examined extracellular CRT exposure in living cells and found that CRT exposure increased 48 h after Ad-CAG-REIC infection (Figure 1F) compared to mock-infected cells. Finally, we examined whether these changes evoked activation of DCs. The mean fluorescence intensity of CD86 molecule expression on DCs increased 24 h after co-cultivation with AB12 mesothelioma cells infected with Ad-REIC (Figure 1G). Taken together, Ad-REIC induced DAMP release in AB12 cells, resulting in the activation of DCs.
AB12 infected with Ad-CAG-REIC or Ad-SGE-REIC acts as a therapeutic cancer vaccine. Next, we evaluated ICD in vivo. The experimental time course is shown in detail in Figure 2A. The results of the four independent experiments are summarized in Figure 2B. Seven days after vaccination, tumour formation was palpable at the vaccination site in the case of vaccination with non-treated AB12 and Ad-CAG-LacZ. In contrast, there was no tumour growth in the case of cisplatin, mitoxantrone, Ad-CAG-REIC, or Ad-SGE-REIC treatment at the vaccination site. Furthermore, mice inoculated with mitoxantrone, Ad-CAG-REIC, and Ad-SGE-REIC–infected AB12 cells showed no tumour growth at tumour challenge sites. Vaccination experiments showed that Ad-REIC infection induced ICD in AB12 cells in vivo. Next, we compared the anti-tumour effect in immunocompetent mice with immunodeficient tumour-bearing mice (Figure 2C, D, and E) by Ad-SGE-REIC for confirmation testing. The comparative analysis revealed that anti-tumour effects were noteworthy in immunocompetent mice models rather than in immunodeficient mice. Since antigens released by dying tumours are efficiently presented by DCs to naive or memory CD8+T cells in the tumour-draining lymph nodes, we examined the change in CD8+ T cells of the memory phenotype in secondary lymphoid tissues (Figure 2F and G). Consequently, the change in memory phenotype occurred, especially in the tumour-draining lymph nodes. In contrast, memory phenotype did not change in other secondary lymphoid tissues, such as the spleen and non-tumour draining lymph nodes. Based on these findings, we concluded that Ad-REIC acts as a bona fide inducer of ICD in murine mesothelioma AB12 cells.
Ad-SGE-REIC inhibits tumour growth in a murine mesothelioma AB1-HA cell model, which is refractory to anti-PD-1 antibodies. As described above, Ad-REIC induced ICD in AB12 mesothelial cells. Therefore, we expected that intratumoural Ad-REIC administration could change from a non-immunogenic tumour environment to an immunologically hot tumour microenvironment. To test this hypothesis, we first selected non-immunogenic murine mesothelioma cells by examining the effect of anti-PD-1 antibody therapy. Treatment with anti-PD-1 antibody completely inhibited AB12 cells at both flank sites, but not in the case of AB1-HA and AB22 mesothelioma cells (Figure 3A and B). Therefore, to determine immune reprogramming by Ad-REIC treatment, we selected murine mesothelioma AB1-HA and AB22 cells as immunologically cold tumour models. Ad-SGE-REIC showed modest growth inhibition of the tumour on the ipsilateral side and a slight inhibition on the contralateral side (Figure 3C and D). These anti-tumour effects were cancelled in the case of immunodeficient mice (Figure 3E and F). We did not observe any anti-tumour effect in the murine mesothelioma AB22 syngeneic mouse model because the murine mesothelioma AB22 cells had low CXADR expression and high DKK3 expression, resulting in weak ER stress induction by Ad-REIC (data not shown).
Ad-SGE-REIC up-regulates the expression of CD8 and its effector molecules in the AB1-HA tumour microenvironment. Based on the results obtained from the study of the AB1-HA syngeneic mouse model, we expected that Ad-SGE-REIC would change the tumour microenvironment. To validate these changes, we analysed the gene expression profiles of the whole tumour injected with Ad-SGE-REIC in the AB1-HA model (Figure 4A). First, we confirmed the up-regulation of the gene expression levels of ER stress markers (XBP1, Grp78, and CRT) by Ad-SGE-REIC compared with the control (Figure 4B). In this model, the expression levels of anti-tumour immune response genes (CD8, CD4, CD11c, IFNg, TNFa, IL-2, perforin, and granzyme B) decreased at day 17 compared to those at day 10 after AB1-HA inoculation (Figure 4C). Next, we analysed the changes in immune-related gene expression in immunologically cold tumours after Ad-SGE-REIC treatment. We examined whether immune cells re-infiltrated the tumour tissue by ICD. We showed that Ad-REIC up-regulated the expression of CD8 genes accompanied by increased lineage-specific gene expression (CD4 and CD11c) and the expression levels of immune effector molecules (IFN-γ, TNF-α, IL-2, perforin, and granzyme B) (Figure 4D). In contrast, gene expression of immunoregulatory immune cells such as regulatory T cells (FOXP3), TGF-b1, and IL-10 and the expression of T-cell exhaustion markers, including PD-1, PD-L1, and CTLA-4, were also increased (Figure 4E). These results showed that Ad-SGE-REIC induced the reprogramming of an immunologically cold tumour microenvironment.
Anti-CTLA-4 antibody therapy augments Ad-SGE-REIC-induced systemic immunity in the AB1-HA syngeneic model. Because co-inhibitory molecules were increased by Ad-SGE-REIC, we tested whether two types of immune checkpoint inhibitors, anti-PD-1 and anti-CTLA-4, potentiate the Ad-SGE-REIC–mediated anti-tumour immune response. We first examined the effect of combination therapy with anti-PD-1 antibody and Ad-SGE-REIC treatment on AB1-HA cells at different dose combinations. In our experimental conditions, we did not observe any difference in the combination of anti-PD-1 treatment (100 μg/mouse) with Ad-SGE-REIC (2×109 IFU/mouse) (Figure 5A and B) compared with each monotherapy or at different dose combinations of anti-PD-1 treatment (200 μg/mouse) and Ad-SGE-REIC (1×109 IFU/mouse) (data not shown). Next, we examined the combined effect of anti-CTLA-4 treatment (100 μg/mouse) with Ad-SGE-REIC treatment (1×109 IFU/mouse). The combination therapy with anti-CTLA-4 treatment (100 μg/mouse) and Ad-SGE-REIC (1×109 IFU/mouse) completely blocked tumour growth in both flanks (Figure 5C and D). We also detected a synergistic effect in another dose combination of anti-CTLA-4 treatment (100 μg/mouse) and Ad-SGE-REIC treatment (2×109 IFU/mouse) (data not shown). Taken together, Ad-SGE-REIC–mediated anti-tumour immune response was enhanced by anti-CTLA-4 immunotherapy.
Discussion
Although Ad-REIC has a potential role in initiating ER stress (6), little is known about the involvement of Ad-REIC in ICD, especially with respect to the emission of DAMPs. In this study, we showed that Ad-REIC induces the release of HMGB-1 and ATP and the translocation of CRT in AB12 murine mesothelioma cells. We further showed that the vaccination effect is elicited by Ad-REIC treatment in vivo, which is a hallmark of ICD. Ad-REIC showed anti-tumour effects and the induction of ER stress markers and immunogenic reprogramming of the tumour microenvironment in murine mesothelioma AB1-HA cells, which are refractory to anti-PD-1 therapy. These changes were potentiated by anti-CTLA-4 antibody therapy. Altogether, these results showed that Ad-REIC acts as an ICD inducer in vivo, raising the possibility that combination therapy with anti-CTLA-4 antibody triggers a potent immune response in subjects with human mesothelioma.
Recently, ICD inducers have been classified into two groups [types I and II (32)]. Type I ICD inducers mainly comprise chemotherapeutic agents that evoke apoptotic cell death through non-ER-associated targets, such as DNA damage or proteasome inhibition, and up-regulate danger signalling via ER stress in an off-target manner. In contrast, type II inducers, such as photodynamic and oncolytic therapies, selectively target the ER to elicit CHOP or JNK-mediated apoptotic cell death and cause the release of danger signals in an ER-dependent manner. Ad-SGE-REIC is considered a type II ICD inducer because of its intensive ER stress–mediated apoptotic cell death. Type II ICD inducers, including photodynamic therapy and oncolytics, show anti-tumour immunity in a syngeneic mouse model (32). Antitumour immune responses induced by both photodynamic therapies (33) and oncolytic virus therapies (34) are potentiated by ICB therapies, including anti-PD-1, anti-PD-L1, and anti-CTLA-4 antibodies. Accordingly, we showed the combinational therapeutic effect of anti-CTLA-4 treatment and Ad-SGE-REIC, a putative ICD type II inhibitor. In our study, the improvement of immune exhaustion by anti-PD-1 treatment was definite, and anti-CTLA-4 antibody monotherapy caused more intensive anti-tumour immunity than anti-PD-1 treatment. These monotherapeutic effects were consistent with those reported by Fear et al. (35). Their study implied that terminally immune-exhausted CD8+ T cells in the tumour microenvironment are non-responsive to anti-PD-1 therapy in murine mesothelioma AB1-HA cells. The authors showed the up-regulation of CTLA-4 molecules on regulatory T cells in the tumour-draining lymph nodes and the tumour microenvironment. Considering the strong monotherapeutic effect of CTLA-4 blockade, regulatory T-cell–mediated trans-endocytosis of B7 molecules of DCs (36) contributes largely to the immune escape of both local and systemic anti-tumour immunity. We inferred that the combination therapy of Ad-SGE-REIC and anti-CTLA-4 antibody enhances the efficient antigen presentation by DCs in the tumour-draining lymph node without being impaired by regulatory T-cell–mediated CTLA-4 dependent transendocytosis of B7 molecules.
These mechanisms can partially explain how Ad-REIC induces anti-tumour immune activation-related phenomena in the bilateral tumour model, as we have previously reported (26, 37). In contrast, REIC protein itself has immune-modulatory functions in vivo (38), such as inducing monocytes into DC differentiation (8). In this study, we could not determine the contribution of REIC protein–mediated anti-tumour immune response to the anti-tumour immune effect by Ad-REIC because REIC-neutralizing antibodies are not available in vivo. The limitations of this study include the focus on malignant mesothelioma cells and the lack of investigation of other types of cancers, such as lung cancer (26) and lymphoma cells (37). Further studies are needed to reveal whether Ad-REIC can induce ICD as a general attribute regarding the emission of DAMPs and show the vaccination effect in other syngeneic mouse models. The exploration of the mechanism requires further investigation, including changes in TILs and maturation of DCs in the draining lymph node when combined with Ad-SGE-REIC and ICB therapy. In this study, changes in the tumour microenvironment were analysed by real-time PCR and only validated at the mRNA level. The lack of validation at the protein level and the lack of cell microscopy data are the major limitations of the present study. These experiments allowed us to analyse the subpopulation of immune cells infiltrating the tumour microenvironment and the interaction between immune cells and cancer cells. Knowing the spatial arrangement and co-localization of these cells will provide stronger support for our hypothesis.
The results obtained from these investigations support the hypothesis that Ad-REIC induces ICD in malignant mesothelioma. Ad-REIC acts as an ICD inducer in vivo, raising the possibility that combination therapy with anti-CTLA-4 antibody triggers a potent immune response in patients with mesothelioma.
Acknowledgements
We thank Dr. Junko Mori (Okayama University) for carefully reading and providing critical comments on this manuscript. This research was supported by the Newly Extended Technology Transfer Program (NexTEP) from the Japan Science and Technology Agency (JST) (grant number JPMJTT14N8).
Footnotes
Authors’ Contributions
KA, NT, and HK conceived the study, KA and HK wrote the manuscript and generated the figures. KA, NT, and NY conducted the experiments. All Authors have read and approved the final manuscript.
Conflicts of Interest
KA and NY were employees of Kyorin Pharmaceutical Co., Ltd. Both authors conducted the study at their discretion and had editorial freedom with respect to the manuscript. The authors will not receive any monetary rewards from the company upon acceptance of the manuscript. Okayama University and Momotaro-Gene Inc. have patents for the SGE system, which was invented by HK. Momotaro-Gene Inc. holds the patents for the Ad-REIC and REIC protein agents and develops the agents as a cancer therapy medicine. HK owns stock in Momotaro-Gene Inc. Okayama University and Momotaro-Gene Inc. are working together to develop the Ad-REIC agent. Okayama University received research funds for joint research. HK is the chief scientific officer of Momotaro-Gene Inc.
- Received August 5, 2021.
- Revision received August 30, 2021.
- Accepted September 2, 2021.
- Copyright © 2021 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.