Synthetic short mRNA prevents metastasis via innate-adaptive immunity

Abstract

Although most cancer deaths are caused by metastasis, there are no effective therapeutic approaches. This study describes the efficacy of a short synthetic mRNA (s-mRNA) designed by the sequence of non-vesicular extracellular IL1β-mRNA found in the pre-metastatic lung of tumor-bearing mice. The administration of s-mRNA inhibits murine lung metastasis by inducing the innate and adaptive immune systems. s-mRNA binds to ZC3H12D, an RNA-binding protein on natural killer cells and cytotoxic T lymphocytes. The ZC3H12D-s-mRNA complex translocated to the nucleus without being involved in translation. This process induces cytolytic activity and cell death in cancer cells without inducing a cytokine storm, and immune cells retain their antitumor activity. Although the antitumor activity of cytotoxic lymphocytes declines as the disease progresses in cancer patients, s-mRNA induces sustained high killing capacities of natural killer cells and cytotoxic T lymphocytes from colon cancer patients. Therefore, s-mRNA could be a breakthrough solution to prevent metastasis.

Fig. 1. Functional synthetic short IL1β-mRNA (s-mRNA) sequences and stability in the tumor-bearing state.

a Summary of metastasis inhibition by systemic s-mRNA-primed NK cells and CTLs. b Detection of core functional sIL1β-mRNA in humans and mice. c Structural formulas of RNase-resistant synthetic nucleotides. d Binding analysis of mRNAs by EMSA. Lane 1, FITC-mouse IL1β-mRNA (50-mer) alone; lane 2, FITC-mouse IL1β-mRNA (50-mer) + ZC36 protein; lane 3, FITC-mouse IL1β-mRNA (50-mer) + ZC36 protein + non-labeled unrelated RNA (50-mer); lane 4, FITC-mouse IL1β-mRNA (50-mer) + ZC36 protein + non-labeled mouse IL1β-mRNA (50-mer); lane 5, FITC-sIL1β-mRNA_OMe_AE alone; lane 6, FITC-sIL1β-mRNA_OMe_AE + ZC36 protein; lane 7, FITC-sIL1β-mRNA_OMe_AE + ZC36 protein + non-labeled unrelated RNA (50-mer); lane 8, FITC-IL1β-mRNA_OMe_AE + ZC36 protein + non-labeled mouse IL1β-mRNA (50-mer). e Migration analyses of ZC3H12D-overexpressing 786-O cells using various synthetic modified IL1β-mRNA. N = 3 independent experimental replicates for IL1β-mRNA, IL1β-mRNA_OMe, IL1β-mRNA_S, IL1β-mRNA_AE, IL1β-mRNA_OMe_AE, and poly A-mRNA_OMe_AE and N = 2 independent experimental replicates for β-actin-mRNA_OMe_AE. One-way ANOVA with Bonferroni correction. Mean ± SEM. f, g FITC-labeled sIL1β-mRNA was incubated with 20% mouse and human serum. sIL1β-mRNA_OMe_AE was more resistant than nonmodified sIL1β-mRNA. Representative electrophoresis gel image (f) and its quantification (g). h Trace scheme of s-mRNA in vivo. Blood and lungs were examined after tail vein injection of FITC-s-mRNA into mice stimulated with TCM. i Detection of fluorescein in mouse serum at 30 min or 48 hr after tail vein injection of FITC-sIL1β-mRNA_OMe_AE (right, serum). The left panels show positive controls after a mixture of PBS or FITC-sIL1β-mRNA_OMe_AE and serum in vitro. j Representative images of FITC-s-mRNA near metastatic PKH26-labeled E0771 cells in the lungs taken 24 hr after injection with either PBS or FITC-sIL1β-mRNA_OMe_AE. Scale bar: 50 μm, 10 μm. Electropherogram of the retrieved mRNA from the lungs 30 min after tail vein injection of FITC-s-mRNA. 0, FITC-sIL1β-mRNA_OMe_AE; 1 and 2, mRNA from the lung without FITC-s-mRNA injection (two mice); 3 and 4, mRNA from the lung with FITC-sIL1β-mRNA_OMe_AE injection (two mice). k Quantification of FITC-s-mRNA signals around E0771 cells in the lungs. The distance between the FITC signal and the metastatic tumor cell was set at 20 μm. N = 30 areas from 3 mice. One-way ANOVA with Bonferroni correction. Mean ± SEM. Source data are provided as a Source Data file.

Fig. 2. Uptake of sIL1β-mRNA and tumoricidal activities of human NK and T cells in vitro.

a Histogram after the uptake of FITC-sIL1β-mRNA_OMe_AE in human CD56⁺CD3^- NK and CD3⁺ T cells derived from PBMCs using flow cytometry. FITC-sPoly A-mRNA_OMe_AE (50-mer) was used as a control. b Quantification of the number of cells that were taken up by s-mRNA-FITC. N = 5 individuals. One-way ANOVA with Bonferroni correction. Mean ± SEM. c, d Quantification of FITC-sIL1β-mRNA_OMe_AE uptake in NK1.1⁺ NK, CD4⁺ T and CD8⁺ T cells from splenocytes of wild-type (WT) (c) and Zc3h12d knockout (KO) mice (d) by FACS. N = 3 mice. One-way ANOVA with Bonferroni correction. Mean ± SEM. e Representative confocal microscopy images of FITC-s-mRNA relocation in the nucleus of human CD56^dimCD3^- NK cells at 1 h after the addition of 10 ng/ml FITC-sIL1β-mRNA_OMe_AE. 10 stacked images. Scale bar: 10 μm. f Quantitative analyses of nuclear FITC signals in human NK cells after treatment with FITC-s-mRNA. N = 13 cells from 3 individuals, none; N = 37 cells from 3 individuals, sIL1β-mRNA_OMe_AE; N = 44 cells from 3 individuals, sPoly A-mRNA_OMe_AE. One-way ANOVA with Bonferroni correction. Mean ± SEM. g Schematic of the tumoricidal assay for NK cells and T cells (see ‘Methods’ for details). h Representative Zombie⁺ cells indicate cells killed by sIL1β-mRNA_OMe_AE-activated CD56^dimCD3^- NK cells. i–k Quantitative analysis of killed tumor cells, DLD-1 (i), Caco-2 (j), and MDA-MB-231 (k) cells. Scale bar: 20 μm. N = 3 independent experimental replicates. One-way ANOVA with Bonferroni correction. Mean ± SEM. l Representative Zombie⁺ dead tumor cells killed by sIL1β-mRNA_OMe_AE-activated CD8⁺ T cells. m Quantitative analysis of tumor cells killed by s-mRNA- or protein-stimulated CD8⁺ T cells cocultured with (left) or without dendritic cells (DCs) (right). Scale bar: 20 μm. N = 3 independent experimental replicates. One-way ANOVA with Bonferroni correction. Mean ± SEM. n Strategy for the tumoricidal assay using s-mRNA-primed CD56^dimCD3^- NK cells or CD8⁺ T cells after multiple rounds of stimulation (see ‘Methods’ for details). o–q Quantitative analysis of tumor cells killed by CD56^dimCD3^- NK cells after single (o), two (p), or three (q) stimulation events. N = 3 independent experimental replicates. One-way ANOVA with Bonferroni correction. Mean ± SEM. r–t Quantitative analysis of tumor cells killed by CD8⁺ T cells after single (r), two (s), or three (t) stimulations. N = 3 independent experimental replicates. One-way ANOVA with Bonferroni correction. Mean ± SEM. Source data are provided as a Source Data file.

Fig. 3. Antimetastatic ability of sIL1β-mRNA_OMe_AE after administration in vivo.

a Schematic of the antimetastatic experiment using E0771 and MC-38 cells. Tumor-bearing mice were treated with intravenous (iv) injection of s-mRNA or PBS, followed by iv injection of PKH26-labeled tumor cells. b Representative image of the primary tumor taken on day 7 after mammary fat pad (mfp) injection of E0771 cells. The white-lined squares indicate 1 cm. c Volumes of primary tumors treated with s-mRNA or PBS on day 7. N = 5 mice, PBS; N = 5 mice, sPoly A-mRNA_OMe_AE; N = 4 mice, 1 μg of sIL1β-mRNA_OMe_AE; N = 4 mice, 10 μg of sIL1β-mRNA_OMe_AE. One-way ANOVA with Bonferroni correction. Mean ± SEM. d Representative images of PKH26-labeled E0771 cells in the lungs on day 7. e Number of E0771 cells in the lungs of mice treated with s-mRNA or PBS. N = 5 mice, PBS; N = 5 mice, sPoly A-mRNA_OMe_AE; N = 4 mice, 1 μg of sIL1β-mRNA_OMe_AE; N = 4 mice, 10 μg of sIL1β-mRNA_OMe_AE. One-way ANOVA with Bonferroni correction. Mean ± SEM. f Number of E0771 cells in the lungs of mice treated with PBS, 1 μg of sIL1β-mRNA_OMe_AE, or 1 μg of full-length IL1β-mRNA. N = 5 mice. One-way ANOVA with Bonferroni correction. Mean ± SEM. g Scheme for the antimetastasis assay, similar to Fig. 3a, in WT and Zc3h12d KO mice. h Number of metastatic cells in the lung of E0771-bearing WT and Zc3h12d KO mice treated with sIL1β-mRNA_OMe_AE or PBS. N = 5 mice. One-way ANOVA with Bonferroni correction. Mean ± SEM. i, j Metastasis of MC-38 cells. Number of metastatic cells in the lung (j) after treatment with sIL1β-mRNA_OMe_AE or PBS. The primary tumor volume is also shown (i). N = 5 mice, PBS; N = 6 mice, sIL1β-mRNA_OMe_AE. Student’s two-sided t test. Mean ± SEM. k Schematic of an antimetastatic experiment using sIL1β-mRNA_OMe_AE in mice with NK cell depletion (l–n), T-cell depletion (o–r), and NK/T-cell double depletion (s) in combination with anti-ASGM-1 or/and anti-CD4/CD8a antibodies. l, m Validation of NK cell depletion among lymphocytes in the blood (l) and lungs (m). N = 5 mice, PBS; N = 6 mice, Anti-ASGM1. Student’s two-sided t test. Mean ± SEM. n Number of metastatic E0771 cells in the lungs of NK cell-depleted mice. N = 5 mice, PBS; N = 6 mice, Anti-ASGM1. Student’s two-sided t test. Mean ± SEM. o, p Analyses of the percentages of CD4⁺ T and CD8a⁺ T cells among TCRβ⁺ lymphocytes in the blood (o) and lung (p). N = 5 mice. Student’s two-sided t test. Mean ± SEM. q Number of metastatic E0771 cells in T-cell-depleted lungs. N = 5 mice, PBS; N = 6 mice, Anti-CD4/CD8a. Student’s two-sided t test. Mean ± SEM. r Number of metastatic E0771 cells in T-cell-depleted lungs. N = 5 mice. One-way ANOVA with Bonferroni correction. Mean ± SEM. s Number of metastatic E0771 cells in the lungs of both PBS-treated control mice and NK/T double-depleted mice. N = 5 mice. Student’s two-sided t test. Mean ± SEM. Source data are provided as a Source Data file.

Fig. 4. Systemic administration of sIL1β-mRNA inhibits macroscopic metastasis without obvious side effects.

a Schematic of screening for immunogenic side effects. The s-mRNA and control PBS groups received intravenous injections 3 times. The IL-12 group received intraperitoneal injections for 5 consecutive days. All groups were dissected 6 hr after the final injections, and their organs and blood were examined. b HE staining of the lung, liver, kidney, and spleen. The necrotic area in the livers of mice treated with IL-12 was increased. Scale bar: 20 µm, 50 μm. c Comparison of blood biochemical values among the 3 groups. N = 5 mice, PBS; N = 6 mice, sIL1β-mRNA_OMe_AE; N = 4 mice, IL-12. One-way ANOVA with Bonferroni correction for multiple comparisons. Mean ± SEM. d Comparison of blood cytokine values among the 3 groups. N = 5 mice, PBS; N = 6 mice, sIL1β-mRNA_OMe_AE; N = 4 mice, IL-12. One-way ANOVA with Bonferroni correction for multiple comparisons. Mean ± SEM. e Comparison of IFNα values among the 4 groups by the ELISA assay. N = 6 mice. One-way ANOVA with Bonferroni correction for multiple comparisons. Mean ± SEM. Source data are provided as a Source Data file.

Fig. 5. Long-term observation models and xenograft model.

a Scheme of preventive metastatic nodule assay. TCM-stimulated mice were pre-treated with s-mRNAs or PBS and then injected intravenously with metastatic tumor cells. The mice were dissected after 2 weeks. b Weights of the body (left) and lungs (right) among three groups. N = 6 mice per group. One-way ANOVA with Bonferroni correction. Mean ± SEM. c Representative photo of lungs with metastatic nodules (red circles). The white lines indicate 1 cm squares. d, e Blockade of metastasis through the use of sIL1β-mRNA. Photos of the lungs were taken 14 days after E0771 injection for each treatment (d). The green arrows show metastatic nodules. Some nodules are present on the dorsal surface. Number of metastatic nodules (e). N = 6 mice per group. One-way ANOVA with Bonferroni correction. Mean ± SEM. f, g IHC analysis of metastatic cells in the lungs using Ki-67 staining. Representative images of Ki-67 signals (f). Scale bar: 50 μm. Number of the Ki-67-positive cells (g). N = 6 mice per group. One-way ANOVA with Bonferroni correction. Mean ± SEM. h Scheme of the orthotopic model using E0771 cells. Primary tumors were removed 14 days after initial orthotopic implantation, and mice were treated with s-mRNAs or PBS. Lung metastases were counted 3 weeks after removal of the primary tumor. i Representative image of the lung with metastatic nodules (red circle). White lines indicate 1 cm squares. j Number of metastatic nodules. N = 5 mice per group. Student’s two-sided t test. Mean ± SEM. k Scheme of the long-term survival model. Primary tumors were removed 14 days after initial orthotopic implantation, and mice were treated with s-mRNAs or PBS. l Representative image of the lung with metastatic nodules (yellow arrows). White lines indicate 1 cm squares. m Survival data were plotted on a Kaplan-Meier survival curve and statistical significance was calculated using the log-rank test. N = 10 mice, PBS; N = 8 mice, sIL1β-mRNA_OMe_AE. n Scheme of the xenograft model using NOD/Shi-scid, IL-2RγKO (NOG) mouse, and MDA-MB-231 cells. CD56⁺CD3^- NK cells or CD8⁺ T cells from a healthy donor were intravenously injected into TCM-stimulated mice, followed by intravenous injection of PKH26-labeled tumor cells. o Number of MDA-MB-231 cells in the lungs of mice treated with s-mRNA or PBS. N = 5 mice. One-way ANOVA with Bonferroni correction. Mean ± SEM. Source data are provided as a Source Data file.

Fig. 6. Induction of Granzyme B in NKs and CTLs after uptake of sIL1β-mRNA into the nucleus.

a Schematic of screening of the genes that responded after the uptake of FITC-sIL1β-mRNA_OMe_AE. FITC-positive CD56⁺CD3^- NK and CD3⁺ T cells were sorted for RNA sequencing. b Summary of the expression profile related to inflammatory cytokines. Data shows RPKM (reads per kilobase million) values of inflammation-related genes. For comparison, ACTB data is added in the top row. c GZMB were upregulated in CD56⁺CD3^- NK and CD8⁺ T cells 0.5 hr after the uptake of FITC-sIL1β-mRNA_OMe_AE. N = 2 biological replicates. d, e IHC analysis of GZMB expression in CD56^dimCD3^- NK cells 6 hr after stimulation with sIL1β-mRNA_OMe_AE. d Representative images of GZMB staining and (e) quantification of GZMB expression. Scale bar: 10 μm. N = 50 cells per group. Student’s two-sided t test. Mean ± SEM. f Schematic of electroporation and GZMB staining. g–j Quantitative analysis of tumor cells (Caco-2, DLD-1, and MDA-MB231) killed by sIL1β-mRNA_OMe_AE-activated CD56^dimCD3^- NK cells or CD8⁺ T cells after electroporation of siRNA -control (g, i), -GZMB (h, j). N = 3 independent experimental replicates. One-way ANOVA with Bonferroni correction was used for multiple comparisons. Mean ± SEM. Source data are provided as a Source Data file.

Fig. 7. Tumoricidal activities of NK and CTL cells from colon cancer patients.

a Schematic of blood collection from colorectal cancer patients. b Summary of the characteristics of patients with colorectal cancer. c–g Quantitative analysis of colon cancer cells (Caco-2 and DLD-1) killed by sIL1β-mRNA_OMe_AE-activated CD56^dimCD3^- NK cells and/or CD8⁺ T cells from patient-1 (c), patient−2 (d), patient-3 (e), patient-4 (f), and patient-5 (g). N = 3 independent experimental replicates. One-way ANOVA with Bonferroni correction was used for multiple comparisons. Mean ± SEM. Source data are provided as a Source Data file.

Fig. 8. Graphical summary of the antimetastatic effect of s-mRNA.

Since the chemically synthesized short mRNA (IL1β-mRNA_OMe_AE) is RNase resistant, it can be used to stimulate NK and CTL via the ZC3H12D receptor in vivo. These immune cells enhance its Granzyme B-dependent tumoricidal activity without inducing inflammatory responses that cause severe side effects.