Dendritic cells (DC) play an essential role in innate immunity and radiation-elicited immune responses. LGP2 is a RIG-I–like receptor involved in cytoplasmic RNA recognition and antiviral responses. Although LGP2 has also been linked to cell survival of both tumor cells and T cells, the role of LGP2 in mediating DC function and antitumor immunity elicited by radiotherapy remains unclear. Here, we report that tumor DCs are linked to the clinical outcome of patients with breast cancer who received radiotherapy, and the presence of DC correlates with gene expression of LGP2 in the tumor microenvironment. In preclinical models, host LGP2 was essential for optimal antitumor control by ionizing radiation (IR). The absence of LGP2 in DC dampened type I IFN production and the priming capacity of DC. In the absence of LGP2, MDA5-mediated activation of type I IFN signaling was abrogated. The MDA5/LGP2 agonist high molecular weight poly I:C improved the antitumor effect of IR. This study reveals a previously undefined role of LGP2 in host immunity and provides a new strategy to improve the efficacy of radiotherapy.

Significance:

These findings reveal an essential role of LGP2 in promoting antitumor immunity after radiotherapy and provide a new strategy to enhance radiotherapy.

Radiotherapy (RT) is widely used as an anticancer treatment (1–3). The effects of ionizing radiation (IR) on tumors include genotoxic effects via the induction of DNA damage and direct killing of tumor cells and tumor-supporting stroma (2–4). Recent studies demonstrate that IR treatment modulates both innate and adaptive antitumor immunity, which are important for the local and systemic response of tumors to IR (3–12). IR is reported to induce cytosolic DNA and to initiate innate immune sensing through the cGAS and STING axis, which leads to expression of type I IFN, dendritic cell (DC) maturation, and induction of a potent adaptive immune response (10, 13, 14). IR is also reported to induce translocation of endogenous RNA to the cytoplasm in tumor cells and may account for some of the direct cytotoxic effects of IR on tumor cells (14).

Cytosolic RNA is recognized by pattern recognition receptors, including RIG-I–like receptors (RLR). The family of RLRs is comprised by three members RIG-I, MDA5, and LGP2, which recognize exogenous cytosolic RNA and initiate antiviral immunity. RIG-I and MDA5 contain aminoterminal tandem caspase activation and recruitment domains (CARD) that mediate activation of downstream signaling to induce the expression of type I IFN and immune-related genes (15). Following RNA binding, RIG-I and MDA5 undergo conformational modifications and signal downstream through direct CARD–CARD-mediated interactions with the mitochondrial-associated adaptor protein MAVS, which then initiates a signaling cascade mediated by transcription factors IRF-3 and NF-κB (15). In contrast, LGP2 (encoded by gene Dhx58) does not contain a CARD domain or other signaling domains and is both a negative and positive regulator of RNA sensing and signaling depending on context (15–22). For example, LGP2 can inhibit RIG-I–mediated signaling by competitive binding viral of dsRNA (21) as well as by preventing conformational modifications of RIG-I (22). However, several studies demonstrate that LGP2 functions as a positive cofactor of RLR signaling of innate immune defenses (18, 20). LGP2-deficient DCs produce less type I IFN and weaker antiviral immune response compared with wild-type (WT) DCs during RNA virus infection (18). LGP2 is also reported to facilitate the MDA5–RNA interaction and MDA5 filament assembly to promote specific activity of MDA5-mediated antiviral response (23).

LGP2 promotes the survival and cytokine production of activated CD8+ T cells in a virus infection model (24). The absence of LGP2 impaired IFNγ expression from T cells and sensitizes activated T cells to apoptosis (24). Recently, LGP2 was found to be a positive master regulator of numerous genes related to immunity and RLR signaling (25). However, whether and how cytosolic RNA sensing pathway involved in antitumor immune response elicited by RT remained to be explored.

We utilized bioinformatic tools to analyze the association between DC enrichment and clinical outcome of patients with breast cancers who received RT. Our results revealed that tumors that contained high levels of DCs have improved overall survival of patients with breast cancer when compared with patients with lower levels of DCs. Interestingly, we found that DC enrichment is correlated with gene expression of LGP2 and MDA5 but not RIG-I. We then employed Dhx58−/− mice, to examine the role of host LGP2 in the radiation-elicited immune response. We demonstrate that LGP2 is essential for type I IFN production in DCs after IR treatment. LGP2 promotes antitumor immunity and the antitumor effects of RT. A combination of MDA5/LGP2 agonist, high molecular weight (HMW) poly I:C enhanced the antitumor effects of IR. Our findings reveal an essential role of LGP2 in promoting RT elicited antitumor immunity and provide a new potential strategy to facilitate therapeutic efficacy by RT.

Mice

Six- to 8-week-old C57BL/6 mice were purchased from Harlan (currently Envigo) and used for the indicated experiments. 2C T-cell receptor (TCR) transgenic mice and OT-1 TCR transgenic mice were purchased from The Jackson Laboratory. Dhx58−/− (LGP2 KO) mice were generous gifts from Dr. Michael Gale, Jr. of the Department of Immunology at the University of Washington, Seattle, Washington. All mice were maintained under specific pathogen-free conditions and used in accordance with the animal experimentation guidelines set by the Institute of Animal Care and Use Committee of the University of Chicago.

Tumor growth and treatment

MC38 cells were obtained from the ATCC. Panc02 cells were kindly provided by Dr. Hans Schreiber (University of Chicago). Panc02 cells were infected by retrovirus with pMFG (SIY)3-Cerulean as described (26, 27). After infection, Panc02-SIY-Cerulean cells were FACS-sorted for low expression of SIY-Cerulean to generate the Panc02-Cerluean-SIYlo (Panc02-SIY) cell line. MC38-SIY cells were selected from a single clone of MC38 tumor cells after being transduced by lentivirus expressing human EGFR (L858)-SIY (10). The cell lines were authenticated by a short tandem repeat profile (IDEXX Bioresearch) within the last 6 months. All cell lines used were routinely screened for mycoplasma contamination using the Lonza MycoAlert Mycoplasma Detection Kit (CN LT07–418). Cells were cultivated and used within 20 passages.

Note that 1 × 106 MC38 tumor cells were s.c. injected into the flank of mice. Panc02-SIY fragment model was established as previously described (11). Tumors were measured and irradiated at 20 Gy as previously described. In vivo Animal Studies Tumor volumes were measured along three orthogonal axes (a, b, and c) and calculated as tumor volume = abc/2. For HMW poly I:C treatment, when MC38 tumor volume reached approximately 150 mm3, a single dose of 20 Gy was administered to the tumor locally to assigned groups, followed by intratumoral injection of HMW poly I:C precomplexed with a lipid carrier (Polyplus In vivo Jet-PEI), on days 0, 3, and 6 following local tumor irradiation. Tumor growth was monitored over time.

In vitro culture and function assay of bone marrow-derived dendritic cells

Single-cell suspensions of bone marrow cells were obtained from WT C57BL/6J and Dhx58−/− mice. The cells were cultured in RPMI-1640 medium containing 10% FBS, supplemented with 20 ng/mL GM-CSF. Fresh media with GM-CSF were added into culture on day 3. Bone marrow-derived dendritic cells (BMDC) were harvested for stimulation assay on day 7. MC38 and MC38-SIY tumor cell lines were grown in DMEM medium containing 10% FBS, at 37°C and 5% CO2. Note that 8 × 106 MC38-SIY cells were plated into 10 cm cell culture dishes overnight, then pretreated with 40 Gy, and incubated for 12 hours. BMDCs were added and cocultured with MC38-SIY cells at the ratio of 1:1 in the presence of fresh GM-CSF for an additional 6 to 8 hours. Subsequently, purified CD11c+ cells with EasySep Mouse CD11c Positive Selection Kit II (STEM CELL) were incubated with isolated CD8+ T cells from naïve 2C mice for 3 days. For IFNβ detection, 1 × 106 cells/mL purified CD11c+ cells (DCs) from coculture were seeded into 96-well plates, and the supernatants were harvested after 3 day of incubation.

Quantitative PCR

Total RNA from cocultured DCs was extracted by using the RNeasy Plus Mini Kit (Qiagen, 74134) and reversed-transcribed with Applied Biosystems High-Capacity cDNA Reverse Transcription Kit (Fisher Scientific, 4368814). Ifnb primers include Ifnb Fw (TGA ACT CCA CCA GCA GAC A) and Infb Rv (ACC ACC ATC CAG GCG TAG); Gapdh primers include Gapdh Fw (AGG TCG GTG TGA ACG GAT TTG) and Gapdh Rv (TGT AGA CCA TGT AGT TGA GGT CA). Real-time qPCR was performed by using Applied Biosystems Power SYBR Green PCR master mix (Fisher Scientific, 43–687–02) according to the manufacturer's instructions. Data were normalized to the RNA level of Gapdh. The 2−ΔΔCt method was used to calculate the relative expression changes.

Measurement of IFNγ-secreting CD8+ T cells by ELISPOT assay

Note that 2 × 104 purified CD11c+ cells were incubated with CD8+ T cells isolated from naïve 2C mice (EasySep Mouse CD8α Positive Selection Kit; STEMCELL) for 3 days at the ratio of 1:10. For tumor-specific CD8+ T-cell functional assay, MC38 tumor cells were exposed to 20 ng/mL murine IFNγ for 24 hours prior to plating with purified CD8+ T from DLN. CD8+ T cells (2 × 105) were incubated with MC38 at the ratio of 10:1 for 48 hours. ELISPOT assays were performed to detect the cytokine spots of IFNγ according to product protocol (Millipore).

ELISA

Cell culture supernatants were obtained from isolated CD11c+ cells after 48-hour incubation with fresh GM-CSF. The concentration of IFNβ was measured with VeriKine-HS Mouse Interferon Beta Serum ELISA Kit (PBL Assay Science) in accordance with the manufacturer's instructions.

Flow cytometric analysis

To obtain single-cell suspensions, tumor tissues were digested with 1 mg/mL Collagenase IV (Sigma) and 0.2 mg/mL DNase I (Sigma) for 45 minutes at 37°C. Single-cell suspensions were incubated with 2.4G2 to block antibody binding to the Fc receptors and then subsequently stained with conjugated antibodies: anti-CD45.2 (clone 104), anti-CD90.2 (clone 30-H12), anti-CD8a (clone 53–6.7), anti-CD11c (clone HL3), anti-CD11b (clone M1/70), anti-Ly6C (clone HK1.4), anti-Ly6G (clone 1A8), anti-CD4 (clone GK1.5), and anti–PD-1 (clone 29F.1A12). For intracellular staining of IFNγ, after surface marker staining, cells were fixed and stained with anti-IFNγ antibody (clone XMG1.2). Antibodies used for cell surface marker staining of BMDCs include anti–PD-L1 (clone MIH5), anti–MHC I (clone AF6–88.5), and anti–MHC II (clone M5/114.15.2). FITC–Annexin V and propidium iodide staining were performed following the manufacturer's instruction of Annexin V Apoptosis Detection Kit (eBioscience). All other purified and fluorescently labeled monoclonal antibodies were purchased from BioLegend. Samples were analyzed on an LSRFortessa Flow Cytometer (BD), and data were analyzed with FlowJo Software (TreeStar).

Database and statistical analysis

Clinical and normalized gene expression data in METABRIC dataset (28, 29) were download from cbioportal (https://www.cbioportal.org/). Triple-negative breast cancer (TNBC) subtypes were defined based on the “ER status,” “PR status,” and “HER2 status,” with all the three stated as “Negative.” Luminal A subtype was based on the information of the “CLAUDIN_SUBTYPE” as “LumA.” In the TNBC and Lumina A subtype of METABRIC, the patients who received RT were selected out for survival analysis if “Radio Therapy” was stated as “Yes.” The gene expression dataset was used to estimate cell subtypes using xCell (30). Gene set enrichment analysis (GSEA) was performed with the software from broad institute (http://software.broadinstitute.org/gsea/index.jsp). The gene sets used for GSEA were obtained from MSigDB (https://www.gsea-msigdb.org/gsea/index.jsp), including all gene sets in the category of “Hallmark” and “KEGG,” as well as RNA response/binding-related gene sets in “GO” Category. Gene sets enriched in DC Hi Group with FDR < 0.05 and Nom P < 0.05 were included in the results table. Normalized gene expression (RNA sequencing) and corresponding clinical data of patients with breast cancer, colon adenocarcinoma, pancreatic adenocarcinoma, melanoma (SKCM), or sarcoma were obtained from The Cancer Genome Atlas (TCGA) through the UCSC Cancer Genomics Browser (31). Additional survival analyses of colon cancer and pancreatic cancer were performed on R2 Genomic platforms (https://hgserver1.amc.nl/cgi-bin/r2/main.cgi) using published studies (32, 33). Survival curves were compared by the log rank (Mantel–Cox) test.

Tumor growth curves were assessed by repeated-measure ANOVA. Differences between two groups were analyzed by a two-tailed unpaired Student t test. All statistics analyses were performed using GraphPad Prism 6.0 or SAS9.4 (SAS Institute Inc.). P < 0.05 denotes differences that are statistically significant.

Tumor DC enrichment is linked to prolonged survival and associated with LGP2 expression in patients with breast cancer treated with RT

To examine whether tumor-infiltrating DCs are associated with the clinical outcome in a population of patients with breast cancer treated with RT, we analyzed gene expression and clinical data obtained from a publicly available METABRIC database (28, 29). The gene expression data were used to compute the enrichment scores of cell types by xCell analysis (30). We analyzed the survival data of patients with Luminal A and TNBC who received RT. In both subsets, the group with high enrichment score of DCs (DC Hi) has a significantly prolonged overall survival than the group with a low score of tumors DCs (DC Lo; Fig. 1A and B). The difference of survival in TNBC subsets appears greater than that in luminal A (HR = 1.928 vs. HR = 1.43). We then further analyzed the gene expression data in the TNBC subset. GSEA revealed that, besides gene sets related to DC functions, gene sets that related to the response to dsRNA and RLR pathway (including LGP2) were also enriched in DC Hi group compared with DC Lo group (Fig. 2A; Supplementary Tables S1 and S2). We further compared the gene expression of cytosolic RNA sensors in these two groups and discovered that both MDA5 and LGP2, but not RIG-I, were highly expressed in tumors containing high level of DCs compared with tumors with lower levels of DCs (Fig. 2B). We found that the expression of LGP2 was correlated with the enrichment scores of DC and CD8+ T cells, respectively (Fig. 2C and D). There is a strong correlation between gene expression of DHX58 and CD8A (Fig. 2E). These data suggest that LGP2 is associated with antitumor immunity in patients with human cancer with RT.

Figure 1.

Tumor DC is linked to the overall survival of patients with breast cancer who received RT. Kaplan–Meier plots of overall survival over time for patients with luminal A breast cancer (A) or TNBC (B) with RT. Data are from the METABRIC public dataset, n = 367 donors (Luminal A) and n = 214 donors (TNBC). Patients in each subset were split into two groups by median value of DC scores in each subset. DC scores were calculated by xCell (see Materials and Methods).

Figure 1.

Tumor DC is linked to the overall survival of patients with breast cancer who received RT. Kaplan–Meier plots of overall survival over time for patients with luminal A breast cancer (A) or TNBC (B) with RT. Data are from the METABRIC public dataset, n = 367 donors (Luminal A) and n = 214 donors (TNBC). Patients in each subset were split into two groups by median value of DC scores in each subset. DC scores were calculated by xCell (see Materials and Methods).

Close modal
Figure 2.

LGP2 is enriched in tumors containing high level of DCs and correlated with CD8+ T cells. The gene expression data of patients with TNBC from METABRIC dataset were used for analysis. A, GSEA plots of the gene expression data of DC Hi and DC Lo group. NES, normalized enrichment score. Wald test with Benjamini–Hochberg's multiple comparison. B, Gene expression of RNA sensors RIG-I (DDX58), MDA5 (IFIH1), and LGP2 (DHX58) expression in DC Hi vs. DC Lo group. Unpaired Student t test. C, Linear regression correlations between LGP2 (DHX58) and DC scores. D, Linear regression correlations between LGP2 (DHX58) and CD8+ T cells were analyzed using Pearson test. E, Linear regression correlations between gene expression of LGP2 (DHX58) and CD8A were analyzed using Pearson test. Enrichment scores for DC and CD8+ T cells were determined by xCell analysis. Correlations were analyzed using Pearson test.

Figure 2.

LGP2 is enriched in tumors containing high level of DCs and correlated with CD8+ T cells. The gene expression data of patients with TNBC from METABRIC dataset were used for analysis. A, GSEA plots of the gene expression data of DC Hi and DC Lo group. NES, normalized enrichment score. Wald test with Benjamini–Hochberg's multiple comparison. B, Gene expression of RNA sensors RIG-I (DDX58), MDA5 (IFIH1), and LGP2 (DHX58) expression in DC Hi vs. DC Lo group. Unpaired Student t test. C, Linear regression correlations between LGP2 (DHX58) and DC scores. D, Linear regression correlations between LGP2 (DHX58) and CD8+ T cells were analyzed using Pearson test. E, Linear regression correlations between gene expression of LGP2 (DHX58) and CD8A were analyzed using Pearson test. Enrichment scores for DC and CD8+ T cells were determined by xCell analysis. Correlations were analyzed using Pearson test.

Close modal

LGP2 in host cells is essential for antitumor effect of RT

To precisely define the role of host LGP2 in antitumor immunity, we grew the murine colon tumor cell MC38 in LGP2-deficient (Dhx58−/−) mice and control (Dhx58+/−) mice. There is no significant difference in tumor growth without treatment. Consistent with our previous studies (10, 12), local IR (20 Gy) delays tumor growth in control mice. The tumor grown in Dhx58−/− mice, however, did not respond to RT compared with control mice (Fig. 3A). Similar findings were also observed with Panc02-SIY fragment model, a model we previously generated (Fig. 3B; ref. 11). CD8+ T-cell accumulation and function are important for the antitumor immunity elicited by RT and are emerging as important determinants of the antitumor effects of IR. We analyzed the cell frequency and function of CD8+ T cells within the tumor microenvironment. IR-induced accumulation of tumor-infiltrating CD8+ T cells is significantly reduced in Dhx58−/− mice compared with that in Dhx58+/− mice (Fig. 3C). Moreover, intracellular staining revealed that IR-induced IFNγ from CD8+ T cells was decreased in Dhx58−/− mice compared with controls (Fig. 3D), suggesting the function of CD8+ T cells is impaired in the absence of LGP2.

Figure 3.

LGP2 is essential for the therapeutic efficacy of IR and accumulation of functional CD8+ T cells in tumor. A, Tumor growth curve of MC38 in LGP2 knockout (Dhx58−/−) and their littermate Dhx58+/− controls. Tumors were left untreated or treated with 20 Gy irradiation. **, P < 0.01, two-way ANOVA. B, Tumor growth curve of Panc02-SIY fragment tumors in Dhx58−/− and their littermate Dhx58+/− controls. Tumors were left untreated or treated with 20 Gy irradiation. Error bars, mean ± SEM. *, P < 0.05, compared with Dhx58+/− IR group (two-way ANOVA). C, CD8+ T-cell percentage among CD45+ cells derived from tumors on day 17 (day 7 post IR). D, Intracellular staining of IFNγ in tumor-infiltrating CD8+ T cell on day 17 (day 7 post IR). Representative histogram (left) and summary of results (right) are presented. *, P < 0.05; **, P < 0.01; N.S., not significant. Unpaired Student t test.

Figure 3.

LGP2 is essential for the therapeutic efficacy of IR and accumulation of functional CD8+ T cells in tumor. A, Tumor growth curve of MC38 in LGP2 knockout (Dhx58−/−) and their littermate Dhx58+/− controls. Tumors were left untreated or treated with 20 Gy irradiation. **, P < 0.01, two-way ANOVA. B, Tumor growth curve of Panc02-SIY fragment tumors in Dhx58−/− and their littermate Dhx58+/− controls. Tumors were left untreated or treated with 20 Gy irradiation. Error bars, mean ± SEM. *, P < 0.05, compared with Dhx58+/− IR group (two-way ANOVA). C, CD8+ T-cell percentage among CD45+ cells derived from tumors on day 17 (day 7 post IR). D, Intracellular staining of IFNγ in tumor-infiltrating CD8+ T cell on day 17 (day 7 post IR). Representative histogram (left) and summary of results (right) are presented. *, P < 0.05; **, P < 0.01; N.S., not significant. Unpaired Student t test.

Close modal

Given that LGP2 was reported to promote survival of activated CD8+ T-cell during RNA virus infection (24), we next examined whether LGP2 is also required for the survival of antigen-specific CD8+ T cells in our tumor model. We harvested CD8+ T cells from the MC38 tumor 3 days after IR and performed Annexin V staining and flow cytometric analysis. Tumor-infiltrating CD8+ T cells from tumors grown in LGP2-deficient mice did not contain a higher frequency of Annexin V+ (apoptotic) cells (Supplementary Fig. S1A). To further investigate the role of LGP2 in T-cell survival, we injected MC38-SIY cells into Dhx58+/− and Dhx58−/− mice and harvested splenocytes 7 days later for SIY peptide restimulation. SIY-specific T cells were then treated with a titration dose of IR, and the apoptotic status was monitored by Annexin V staining. No difference was observed in the IR-induced apoptosis between WT and Dhx58−/− SIY-specific CD8+ T cells (Supplementary Fig. S1B). These results suggested that in the tumor model the absence of LGP2 does not affect T-cell survival.

LGP2 is important for IFNβ induction and cross-priming capability of DC following irradiation

Given that DCs play important roles in cytosolic nucleic acid sensing and RT-elicited antitumor immunity, we next examined whether the DC phenotype and function are altered in the absence of LGP2. We conducted an in vitro cross-priming assay with BMDCs from WT and LGP2-deficient mice. The phenotype and apoptotic status of BMDCs were characterized by flow cytometry. No differences were observed in the expression of PD-L1, MHC I, and MHC II (Supplementary Fig. S2A) or apoptosis of BMDCs (Supplementary Fig. S2B) between groups. The cross-priming capability of WT DCs was enhanced following stimulation with irradiated tumor cells, whereas the absence of LGP2 in DCs dampened DCs' capability to cross-prime naïve T cells (Fig. 4A). To determine whether exogenous IFNβ treatment rescues the function of LGP2-deficient BMDCs, we added IFNβ to a coculture system of BMDCs and irradiated tumor cells. Our results demonstrate that the ability of LGP2-deficient BMDCs to cross-prime T cells is restored with the addition of exogenous IFNβ (Fig. 4B).

Figure 4.

LGP2 is essential for IFNβ production and priming capability of DC. A, WT and LGP2 KO (Dhx58−/−) BMDCs were cocultured with 40 Gy pretreated or untreated MC38-SIY cells. Subsequent purified CD11c+ cells were incubated with naïve CD8+ 2C T cells and followed by ELISPOT detection of IFNγ production. B, WT and Dhx58−/− BMDCs were cocultured with 40 Gy pretreated MC38-SIY cells, with or without additional 20 ng of recombinant mouse IFNβ. Subsequent purified CD11c+ cells were incubated with naïve CD8+ 2C T cells and followed by ELISPOT detection of IFNγ production. C, WT and Dhx58−/− BMDCs were cocultured with 40 Gy–pretreated or untreated MC38-SIY cells. Subsequent purified CD11c+ cells were incubated for additional 2 days, and the supernatants were collected to measure IFNβ by ELISA assay. D, 40 Gy pretreated or untreated MC38-SIY cells were cultured for 12 hours. The resulting supernatants were collected and added (1:5) into culture medium of WT or Dhx58−/− BMDCs. BMDCs were incubated for additional 2 days, and the supernatants were collected to measure IFNβ by ELISA assay. *, P < 0.05; **, P < 0.01; N.S., not significant. Unpaired Student t test.

Figure 4.

LGP2 is essential for IFNβ production and priming capability of DC. A, WT and LGP2 KO (Dhx58−/−) BMDCs were cocultured with 40 Gy pretreated or untreated MC38-SIY cells. Subsequent purified CD11c+ cells were incubated with naïve CD8+ 2C T cells and followed by ELISPOT detection of IFNγ production. B, WT and Dhx58−/− BMDCs were cocultured with 40 Gy pretreated MC38-SIY cells, with or without additional 20 ng of recombinant mouse IFNβ. Subsequent purified CD11c+ cells were incubated with naïve CD8+ 2C T cells and followed by ELISPOT detection of IFNγ production. C, WT and Dhx58−/− BMDCs were cocultured with 40 Gy–pretreated or untreated MC38-SIY cells. Subsequent purified CD11c+ cells were incubated for additional 2 days, and the supernatants were collected to measure IFNβ by ELISA assay. D, 40 Gy pretreated or untreated MC38-SIY cells were cultured for 12 hours. The resulting supernatants were collected and added (1:5) into culture medium of WT or Dhx58−/− BMDCs. BMDCs were incubated for additional 2 days, and the supernatants were collected to measure IFNβ by ELISA assay. *, P < 0.05; **, P < 0.01; N.S., not significant. Unpaired Student t test.

Close modal

We then sought to determine whether the absence of LGP2 impairs IFNβ induction in DCs after being cocultured with irradiated tumor cells. IFNβ production in WT BMDCs was elevated by being cocultured with irradiated tumor cells. The induction of IFNβ was significantly impaired in LGP2-deficient BMDCs (Fig. 4C). Next, we examined whether the IFNβ induction was mediated by the secreted factors from the irradiated tumor cells. The supernatant from culture medium of irradiated or untreated tumor cells failed to induce IFNβ in BMDCs (Fig. 4D), suggesting the induction of IFNβ requires cell–cell contact between tumor cells and DCs. These results suggested LGP2 is essential for DCs to sense stimuli from irradiated tumor cells and for their capability to prime T cells.

HMW poly I:C administration increases the therapeutic efficacy of IR in an LGP2-dependent manner

Given that LGP2 is known as an RNA sensor, we then investigated whether LGP2-deficient DCs confer a differential response to RNA agonists. We treated the WT and LGP2-deficient DCs with RIG-I or MDA5-specific agonists. Both WT and LGP2-deficient BMDCs produced high levels of IFNβ following stimulation with the RIG-I agonist, 5′ppp-dsRNA (Fig. 5A). In contrast, after treatment with an MDA5 agonist, HMW poly I:C, LGP2-deficient DCs produced less IFNβ compared with WT BMDCs (Fig. 5B and C). Activation of MDA-mediated signaling promotes an antiviral and antitumor immune response (15). To test whether HMW poly I:C, an MDA5 agonist, can improve the therapeutic efficacy of IR, we administrated the synthetic HMW poly I:C with cytosolic delivering reagent intratumorally following irradiation. Treatment of HMW poly I:C alone modestly inhibited tumor growth of WT mice (Fig. 5D) but had no antitumor effects on tumors in LGP2-deficient mice (Fig. 5E). Treatment with the combination of IR and three doses of 10 μg of HMW poly I:C demonstrates better suppression of tumor growth compared with IR alone or HMW poly I:C alone, suggesting HMW poly I:C potentiates the antitumor effect of IR (Fig. 5F). Interestingly, the therapeutic effect of this combination was abrogated in LGP2-decificent mice (Fig. 5F). The results indicate that HMW poly I:C promotes the therapeutic efficacy of irradiation, and the combination effect is dependent on the presence of LGP2 in host cells.

Figure 5.

HMW poly I:C promotes IFN production and tumor control of RT in LGP2-dependent manner. A, WT and Dhx58−/− BMDCs were treated with 5′ppp-dsRNA at indicated concentration for 24 hours. Resulting supernatants were collected to measure IFNβ by ELISA assay. B, WT and Dhx58−/− BMDCs were treated with 2 μg/mL HMW poly I:C concentration for 24 hours. Resulting supernatants were collected to measure IFNβ by ELISA assay. C, WT and Dhx58−/− BMDCs were treated with HMW poly I:C concentration for 6 hours. BMDCs were collected for qPCR measurement of IFNβ mRNA. A–C, Data are represented as mean ± SEM. *, P < 0.05; **, P < 0.01, unpaired Student t test. D–F, MC38 tumors were established in WT and Dhx58−/− mice. D, Tumors grown in WT mice were intratumorally injected with a mixture of 10 μg HMW poly I:C or left untreated on day 7. E, Tumors grown in Dhx58−/− mice were intratumorally injected with a mixture of 10 μg HMW poly I:C or left untreated on day 7. F, Tumors were intratumorally injected with a mixture of 10 μg HMW poly I:C and in vivo transfect reagent JET-PEI on days 0, 3, and 6 after IR treatment. Data are represented as mean ± SEM. **, P < 0.01, ANOVA test.

Figure 5.

HMW poly I:C promotes IFN production and tumor control of RT in LGP2-dependent manner. A, WT and Dhx58−/− BMDCs were treated with 5′ppp-dsRNA at indicated concentration for 24 hours. Resulting supernatants were collected to measure IFNβ by ELISA assay. B, WT and Dhx58−/− BMDCs were treated with 2 μg/mL HMW poly I:C concentration for 24 hours. Resulting supernatants were collected to measure IFNβ by ELISA assay. C, WT and Dhx58−/− BMDCs were treated with HMW poly I:C concentration for 6 hours. BMDCs were collected for qPCR measurement of IFNβ mRNA. A–C, Data are represented as mean ± SEM. *, P < 0.05; **, P < 0.01, unpaired Student t test. D–F, MC38 tumors were established in WT and Dhx58−/− mice. D, Tumors grown in WT mice were intratumorally injected with a mixture of 10 μg HMW poly I:C or left untreated on day 7. E, Tumors grown in Dhx58−/− mice were intratumorally injected with a mixture of 10 μg HMW poly I:C or left untreated on day 7. F, Tumors were intratumorally injected with a mixture of 10 μg HMW poly I:C and in vivo transfect reagent JET-PEI on days 0, 3, and 6 after IR treatment. Data are represented as mean ± SEM. **, P < 0.01, ANOVA test.

Close modal

IR-induced innate immune response relies on the production of type I IFN, which facilitates DC maturation and the priming of cytotoxic T cells. Type I IFN receptor–deficient mice generate impaired antitumor immunity and fail to respond to RT (8). Mechanisms by which type I IFN is induced following radiation have been intensely studied (3, 4, 13). These studies showed that IR treatment of tumor initiated cGAS-STING–mediated cytosolic DNA sensing pathways, which activated the transcription factor IRF3 and NF-κB, resulting in elevated expression of type I IFNs (10). In current study, we demonstrate that LGP2 in host cells is essential for the antitumor effects of RT. Following stimulation with irradiated tumor cells, LGP2 knockout DCs are deficient in the upregulation of IFNβ production and priming of CD8+ T cells, whereas addition of exogenous IFNβ rescued the defect. Interestingly, human cancer database analysis suggested that DC is correlated with LGP2 expression and predicted better prognosis after RT in patients with TNBC. Our results revealed a previously unidentified role of host LGP2 in antitumor immunity and provided a new potential method to improve RT efficacy by combination of MDA5/LGP2 agonist(s) and IR.

RLRs are the major receptors sensing cytosolic RNA. The role of RIG-I, MDA5, and LGP2 has been well studied in virus infection models (15). RIG-I and MDA5 have differential recognition of viruses, which is attributed to their distinct preferences for RNA ligands. RIG-I preferentially senses 5′ triphosphated dsRNA (5′ppp-dsRNA) and short chain dsRNA, whereas MDA5 binds to long dsRNA, such as HMW poly I:C (17, 34). Following ligand encounter, RIG-I and MDA5 are recruited to adaptor protein MAVS and activate a series of downstream signaling events that lead to type I IFN production. The studies of LGP2 during viral infection have generated varied results in different contexts, which relate to the RNA properties of the virus (18, 20–22). Emerging evidence supports the concept that LGP2 inhibits RIG-I–mediated signaling and enhances MDA5-initiated signals (35). For example, LGP2 is reported to suppress RIG-I signaling through prevention of RIG-I binding to mitochondrial MAVS (21), whereas other structural and functional studies showed that LGP2, through cooperative binding, facilitates the nucleation of MDA5 oligomerization on dsRNA and promotes MDA5 signaling (23). We previously reported that LGP2 acted as a negative feedback molecule in some tumor cells to inhibit radiation-enhanced and RIG-I–mediated IFNβ expression (36). In the current study, we also sought to determine if LGP2 modulates RIG-I–mediated signaling in DC. Interestingly, LGP2-deficient BMDCs responded to a RIG-I–specific agonist in a similar pattern as WT BMDCs. In contrast, LGP2-deficient BMDCs respond poorly to a MDA5 agonist HMW poly I:C, suggesting that LGP2 partner with MDA5 for DC maturation within this setting. Noteworthy, in patients with TNBC, we also observed that both MDA5 and LGP2, but not RIG-I, were enriched in tumors containing high level of DCs.

Target activation of RLRs has been considered as a promising strategy for tumor immunotherapy (15, 36–39). An RIG-I agonist, 5′ppp-dsRNA has been reported to induce apoptosis in human melanoma cells (38). MDA5-based immunotherapy by intratumoral injection of HMW poly I:C prolonged survival of mice bearing murine pancreatic cancer (40). Most of these studies focus on direct targeting of tumor cells via induction of cell apoptosis. In our studies, however, HMW poly I:C alone did not demonstrate a significant therapeutic effect in controlling tumor growth of murine colon cancer. Previous studies reported that poly I:C in combination with radiation improved tumor control (41–43). We similarly demonstrated that combination of HMW poly I:C and IR achieved a greater control of tumors than IR or HMW poly I:C alone in WT mice. Though we did not rule out possible contribution from cell apoptosis of tumor cells potentially induced by the injection of poly I:C, it is convincing that LGP2 from host cells is critical for the combination treatment of IR and poly I:C, as this combination fails to control tumor growth well in LGP2-deficient mice. Yoshida and colleagues reported that, in their experimental systems, the efficacy of poly I:C and RT was mainly mediated by TLR3 but has minimal contribution from the MDA5-MAVS pathway (42). However, there are several differences between the work by Yoshida and colleagues and this study. First, Yoshida and colleagues administrated poly I:C intraperitoneally, and we injected poly I:C intratumorally. Second, we delivered the poly I:C together with in vivo-jetPEI, which is a transfection reagent designed for the delivery of nucleic acids in animal models. These differences may lead to different level of poly I:C presence in the cytosol and exert differential effects on the MDA5/LGP2-MAVS pathway. In combination with our in vitro experiment, our results demonstrate that HMW poly I:C can improve RT efficacy in a LGP2-dependent manner. Although we used poly I:C as a pharmacologic validation of our findings, there are several clinical trials that suggest poly I:C may be useful clinically (44, 45). These early data taken together with our results suggest that LGP2 agonists may be effective in enhancing RT especially if these agents can be directed toward the antigen-presenting cells.

Recent studies revealed that LGP2 has broader and more complex biological functions beyond RNA recognition (24, 25, 46, 47). LGP2 can either negatively or positively regulate immune signaling, depending on the biological context (25, 46). Parisien and colleagues indicated that LGP2 acts downstream of MAVS and suppresses TRAF-mediated signaling in HEK293 and MEF cells (46). By contrast, Liu and colleagues revealed that LGP2 functions as positive master regulator for several innate immune genes, including IL6 and CXCL10, in HEp-2 cells (25). These different roles of LGP2 likely are attributed to different cells and stimuli utilized in these studies. We also examined the ubiquitination and level of TRAFs in BMDCs after coculture with irradiated tumor cells or HMW poly I:C but did not observe differences between LGP2-deficient and WT DCs. LGP2 also promoted cell survival and IFNγ production of antigen-specific CD8+ T cells during RNA virus infection (24). In our tumor model, the results did show LGP2 deficiency reduced IFNγ production from tumor-infiltrating CD8+ T cells. However, we did not observe any significant difference regarding the apoptosis of CD8+ T cells following IR exposure in vivo and in vitro. These results, together with our data from T-cell priming experiments (Fig. 4), suggest that LGP2 modulates function of tumor-infiltrating CD8+ T cells via regulating priming capability of DCs but does not directly affect T-cell survival in the response to RT. Whether other LGP2 modulates function of other immune cells (e.g., macrophage) and affects the response to RT needs future investigation.

DCs are important in mediating innate and adaptive immunity in the tumor microenvironments. Some DC signatures were positively correlated with positive outcome in several cancer types (48, 49). In this study, we focus the subsets of patients with breast cancer who received RT. We observed that proportions of patients with TNBC who received chemotherapy are similar between the two groups (71/107 in DC Hi vs. 65/107 in DC Lo). We found that DC enrichment is associated the prognosis after RT in both luminal A and triple-negative subsets of breast cancer patients (TNBC). Interestingly, we noticed that in TNBC, DC scores has a greater association with the prognosis, suggesting the role of DCs in different subsets of breast cancer may vary. In TNBC tumors, we revealed that LGP2 is correlated with both DC and CD8+ T-cell scores, suggesting an association between LGP2 and the antitumor immunity in human cancers.

To evaluate if LGP2 expression alone can predict prognosis, we analyzed the survival of patients with breast cancer. No significant difference was observed between patients with high LGP2 expression and the group with low LGP2 expression in either Luminal A or TNBC subsets who received RT (Supplementary Fig. S3A and S3B). We further analyzed other cancer types from TCGA and other public datasets. In an analysis of the TCGA database, we found that, in several human cancer types (including melanoma and sarcoma), relatively greater expression of LGP2 predicted prolonged overall survival of patients compared with patients with lower LGP2 expression (Supplementary Fig. S4A and S4B). Furthermore, our results showed that high expression of LGP2 is correlated several signature transcription factors and gene of DC (Supplementary Fig. S5A–S5C), as well as functional markers of CD8+ T cells (Supplementary Fig. S5D–S5F). We also observed some mixed results in some other cancer types. LGP2 was negatively correlated with overall survival in the colorectal cancer datasets from TCGA, but positively correlated with survival in another publicly available dataset (Supplementary Fig. S6A and S6B). Similarly, LGP2 did not correlate with overall survival in pancreatic cancers in the dataset from TCGA but have a positive correlation with survival of patients in another dataset (Supplementary Fig. S6C and S6D).

The mixed results in the correlation analysis between LGP2 expression and clinical outcome of human cancers may be partially attributed to the nature of cancer heterogeneity as well as differences of tumor–host relationships among individuals. In addition, it is also acknowledged that a single-gene expression may be of limited prognostic value compared with gene sets or gene signatures with various genes. One of the reasons is that many genes might have distinct roles in different cell populations inside the tumor microenvironment. We have previously reported in some tumor cells that the absence of LGP2 enhances radiation-induced killing (36). It is noteworthy that LGP2 level in these analyses is the average level in bulk tumor tissue, which is also a limitation of our current database analysis. We speculate that when LGP2 expression in tumor tissue is mainly contributed by immune cells (such as DCs), high levels of LGP2 expression predict prolonged survival and an increased response to RT, whereas if LGP2 is expressed principally in tumor cells, this may be associated with poor outcome. These data point to the complexity of LGP2 function in the face of radiation and potential differences between the host and tumor function of the RNA cytosolic sensors.

In conclusion, we reported here that host LGP2 is required for the function of DCs and the therapeutic effect of RT. Our data demonstrated that LGP2 is a positive regulator of MDA5-mediated innate immune sensing and type I IFN by DCs in response to irradiated tumor cells. Moreover, a synthetic MDA5/LGP2 agonist HMW poly I:C treatment improves the efficacy of RT. In addition, high-level expression of LGP2 was associated with DC enrichment, which predicted prolonged survival and response to RT in patients with breast cancer. Our findings provided novel insights and potential strategy to improve RT.

R.R. Weichselbaum reports other compensation from Boost Therapeutics, from Immvira LLC, from Reflexion Pharmaceuticals, from Coordination Pharmaceuticals, from Magi Therapeutics, from Oncosenescence, from AstraZeneca, from Aettis Inc, from Coordination Pharmaceuticals, from Genus, from Merck Serono S.A., from Nano Proteagen, from NKMax America Inc, from Shuttle Pharmaceuticals, and from Highlight Therapeutics, S.L., grants from Varian, grants from Regeneron, and personal fees and other compensation from Boehringer Ingelheim Ltd. outside the submitted work; in addition, R.R. Weichselbaum has a patent (Methods and Kits for Diagnosis and Triage of Patients With Colorectal Liver Metastases) pending. No disclosures were reported by the other authors.

W. Zheng: Conceptualization, data curation, validation, investigation, methodology, writing-original draft, project administration, writing-review and editing. D.R.E. Ranoa: Investigation, methodology. X. Huang: Investigation, methodology. Y. Hou: Investigation, methodology. K. Yang: Investigation, methodology. E.C. Poli: Investigation, methodology. M.A. Beckett: Investigation, methodology. Y.-X. Fu: Resources, supervision, funding acquisition, project administration, writing-review and editing. R.R. Weichselbaum: Conceptualization, resources, supervision, funding acquisition, project administration, writing-review and editing.

We thank Rolando Torres for technical assistance and members in Weichselbaum lab for helpful discussions. This work was supported in part by a generous gift from The Foglia Foundation and an endowment from the Ludwig Cancer Research Foundation to R.R. Weichselbaum. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the article.

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

1.
Begg
AC
,
Stewart
FA
,
Vens
C
. 
Strategies to improve radiotherapy with targeted drugs
.
Nat Rev Cancer
2011
;
11
:
239
53
.
2.
Liauw
SL
,
Connell
PP
,
Weichselbaum
RR
. 
New paradigms and future challenges in radiation oncology: an update of biological targets and technology
.
Sci Transl Med
2013
;
5
:
173sr2
.
3.
Spiotto
M
,
Fu
YX
,
Weichselbaum
RR
. 
The intersection of radiotherapy and immunotherapy: mechanisms and clinical implications
.
Sci Immunol
2016
;
1
:
EAAG126
.
4.
Weichselbaum
RR
,
Liang
H
,
Deng
L
,
Fu
YX
. 
Radiotherapy and immunotherapy: a beneficial liaison?
Nat Rev Clin Oncol
2017
;
14
:
365
79
.
5.
Twyman-Saint Victor
C
,
Rech
AJ
,
Maity
A
,
Rengan
R
,
Pauken
KE
,
Stelekati
E
, et al
Radiation and dual checkpoint blockade activate non-redundant immune mechanisms in cancer
.
Nature
2015
;
520
:
373
7
.
6.
Apetoh
L
,
Ghiringhelli
F
,
Tesniere
A
,
Obeid
M
,
Ortiz
C
,
Criollo
A
, et al
Toll-like receptor 4-dependent contribution of the immune system to anticancer chemotherapy and radiotherapy
.
Nat Med
2007
;
13
:
1050
9
.
7.
Lee
Y
,
Auh
SL
,
Wang
Y
,
Burnette
B
,
Wang
Y
,
Meng
Y
, et al
Therapeutic effects of ablative radiation on local tumor require CD8+ T cells: changing strategies for cancer treatment
.
Blood
2009
;
114
:
589
95
.
8.
Burnette
BC
,
Liang
H
,
Lee
Y
,
Chlewicki
L
,
Khodarev
NN
,
Weichselbaum
RR
, et al
The efficacy of radiotherapy relies upon induction of type i interferon-dependent innate and adaptive immunity
.
Cancer Res
2011
;
71
:
2488
96
.
9.
Deng
L
,
Liang
H
,
Burnette
B
,
Beckett
M
,
Darga
T
,
Weichselbaum
RR
, et al
Irradiation and anti-PD-L1 treatment synergistically promote antitumor immunity in mice
.
J Clin Invest
2014
;
124
:
687
95
.
10.
Deng
L
,
Liang
H
,
Xu
M
,
Yang
X
,
Burnette
B
,
Arina
A
, et al
STING-dependent cytosolic DNA sensing promotes radiation-induced type I Interferon-dependent antitumor immunity in immunogenic tumors
.
Immunity
2014
;
41
:
843
52
.
11.
Zheng
W
,
Skowron
KB
,
Namm
JP
,
Burnette
B
,
Fernandez
C
,
Arina
A
, et al
Combination of radiotherapy and vaccination overcomes checkpoint blockade resistance
.
Oncotarget
2016
;
7
:
43039
51
.
12.
Liang
H
,
Deng
L
,
Hou
Y
,
Meng
X
,
Huang
X
,
Rao
E
, et al
Host STING-dependent MDSC mobilization drives extrinsic radiation resistance
.
Nat Commun
2017
;
8
:
1736
.
13.
Deng
L
,
Liang
H
,
Fu
S
,
Weichselbaum
RR
,
Fu
YX
. 
From DNA damage to nucleic acid sensing: a strategy to enhance radiation therapy
.
Clin Cancer Res
2016
;
22
:
20
5
.
14.
Ranoa
DR
,
Parekh
AD
,
Pitroda
SP
,
Huang
X
,
Darga
T
,
Wong
AC
, et al
Cancer therapies activate RIG-I-like receptor pathway through endogenous non-coding RNAs
.
Oncotarget
2016
;
7
:
26496
515
.
15.
Chow
KT
,
Gale
M
 Jr
,
Loo
YM
. 
RIG-I and other RNA sensors in antiviral immunity
.
Annu Rev Immunol
2018
;
36
:
667
94
.
16.
Wu
J
,
Chen
ZJ
. 
Innate immune sensing and signaling of cytosolic nucleic acids
.
Annu Rev Immunol
2014
;
32
:
461
88
.
17.
Rodriguez
KR
,
Bruns
AM
,
Horvath
CM
. 
MDA5 and LGP2: accomplices and antagonists of antiviral signal transduction
.
J Virol
2014
;
88
:
8194
200
.
18.
Satoh
T
,
Kato
H
,
Kumagai
Y
,
Yoneyama
M
,
Sato
S
,
Matsushita
K
, et al
LGP2 is a positive regulator of RIG-I- and MDA5-mediated antiviral responses
.
PNAS
2010
;
107
:
1512
7
.
19.
Li
X
,
Ranjith-Kumar
CT
,
Brooks
MT
,
Dharmaiah
S
,
Herr
AB
,
Kao
C
, et al
The RIG-I-like receptor LGP2 recognizes the termini of double-stranded RNA
.
J Biol Chem
2009
;
284
:
13881
91
.
20.
Venkataraman
T
,
Valdes
M
,
Elsby
R
,
Kakuta
S
,
Caceres
G
,
Saijo
S
, et al
Loss of DExD/H box RNA helicase LGP2 manifests disparate antiviral responses
.
J Immunol
2007
;
178
:
6444
55
.
21.
Rothenfusser
S
,
Goutagny
N
,
DiPerna
G
,
Gong
M
,
Monks
BG
,
Schoenemeyer
A
, et al
The RNA helicase Lgp2 inhibits TLR-independent sensing of viral replication by retinoic acid-inducible gene-I
.
J Immunol
2005
;
175
:
5260
8
.
22.
Saito
T
,
Hirai
R
,
Loo
YM
,
Owen
D
,
Johnson
CL
,
Sinha
SC
, et al
Regulation of innate antiviral defenses through a shared repressor domain in RIG-I and LGP2
.
PNAS
2007
;
104
:
582
7
.
23.
Bruns
AM
,
Leser
GP
,
Lamb
RA
,
Horvath
CM
. 
The innate immune sensor LGP2 activates antiviral signaling by regulating MDA5-RNA interaction and filament assembly
.
Mol Cell
2014
;
55
:
771
81
.
24.
Suthar
MS
,
Ramos
HJ
,
Brassil
MM
,
Netland
J
,
Chappell
CP
,
Blahnik
G
, et al
The RIG-I-like receptor LGP2 controls CD8(+) T cell survival and fitness
.
Immunity
2012
;
37
:
235
48
.
25.
Liu
Y
,
Qu
L
,
Liu
Y
,
Roizman
B
,
Zhou
GG
. 
PUM1 is a biphasic negative regulator of innate immunity genes by suppressing LGP2
.
PNAS
2017
;
114
:
E6902
E11
.
26.
Schietinger
A
,
Arina
A
,
Liu
RB
,
Wells
S
,
Huang
J
,
Engels
B
, et al
Longitudinal confocal microscopy imaging of solid tumor destruction following adoptive T cell transfer
.
Oncoimmunology
2013
;
2
:
e26677
.
27.
Engels
B
,
Engelhard
VH
,
Sidney
J
,
Sette
A
,
Binder
DC
,
Liu
RB
, et al
Relapse or eradication of cancer is predicted by peptide-major histocompatibility complex affinity
.
Cancer Cell
2013
;
23
:
516
26
.
28.
Pereira
B
,
Chin
SF
,
Rueda
OM
,
Vollan
HK
,
Provenzano
E
,
Bardwell
HA
, et al
The somatic mutation profiles of 2,433 breast cancers refines their genomic and transcriptomic landscapes
.
Nat Commun
2016
;
7
:
11479
.
29.
Curtis
C
,
Shah
SP
,
Chin
SF
,
Turashvili
G
,
Rueda
OM
,
Dunning
MJ
, et al
The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups
.
Nature
2012
;
486
:
346
52
.
30.
Aran
D
,
Hu
Z
,
Butte
AJ
. 
xCell: digitally portraying the tissue cellular heterogeneity landscape
.
Genome Biol
2017
;
18
:
220
.
31.
Goldman
M
,
Craft
B
,
Swatloski
T
,
Ellrott
K
,
Cline
M
,
Diekhans
M
, et al
The UCSC cancer genomics browser: update 2013
.
Nucleic Acids Res
2013
;
41
:
D949
54
.
32.
Marisa
L
,
de Reynies
A
,
Duval
A
,
Selves
J
,
Gaub
MP
,
Vescovo
L
, et al
Gene expression classification of colon cancer into molecular subtypes: characterization, validation, and prognostic value
.
PLoS Med
2013
;
10
:
e1001453
.
33.
Stratford
JK
,
Bentrem
DJ
,
Anderson
JM
,
Fan
C
,
Volmar
KA
,
Marron
JS
, et al
A six-gene signature predicts survival of patients with localized pancreatic ductal adenocarcinoma
.
PLoS Med
2010
;
7
:
e1000307
.
34.
Reikine
S
,
Nguyen
JB
,
Modis
Y
. 
Pattern recognition and signaling mechanisms of RIG-I and MDA5
.
Front Immunol
2014
;
5
:
342
.
35.
Rehwinkel
J
,
Gack
MU
. 
RIG-I-like receptors: their regulation and roles in RNA sensing
.
Nat Rev Immunol
2020
;
20
:
537
51
.
36.
Widau
RC
,
Parekh
AD
,
Ranck
MC
,
Golden
DW
,
Kumar
KA
,
Sood
RF
, et al
RIG-I-like receptor LGP2 protects tumor cells from ionizing radiation
.
Proc Natl Acad Sci U S A
2014
;
111
:
E484
91
.
37.
Elion
DL
,
Jacobson
ME
,
Hicks
DJ
,
Rahman
B
,
Sanchez
V
,
Gonzales-Ericsson
PI
, et al
Therapeutically active RIG-I agonist induces immunogenic tumor cell killing in breast cancers
.
Cancer Res
2018
;
78
:
6183
95
.
38.
Besch
R
,
Poeck
H
,
Hohenauer
T
,
Senft
D
,
Hacker
G
,
Berking
C
, et al
Proapoptotic signaling induced by RIG-I and MDA-5 results in type I interferon-independent apoptosis in human melanoma cells
.
J Clin Invest
2009
;
119
:
2399
411
.
39.
Kasumba
DM
,
Grandvaux
N
. 
Therapeutic targeting of RIG-I and MDA5 might not lead to the same rome
.
Trends Pharmacol Sci
2019
;
40
:
116
27
.
40.
Duewell
P
,
Beller
E
,
Kirchleitner
SV
,
Adunka
T
,
Bourhis
H
,
Siveke
J
, et al
Targeted activation of melanoma differentiation-associated protein 5 (MDA5) for immunotherapy of pancreatic carcinoma
.
Oncoimmunology
2015
;
4
:
e1029698
.
41.
Blair
TC
,
Bambina
S
,
Alice
AF
,
Kramer
GF
,
Medler
TR
,
Baird
JR
, et al
Dendritic cell maturation defines immunological responsiveness of tumors to radiation therapy
.
J Immunol
2020
;
204
:
3416
24
.
42.
Yoshida
S
,
Shime
H
,
Takeda
Y
,
Nam
JM
,
Takashima
K
,
Matsumoto
M
, et al
Toll-like receptor 3 signal augments radiation-induced tumor growth retardation in a murine model
.
Cancer Sci
2018
;
109
:
956
65
.
43.
Le
UM
,
Kaurin
DG
,
Sloat
BR
,
Yanasarn
N
,
Cui
Z
. 
Localized irradiation of tumors prior to synthetic dsRNA therapy enhanced the resultant anti-tumor activity
.
Radiother Oncol
2009
;
90
:
273
9
.
44.
Hammerich
L
,
Marron
TU
,
Upadhyay
R
,
Svensson-Arvelund
J
,
Dhainaut
M
,
Hussein
S
, et al
Systemic clinical tumor regressions and potentiation of PD1 blockade with in situ vaccination
.
Nat Med
2019
;
25
:
814
24
.
45.
Rodriguez-Ruiz
ME
,
Perez-Gracia
JL
,
Rodriguez
I
,
Alfaro
C
,
Onate
C
,
Perez
G
, et al
Combined immunotherapy encompassing intratumoral poly-ICLC, dendritic-cell vaccination and radiotherapy in advanced cancer patients
.
Ann Oncol
2018
;
29
:
1312
9
.
46.
Parisien
JP
,
Lenoir
JJ
,
Mandhana
R
,
Rodriguez
KR
,
Qian
K
,
Bruns
AM
, et al
RNA sensor LGP2 inhibits TRAF ubiquitin ligase to negatively regulate innate immune signaling
.
EMBO Rep
2018
;
19
:
e45176
.
47.
Zevini
A
,
Olagnier
D
,
Hiscott
J
. 
Crosstalk between cytoplasmic RIG-I and STING sensing pathways
.
Trends Immunol
2017
;
38
:
194
205
.
48.
Broz
ML
,
Binnewies
M
,
Boldajipour
B
,
Nelson
AE
,
Pollack
JL
,
Erle
DJ
, et al
Dissecting the tumor myeloid compartment reveals rare activating antigen-presenting cells critical for T cell immunity
.
Cancer Cell
2014
;
26
:
938
.
49.
Michea
P
,
Noel
F
,
Zakine
E
,
Czerwinska
U
,
Sirven
P
,
Abouzid
O
, et al
Adjustment of dendritic cells to the breast-cancer microenvironment is subset specific
.
Nat Immunol
2018
;
19
:
885
97
.