Reovirus combined with a STING agonist enhances anti-tumor immunity in a mouse model of colorectal cancer

Reovirus, a naturally occurring oncolytic virus, initiates the lysis of tumor cells while simultaneously releasing tumor antigens or proapoptotic cytokines in the tumor microenvironment to augment anticancer immunity. However, reovirus has developed a strategy to evade antiviral immunity via its inhibitory effect on interferon production, which negatively affects the induction of antitumor immune responses. The mammalian adaptor protein Stimulator of Interferon Genes (STING) was identified as a key regulator that orchestrates immune responses by sensing cytosolic DNA derived from pathogens or tumors, resulting in the production of type I interferon. Recent studies reported the role of STING in innate immune responses to RNA viruses leading to the restriction of RNA virus replication. In the current study, we found that reovirus had a reciprocal reaction with a STING agonist regarding type I interferon responses in vitro; however, we found that the combination of reovirus and STING agonist enhanced anti-tumor immunity by enhancing cytotoxic T cell trafficking into tumors, leading to significant tumor regression and survival benefit in a syngeneic colorectal cancer model. Our data indicate the combination of reovirus and a STING agonist to enhance inflammation in the tumor microenvironment might be a strategy to improve oncolytic reovirus immunotherapy.


Introduction
Human oncolytic reovirus is a non-enveloped doublestranded RNA virus composed of an outer and inner protein shell, which forms a 20-sided icosahedral capsid.Although reovirus infections are mild and sometimes subclinical, reovirus displays selective oncolytic activity against transformed and malignant cells [1,2].Previous studies on the mechanisms involved in how reovirus kills tumor cells suggest Ras pathway activation is a key determinant of viral replication and subsequent oncolysis [3,4].We investigated the efficacy of reovirus against gastrointestinal cancer and reported the strong antitumor activity of reovirus alone or in combination with anticancer drugs against gastric cancer cells [5].Additionally, we found the antitumor efficacy of reovirus involved oncolytic effects of the reovirus as well as its immunostimulatory effects [6].Multiple preclinical cancer models also demonstrated the functionality of reovirus as an immunotherapeutic agent [7][8][9] and the enhanced efficacy of reovirus in combination with immunotherapeutic agents including PD-1/PD-L1 blocking agents [10][11][12][13].
Reovirus has an intrinsic preference for propagation in KRAS-mutated cancer cells, and it has specific mechanisms to escape from type I interferon (IFN) antiviral responses [14].A recent study reported that reovirus inhibited the production of type I IFNs by blocking IRF3 translocation into the nucleus [15].Inhibition of type I IFNs by reovirus promoted its infection of cancer cells and its replication, which killed cancer cells, although it simultaneously suppressed type I IFN-induced antitumor immunity.Type I IFNs are key regulators that facilitate the antitumor immunity and activity of conventional cell death-inducing therapies by enhancing the activity of tumor infiltrating lymphocytes; therefore, the anti-tumor activity of reovirus might be partially opposed by its strategy for propagation in cancer cells.
Stimulator of interferon gene (STING) is a cytosolic DNA sensor involved in immune defense against pathogens such as bacteria and viruses.STING is activated by cyclic GMP-AMP (cGAMP) produced by the binding of cyclic GMP-AMP synthase (cGAS) to aberrant cytoplasmic double-stranded DNA (dsDNA) originating from intracellular pathogens, which induces the expression of type I IFNs.The cGAS-STING pathway is also stimulated by DNA leakage from cancer cells or by cytosolic fragments of DNA caused by irradiation or anticancer agents that break DNA strands.STING is a potent target for cancer immunotherapy and the compounds that modulate STING have been developed for cancer treatment [16,17].
Although reovirus induces anti-tumor immunity in addition to direct tumor killing activity, it also represses type I IFN responses that modulate immune responses that eliminate cancer cells.Thus, we hypothesized that activation of the STING axis counteracts the suppression of type I IFN responses by reovirus, and that combination treatment with reovirus and a STING agonist promotes anticancer immunity by improving the tumor microenvironment.In this study, we investigated the effects of reovirus combined with a STING agonist on antitumor immunity.

Reagents
ADU-S100 (MIW815), a cyclic dinucleotide (CDN) agonist of STING, was purchased from CHEMIETEK (Indianapolis, IN), dissolved in double distilled water, and then stored at − 20 °C until use.Reovirus was provided by the University of Calgary, Department of Molecular Biology and Oncology (Canada).Reovirus was divided into aliquots and stored in the dark at neat concentrations in phosphate buffered saline (PBS) at − 80 °C, and used it within 2 months [18].

Cell culture
The murine colon cancer cell lines used for the experiments were CT26, MC38, and CMT93.CT26 and CMT93 cells were purchased from ATCC (Manassas, FL), and MC38 was purchased from Kerafast (Boston, MA).CT26, MC38, and CMT93 cells were cultured in low glucose Dulbecco's Modified Eagle's Medium supplemented with 10% fetal bovine serum and maintained at 37 °C with 5% CO 2 .

Cell proliferation assay
Cell proliferation was analyzed using the WST-8 cell proliferation assay.CT26 cells were seeded into 96-well culture plates at a concentration of 500 cells/100 μl/well and incubated for 8 h.Cells were then treated with reovirus at a multiplicity of infection (MOI) of 3 or 10. Cell proliferation was evaluated using the Cell Counting Kit-8 (Dojindo Laboratories, Kumamoto, Japan) according to the manufacturer's protocol, and the absorption at 450 nm was measured with a microplate spectrophotometer (SPECTRA MAX340, Molecular Devices, San Jose, CA).We determine the efficacy of reovirus by adjusting the OD values of the treated cells relative to the control after 72 h of treatment.

Quantitative reverse transcription-polymerase chain reaction
To investigate differences in gene expression induced by reovirus infection and ADU-S100 administration, reverse transcription-polymerase chain reaction (RT-PCR) was performed.We assessed the mRNA expression profiles of CT26 cells treated with reovirus at 10 MOI and 10 µM ADU-S100 for 1, 2, and 6 h.Real-time quantitative RT-PCR analyses were performed using an ABI 7900 HT Fast Real-Time PCR system (Applied Biosystems, Foster City, CA) according to the manufacturer's recommendations, and the calculation was performed using the delta delta Ct method.The primer sequences are as follows: IFNβ1 forward primer: 5′-GGC ATC AAC TGA CAG GTC TT-3′; reverse primer: 5′-ACT CAT GAA GTA CAA CAG CTACG-3′.Specific to GAPDH forward primer: 5′-TGA GCA AGA GAG GCC CTA TC-3′; reverse primer: 5′-AGG CCC CTC CTG TTA TTA TG-3′ (Applied Biosystems) served as cellular RNA controls.

Detection of ICD markers
CT26 cells seeded in 24-well plates (3 × 10 4 cells/well) were treated with 10 MOI Reovirus, and cell supernatants collected 48 h and 72 h.Cell culture supernatants were centrifuged at 10,000 rpm for 3 min at 4 ℃.HMGB1 released in culture supernatants was detected using an HMGB1 Detection ELISA kit (# 6010, Chondrex, Woodinville, WA).CT26 cells seeded in 96-well plates at 500 cells/well and incubated overnight were treated with 10 MOI Reovirus for 48 h and 72 h.Extracellular ATP was measured with a RealTime-Glo™ Extracellular ATP Assay (GA5010, Promega, Madison、WI).
Tumor samples extracted from mice were fixed in 10% neutral buffered formalin for 24 h.Formalin-fixed and paraffin-embedded Sects.(4 µm) were used.After deparaffinization, slides submerged in 10 mM citrate buffer were heated in a microwave until boiling followed by 10 min at a sub-boiling temperature (95-98 °C).Slides were cooled for 30 min.After pre-incubation with 3% hydrogen peroxide for 10 min to block endogenous peroxidase activity, sections were rinsed in PBS, and blocked with 100 µl of preferred 10% normal goat serum (Nichirei, Tokyo, Japan) for 10 min at room temperature.Primary rabbit anti-reovirus polyclonal antibody (kindly gifted from Dr. Randal N. Johnston) were incubated overnight at 4 °C, protected from light.Sections were rinsed and incubated sequentially with secondary antibody (Alexa Fluor 488 goat anti-rabbit IgG).Sections were then washed with PBS, and protected from light.All sections were counterstained with DAPI (Biotium).Images were taken with a FV3000 Confocal Laser Scanning Microscope (Olympus, Tokyo, Japan).

Mouse tumor models
Female mice (BALB/cCr Slc), 6-8 weeks of age, with a body weight of 18-22 g, were obtained from Japan SLC, Inc. (Shizuoka, Japan).Syngeneic mouse models were established by subcutaneously implanting 5 × 10 5 CT26 cells in 100 μl of PBS into the right flank or both flanks of mice.Mice were then treated with ADU-S100 and/or reovirus.Mice were divided into four groups; control, ADU-S100 IT injection, reovirus IT injection, and ADU-S100 and reovirus groups.Some mice were treated with one IT injection containing 10 μg/50 μL of ADU-S100.Other mice were treated with three IT injections of reovirus at a dose of 3 × 10 8 plaque-forming units (pfu) per mouse in 50 μL on alternate days.The dose of reovirus was chosen based on our previous study [5,6].The maximum tumor diameter (L) and the diameter at right angles to that axis (W) were measured twice a week using calipers, and the volume was calculated according to the formula, (L × W 2 )/2.Tumor size was measured and anti-tumor activity of reovirus or ADU-S100 was evaluated until the tumor volume reached 2 000 mm 3 .For histological and FACS analyses (see below), three days after the IT injection of anticancer agents, transplanted tumors were excised and fixed in formalin after euthanasia.No adverse effects, including weight loss or infection, were observed throughout the treatment period.The number of mice used in each experiment is denoted as Supplemental Table 1.The procedures in these experiments were approved by the Nagoya City University Center for Experimental Animal Science, and the mice were cared for according to the guidelines of the Nagoya City University for Animal Experiments.

Flow cytometry
Excised tumors were dissected into smaller fragments using scalpels and scissors, and further dissociated into single cell suspensions using the GentleMACS Octo Dissociator with Heaters (Miltenyi Biotec, San Diego, CA).For each sample, 1 × 10 6 cells were treated with mouse Fc block (#553142, BD Biosciences, Franklin Lakes, NJ) and then stained with Fixable Viability Stain 780 (#565388, BD Biosciences) and five different labeling antibodies.Anti-CD3, anti-CD4, anti-CD8, anti-CD25, and anti-FoxP3 were purchased from BD Biosciences (San Jose, CA).Data were acquired on a BD FACSCanto II.

Multiplex gene expression analysis
Total RNA was extracted from the tumor of the mice three days after treatment with reovirus, ADU-S100, or the combination of reovirus and ADU-S100 (RNeasy Lipid Tissue Mini Kit, Qiagen, Hilden, Germany) according to the manufacturer's instructions.Samples with RNA integrity values > 7.0 (4200 TapeStation System, Agilent, Santa Clara, CA) were included for multiplex gene expression analysis.Data were collected on a NanoString nCounter digital analyzer using the nCounter Mouse PanCancer Immune Profiling Panel (Nanostring Technologies, Seattle, WA) according to the manufacturer´s instructions.Data analysis was performed by the nSolver analysis software (Nanostring Technologies) normalized to the internally provided control genes set.

Statistical analysis
All statistical analyses were performed using GraphPad Prism version 8.4.2 (GraphPad Software, La Jolla, CA).The Student's t-test was used when two independent groups were compared, and multiple comparisons between groups were analyzed by one-way ANOVA with Tukey's multiple comparisons test or two-way ANOVA.Survival data were analyzed by the log-rank (Mantel-Cox) test.P-values less than 0.05 were considered to indicate statistical significance.

Significance of STING pathway on reovirus infection in murine colorectal cancer cell lines
We examined the expression of a cytosolic DNA sensor, cGAS and STING, in murine colorectal cancer cell lines including CT26, CMT93, and MC38.Western blot analysis demonstrated that cGAS was expressed in CT26 and MC38.STING showed varied expression in all cell lines examined (Fig. 1A).Next, we examined the activation of STING signaling responsible for type I IFN regulation in CT26 stimulated by reovirus alone or in combination with ADU-S100 by evaluating the phosphorylation of IRF-3, a downstream factor of STING signaling, using western blot analysis.No activation of STING signals was detected in CT26 cells treated with reovirus.In contrast, STING agonist, ADU-S100 significantly induced the phosphorylation of IRF3; however, this effect was attenuated when ADU-S100 was administered with the reovirus (Fig. 1B, C).We performed immunofluorescence staining to examine the translocation of IRF3 in CT26 cells treated with reovirus.One hour after treatment with ADU-S100, IRF3 localized into the nucleus confirming the western blot analysis.Although reovirus alone did not induce the translocation of IRF3 into the nucleus, IRF3 was detected in the nucleus of cells treated with reovirus combined with ADU-S100 (Fig. 1D).Quantitative RT-PCR revealed that CT26 cells produced IFN-β in a concentration-dependent manner in response to ADU-S100 (Supplementary Fig. 1).Reovirus significantly suppressed the expression of IFN-β enhanced by ADU-S100 in CT26 cells (Fig. 1E).Reovirus also suppresses the effect of ADU-S100 against IFN responses in MC 38 cells (Supplementary Fig. 2).These results indicate that reovirus infection is not sensed by the STING signaling pathway, and the ability to produce type I IFN, which is the fundamental mechanism of the anticancer activity of STING agonist, was partially canceled by reovirus when they are introduced simultaneously in vitro.

Reovirus inhibits the growth of murine colorectal cancer cell line, CT26, and induces immunogenic cell death
We investigated the cytotoxic effect of reovirus against CT26 using a cell viability assay.After treatment with reovirus for three days, proliferation was significantly decreased in CT26 in a concentration dependent manner (Fig. 2A).A STING agonist, ADU-S100, had no significant additive effect on the cytotoxicity of the reovirus against CT26 in vitro (Fig. 2B).
We measured the concentration of ATP in the culture media of CT26 cells treated with reovirus and the HGMB-1 released from CT26 cells after exposure to reovirus to evaluate the immunogenic cell death induced by reovirus in CT26 cells.The concentration of extracellular ATP in culture media of CT26 treated with reovirus was significantly higher than that in culture media of CT26 without reovirus treatment (Fig. 2C).A higher concentration of HGMB-1 was detected in the culture media congaing CT26 treated reovirus (Fig. 2D).These results demonstrate that reovirus induces immunogenic cell death in CT26.

Efficacy of reovirus in combination with a STING agonist on the growth of CT26 cells in vivo
To investigate whether reovirus and ADU-S100 synergistically induced anti-tumor efficacy against colorectal cancer (CRC) in immunogenic settings, we used a syngeneic murine model of CRC.To analyze the anti-tumor immune response, CT26 cells were subcutaneously injected into BALB/c mice.The tumor size was measured, and the tumor volume was calculated after the intratumoral (IT) injection of reovirus, ADU-S100, or reovirus and ADU-S100.Reovirus IT injection significantly inhibited tumor growth compared with controls.IT treatment with ADU-S100 also inhibited tumor growth, and the combination of reovirus and ADU-S100 demonstrated the strongest anti-tumor activity, and ADU-S100 showed a significant additive effect on the anti-tumor activity of the reovirus (Fig. 3A).Additionally, the combination treatment significantly improved the survival of tumor-bearing mice compared with controls (Fig. 3B).In the combination treatment group, four of eight mice achieved complete regression of the tumor.Histopathological analyses (Supplementary Fig. 3) revealed that the combination therapy significantly decreased the number of Ki-67-positive cells., increased the number of cleaved caspase-3-positive cells, and increased the number of granzyme B-positive cells in tumors compared to controls (Supplementary Fig. 4).These results suggest that combining reovirus and ADU-S100 promotes an anti-tumor effect in tumor-bearing mice through enhanced anti-tumor immunity.

Combined reovirus and ADU-S100 control tumor growth in a dual flank tumor model
To investigate the role of immune responses in primary tumor growth inhibition, we performed a dual flank study using an immunocompetent syngeneic model.In the dual flank syngeneic mouse tumor model, CT26 cells were implanted to both sides of a mouse and then one tumor was injected with reovirus, ADU-S100, or reovirus and 1 3 ADU-S100, while the other tumor remained untreated.The combination reovirus and ADU-S100 treatment induced significant regression of the treated tumor and the untreated distant tumor, suggesting a systemic immune response was activated by the combination treatment in a dual flunk tumor model (Fig. 4A).In contrast, IT injection of reovirus or ADU-S100 could not significantly delayed tumor growth in the untreated distal tumor.The combination treatment also significantly prolonged the survival of dual tumorbearing mice (Fig. 4B).Furthermore, we found the survival of tumor-bearing mice was determined by the growth of the untreated tumor in this model.These results demonstrate the abscopal effect mediated by tumor immunity induced by the focal IT injection of reovirus in combination with ADU-S100, and that control of the untreated tumor is essential for achieving long-term survival.

Microenvironment of tumors treated with reovirus and a STING agonist
To investigate whether the STING agonist restricted the propagation of reovirus in vivo, we performed immunofluorescence staining of reovirus protein in tumor samples extracted from mice.Three days after reovirus IT injection, reovirus protein was detected in the reovirus IT injected tumors but not in untreated distal tumor grafts (Fig. 5A).In addition to that, in tumors treated with the IT injection of reovirus followed by ADU-S100, the ratios of reovirus protein positive cells were almost the same as those in tumors injected with reovirus alone (Fig. 5B).These results indicate that pretreatment with a STING agonist did not restrict the propagation of reovirus in the tumor despite the activation of STING signaling, and that reovirus infection had not spread to the untreated distal tumor within three days after reovirus IT injection.Immunohistochemical staining demonstrated that a combination reovirus and ADU-S100 treatment induced significant apoptotic cell death (Fig. 5C,  D).Enhanced granzyme B staining and upregulation of IFN-β mRNA expression in untreated tumors of the combination group indicated the combination IT injection boosted cytotoxic immune responses (Fig. 5E, F).These results suggest the combination reovirus and ADU-S100 treatment enhanced antitumoral immune responses via type I IFNs without reovirus elimination.
To obtain an overview of the immune cell profile within the tumor microenvironment with each treatment, the expression of cancer immunology-related genes of the tumors was analyzed using the NanoString nCounter Pan-Cancer Immune Profiling.Gene clusters involved in immune cell functions, including adaptive immune cells, T cell function, and interferon responses, were found to be upregulated in the distal tumors of mice treated with reovirus combined with ADU-S100 compared to those in control (Fig. 6A).In contrast, a remarkable upregulation of cluster genes related to immune functions could not be detected in the distal tumors in the mice with reovirus or ADU-S100 single treatment.We also identified tumor-infiltrating lymphocytes using the NanoString nCounter data analysis and found that cytotoxic cells and CD8 cells scores were upregulated in the distal tumor in mice treated with the combination of reovirus and ADU-S100 compared to the distal tumor in mice treated with reovirus or ADU-S100 alone (Fig. 6B).These results imply that the enhanced anti-tumor activity in the tumor with a combination treatment reovirus and ADU-S100 was induced by accumulation of cytotoxic immune cells within tumor microenvironment.
To examine the profile of immune cells in the tumor microenvironment that might promote enhanced anti-tumor immunity in the combination group, we performed immunohistochemical staining of tumors treated with reovirus alone or in combination with ADU-S100.The proportions of M1 and M2 tumor-associated macrophages (TAMs) infiltrated in tumors were evaluated by determining M1 marker inducible nitric oxide synthesis (iNOS) and the M2 marker arginase-I with immunohistochemical staining.The expression of iNOS was not significantly upregulated in the tumor tissue of mice treated with the reovirus alone or in combination with ADU-S100 (Fig. 7A).The expression of arginase I was significantly downregulated in all treatment groups as compared with controls, and we detected the significant difference in the expression of arginase I between the mice treated with the reovirus alone and the mice treated with reovirus in combination with ADU-S100 (Fig. 7B).From these results, the polarization of TAMs seems to be partially involved in the anti-tumor immunity enhanced by the combination treatment.To further verify the immunotherapeutic effect of the T cell infiltration in tumor tissues of mice receiving the combination treatment, we also performed the tumor-infiltrating lymphocytes profiling in the tumor tissues of mice with the treatment.The number of CD8 + T cells was significantly increased in the treated and untreated tumors in all treatment groups compared with controls (Fig. 7D).By contrast, decreased numbers of CD4 + T cells were observed in the treated and untreated tumors of all treated groups (Fig. 7C).Next, we investigated the profile of immune cells in the tumor microenvironment using fluorescence-activated cell sorting (FACS).Tumor samples were collected and dissociated three days after IT injection of reovirus with or without pretreatment with ADU-S100, and subjected Fig. 1 The expression of cGAS/STING in murine colorectal cancer cell lines and the activation of STING signaling in CT26 cells.A Expressions of cGAS and STING protein in murine colon cancer cell lines evaluated by western blot analysis.β-actin is shown as a loading control.B Protein extracts from CT26 cells incubated in culture media supplemented with 50 μg ADU-S100, reovirus at a multiplicity of infection (MOI) of 10, or a combination of them were immunoblotted with anti-phospho-NAK/TBK1, anti-NAK/TBK1, anti-phospho-IRF3, and anti-IRF3 antibodies.β-actin is shown as a loading control.C The ratio of quantified P-IRF-3 and IRF-3 of western blot images of CT26 cells exposure to 50 mg ADU-S100 and/or reovirus at an MOI of 10 for 2 h were quantified using ImageJ software, n = 3. *p < 0.05, **p < 0.005, vs control.D Nuclear translocation of IRF3 in CT26 cells incubated in culture media supplemented with or without 50 μg ADU-S100 or reovirus at an MOI of 10 was evaluated by fluorescence immunostaining using an anti-IRF3 antibody (green) merged DAPI stained nuclei (blue).Scale bars: 10 μm.(E) Expression of IFN-β mRNA in CT26 cell lines incubated in culture media supplemented with or without reovirus at an MOI of 10 and/or 10 μg ADU-S100 for 2 h was evaluated by RT-PCR, n = 5. **p < 0.005, ***p < 0.001, ****p < 0.0001 ◂ 1 3 to FACS analysis.Significantly lower CD4:CD8 ratios of T cells were observed in the treated and untreated tumors from CT26 tumor-bearing mice treated with the combination reovirus and ADU-S100 IT injection compared with untreated controls, whereas no significant differences in the CD4:CD8 ratio were observed between the control and other treatment groups (Fig. 7E).We detect a significant reduction in the ratio of Foxp3 + CD45 + CD4 + CD3 + regulatory T cells in tumors treated with reovirus alone or the combination reovirus and AUD-S100 compared to the control (Fig. 7F).Taken together, local treatment with reovirus in combination with a STING agonist increased the levels of infiltrating CD8 + T cells in the treated tumor and untreated distal tumor.The significant tumor regression and survival benefit observed in combination treated mice were mediated by an inflamed tumor microenvironment associated with the increased numbers of infiltrating cytotoxic T cells.

Discussion
CRC is the third most diagnosed cancer worldwide according to the database of GLOBOCAN 2020 [20], and in Japan, it is a leading cause of cancer death.Although the outcome of inoperable CRC has improved after the development of biological therapies, it is still difficult to control CRC.In addition to the limited efficacy of conventional treatment for CRC, numbers of elderly CRC patients, who are intolerant to chemotherapy toxicity, are increasing.Under these circumstances, cancer immunotherapy has emerged as a promising next-generation cancer therapy to overcome obstacles related to conventional cancer treatment.Although cancer immunotherapy for microsatellite instability (MSI)high CRC using immune checkpoint inhibitors was recently approved [21], the proportion of MSI-high CRC is approximately 15%-16% of all CRC cases [22,23].Therefore, the Fig. 2 The efficacy of reovirus against murine colon cancer cell line, CT26.A Murine colon cancer cell line, CT26 cells were plated in 96-well plates, treated with reovirus, at a multiplicity of infection (MOI) of 3 or 10 and cell proliferation was measured by WST-8 assay, n = 5, ***p < 0.001, ****p < 0.0001, vs control.B CT26 were plated in 96-well plates, treated with reovirus at an MOI of 10 with or without 10 mg ADU-S100, and cell proliferation was measured by WST-8 assay, n = 5, ****p < 0.0001, ns, not significant.C CT26 cells were plated in 96-well plates, treated with 10 MOI reovirus for 48 h and 72 h, and extracellular ATP concentration in culture media was measured, n = 5. ****p < 0.0001 vs control.D CT26 cells were plated in 24-well plates, treated with 10 MOI reovirus for 48 h and 72 h, and HMGB-1 concentration in culture media was measured, n = 5. ****p < 0.0001 vs control.Data are presented as the means ± SD development of novel cancer immunotherapies effective for CRC irrespective of MSI status is indispensable to improving the outcome of CRC patients.
The oncolytic effect of reovirus against human CRC cell lines was previously reported [24] and another study demonstrated that CRC is a viable target for reovirus therapy mediated by direct and innate immune killing [25].As reported previously, reovirus induces immunogenic cell death [26] which might confer the utility of reovirus as a potential anti-tumor immune agent, and we also found that reovirus induced immunogenic cell death in CT26 (Fig. 2C, D).Preclinical studies using genetically modified reovirus [27,28] and combination treatment of reovirus with other anticancer agents [5,29,30] reported enhanced anti-tumor activity and tumor targeting of reovirus.Based on these studies, multiple phase I/II clinical trials of reovirus alone or in combination with chemotherapy against CRC have been conducted [31].Of note, a phase I study of reovirus against metastatic CRC investigated tumor immune responses, and the analysis of blood and tumor samples from patients revealed that reovirus had an immunomodulatory effect that enhanced the destruction of tumor cells in addition to its oncolytic effect against tumor cells [32].
In this study, we used STING expressing murine colorectal CT26 cancer cells that are sensitive to reovirus (Figs.1A,  2A).CT26 cells are used as a non-hypermutated/MSS model without mutations in mismatch repair genes [33].Evaluation of the tumor-immune profile of murine syngeneic tumor models demonstrated CT26 cells are moderately immunogenic, which may explain why CT26 cells are not sensitive to PD-1 based immunotherapy [34].Indeed, our study demonstrated a low level of CD8 + T cell infiltration within the non-treated CT26 tumor microenvironment (Fig. 6B, 7DE).Regarding the immunogenic profile of tumors, the CT26 syngeneic mouse model is appropriate for evaluating the efficacy of the combination reovirus and STING agonist treatment of CRC without an inflamed tumor microenvironment.
In the current study, we selected ADU-S100 as a STING agonist, because it is stable and activates all five human STING alleles [16].ADU-S100 induced the phosphorylation of IRF3 and its translocation into the nucleus of CT26 cells that express STING (Fig. 1B, C).It also strongly enhanced IFN-β mRNA expression in CT26 cells, which was concentration dependent (Supplementary Fig. 1).A previous study reported that the IT injection of ADU-S100 induced antitumor immune responses by inflaming the tumor microenvironment [35].Sivick et al. optimized the dose of ADU-S100 IT required to induce the local activation and systemic expansion of tumor-specific CD8 + T cells [36].In that study, a high dose of ADS-S100 ablated the IT injected tumor but diminished systemic tumor-specific T cell responses that are required for durable anti-tumor immunity.Therefore, we used a low dose, 10 mM, of ADU-S100, which enhanced systemic antitumor immunity.The IT injection Fig. 3 Tumor growth in a syngeneic mouse colon cancer model treated with reovirus with or without ADU-S100.A Experimental treatment timeline of subcutaneous CRC tumors in BALB/c mice.5.0 × 10 6 CT26 cells were subcutaneously (s.c.) injected into the right flank of BALB/c mice (5 per group).ADU-S100 IT injection was performed 9 days after transplantation of tumor cells in mice.IT treatment with reovirus at 3.0 × 10 8 pfu was conducted with or without administration of 10 μg ADU-S100 on days 12, 14, and 16 after transplantation of tumor cells in mice.B Tumor growth was measured twice a week after IT injection of reovirus with or without ADU-S100, * p < 0.05, **p < 0.005, ***p < 0.001, ****p < 0.0001.Data are presented as the means ± SD of ADU-S100 demonstrated anti-tumor efficacy in CT26bearing mice regardless of the single or dual flank model (Fig. 3A, 4A).However, the low dose IT injection of ADU-S100 did not significantly prolong survival in double tumorbearing mice compared with controls (Fig. 4B).Consistent with a previous study [36], our data suggest the efficacy of the IT injection of ADU-S100 alone might be limited as an inducer of systemic antitumor immunity.
Mammalian cells express cellular pattern recognition receptors, which detect the presence of pathogens or molecules released by damaged cells, resulting in the induction of type I IFN responses that restrict virus replication and dissemination [37].Reovirus has developed several mechanisms to attenuate type I IFN responses through inhibiting RNA sensors, including retinoic acid-inducible gene-I, melanoma differentiation-associated protein 5, and protein kinase R allowing it to evade the innate immune system [14].It is not well known about the involvement of the STING axis in immune responses induced by RNA viruses, including reovirus.STING is a key DNA sensing molecule that induces type I IFN production in response to DNA viruses and aberrant cytosolic double-stranded DNA originating from cancer cells [38,39].Previous studies demonstrated that the downstream signaling components of the RNA and DNA sensing pathways are physically and functionally interconnected [40,41].Recent studies also reported the role of STING in innate immune responses to RNA viruses, where it was required to restrict the replication of diverse RNA viruses including reovirus [42].In this study, we did not find the activation of the STING pathway in CRC cells exposed to reovirus (Fig. 1B, C).In addition to that, we demonstrated no significant upregulation of IFN-β mRNA expression in CT26 cells infected with reovirus and the inhibitory effect of the reovirus on IFN-β mRNA expression induced by the STING agonist (Fig. 1D).These results indicate that reovirus is not sensed by the STING signal, and reovirus and a STING agonist show a contradictory effect on type I IFN responses in our in vitro study.In our mouse model, we intratumorally injected reovirus combined with a STING agonist to CT26-bearing mice.Therefore, the interaction between reovirus and the STING agonist occurred within the tumor microenvironment.However, IFN-β mRNA expression was significantly upregulated in injected and distal tumors treated with the combination of reovirus and the STING agonist compared to control in CT26 syngeneic tumor model (Fig. 5F).Furthermore, we did not observe a difference in the expression of reovirus protein between tumors treated with an IT injection of reovirus Fig. 6 Transcriptome analysis of immunological effects within the tumor microenvironment with administration of reovirus combined with ADU-S100.A Total RNA of the tumors treated with an IT injection of reovirus at 3.0 × 10 8 pfu with or without 10 μg ADU-S100 was extracted, and multiplex gene expression analysis of isolated RNA was performed using the nConnter Mouse PanCancer Immune Profiling panel and the nSolver software.Heatmap represent the changes in the expression of immune function genes sets in the tumors with each treatment compared to control, n = 3. B-F Cell type differentiation scores obtained by analyzing isolated RNA from the tumors treated with an IT injection of reovirus at an MOI of 10 with or without 10 μg ADU-S100 are indicated, n = 3 alone or reovirus combined with ADU-S100 (Fig. 5A, B), indicating pretreatment with ADU-S100 does not restrict reovirus propagation within tumors during the limited observational period in this study.Additionally, our data indicate that the combination treatment did not increase the number of CD8 + T cells in the injected tumors compared to that in the injected tumors with a single therapy (Fig. 6B, 7D).Non-IFN-β mediated CD8 + T recruitment, including tumorassociated antigens and TNF-α induced by reovirus [8,43], may contribute to the status of CD8 + T cells in the injected tumors treated with reovirus alone, and the combination treatment could not take the advantage over STING agonist single therapy in terms of inflaming the tumor microenvironment in the injected tumor sites.In contrast, in distal tumor sites, the combination treatment of reovirus and ADU-S100 induced the accumulation of CD8 + T cells (Fig. 6E, 7D, 7E), which may account for the potent antitumor immune response of the combination therapy.IFN-β upregulation, high DC score (Fig. 6C), and cytotoxic cells score (Fig. 6F) in the distal tumor sites also support the significant role of CD8 + T cell-dependent antitumor immunity induced by combination therapy in our experimental model.However, gene expression analysis revealed several factors involved in the induction of an inflamed tumor microenvironment, independent of CD8 + T cell-mediated anti-tumor immunity (Fig. 6A).Therefore, additional studies are required to clarify the exact mechanism by which combination therapy boosts antitumor activity.
STING expression in cancer cells varies and is related to their sensitivity to DNA-damaging agents, including irradiation and cisplatin [44].It also defines the immunogenicity of tumors, which is associated with susceptibility to immunotherapy [45], and inactivation of STING signal in cancer cells to evade anti-cancer immunity has been reported in lung cancer [46].These reports indicate that the effect of STING agonists as anticancer agents is dependent on endogenous STING expression in cancer cells.However, it has also been reported that STING agonists show antitumor effects via activation of STING signaling in immune cells involved in antitumor immunity [16].Although our experimental model, in which we employed the STING-expressing mouse CRC cell line, is insufficient to investigate the effect of STING agonists in combination with reovirus on cancer immunity in terms of STING expression in cancer cells, the combination treatment has anti-tumor activity against STING expressing cancer, and it may activate antitumor immunity irrespective of STING expression in cancer.
A previous study reported that reovirus antigens, including the reovirus genome, released from tumors after reovirus treatment induce anti-tumor immunity by inhibiting the immunosuppressive activity of myeloid-delivered suppressor cells [47].In other studies using inactivated reovirus, anti-tumor immunity was primed in tumor-bearing immunocompetent mice independent of viral genome replication [9,48], and it was speculated that components of inactivated reoviruses, including dsRNA might trigger an immune response by being recognized by pattern recognition receptors or toll-like receptors.In contrast, inactivated reovirus, which contain reovirus antigens and genomes, shows no activation of anti-tumor immunity in some reports [7,49].The discrepancy among reports might be partially explained by cell type, mouse model, and administration route of the reovirus.In the current study, we did not assess the effect of inactivated reovirus on tumor immune responses.Further studies are needed to determine whether the reovirus component activates tumor immunity when used in combination with STING agonist.
In conclusion, we demonstrated oncolytic reovirusinduced tumor regression in a CT26 syngeneic tumor model, and that antitumor immunity was enhanced by combination treatment of reovirus and a STING agonist.This boosted antitumor immunity was partially mediated by IFN-induced T cell activation.Although further investigation is required to clarify the mechanism underlying STING agonist synergy with reovirus to provoke immune responses in the context of anti-tumor activity, our strategy using the combination treatment of reovirus and a STING agonist might have potential for immunotherapy against CRC.

Fig. 4
Fig. 4 Tumor growth in the CT26 cell double subcutaneous injection syngeneic mouse model.A Experimental treatment timeline of double subcutaneous CRC tumors in BALB/c mice.5.0 × 10 6 CT26 cells were subcutaneously injected into the right and left flank of BALB/c mice. 10 μg ADU-S100 IT injection was performed 9 days after transplantation of tumor cells (in the right flank tumor), and IT treatment with reovirus at 3.0 × 10 8 pfu was conducted with or without administration of ADU-S100 on days 12, 14, and 16 after transplantation of tumor cells in mice (in the right flank tumor).B 5.0 × 10 6 CT26 were subcutaneously injected into both flanks of BALB/c mice (5 per group).ADU-S100 IT injection was performed nine days after transplantation of tumor cells in mice.IT treatment with reovirus was administered with or without ADU-S100 IT injection on days 12, 14, and 16 after transplantation of tumor cells in mice.* p < 0.05, **p < 0.005, ***p < 0.001.Data are presented as the means ± SD.C Kaplan-Meier survival curve of the different treatment groups (5 mice per group).Mice were euthanized when the treated tumor or the untreated tumor volume reached 2 000 mm 3 , ****p < 0.0001 vs control