Vesicle-Based Compositions and Uses Thereof

Abstract

Vesicle-based compositions and their uses are disclosed. Also disclosed is a composition comprising vesicles, such as red blood cell-derived extracellular vesicles (RBCEVs), for the delivery of retinoic acid-inducible gene I receptor (RIG-I) agonists to cells. Also disclosed is a method for treating disease by administering the vesicle-based compositions. TIFF2025510600000046.tif33139

Description

[Technical field]

[0001] FIELD OF THEINVENTION The present invention relates generally to the field of molecular biology. In particular, the present invention relates to vesicle-based compositions and uses thereof. [Background technology]

[0002] 2. Background of the Invention Many therapeutic approaches, including RNA oligonucleotides, small interfering RNA (siRNA) and antisense oligonucleotides, have been developed to target many disease genes that were once considered undruggable. Despite their potential, the clinical application of such therapeutic approaches continues to be limited by the current inefficient delivery of these molecules to target cells as well as toxicity, which can induce organ damage and cause poor survival.

[0003] Lipid-based nanoparticles, for example, have been developed to deliver siRNA for cancer treatment, and little toxicity was observed. However, when lipid-based nanoparticles were used in combination with conventional siRNA, off-target effects were observed, resulting in the accumulation of siRNA in the wrong cells and reducing its effectiveness as an anti-cancer therapy.

[0004] There is therefore a need to provide improved compositions that address at least some of the above problems. Summary of the Invention

[0005] overview In one aspect, the present disclosure refers to a composition comprising a vesicle for delivery of a retinoic acid-inducible gene I receptor (RIG-I) agonist to a cell.

[0006] In another aspect, the present disclosure refers to a composition comprising: red blood cell-derived extracellular vesicles (RBCEVs); an immunomodulatory RNA (immRNA) comprising or consisting of the sequence TIFF2025510600000002.tif4129, further comprising a 5' triphosphate cap; and / or An antisense oligonucleotide (ASO) comprising or consisting of the sequence TIFF2025510600000003.tif4134 and further comprising a 5' triphosphate cap.

[0007] In another aspect, the present disclosure refers to a composition comprising: lipid nanoparticles; an immune modulatory RNA (immRNA), comprising TIFF2025510600000004.tif4129, and further comprising a 5' triphosphate cap; and / or KRAS-ASO.

[0008] In another aspect, the present disclosure refers to a method of delivering a retinoic acid-inducible gene I receptor (RIG-I) agonist to a cell, comprising administering to the cell a pharma- ceutical effective amount of a composition disclosed herein, a pharmaceutical composition disclosed herein, or a kit disclosed herein.

[0009] In another aspect, the present disclosure refers to a method of treating a disease comprising administering to a subject in need thereof a pharma- ceutical effective amount of a composition disclosed herein, a pharmaceutical composition disclosed herein, or a kit disclosed herein. [Brief description of the drawings]

[0010] The present invention will be better understood by reference to the detailed description when considered in conjunction with the following non-limiting examples and the accompanying drawings.

[0011] [Figure 1-1]FIG. 1A is a schematic diagram showing fluorescence-activated cell sorting (FACS) analysis of RBCEVs loaded with FAM-labeled antisense oligonucleotides (ASOs) using electroporation, Exo-Fect, and REG1. As a negative control, red blood cell-derived extracellular vesicles (RBCEVs) were incubated with FAM-ASOs without electroporation or loading reagents. FIG. 1B is a schematic diagram showing the average intensity of FAM in RBCEVs loaded with FAM-labeled ASOs using electroporation, Exo-Fect, and REG1 (n = 3) compared to the extracellular vesicles (EVs) only condition for statistical analysis. FIG. 1C is a schematic diagram showing transmission electron microscopy images of RBCEVs loaded with FAM-ASOs. Scale bar, 200 nm. FIG. 1D is a schematic diagram showing the size distribution of FAM-ASO-loaded RBCEVs determined using a ZetaView® nanoparticle tracking analyzer. FIG. 1E is a schematic diagram showing representative immunofluorescence images of CA1a cells that have taken up FAM-ASOs in RBCEVs. Nuclei were stained with Hoechst (blue). Plasma membranes were stained with CellMask Deep Red (red). Scale bar, 20 μm. Figure 1F is a schematic showing the average intensity of FAM in CA1a cells as in (E) compared to untreated control for statistical analysis. Figure 1G is a schematic showing FACS analysis of CA1a cells loaded with FAM-ASO-loaded RBCEVs (n = 3) compared to untreated control for statistical analysis. Figure 1H is a schematic showing the viability of human breast cancer CA1a cells treated with RBCEVs loaded with 0.05 μg / μl NC-ASO at different time points. Viable cells were quantified by normalizing absorbance readings to untreated control at each time point (n = 3). Figure 1I is a schematic showing the loading efficiency of miR-125b ASO in RBCEVs with Exo-Fect and REG1 as determined using gel electrophoresis (n = 3).Figure 1J is a schematic showing quantitative polymerase chain reaction (qPCR) analysis of miR-125b fold change (FC) normalized to U6B compared to the EV-only condition in CA1a cells (n = 3) treated with different doses of RBCEVs containing miR-125b ASO. All bar graphs represent mean ± SEM. Ns, not significant; *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001 as determined by two-tailed Student's t-test. Figure 1: Multiple methods can be used to load small RNAs into RBCEVs for delivery of RNA to cancer cells. [Figure 1-2] See description of Figure 1-1. [Figure 1-3] See description of Figure 1-1. [Figure 1-4] See description of Figure 1-1. [Figure 1-5] See description of Figure 1-1. [Figure 1-6] See description of Figure 1-1. [Figure 2-1]Figure 2A is a schematic showing the design of immune-modulatory RNA (immRNA) with 5' triphosphate (ppp). Figure 2B-D are schematics showing qPCR analysis of mRNA encoding RIG-I (Ddx58) relative to Gapdh and its downstream effector in mouse breast cancer 4T1 cells (B), human breast cancer CA1a cells (C) and human lung cancer H358 cells (D) treated for 24 h with 0.1 μg / μL unloaded RBCEVs, NC-RNA-loaded RBCEVs and immRNA-loaded RBCEVs (n = 4, REG1 RNA loaded). Figure 2E is a schematic showing the average luciferase activity in A549-Dual™ and A549-Dual™ RIG-I- / - cells treated with PBS, 0.05 μg / μL unloaded RBCEVs, NC-RNA-loaded RBCEVs and immRNA-loaded RBCEVs for 24 and 48 hours (n = 5-9). Figure 2F is a schematic showing multiplex immunoassay analysis of cytokines in conditioned medium of 4T1 cells treated with 0.1 μg / μL unloaded RBCEVs, NC-RNA-loaded RBCEVs and immRNA-loaded RBCEVs for 48 hours (n = 3). Figure 2G-I are schematic diagrams showing FACS analysis of the mean percentage of ANXV+PI+ population in 4T1 cells (G), CA1a cells (H) and H358 cells (I) after 72 h treatment with 0.1 μg / μL unloaded RBCEVs, NC-RNA-loaded RBCEVs and immRNA-loaded RBCEVs (n = 4). All bar graphs represent the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001 as determined by two-tailed Student's t-test. Figure 2: RBCEVs deliver immunomodulatory RNA to activate the RIG-I pathway and induce immunogenic cell death of cancer cells. [Figure 2-2] See description of Figure 2-1. [Figure 3-1]Figure 3A is a schematic showing the miR-125b sequence and the design of ASOs against miR-125b with and without 5' triphosphate. The seed sequence of miR-125b is colored in red and underlined. A triphosphate group (PPP) was added to the 5' end of 3p-125b-ASO and the first four nucleotides were replaced with GG. A short sequence (gray circle) was added to the 3' end of 3p-125b-ASO during IVT. Figure 3B is a schematic showing qPCR analysis of miR-125b compared to snoRNA234 in 4T1 cells treated for 24 h with 0.1 μg / μL unloaded or NC-RNA-loaded, 125b-ASO-loaded and 3p-125b-ASO-loaded RBCEVs (n = 4, RNA loaded with REG1). FIG. 3C is a schematic diagram showing qPCR analysis of Ddx58 and its downstream effectors relative to Gapdh in 4T1 cells treated with RBCEVs as in (B). FIG. 3D is a schematic diagram showing qPCR analysis of miR-125b relative to U6B in CA1a cells treated with RBCEVs as in (B). FIG. 3E is a schematic diagram showing qPCR analysis of DDX58 and its downstream effectors relative to GAPDH in CA1a cells treated with RBCEVs as in (B). FIG. 3F is a schematic diagram showing qPCR analysis of miR-125b relative to U6B in H358 cells treated with RBCEVs as in (B). FIG. 3G is a schematic diagram showing qPCR analysis of DDX58 and its downstream effectors relative to GAPDH in H358 cells treated with RBCEVs as in (B). Figure 3H is a schematic showing the average luciferase activity in A549-Dual™ and A549-Dual™ RIG-I- / - cells treated with PBS, 0.05 μg / μL unloaded RBCEVs, NC-RNA-loaded RBCEVs and 3p-125b-ASO-loaded RBCEVs for 24 and 48 h (n = 5-8). Figure 3I is a schematic showing multiplex immunoassay analysis of cytokines in conditioned medium of 4T1 cells treated with 0.1 μg / μL unloaded RBCEVs, NC-RNA-loaded RBCEVs and 3p-125b-ASO-loaded RBCEVs for 48 h (n = 3).Figure 3J-L are schematic diagrams showing FACS analysis revealing the mean percentage of ANXV+PI+ population in 4T1 cells (J), CA1a cells (K) and H358 cells (L) after 72 h treatment with 0.1 μg / μL of unloaded RBCEVs, NC-RNA-loaded RBCEVs, 125b-ASO-loaded RBCEVs and 3p-125b-ASO-loaded RBCEVs (n = 4). All bar graphs represent the mean ± SEM. **P < 0.01, ***P < 0.001 and ****P < 0.0001, determined by two-tailed Student's t-test. Figure 3: RBCEVs deliver bifunctional ASOs to inhibit oncogenic miR-125b and simultaneously activate the RIG-I pathway leading to cell death in cancer cells. [Figure 3-2] See description of Figure 3-1. [Figure 3-3] See description of Figure 3-1. [Figure 4-1]Figure 4A is a schematic showing the treatment of murine 4T1 mammary tumors in BALB / c mice by intratumoral delivery of immRNA in RBCEVs. Figure 4B is a schematic showing the volume of 4T1 tumors injected intratumorally every 3 days with 2.5 mg / kg RBCEVs containing immRNA or NC RNA (n = 6 mice). Figure 4C is a schematic showing qPCR analysis of RIG-I pathway gene expression relative to Gapdh in untreated and treated 4T1 tumors (n = 6 mice). Figure 4D is a schematic showing flow cytometry analysis of immune cells in untreated and treated 4T1 tumors presented as the average percentage of each subset in total cells (n = 6 mice). NEUT, neutrophils; NK, natural killer cells; MΦ, macrophages; DC, dendritic cells. Figure 4E is a schematic showing ELISA quantification of IFNβ in serum of mice bearing 4T1 tumors (n = 4 mice). FIG. 4F is a schematic showing qPCR analysis of cytokine gene expression relative to Gapdh in untreated and treated 4T1 tumors (n = 4 mice). FIG. 4G is a schematic showing representative images of TUNEL staining (orange fluorescence) of untreated and treated 4T1 tumor sections. Cancer-associated fibroblasts (green) were stained with anti-αSMA antibody. Nuclei were stained with Hoechst (blue). Scale bar, 50 μm. T, tumor; S, stroma. FIG. 4H is a schematic showing the average intensity per nuclear area of ​​TUNEL staining signal (n = 4 mice). FIG. 4I is a schematic showing treatment of human CA1a mammary tumors in NSG-SGM3 mice by intratumoral delivery of immRNA-loaded RBCEVs. FIG. 4J is a schematic showing the volume of CA1a tumors injected intratumorally every 3 days with 2.5 mg / kg RBCEVs loaded with immRNA or NC RNA (n = 6 mice). Figure 4K is a schematic showing qPCR analysis of RIG-I pathway gene expression relative to GAPDH in untreated and treated CA1a tumors (n = 6 mice). Figure 4L is a schematic showing flow cytometry analysis of immune cells in untreated and treated CA1a tumors (n = 6 mice). All bar graphs represent the mean ± SEM.*P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001 as determined by two-tailed Student's t test. Figure 4: Intratumoral delivery of immRNA suppresses breast cancer growth by inducing RIG-I-mediated immune responses. [Figure 4-2] See description of Figure 4-1. [Figure 4-3] See description of Figure 4-1. [Figure 4-4] See description of Figure 4-1. [Figure 5-1]Figure 5A is a schematic diagram showing the treatment of murine 4T1 mammary tumors in BALB / c mice by intratumoral delivery of 3p-125b-ASO in RBCEVs. Figure 5B is a schematic diagram showing the volume of 4T1 tumors injected intratumorally every 3 days with 5 mg / kg RBCEVs containing 3p-125b-ASO or NC RNA (n = 6 mice). Figure 5C is a schematic diagram showing qPCR analysis of miR-125b compared to sno234 in untreated and treated 4T1 tumors (n = 6 mice). Figure 5D is a schematic diagram showing qPCR analysis of RIG-I pathway gene expression compared to Gapdh in untreated and treated 4T1 tumors (n = 6 mice). Figure 5E is a schematic diagram showing flow cytometry analysis of immune cells in untreated and treated 4T1 tumors (n = 5 mice). NEUT, neutrophils; NK, natural killer cells; MΦ, macrophages; DC, dendritic cells. Figure 5F is a schematic showing multiplex immunoassay analysis of cytokine concentrations (conc.) in mouse tumor lysates (n = 3 mice). Figure 5G is a schematic showing qPCR analysis of cytokine gene expression relative to Gapdh in untreated and treated 4T1 tumors (n = 5 mice). Figure 5H is a schematic showing representative images of TUNEL staining (orange fluorescence) of untreated and treated 4T1 tumor sections. Cancer-associated fibroblasts (green) were stained with anti-αSMA antibody. Nuclei were stained with Hoechst (blue). Scale bar, 50 μm. T, tumor; S, stroma. Figure 5I is a schematic showing the mean intensity per nuclear area of ​​TUNEL staining signal (n = 4 mice). All bar graphs represent the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001 as determined by two-tailed Student's t test. Figure 5: Intratumoral delivery of bifunctional ASOs inhibits breast cancer growth by inducing apoptosis and RIG-I-mediated immune responses. [Figure 5-2] See description of Figure 5-1. [Figure 6-1]Figure 6A is a schematic showing RBCEV modification: RBCEVs were conjugated with anti-EGFR nanobodies and loaded with immRNA. Figure 6B is a schematic showing FACS analysis of 4T1-hEGFR and parental 4T1 cells treated with 0.02 μg / μL of uncoated, control (Ctrl)-VHH-coated, and EGFR-VHH-coated CFSE-labeled RBCEVs (n = 3). Figure 6C is a schematic showing FACS analysis of 4T1-hEGFR and parental 4T1 cells treated with 0.02 μg / μL of uncoated, Ctrl-VHH-coated, and EGFR-VHH-coated FAM-NC-ASO-loaded RBCEVs (n = 3). FIG. 6D is a schematic showing qPCR analysis of RIG-I pathway gene expression relative to Gapdh in 4T1-hEGFR and parental 4T1 cells treated with 0.1 μg / μL uncoated, Ctrl-VHH-coated, and EGFR-VHH-coated immRNA-containing RBCEVs (n = 3). FIG. 6E is a schematic showing intrapulmonary delivery of EGFR-targeted CFSE-RBCEVs in mice bearing metastatic 4T1 breast cancer. FIG. 6F is a schematic showing FACS analysis of CFSE signal in tumor and immune cells in lungs of mice treated with 25 mg / kg uncoated, Ctrl-VHH-coated, and EGFR-VHH-coated CFSE-RBCEVs (n = 3 mice). NEUT, neutrophils; MΦ, macrophages; DC, dendritic cells. All bars represent mean ± SEM. **P < 0.01 and ***P < 0.001 as determined by two-tailed Student's t test. Figure 6: Conjugation of RBCEVs with EGFR-binding nanobodies promotes specific delivery of imMRNA to metastatic breast cancer cells. [Figure 6-2] See description of Figure 6-1. [Figure 6-3] See description of Figure 6-1. [Figure 7-1]FIG. 7A is a schematic showing intrapulmonary treatment of mice bearing breast cancer metastases. FIG. 7B is a schematic showing flow cytometry analysis of lung tumor burden using anti-human EGFR antibody after 5 treatments. FIG. 7C is a schematic showing the average percentage of hEGFR-positive tumor cells in the lungs as shown in (B) (n = 4 mice). FIG. 7D is a schematic showing the average percentage of metastatic area in lungs treated with 25 mg / kg uncoated, Ctrl-VHH-coated, and EGFR-VHH-coated immRNA-containing RBCEVs after 5 treatments (n = 4 mice). FIG. 7E is a schematic showing representative hematoxylin and eosin (H&E) stained lung sections from mice as in (D). Scale bar, 50 μm. FIG. 7F is a schematic showing qPCR analysis of RIG-I pathway gene expression relative to Gapdh in lungs of mice as in (A) (n = 4 mice). Figure 7G is a schematic showing FACS analysis of immune cells in mouse lungs as in (A) (n = 4 mice). Figure 7H is a schematic showing representative images of TUNEL staining (orange fluorescence) of mouse lung sections as in (A). Nuclei were stained with Hoechst (blue). Scale bar, 20 μm. Figure 7I is a schematic showing the average intensity per nuclear area of ​​TUNEL staining signal (n = 4 mice). Figure 7J is a schematic showing multiplex immunoassay of cytokines in lung homogenates of mice bearing 4T1-hEGFR metastatic tumors (n = 3 mice). All bar graphs represent the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001, as determined by two-tailed Student's t-test. Figure 7: Pulmonary delivery of immRNA using RBCEVs targeting EGFR suppresses breast cancer metastasis in the lung. [Figure 7-2] See description of Figure 7-1. [Figure 7-3] See description of Figure 7-1. [Figure 8-1]FIG. 8A is a schematic diagram showing flow cytometric analysis of DC and macrophage activation and polarization in mouse lungs as in FIG. 7A (n = 3-4 mice). MΦ, macrophages. FIG. 8B is a schematic diagram showing the average percentage of cells as shown in (A) (n = 3-4 mice). FIG. 8C is a schematic diagram showing flow cytometric analysis of CD69 and granzyme B in CD8+ tumor-infiltrating lymphocytes (TILs) from mouse lungs as in FIG. 7A (n = 3-4 mice). FIG. 8D is a schematic diagram showing the average percentage of CD69 and granzyme B in CD8+ TILs as shown in (C). FIG. 8E is a schematic diagram showing ELISA quantification of IFNγ from tumor-specific CD8+ TILs in culture supernatants at 48 h (n = 3-4 mice). FIG. 8F is a schematic diagram showing tumor-specific cytotoxic T lymphocyte (CTL) activity of CD8+ TILs (n = 3-4 mice). All bars represent the mean ± SEM. ns, not significant, *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001 as determined by two-tailed Student's t-test. Figure 8: EGFR-targeted immRNA-loaded RBCEVs induce DC activation, promote M1 macrophage polarization and enhance tumor-specific CD8+ T cell activity. [Figure 8-2] See description of Figure 8-1. [Figure 9-1]Figure 9A is a schematic diagram showing the design and chemical modification of ASO targeting KRAS G12D and G12V. Figure 9B is a schematic diagram showing the knockdown of KRAS in A427 cells and AsPC-1 cells carrying KRAS G12D mutant and H1975 cells and DU145 cells carrying wild-type KRAS after treatment with RBCEV loaded with KRAS G12D ASO, as determined by Western blot (n=3). Figure 9C is a schematic diagram showing the knockdown of KRAS in A427 cells carrying KRAS G12D mutant and H1975 cells and DU145 cells with wild-type KRAS after treatment with RBCEV loaded with KRAS G12D ASO, as determined by qPCR analysis (n=3). Figure 9D is a schematic diagram showing the knockdown of KRAS in AsPC-1 cells carrying KRAS G12D mutant and H1975 and DU145 cells carrying wild-type KRAS after treatment with RBCEV loaded with KRAS G12D ASO, as determined by qPCR analysis (n=3). Figure 9E is a schematic diagram showing the change in cell viability caused by RBCEV loaded with KRAS G12D ASO2 in A427 cells, as determined by CCK-8 assay (n=5). Figure 9F is a schematic diagram showing the change in cell viability caused by RBCEV loaded with KRAS G12D ASO2 in AsPC-1 cells, as determined by CCK-8 assay (n=5). Figure 9G is a schematic diagram showing the knockdown of KRAS in H441 cells carrying KRAS G12V mutant and H1975 cells carrying wild-type KRAS after treatment with RBCEV loaded with KRAS G12V ASO as determined by Western blot (n=3). Figure 9H is a schematic diagram showing the knockdown of KRAS in H441 cells carrying KRAS G12V mutant and H1975 cells carrying wild-type KRAS after treatment with RBCEV loaded with KRAS G12V ASO as determined by qPCR analysis (n=3).Figure 9I is a schematic diagram showing the reduction in cell viability caused by RBCEVs loaded with KRASG12V ASO2 in A427 and H1975 cells as determined by CCK-8 assay (n=5). All bars represent the mean ± SD. ns, *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001 as determined by two-tailed Student's t-test. [Figure 9-2] See description of Figure 9-1. [Figure 9-3] See description of Figure 9-1. [Figure 10] FIG. 1 is a schematic diagram showing delivery of RIG-I agonists using RBCEVs for anti-cancer immunotherapy. [Figure 11-1] Figure 11A is a schematic showing Western blots of ALIX, TSG101, GPA, HBA, β-actin and GAPDH in RBCs and RBCEVs. Figure 11B is a schematic showing single EV FACS analysis of GPA in RBCEVs. Gating strategy used to obtain a distinct population of RBCEVs from background noise. Figure 11C is a schematic showing flow cytometry analysis of EpCAM expression in mouse lung epithelial cells and α-EGFR-VHH binding affinity to multiple cells. Figure 11: Characterization of RBCEVs, mouse lung epithelial cells and α-EGFR-VHH binding affinity. [Figure 11-2] See description of Figure 11-1. [Figure 12-1]Figure 12A-D are schematic diagrams showing qPCR analysis of DDX58 and its downstream effectors relative to GAPDH in MDA-MB-468 (A), MDA-MB-231 (B), MCF10A (C) and mouse lung epithelial cells (D) treated with 0.1 μg / μL unloaded RBCEVs or NC-RNA-loaded RBCEVs, immRNA-loaded RBCEVs for 24 h (n = 3-4, RNA loaded with REG1). ND, not detected. FIG. 12E is a schematic showing representative flow cytometry plots of ANXV / PI staining in 4T1, CA1a, H358, MDA-MB-468 (MB468), MDA-MB-231 (MB231), MCF10A and mouse lung epithelial (mLE) cells treated with 0.1 μg / μL of unloaded RBCEVs or NC RNA-loaded RBCEVs and immRNA-loaded RBCEVs for 72 hours. FIG. 12F is a schematic showing the average percentage of ANXV+PI+ population in MB468, MB231, MCF10A and mLE cells treated with RBCEVs as in (E) (n = 3). All bar graphs represent the mean ± SEM. *P < 0.05, **P < 0.01 and ****P < 0.0001 as determined by two-tailed Student's t-test. FIG. 12: Effect of immRNA-loaded RBCEVs on cancer and non-malignant cells. [Figure 12-2] See description of Figure 12-1. [Figure 12-3] See description of Figure 12-1. [Figure 13-1]Figure 13A is a schematic showing the secondary structure of 3p-125b-ASO. Figure 13B-D are schematics showing qPCR analysis of DDX58 and its downstream effectors relative to GAPDH in MDA-MB-231 (B), MCF10A (C) and mouse lung epithelial cells (D) treated with 0.1 μg / μL of unloaded or NCRNA-loaded RBCEVs, 125b-ASO-loaded RBCEVs and 3p-125b-ASO-loaded RBCEVs for 24 h (n = 3-4, RNA loaded with REG1). ND, not detected. Figure 13E is a schematic showing representative flow cytometry plots of ANXV / PI staining in 4T1, CA1a, H358, MDA-MB-231 (MB231), MCF10A and mouse lung epithelial (mLE) cells treated with 0.1 μg / μL of unloaded or NC RNA-loaded RBCEVs, 125b-ASO-loaded RBCEVs and 3p-125b-ASO-loaded RBCEVs for 72 hours. Figure 13F is a schematic showing the average percentage of ANXV+PI+ population in MB231, MCF10A and mLE cells treated with RBCEVs as in (E) (n = 3). Figure 13G is a schematic showing representative flow cytometry plots of ANXV / PI staining in 4T1 and CA1a cells after 72 hours of treatment with combined 3p-125-ASO-loaded and immRNA-loaded RBCEVs. Figure 13H is a schematic showing qPCR analysis of RIG-I pathway gene expression in 4T1 and CA1a cells after 24 h treatment with combined 3p-125-ASO and immRNA-loaded RBCEVs. All bar graphs represent mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001, determined by two-tailed Student's t-test. Figure 13: Effect of 3p-125b-ASO-loaded RBCEVs on cancer and non-malignant cells. [Figure 13-2] See description of Figure 13-1. [Figure 13-3] See description of Figure 13-1. [Figure 13-4] See description of Figure 13-1. [Figure 14-1]FIG. 14A is a schematic diagram showing flow cytometry analysis of immune cells in 4T1 mammary tumors treated with RBCEVs loaded with immRNA as in FIG. 4. Cells were gated based on FSC-A and SSC-A to exclude debris and dead cells. Single cells were further gated based on FSC-W and FSC-H to exclude doublets and aggregates. Live cells were gated from the singlet population based on Sytox™ Blue negativity. Fluorescence of immune cell phenotypic markers was then gated based on fluorescence channel and subjected to viSNE analysis. FIG. 14B is a schematic showing survival of mice bearing orthotopic 4T1 tumors treated with 2.5 mg / kg NC-RNA-RBCEV(it), 2.5 mg / kg immRNA-RBCEV(it), 2 mg / kg anti-PD-L1 antibody (intraperitoneally (ip)), and 2.5 mg / kg immRNA-RBCEV(it) in combination with 2 mg / kg anti-PD-L1 antibody (ip) (n = 2-4 mice). FIG. 14C is a schematic showing tumor growth in mice treated as in (B) (n = 2-4 mice). FIG. 14D is a schematic showing ELISA quantification of IFNβ in serum of mice bearing 4T1 tumors treated as in (B) (n = 2-4 mice). FIG. 14E is a schematic showing representative viSNE plots of immune cells in CA1a mammary tumors treated with immRNA-loaded RBCEVs as in FIG. 4. Figure 14F is a schematic diagram showing the volume of MDA-MB-468 tumors injected intratumorally every 3 days with 2.5 mg / kg RBCEVs containing immRNA or NC RNA (n = 3-4 mice). Figure 14G is a schematic diagram showing qPCR analysis of RIG-I pathway gene expression relative to GAPDH in untreated and treated MDA-MB-468 tumors (n = 3-4 mice). All bar graphs represent the mean ± SEM. **P < 0.01, ***P < 0.001 and ****P < 0.0001 as determined by two-tailed Student's t-test. Figure 14: Therapeutic efficacy of immRNA-loaded RBCEVs in mammary tumors. [Figure 14-2] See description of Figure 14-1. [Figure 14-3] See description of Figure 14-1. [Figure 15] Schematic diagram showing flow cytometry analysis with representative viSNE plots of immune cells in 4T1 mammary tumors treated with 3p-125b-ASO-loaded RBCEVs. [Figure 16A] Schematic diagram showing qPCR analysis of RIG-I pathway gene expression relative to Gapdh in 4T1-hEGFR and parental 4T1 cells treated with 0.1 μg / μL uncoated, Ctrl-VHH-coated, and EGFR-VHH-coated immRNA-containing RBCEVs (n = 3). All bar graphs represent mean ± SEM. **P < 0.01 and ***P < 0.001, determined by two-tailed Student's t-test. [Figure 16B] Schematic diagram showing representative viSNE plots of EGFR-VHH-CFSE-RBCEV biodistribution in each cell type in lungs bearing 4T1-hEGFR metastatic tumors. Figure 16: EGFR-VHH-RBCEVs target 4T1-hEGFR cells in vitro and 4T1-hEGFR lung metastatic tumors in vivo. [Figure 17-1] Figure 17A is a schematic showing representative viSNE plots of each cell type in 4T1 metastatic tumor-bearing lungs treated with EGFR-VHH-immRNA-RBCEV. Figure 17B is a schematic showing qPCR analysis of cytokine gene expression relative to Gapdh in lungs of 4T1-hEGFR metastatic tumor-bearing mice (n = 4 mice). All bar graphs represent mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001 as determined by two-tailed Student's t-test. Figure 17: Flow cytometry analysis of immune cells in 4T1 metastatic tumor-bearing lungs treated with EGFR-VHH-immRNA-RBCEV. [Figure 17-2] See description of Figure 17-1. [Figure 17-3] See description of Figure 17-1. [Figure 18-1]Figure 18A is a schematic diagram showing the design and chemical modification of ASO targeting EGFR L858R and T790M. Figure 18B is a schematic diagram showing the knockdown of EGFR in H1975 cells carrying EGFR L858R / T790M mutants and A549 cells carrying wild-type EGFR after treatment with RBCEV loaded with EGFR L858R ASO, as determined by Western blot (n=3). Figure 18C is a schematic diagram showing the knockdown of EGFR in H1975 cells carrying EGFR L858R / T790M mutants and A549 cells carrying wild-type EGFR after treatment with RBCEV loaded with EGFR L858R ASO, as determined by qPCR analysis (n=3). Figure 18D is a schematic diagram showing the reduction in cell viability caused by RBCEV loaded with EGFR L858R ASO4 in H1975 cells and A549 cells as determined by CCK-8 assay (n=5). Figure 18E is a schematic diagram showing the knockdown of EGFR in H1975 cells carrying EGFR L858R / T790M mutants and A549 cells carrying wild-type EGFR after treatment with RBCEV loaded with EGFR T790M ASO as determined by Western blot (n=3). Figure 18F is a schematic diagram showing the knockdown of EGFR in H1975 cells carrying EGFR L858R / T790M mutants and A549 cells carrying wild-type EGFR after treatment with RBCEV loaded with EGFR T790M ASO as determined by qPCR analysis (n=3). Figure 18G is a schematic diagram showing the reduction in cell viability caused by RBCEVs loaded with EGFR T790M ASO4 in H1975 and A549 cells as determined by CCK-8 assay (n=5). All bars represent the mean ± SD. ns, *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001 as determined by two-tailed Student's t-test. [Figure 18-2] See description of Figure 18-1. [Figure 18-3] See description of Figure 18-1. [Figure 19-1] FIG. 19A is a schematic diagram showing qPCR analysis of DDX58 and IFN-β expression in A427 cells after treatment with different RBCEV formulations. FIG. 19B is a schematic diagram showing qPCR analysis of DDX58 and IFN-β expression in CT26 cells after treatment with different RBCEV formulations. FIG. 19C is a schematic diagram showing qPCR analysis of DDX58 and IFN-β expression in AsPC-1 cells after treatment with different RBCEV formulations. FIG. 19D is a schematic diagram showing qPCR analysis of DDX58 and IFN-β expression in H1975 cells after treatment with different RBCEV formulations. FIG. 19E is a schematic diagram showing cell viability of A427 cells after treatment with different RBCEV formulations as determined by CCK-8 assay. FIG. 19F is a schematic diagram showing cell viability of CT26 cells after treatment with different RBCEV formulations as determined by CCK-8 assay. FIG. 19G is a schematic diagram showing cell viability of AsPC-1 cells after treatment with different RBCEV formulations as determined by CCK-8 assay. FIG. 19H is a schematic diagram showing cell viability of H1975 cells after treatment with different RBCEV formulations as determined by CCK-8 assay. FIG. 19I is a schematic diagram showing expression of DDX58 and IFN-β in AsPC-1 cells after treatment with different LNP formulations as determined by qPCR. FIG. 19J is a schematic diagram showing cell viability of AsPC-1 cells after treatment with different RBCEV formulations as determined by CCK-8 assay. FIG. 19K is a schematic diagram showing analysis of activation of THP-1 monocytic cells after incubation with cancer cell culture supernatant. FIG. 19L is a schematic diagram showing flow cytometry analysis of CD86 expression in THP-1 cells after incubation with pretreated pancreatic AsPC-1 cancer cells. All bar graphs represent mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001, determined by one-way ANOVA test. [Figure 19-2] See description of Figure 19-1. [Figure 20]FIG. 20A is a schematic showing treatment of mCherry-luciferase-expressing H1975 lung tumors in NSGS mice by intratracheal administration of NC ASO or EGFR L858R ASO in RBCEVs. FIG. 20B is a schematic showing tumor progression in mice quantified using mean bioluminescence signal after treatment with EV-NC ASO or EV-L858R ASO over time. Graphs present mean ± SEM (n = 5). **p < 0.01, determined by two-way ANOVA test. FIG. 20C is a schematic showing representative bioluminescence imaging of xenografted mice treated with EV-NC ASO or EV-L858R ASO over time. FIG. 20D is a schematic showing representative IHC images of EGFR expression in tumor sections from treated mice at the end of the study. Scale bar, 100 μm. [Figure 21] FIG. 21A is a schematic diagram showing tumor volumes in BALB / c mice injected with CT26 cells and treated with RBCEVs loaded or not with KRAS-G12D ASO and / or immRNA by intratumoral injection every 2 days. Graphs represent mean ± SEM (n = 3). *P < 0.05, **P < 0.01, determined by two-way ANOVA test. FIG. 21B is a schematic diagram showing tumor weights of RBCEV-treated mice at the end point. Graphs represent mean ± SEM (n = 3). *P < 0.05, determined by two-tailed Student's t test. FIG. 21C is a schematic diagram showing representative H&E staining of tumors harvested from treated mice. Scale bar, 100 μm. FIG. 21D is a schematic diagram showing representative TUNEL staining of tumor sections from treated mice at the end of the study. Scale bar, 100 μm. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] definition As used herein, the term "lipid" refers to a group of organic compounds that contain carbon, hydrogen and oxygen atoms.Examples of lipids can be, but are not limited to, eicosanoids, monoglycerides, diglycerides, triglycerides, phospholipids, sphingolipids, terpenes, prenols, fatty acids, waxes, steroids or phospholipids.Further examples of lipids can be, but are not limited to, cationic lipids, ionizable lipids, PEG-lipids, phosphatidylcholine, phosphatidylethanolamine or cholesterol.

[0013] As used herein, the term "cationic lipid" refers to a lipid comprising a hydrophilic head and a hydrophobic tail, where a linker, which can be, but is not limited to, an ether, ester, or amide, connects the hydrophilic head and the hydrophobic tail.

[0014] As used herein, the term "ionizable lipid" refers to a lipid derived from a cationic lipid in which the quaternary ammonium head of the cationic lipid has been replaced with a titratable moiety and is therefore neutral at physiological pH but positively charged at low pH.

[0015] As used herein, the terms "PEG-lipid" and "PEG-modified lipid" are used interchangeably and refer to lipid attached to polyethylene glycol (PEG), with one end of the PEG chain attached to the hydrophobic tail of lipid via a linker.The molecular weight of the PEG linked to lipid can be, but is not limited to, 750, 2000, 5000 or 8000.PEG can be linked to lipid, such as lipid of lipid vesicle or lipid nanoparticle, to extend the blood circulation time of the lipid vesicle.

[0016] As used herein, the term "phospholipid" refers to a lipid that includes a glycerol backbone, a fatty acid tail, and a phosphate group, which can be modified with an organic molecule, including but not limited to, choline, ethanolamine, or serine. Examples of phospholipids can be, but are not limited to, phosphatidylcholine, phosphatidylethanolamine, phosphatidylinositol, or phosphatidylserine.

[0017] As used herein, the term "steroid" refers to a lipid characterized by four carbon rings. Examples of steroids can be, but are not limited to, corticosteroids, progestogens, androgens, estrogens, cholesterol, phytosterols, or ergosterol.

[0018] As used herein, the term "vesicle" refers to a structure that includes a lipid layer that surrounds a liquid or cytoplasm. Vesicles can be made of either a single lipid layer or a lipid bilayer. Vesicles can be naturally occurring or artificially prepared. Examples of vesicles can be, but are not limited to, liposomes, lysosomes, transport vesicles, secretory vesicles, or extracellular vesicles. In the present disclosure, in one example, the vesicle contains or encapsulates a retinoic acid-inducible gene I receptor (RIG-I) agonist that is to be delivered to a cell as defined herein.

[0019] As used herein, the term "extracellular vesicles" or "EVs" refers to cell-derived membrane vesicles secreted by cells into the extracellular space. Extracellular vesicles play an important role in intercellular communication between cells, and the contents of extracellular vesicles can contain specific molecules that can serve as biomarkers or function as mediators of various physiological processes. Extracellular vesicles can be derived from various cells, such as, but not limited to, red blood cells, cancer cells, fibroblasts, epithelial cells, endothelial cells, immune cells, or platelets. Thus, examples of extracellular vesicles can be, but are not limited to, red blood cell-derived extracellular vesicles, milk-derived extracellular vesicles, plasmid-derived extracellular vesicles, or cancer cell-derived extracellular vesicles. In one example, these extracellular vesicles are used to deliver retinoic acid-inducible gene I receptor (RIG-I) agonists to cells, as described herein.

[0020] As used herein, the term "red blood cell derived extracellular vesicles" or "RBCEVs" refers to extracellular vesicles derived from red blood cells and having a diameter of about 120-200 nm. RBCEVs can be purified and purified RBCEVs can have an increase in a marker, which can be, but is not limited to, hemoglobin A (HBA), ALIX or TSG101m. Purified RBCEVs can have a reduction in a marker, which can be, but is not limited to, the cytoskeletal protein β-actin. RBCEVs can be derived from any red blood cell. When derived from human red blood cells, RBCEVs can have common features, which can be, but are not limited to, a lipid membrane, a size of 100-300 nm, an increase in glycophorin A (GPA), or an increase in intraluminal hemoglobin.

[0021] As used herein, the term "lipid nanoparticle" or "LNP" refers to a spherical vesicle made of lipid that is positively charged at low pH and neutral at physiological pH. Lipid nanoparticles can include, for example, but are not limited to, one or more cationic lipids, non-cationic lipids and / or PEG-modified lipids. In one example, lipid nanoparticles include PEG2000-modified lipids. In another example, LNPs are used to deliver retinoic acid-inducible gene I receptor (RIG-I) agonists to cells, as described herein.

[0022] As used herein, the term "retinoic acid-inducible gene I receptor (RIG-I)" refers to a protein that functions as a cytoplasmic RNA sensor, recognizing RNA sequences that have a 5' triphosphate moiety and binding with high affinity to short double-stranded RNA.

[0023] As used herein, the term "antagonist" refers to a chemical entity that binds to a target receptor and inhibits downstream effector mechanisms. Inhibition of downstream effector mechanisms by an antagonist can be reversible or irreversible.

[0024] As used herein, the term "agonist" refers to a chemical substance that activates target receptor and activates downstream effector mechanism, resulting in a biological response. Agonists can be either endogenous or exogenous. In one example described herein, retinoic acid-inducible gene I receptor (RIG-I) agonists are exogenous.

[0025] As used herein, the term "retinoic acid-inducible gene I receptor (RIG-I) agonist" or "RIG-I agonist" refers to a substance that binds to the RIG-I gene or protein (direct agonist), or to a gene or protein upstream of the RIG-I gene or protein (indirect agonist), resulting in the activation of RIG-I and producing a biological response. Examples of RIG-I direct agonists include, but are not limited to, 3p-125b-ASO and immRNA. Examples of RIG-I indirect agonists include, but are not limited to, KRAS-G12D ASO. KRAS-G12D does not bind to RIG-I, but can synergize with immRNA to promote the activation of RIG-I.

[0026] As used herein, the term "immunomodulatory RNA" or "immRNA" refers to the RNA that has a short hairpin RNA (shRNA) structure that binds to target gene.For example, immRNA binds to the gene of interest to activate retinoic acid-inducible gene I receptor (RIG-I). The immRNA can comprise about 1-30 nucleotides, or about 7-11 nucleotides, about 10-14 nucleotides, about 13-17 nucleotides, about 16-20 nucleotides, about 19-23 nucleotides, about 22-26 nucleotides, about 25-30 nucleotides, or about 1, about 2, about 3, about 4, about 5, about 6, about 7, about 8, about 9, about 10, about 11, about 12, about 13, about 14, about 15, about 16, about 17, about 18, about 19, about 20, about 21, about 22, about 23, about 24, about 25, about 26, about 27, about 28, about 29, or about 30 nucleotides.

[0027] As used herein, the term "antisense oligonucleotide" or "ASO" refers to a single-stranded deoxyribonucleotide that is complementary to an RNA target, such as, but not limited to, mRNA, microRNA, or non-long coding RNA. Antisense oligonucleotides silence (block) target genes. For example, antisense oligonucleotides bind to a gene of interest to silence one or more cancer genes. In another example, modified ASOs can activate a gene of interest, such as retinoic acid-inducible gene I receptor (RIG-I). Exemplary modified ASOs can include, but are not limited to, ASOs with a 5' triphosphate cap that binds to miR-125b and / or RIG-I. The antisense oligonucleotide may be about 1 to 30 nucleotides, or about 1 to 5 nucleotides, about 4 to 8 nucleotides, about 7 to 11 nucleotides, about 10 to 14 nucleotides, about 13 to 17 nucleotides, about 16 to 20 nucleotides, about 19 to 23 nucleotides, about 22 to 26 nucleotides, about 25 to 30 nucleotides, or about 1, about 2, about 3, about 4, about 5, about 6, about 7, about 8, about 9, about 10, about 11, about 12, about 13, about 14, about 15, about 16, about 17, about 18, about 19, about 20, about 21, about 22, about 23, about 24, about 25, about 26, about 27, about 28, about 29, or about 30 nucleotides. may include.

[0028] As used herein, the term "siRNA," also known in the art as "small interfering RNA," refers to a double-stranded RNA that is about 20-24 nucleotides in length. siRNAs bind to target mRNA through complementary base pairing and disrupt gene expression by degrading the target mRNA.

[0029] As used herein, the term "antibody" refers to an immunoglobulin molecule capable of binding to a specific epitope on an antigen. Antibodies can be composed of a polyclonal mixture or can be monoclonal in nature. Furthermore, antibodies can be whole immunoglobulins derived from natural or recombinant sources. The antibodies of the present disclosure can exist in various forms, for example, as complete antibodies or as antibody fragments or other immunologically active fragments thereof, such as complementarity determining regions. Antibodies can exist as antibody fragments having functional antigen-binding domains, i.e., heavy and light chain variable domains.

[0030] As used herein, the term "monoclonal antibody" refers to an antibody that recognizes and binds to a single epitope.

[0031] As used herein, the term "single chain antibody" is used interchangeably with the terms "single chain Fv" and "scFv" and refers to an antibody or immunoglobulin that contains the variable regions from both the heavy and light chains connected by a peptide linker and lacks constant regions.

[0032] As used herein, the term "nanobody" refers to an antibody that has only the variable domain of the heavy chain and lacks the light chain. However, the antigen binding ability of nanobody remains similar to that of conventional antibody and can bind to the target. Examples of nanobody can be, but are not limited to, EGFR, HER1, HER2, HER3, human growth factor (HGF), CXCR4, PD-L1, CD49a, CD49b, CD49c, CD49d, CD49e, CD49f, ITGA7, ITGA8, ITGA9, ITGA10, ITGA11, CD11D, CD103, CD11a, CD11b, CD51, CD41, CD11c, CD29, CD18, CD61, CD104, ITGB5, ITGB6, ITGB, ITGB8, FOLR1, FOLR2 or FOLR3 binding nanobody.

[0033] As used herein, the term "cancer marker", also known in the art as "tumor marker", refers to any molecule that can be found at an increased level in subjects suffering from cancer compared to the average level. Cancer markers vary from cancer to cancer and are well known to those skilled in the art who will know which molecules can be found at an increased level in subjects suffering from cancer. Exemplary molecules may be, but are not limited to, alpha-fetoprotein (AFP), immunoglobulins, epidermal growth factor receptor (EGFR), HER1, HER2, HER3, human growth factor (HGF), CXCR4, PD-L1, one or more integrin proteins (e.g., CD49a, CD49b, CD49c, CD49d, CD49e, CD49f, ITGA7, ITGA8, ITGA9, ITGA10, ITGA11, CD11D, CD103, CD11a, CD11b, CD51, CD41, CD11c, CD29, CD18, CD61, CD104, ITGB5, ITGB6, ITGB or ITGB8), or folate receptors (e.g., FOLR1, FOLR2 or FOLR3). EGFR may be found at increased levels in cancers such as, but not limited to, breast cancer, lung cancer, pancreatic cancer, colorectal cancer, or prostate cancer. HER1 / HER2 / HER3 can be found at increased levels in cancers such as, but not limited to, breast, bladder, pancreatic, ovarian, or gastric cancer.

[0034] Detailed Description In recent decades, there have been great strides in the development of immunotherapy, beginning with the landmark approval of the antitumor cytokine interferon-α2 by the US Food and Drug Administration (FDA).

[0035] RIG-I is a cytoplasmic RNA sensor that recognizes RNA sequences with 5' triphosphate moieties and binds to short double-stranded RNA with high affinity. Although the role of RIG-I as a prognostic marker varies across cancer types, its activation consistently induces apoptotic tumor cell death, increases immune cell infiltration into tumors, enhances their anticancer activity, and reduces their immunosuppressive activity.

[0036] Intratumoral delivery of RIG-I agonists induces apoptosis of pancreatic cancer cells in a type I IFN-dependent manner and enhances effective cross-presentation of tumor-associated antigens by dendritic cells to CD8+ T cells, thus leading to tumor regression, for example, in mice bearing pancreatic cancer.

[0037] As RNA oligonucleotides, including small interfering RNA (siRNA) and antisense oligonucleotides, have been developed to target disease genes, combining gene knockdown with RIG-I activation is one approach for anti-cancer therapy.

[0038] However, the clinical application of RNA therapeutics continues to be limited by the current delivery of these molecules to target cells. One of the most widely used delivery vehicles for RIGI agonists is the polymeric in vivo-jetPEI, whose safety has been reported with contrasting reports. in vivo-jetPEI has a tendency to induce liver damage and is associated with poor survival in mice, an effect attributed to toxicity. Therefore, the development of delivery methods is essential for RNA therapeutics.

[0039] Considering the above problems, there is a need to provide a composition that can be used to deliver retinoic acid-inducible gene I receptor (RIG-I) agonists to cells. Effective delivery of retinoic acid-inducible gene I receptor (RIG-I) agonists would allow for the treatment of diseases.

[0040] Thus, the inventors of the present disclosure have discovered a composition comprising a vesicle for delivery of a retinoic acid-inducible gene I receptor (RIG-I) agonist to a cell.

[0041] As used herein, a retinoic acid-inducible gene I receptor (RIG-I) agonist can be an immunomodulatory RNA (immRNA), an antisense oligonucleotide (ASO), a small interfering RNA (siRNA), or any combination thereof.

[0042] Immunomodulatory RNA (immRNA) can be delivered to cancer cells to induce immunogenic cell death of cancer cells by activation of the RIG-I pathway (Figures 2 and 12). Thus, in one example, a retinoic acid-inducible gene I receptor (RIG-I) agonist can be an immRNA. In another example, an immRNA can be a polypeptide having the sequence Includes TIFF2025510600000005.tif4128.

[0043] Antisense oligonucleotides (ASOs) can also be delivered to cancer cells to induce cell death of the cancer cells (Figures 3 and 9). Furthermore, ASOs targeting KRAS-G12D and / or KRAS-G12V can synergize with RIG-I agonists to induce cancer cell death (Figure 9). Thus, in one example, the retinoic acid-inducible gene I receptor (RIG-I) agonist can be an antisense oligonucleotide (ASO). In another example, the antisense oligonucleotide (ASO) can be a miRNA-125b-ASO, a KRAS-ASO, an ASO targeting EGFR, or any combination thereof. In one example, the antisense oligonucleotide (ASO) is a miRNA-125b-ASO. In a further example, the antisense oligonucleotide (ASO) is a polypeptide comprising the sequence The miRNA-125b-ASO comprises TIFF2025510600000006.tif4128. In another example, the antisense oligonucleotide (ASO) is a KRAS-ASO. In a further example, the retinoic KRAS-ASO can be a KRAS-G12D ASO, a KRAS-G12V ASO, a KRAS-G12C ASO or a KRAS-G12S ASO. In yet another example, the KRAS-ASO is a KRAS-G12D ASO that can comprise or consist of any one of the sequences in the following table. TIFF2025510600000007.tif67170

[0044] In yet another example, the KRAS-ASO is a KRAS-G12V ASO that can include or consist of any one of the sequences in the table below. TIFF2025510600000008.tif50170

[0045] In another example, the antisense oligonucleotide (ASO) is the ASO that targets EGFR.In another example, the antisense oligonucleotide (ASO) is the ASO that targets EGFR L858R.In yet another example, the antisense oligonucleotide (ASO) is the ASO that targets EGFR L858R, and can comprise any one of the sequences in the following table or consist of any one of the sequences in the following table. TIFF2025510600000009.tif59170

[0046] In another example, the antisense oligonucleotide (ASO) is an ASO that targets EGFR T790M. In yet another example, the antisense oligonucleotide (ASO) is an ASO that targets EGFR T790M that can include or consist of any one of the sequences in the following table. TIFF2025510600000010.tif59170

[0047] Modifications to retinoic acid-inducible gene I receptor (RIG-I) agonists can activate multiple targets (Figures 3 and 13). Thus, in one example, the immunomodulatory RNA (immRNA) includes modifications at the 5' and / or 3' end. In another example, the modification can be, but is not limited to, a 5' triphosphate cap, phosphorylation, methylation, adenylation, or any combination thereof. In a particular example, the modification is a 5' triphosphate cap. Thus, in one example, the modified immunomodulatory RNA (immRNA) comprises the sequence Includes TIFF2025510600000011.tif4128.

[0048] In another example, the antisense oligonucleotide (ASO) comprises a modification at the 5' and / or 3' end. Suitable modifications can be, but are not limited to, a 5' triphosphate cap, phosphorylation, methylation, adenylation, locked nucleic acid, phosphorothioate, 2-methoxyethyl, and combinations thereof. In a particular example, the modification is a 5' triphosphate cap. Thus, in one example, the modified antisense oligonucleotide (ASO) can be a modified miRNA-125b-ASO, a KRAS-ASO, an ASO targeting EGFR, or any combination thereof. In one example, the modified antisense oligonucleotide (ASO) comprises the sequence The modified miRNA-125b-ASO comprises TIFF2025510600000012.tif4136. In yet another example, the modified antisense oligonucleotide (ASO) is a modified KRAS-ASO. In a further example, the modified KRAS-ASO can be a modified KRAS-G12D ASO, a modified KRAS-G12V ASO, a modified KRAS-G12C ASO, or a modified KRAS-G12S ASO. In one example, the modified KRAS-ASO is a modified KRAS-G12D ASO that can comprise any one of the sequences in the following table or consist of any one of the sequences in the following table. TIFF2025510600000013.tif109170

[0049] In another example, the modified KRAS-ASO is a modified KRAS-G12V ASO that can include or consist of any one of the sequences in the table below. TIFF2025510600000014.tif66170

[0050] In another example, the modified antisense oligonucleotide (ASO) is a modified EGFR targeting ASO.In another example, the modified antisense oligonucleotide (ASO) is a modified EGFR L858R targeting ASO.In yet another example, the modified antisense oligonucleotide (ASO) is a modified EGFR L858R targeting ASO that can include any one of the sequences in the following table or can consist of any one of the sequences in the following table. TIFF2025510600000015.tif83170

[0051] In another example, the modified antisense oligonucleotide (ASO) is a modified EGFR targeting ASO.In another example, the modified antisense oligonucleotide (ASO) is a modified EGFR T790M targeting ASO.In yet another example, the modified antisense oligonucleotide (ASO) is a modified EGFR T790M targeting ASO that can include any one of the sequences in the following table or can consist of any one of the sequences in the following table. TIFF2025510600000016.tif91170, where +: Locked Nucleic Acid (LNA) * : Phosphorothioated (PS) m: 2'-O-methoxyethyl (2'-MOE)

[0052] Extracellular vesicles (EVs) derived from red blood cells (RBCs) are non-immunogenic and non-carcinogenic. The absence of DNA in RBCs eliminates the risk of horizontal gene transfer by RBCEVs, a known risk of using nucleated cell-derived EVs for nucleic acid delivery, and immunogenicity can be abrogated by blood type matching. Furthermore, there is no observable toxicity associated with RBCEV treatment. Therefore, extracellular vesicles (EVs) for delivery of RIG-I offer multiple advantages over existing RIG-I agonist delivery systems due to their limited immunogenicity.

[0053] Thus, the compositions described herein can include extracellular vesicles or lipid nanoparticles for delivery of retinoic acid-inducible gene I receptor (RIG-I) agonists (FIGS. 1 and 11). Thus, in one example, the extracellular vesicles can be, but are not limited to, red blood cell-derived extracellular vesicles (RBCEVs), milk-derived EVs, plasmid-derived EVs, and cancer cell-derived EVs. In a further example, the extracellular vesicles are red blood cell-derived extracellular vesicles (RBCEVs). In another example, the lipid nanoparticles include lipids, which can be, but are not limited to, cationic lipids, ionizable lipids, PEG-lipids, phospholipids, phosphatidylcholine, phosphatidylethanolamine, or cholesterol. In one example, the lipid nanoparticles include PEG2000-modified lipids.

[0054] Specific targeting of cancer cells is a key feature of EV-based drug delivery to increase therapeutic efficacy and reduce toxicity. In the present invention, RBCEVs are conjugated to antibodies to promote the uptake of extracellular vesicles by target cells (Figures 6 and 16). Thus, in one example, the antibody can be, but is not limited to, a nanobody, a monoclonal antibody, or a single-chain antibody. In another example, the antibody is a nanobody. In yet another example, the antibody is a nanobody that binds to a cancer marker. Thus, in one example, the cancer marker can be, but is not limited to, epidermal growth factor receptor (EGFR), HER1, HER2, HER3, human growth factor (HGF), CXCR4, PD-L1, one or more integrin proteins (e.g., CD49a, CD49b, CD49c, CD49d, CD49e, CD49f, ITGA7, ITGA8, ITGA9, ITGA10, ITGA11, CD11D, CD103, CD11a, CD11b, CD51, CD41, CD11c, CD29, CD18, CD61, CD104, ITGB5, ITGB6, ITGB or ITGB8), one or more folate receptors (e.g., FOLR1, FOLR2 or FOLR3), or a combination thereof.

[0055] In the present disclosure, novel compositions of EVs and RIG-I agonists are provided that can activate RIG-I, induce anti-tumor immunity, particularly type I interferon-mediated anti-tumor polarization, and target metastatic EGFR-positive cancers.

[0056] Thus, in one example, disclosed herein is a composition comprising: red blood cell-derived extracellular vesicles (RBCEVs); and / or an immunomodulatory RNA (immRNA) comprising or consisting of the sequence of TIFF2025510600000017.tif4138, further comprising a 5' triphosphate cap; and / or An antisense oligonucleotide (ASO) comprising or consisting of the sequence TIFF2025510600000018.tif4130, further comprising a 5' triphosphate cap (e.g., 3p-125b-ASO).

[0057] In another example, disclosed herein is a composition comprising: lipid nanoparticles; and / or an immunomodulatory RNA (immRNA) comprising or consisting of the sequence of TIFF2025510600000019.tif4128, further comprising a 5' triphosphate cap; and / or KRAS-ASO.

[0058] The composition disclosed herein can be formulated to be suitable for administration.Therefore, in one example, disclosed herein is a pharmaceutical composition comprising the composition disclosed herein and pharma-ceutically acceptable excipient.Suitable excipient is well known to those skilled in the art, and can be, but is not limited to, polymer, lactose, sucrose, or sodium starch glycolate.

[0059] The compositions or pharmaceutical compositions disclosed herein can be packaged in a kit for user convenience. Such kits can include the compositions disclosed herein or the pharmaceutical compositions disclosed herein. Additional components of the kit can include other components, including, but not limited to, anti-cancer drugs or buffers. Suitable anti-cancer drugs can be, but are not limited to, anti-PD1 antibodies, anti-cancer gene ASOs, anti-cancer cytokines, and combinations thereof.

[0060] Effective immunomodulation of the tumor microenvironment and antitumor activity was achieved without observable side effects using the compositions described herein, suggesting that the compositions described herein are advantageous for clinical application. For example, treatment with EVs loaded with RIG-I agonists was validated in breast cancer and lung metastasis models (Figures 4, 5, 7 and 17).

[0061] In another example, delivery of RIG-I agonists and antisense oligonucleotides (ASOs) was demonstrated to induce activation of the RIG-I pathway, production of type I-IFN, and apoptosis of cancer cells. In addition, immRNA also exerts synergistic effects with KRAS-G12D ASO and KRAS-G12V ASO in suppressing the viability of cancer cells.

[0062] Thus, in one example, disclosed herein is a method for delivering RIG-I agonist using extracellular vesicle.In another example, disclosed herein is a method for providing / delivering anti-cancer therapy or anti-cancer immunotherapy to a subject in need thereof.

[0063] The delivery method can be carried out by administering the composition, pharmaceutical composition or kit components of the present disclosure. In one example, the composition disclosed herein or the pharmaceutical composition disclosed herein can be administered together with an anti-cancer drug. The anti-cancer drug can be, but is not limited to, an anti-PD1 antibody, one or more anti-cancer gene ASOs, and one or more anti-cancer cytokines. In another example, the anti-cancer drug can be administered simultaneously with or subsequent to the composition described herein. Suitable modes of administration are well known to those skilled in the art. Thus, in one example, the administration method can be, but is not limited to, intrapulmonary, intranasal, intratracheal, subcutaneous, intravenous, intramuscular, intraarticular, intrasynovial, intrasternal, intrathecal, intrahepatic, intracranial injection, infusion technique, topical, oral, rectal, nasal, buccal, and vaginal. In one example, the administration method is intrapulmonary. In another example, the administration method is intratumoral.

[0064] Intratumoral administration of RBCEVs containing or encapsulating RIG-I agonists and antisense oligonucleotides (ASOs) inhibited tumor growth and induced high levels of type I IFN in the tumor microenvironment, immune cell activation, and significant apoptosis of tumor cells, conferring potent antitumor efficacy through targeted delivery (Figures 7, 8, and 17).

[0065] Thus, disclosed herein are methods of treating a disease using the disclosed compositions, pharmaceutical compositions, or kits in a subject in need thereof.

[0066] In one example, the disease is cancer.In another example, the cancer can be, but is not limited to, breast cancer, lung cancer, pancreatic cancer, brain cancer, head and neck cancer, liver cancer, stomach cancer, colon cancer, prostate cancer, cervical cancer and bone cancer.In another example, the cancer is EGFR positive breast cancer.In another example, the cancer is EGFR positive lung cancer.In another example, the cancer is lung metastasis cancer.In another example, the cancer is pancreatic cancer.

[0067] In yet another example, the disease is a pulmonary disease or disorder. In another example, the pulmonary disease or disorder can be, but is not limited to, asthma or pneumonia.

[0068] Also disclosed herein is a composition, pharmaceutical composition or kit for use in therapy or for use in disease treatment.In one example, disclosed herein is a composition, pharmaceutical composition or kit for use in anti-cancer therapy or anti-cancer immunotherapy.

[0069] In another example, disclosed herein is the use of a composition, pharmaceutical composition, or kit of the present disclosure in the manufacture of a medicament for treating a disease.

[0070] As used in this application, the singular forms "a," "an," and "the" include plural references unless the context clearly dictates otherwise. For example, the term "a genetic marker" includes a plurality of genetic markers, including mixtures and combinations thereof.

[0071] As used herein, the terms "increase" and "decrease" refer to the relative change in a selected trait or characteristic in a subset of a population compared to the same trait or characteristic present in the entire population. Thus, an increase indicates a change on a positive scale, and a decrease indicates a change on a negative scale. The term "change" as used herein also refers to the difference in a selected trait or characteristic of an isolated subset of a population compared to the same trait or characteristic in the population as a whole. However, this term does not involve an assessment of the difference seen.

[0072] As used herein, the term "about" in the context of a concentration of a substance, a size of a substance, a length of time, or other stated value means + / - 5% of the stated value, or + / - 4% of the stated value, or + / - 3% of the stated value, or + / - 2% of the stated value, or + / - 1% of the stated value, or + / - 0.5% of the stated value.

[0073] Throughout this disclosure, certain embodiments may be disclosed in a range format. It should be understood that the description in range format is merely for convenience and brevity, and should not be construed as an inflexible limitation on the disclosed range. Thus, the description of a range should be considered to have specifically disclosed all possible subranges, as well as individual numerical values ​​within the range. For example, the description of a range such as 1 to 6 should be considered to have specifically disclosed subranges such as 1 to 3, 1 to 4, 1 to 5, 2 to 4, 2 to 6, 3 to 6, etc., and individual numerical values ​​within the range, for example, 1, 2, 3, 4, 5, and 6. This applies regardless of the breadth of the range.

[0074] The invention illustratively described herein may be suitably practiced in the absence of any element or limitation not specifically disclosed herein. Thus, for example, terms such as "comprising", "including", "containing" and the like are to be interpreted broadly without limitation. Furthermore, the terms and expressions employed herein are used as terms of description and not as terms of limitation, and the use of such terms and expressions is not intended to exclude any equivalents of the features shown and described or portions thereof, but is recognized as allowing various modifications within the scope of the invention claimed. Thus, although the invention has been specifically disclosed by preferred embodiments and optional features, it is to be understood that modifications and variations of the invention embodied therein disclosed herein may be resorted to by those skilled in the art, and such modifications and variations are considered to be within the scope of the invention.

[0075] The present invention has been described broadly and generally herein. Each of the narrower species and subgeneric groupings falling within the generic disclosure also forms part of the present invention. This includes the generic description of the invention with a proviso or negative limitation removing any subject matter from the genus, regardless of whether the removed material is specifically recited herein.

[0076] Other embodiments are within the scope of the following claims and non-limiting examples. Moreover, when features or aspects of the invention are described in terms of a Markush group, those of skill in the art will recognize that the invention is also thereby described in terms of any individual members or subgroups of members of the Markush group. EXAMPLES

[0077] Experimental Section material and method cell culture Mouse breast cancer 4T1 cell line and human breast cancer MCF10CA1a (CA1a) cell line were purchased from Karmanos Cancer Institute (Wayne State University, USA). Human breast cancer MDA-MB-468 and MDA-MB-231 cell lines, human non-transformed mammary epithelial cell line MCF10A, human lung cancer NCI- H358 (H358) cell line, H1975, H441 and A427 cell lines, human pancreatic cancer AsPC-1 cells and human prostate cancer DU145 cell line, human monocytic THP-1 cells, and mouse colon cancer CT26 cell line were obtained from American Type Culture Collection (ATCC, USA). Human lung epithelial cancer reporter cell lines A549-Dual™ and A549-Dual™ RIGI - / -were purchased from InvivoGen, USA. All cell lines except MCF10A cells were cultured in Dulbecco's Modified Eagle's Medium-DMEM (Thermo Fisher Scientific, USA) supplemented with 10% fetal bovine serum (FBS) (Biosera, France), 1x penicillin-streptomycin (Thermo Fisher Scientific), and 5 μg / ml Plasmocin™ prophylactic (InvivoGen, USA) at 37°C in a humidified incubator with 5% CO2. MCF10A cells were cultured in MEBM medium supplemented with supplements from the MEGM kit (Lonza, UK) and 100 ng / ml cholera toxin (Sigma, USA).

[0078] Isolation of mouse primary lung epithelial cells Mouse lung epithelial cells were isolated from female BALB / c mice as previously described (Kasinski & Slack, 2013). Briefly, 7-week-old mice were euthanized. Mouse lungs were removed and dissociated in RPMI medium containing 10% FBS and 5 mg / ml collagenase IV (Thermo Fisher Scientific) using a GentleMACS dissociator (Miltenyi Biotec, Germany). Cells were filtered through a 70 μm strainer, washed, and seeded in 10 cm plates. Fibroblasts are more sensitive to trypsin and can be removed while epithelial cells are attached for a long time. During the process of epithelial cell selection, removing the first cells that start to detach from the plate by trypsinization helps to increase the purity of epithelial cells. EpCAM expression of epithelial cells was evaluated using flow cytometry after three rounds of selection.

[0079] Purification of RBCEVs Blood samples from healthy donors with informed consent were obtained from the Health Sciences Authority of Singapore and the Hong Kong Red Cross. RBCs were separated from whole blood, and RBCEVs were purified from RBCs as previously described (Pham et al., 2021). Briefly, RBCs were separated from plasma by centrifugation at 1,000 × g for 8 min at 4 °C and washed three times with PBS (1,000 × g for 8 min at 4 °C). Leukocytes were completely removed by leukocyte reduction filters (Nigale, China). Isolated RBCs were collected in Nigale buffer (0.2 g / L citric acid, 1.5 g / L sodium citrate, 7.93 g / L glucose, 0.94 g / L sodium dihydrogen phosphate, 0.14 g / L adenine, 4.97 g / L sodium chloride, 14.57 g / L mannitol), diluted in PBS containing 0.1 mg / mL calcium chloride, and incubated overnight at 37°C and 5% CO2 with 10 μM calcium ionophore (Sigma, USA). RBCs and cell debris were pelleted, and the supernatant was collected by centrifugation at increasing speeds (600 × g for 20 min, 1,600 × g for 15 min, and 3,260 × g for 15 min). The supernatant was filtered through a 0.45 μm membrane and then ultracentrifuged at 50,000 × g for 70 min at 4°C in an SW32 rotor (Beckman Coulter, USA). The pellet was resuspended in 1 mL of PBS and then loaded onto a 60% sucrose cushion and ultracentrifuged at 50,000 × g for 16 h at 4 ° C. RBCEVs were collected at the interface and washed with PBS at 50,000 × g for 70 min at 4 ° C. Purified RBCEVs were stored at -80 ° C in PBS containing 4 % trehalose.

[0080] Characterization of RBCEVs The size distribution and concentration of RBCEVs were quantified using a ZetaView® nanoparticle tracking analyzer (Particle Metrix, Germany). Because hemoglobin is the main component of RBCEVs, the amount of RBCEVs is indicated by the amount of hemoglobin throughout this study. The hemoglobin content of RBCEVs was quantified using a NanoDrop™ 2000 spectrophotometer (Thermo Fisher Scientific).

[0081] Western blot analysis RBCEVs were lysed with RIPA buffer (Thermo Fisher Scientific) supplemented with protease inhibitors (Biotool, USA) for 5 min on ice. Cells were lysed for 30 min on ice. Protein concentration was measured by Pierce™ BCA assay (Life Technologies, USA) using BSA (New England Biolabs, UK) concentration as standard. A total of 50 μg of protein lysate was loaded onto a 10% polyacrylamide gel along with Precision Plus Protein™ Kaleidoscope™ prestained protein standards (Bio-Rad, USA). Proteins were transferred to an Immobilon-P polyvinylidene difluoride membrane (Merck Millipore, USA), blocked with 5% milk in Tris-buffered saline containing 0.1% Tween-20 (TBST) for 1 h, and then incubated with primary antibodies: anti-ALIX (Santa Cruz, USA, 1:1,000 dilution), anti-TSG101 (Santa Cruz, USA, 1:1,000 dilution), anti-human GAPDH (Santa Cruz, USA, 1:1,000 dilution), anti-human GPA (Santa Cruz, USA, 1:500 dilution), anti-human HBA (Santa Cruz, USA, 1:1,000 dilution), anti-human KRAS (Invitrogen, USA, 1:1,000 dilution), and anti-EGFR (Santa Cruz, USA, 1:1,000 dilution). The blots were incubated overnight at 4° C. with either anti-human β-actin antibody (CST, USA, 1:1,000 dilution) or anti-human β-actin antibody (CST, USA, 1:1,000 dilution). The membranes were washed three times with TBST and then incubated with HRP-conjugated secondary antibody (Jackson ImmunoResearch, USA, 1:10,000 dilution) for 1 h at room temperature. The blots were imaged using the ChemiDoc™ gel documentation system (Bio-Rad).

[0082] Transmission electron microscopy RBCEVs were fixed with 2% paraformaldehyde for 10 min and mounted on glow-discharged copper grids (200 mesh, coated with formvar carbon film). Negative staining of RBCEVs was performed by incubating the RBCEVs on the grids with 3% uranyl acetate for 5 min. This was followed by a quick wash with distilled water to remove excess dye. The grids were air-dried for 10 min and then imaged at 100 kV using a Tecnai G2 transmission electron microscope (FEI).

[0083] Single EV flow cytometry Single EV flow cytometry was performed using a CytoFLEX LX flow cytometer (Beckman Coulter, USA). RBCEVs were stained with anti-human GPA-FITC antibody (BioLegend, USA) for 3 h at 4 °C, followed by two washing steps with sterile 0.2 μm-filtered PBS to remove unbound antibody. 3.9 × 10 3 ~2.5×10 5 It has previously been shown that EV suspensions in the range of EVs / μL allow accurate analysis of single EVs ( Pham et al., 2021 ). Therefore, stained RBCEVs were collected at 3 × 10 in 0.2 μm-filtered PBS. 4 EVs were diluted to a concentration of 100 ng / µL. RBCEVs were gated out from background noise using the violet side scatter (VSSC) channel acquired via a 405 nm laser (Figure 11B). Reference 100 nm beads were used to verify gating of the EV population. Data were acquired using the following settings: FSC 138, SSC 180, FITC 3000, VSSC 800, with the trigger signal (VSSC) threshold manually set to 4000. The flow rate was maintained at 30 µL / min (achieving an event rate of approximately 5000 events / sec) and the discard rate was kept below 5% for the duration of the experiment.

[0084] RNA oligonucleotide sequences and modifications immRNA TIFF2025510600000020.tif4129 and 5' triphosphate 125b-ASO (3p-125b-ASO) TIFF2025510600000021.tif4141 was synthesized by in vitro transcription as described in the next section. Anti-miR-125b ASO (125b-ASO) TIFF2025510600000022.tif4128, negative control RNA (NC RNA) TIFF2025510600000023.tif4128 and FAM-labeled NC-ASO TIFF2025510600000024.tif4128 was synthesized by Shanghai GenePharma (Shanghai, China) with 2′ O-methyl modifications on all ribonucleotides.

[0085] Cancer gene-targeting ASOs with 2'-O-methoxyethyl (MOE), phosphorothioate (PS), and locked nucleic acid (LNA) modifications (Figures 9A and 18A) were synthesized by Integrated DNA Technologies (USA).

[0086] In vitro transcription (IVT) of RNA iMRNA was prepared according to the protocol of Yong et al., 2019. Briefly, RNA was incubated with T7 RNA polymerase and a pair of primers. The IVT reactions were in vitro transcribed using TIFF2025510600000025.tif26168. The reactions contained 40 mM Tris-HCl pH 8.0, 30 mM MgCl2, 2 mM spermidine, 10 mM DTT, 0.01% Triton-X100, 5 mM GTP and 4 mM NTPs (CTP, ATP and UTP), 1 μM annealed DNA template, 400-600 nM T7 RNA polymerase, and 0.2 U / mL thermostable inorganic pyrophosphatase at 37 °C overnight. The IVT reactions were stopped and RNA was extracted using phenol:chloroform:isoamyl alcohol (25:24:1). The RNA was precipitated overnight at -80 °C with 3 volumes of 95% ethanol in the presence of 0.1% (v / v) sodium acetate. The target RNA was further separated using a Hi-TrapQ HP column, and the expected RNA bands were excised from 20% denaturing urea-PAGE. TIFF2025510600000026.tif26168 and a pair of primers 3p-125b-ASO was prepared as described above using TIFF2025510600000027.tif19168. The quality of im mRNA and 3p-125b-ASO was assessed again by 20% denaturing urea-PAGE and by IFN activity assay prior to further experiments.

[0087] To verify the sequence of 3p-125b-ASO, the ASO was incubated with RNA 5' pyrophosphohydrolase (RppH) (New England Biolabs) for 1-2 h at 37°C. RNA was analyzed using a BioAnalyzer 2100 (Agilent, USA) and converted to cDNA using the NEBNext Small RNA Library Preparation Kit (New England Biolabs) according to the manufacturer's instructions. Libraries were sequenced using an Illumina HiSeq 2500 system. Sequencing reads were processed using Geneious Prime to amplify 3' adapter sequences. TIFF2025510600000028.tif4128 was trimmed. Sequences were aligned to the miR-125b ASO using Bowtie.

[0088] RNA loading into RBCEVs 50 μg of RBCEVs were transfected with 1 μg of RNA using REG1 EV Loading Reagent (Carmine Therapeutics, USA) or Exo-Fect™ Exosome Transfection Kit (System Biosciences, Canada) according to the manufacturer's protocol. Free RNA and transfection reagent were then washed away by centrifugation at 21,000 × g for 30 min. For electroporation, 50 μg of RBCEVs were electroporated with 1 μg of RNA at 250 V using an exponential program with a fixed capacitance of 100 μF in a 0.4 cm cuvette using a GenePulser Xcell electroporator (Bio-Rad).

[0089] RNA loading into lipid nanoparticles (LNPs) RNA-loaded LNPs were prepared by the dilution method. Briefly, 10 μg of NC-ASO, KRAS G12D ASO, immRNA, NC-ASO-immRNA or KRAS G12D ASO-immRNA in 500 μl of HEPES buffer (10 mM, pH 4.0) was added to 500 μl of ethanol containing 185 nmol of DOTAP, 500 nmol of DOPE, 300 nmol of cholesterol and 15 nmol of DSPE-mPEG while vortexing. It was then added to 4.5 ml of HEPES (10 mM, pH 4.0) solution to dilute the ethanol concentration. The resulting solution was centrifuged (4000 g, 30 min, 20°C) using an Amicon Ultra 100K tube. The LNPs were then buffer exchanged and concentrated by adding 4.0 ml of PBS pH 7.4 to the tube and centrifuged again at 4000 g, 30 min, 20°C.

[0090] RNA loading into lipid nanoparticles (LNPs) 1,2-Dioleoyl-3-trimethylammonium-propane (chloride salt) (DOTAP), (1,2-Dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE), 1,2-Distearoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000] (ammonium salt)) (DSPE-mPEG2000), and cholesterol were purchased from Avanti Polar Lipids (Alabama, USA). RNA-loaded LNPs were prepared by the dilution method. Briefly, 10 μg of NC-ASO, KRAS G12D ASO, immRNA, NC-ASO-immRNA or KRAS G12D ASO-immRNA in 500 μl of HEPES buffer (10 mM, pH 4.0) was added to 500 μl of ethanol containing 185 nmol of DOTAP, 500 nmol of DOPE, 300 nmol of cholesterol and 15 nmol of DSPE-mPEG while vortexing. It was then added to 4.5 ml of HEPES (10 mM, pH 4.0) solution to dilute the ethanol concentration. The resulting solution was centrifuged (4000 × g, 30 min, 20 ° C) using an Amicon Ultra 100K tube. The LNPs were then buffer exchanged and concentrated by adding 4.0 ml of PBS pH 7.4 to the tube and centrifuged again at 4000 g, 30 min, 20 ° C.

[0091] Quantification of loaded RNA in RBCEVs After loading of RNA, RBCEVs or LNPs were incubated with 1% Triton-X (Sigma) for 5 min at room temperature, then incubated with heparin sulfate at a final concentration of 20 mg / mL for 1 h at 37° C. After incubation, the mixtures were loaded onto a 2% Tris-acetate-EDTA agarose gel with GelRed® nucleic acid gel stain (Sigma), separated at 100 V for 60 min, and visualized with a ChemiDoc™ gel documentation system. Fluorescence intensity of bands was quantified using ImageJ v1.8.0.

[0092] Conjugation of RBCEVs with anti-human EGFR nanobodies A method for RBCEV surface functionalization using OaAEP1 ligase was previously developed to ligate peptides with a ligase binding motif (NGL) at the C-terminus to the RBCEV surface (Pham et al., 2021). For OaAEP1-mediated peptide ligation, RBCEVs were incubated for 3 h at 25 °C in a solution containing a final concentration of 2 μM ligase and 500 μL of biotinylated linker peptide (Biotin-TRNGL, GL Biochem, China). After incubation, RBCEVs were washed three times with PBS using centrifugation at 21,000 × g for 20 min. Then, biotin-TRNGL-ligated RBCEVs were incubated with streptavidin (SA) (Abcam, UK) at a final concentration of 0.1 mg / mL for 2 h at room temperature. SA-biotin-RBCEVs were then washed as previously described to remove free unbound streptavidin. For further functionalization, the sequences of anti-human EGFR nanobody (α-EGFR-VHH) and anti-mCherry nanobody (α-mCherry-VHH) were cloned with epitope tags for detection and purification. DNA encoding VHH was synthesized by Guangzhou IGE Biotechnology Ltd (China) and inserted into pET32(a+) plasmid behind a T7 promoter. Nanobodies were expressed and purified as previously described (Pham et al., 2021). After purification, α-EGFR-VHH or α-mCherry-VHH were biotinylated using the Type B-Lightning-Link® Biotinylation Kit (Abcam) according to the manufacturer's instructions. Biotin-hEGFR-VHH or biotin-mCherry-VHH were then incubated with SA-biotin-RBCEV for 6 h at 4 °C and washed twice with PBS. After washing, uncoated or VHH-coated RBCEVs were incubated with 20 μM CFSE (Life Technologies) for 1 h at 37°C and washed twice with PBS.After washing, the labeled RBCEVs were washed once using centrifugation, loaded onto a size-exclusion chromatography (SEC) column (Izon, New Zealand) and eluted with PBS to wash away unbound dye. RBCEVs collected from SEC fractions 7–9 were washed twice with PBS by centrifugation at 21,000 × g for 15 min at 4°C. For in vivo treatment, uncoated or VHH-coated RBCEVs were loaded with immRNA using REG1 as previously described.

[0093] In vitro treatment of cancer cells with RBCEVs A total of 50,000 4T1, CA1a, or H358 cells were seeded in 24-well plates prior to incubation with 0.1 μg / μL immRNA-loaded RBCEVs, 3p-125b-ASO-loaded RBCEVs, 125b-ASO-loaded RBCEVs, or NC-RNA-loaded RBCEVs for 24–72 h in a humidified incubator at 37 °C and 5% CO2. Cells were then harvested for immunofluorescence imaging, RT-qPCR, or flow cytometry analysis. For dose-response assays, 50,000 CA1a cells were incubated with 0.2, 0.1, 0.05, 0.025, and 0.0125 μg / μL 125b-ASO-loaded RBCEVs for 24 h and then harvested for RT-qPCR analysis. For EV-targeted delivery assays, 100,000 4T1-hEGFR cells were incubated with 0.02 μg / μL of uncoated or VHH-coated CFSE-labeled or FAM-NC-ASO-loaded RBCEVs for 2 h and then collected for flow cytometry analysis. 100,000 4T1-hEGFR cells were incubated with 0.1 μg / μL of uncoated or VHH-coated immRNA-RBCEVs for 24 h and then collected for RT-qPCR analysis.

[0094] To examine the knockdown efficacy of KRAS G12D, KRAS G12V, EGFR L858R, or EGFR T790M-targeting ASOs, 50,000 cells were seeded into each well of a 12-well plate and cultured overnight in a humidified incubator at 37°C and 5% CO2. The cells were then treated with 100 nM NC ASO or oncogene-targeting ASO-loaded RBCEVs. After 48 h, the cells were harvested for RT-qPCR and Western blot analysis.

[0095] Cell viability assay A total of 10,000 CA1a cells were seeded in 96-well plates before incubation with 0.05 μg / μL NC-ASO-loaded RBCEVs. After 24, 48, and 72 h, cells were incubated with 10% (v / v) CCK-8 reagent (Biosharp, China) for 2 h at 37 °C and 5% CO2. The absorbance was measured at 450 nm using a microplate reader (Tecan, Switzerland).

[0096] To evaluate the antitumor effect of the combination of KRAS or EGFR ASO and immRNA, 5,000 AsPC-1 cells were seeded into each well of a 96-well plate and then incubated with 0.1 μg / μl of NC-ASO-, targeted-, immRNA-, NC-ASO-immRNA or KRAS G12D ASO-immRNA-loaded RBCEVs or lipid nanoparticles for 48 hours. The cells were then incubated with 10% (v / v) CCK-8 reagent and the absorbance was measured as described above. Cell viability was determined according to the following formula: Cell viability (%) = [(OD450 (サンプル) - OD450 (ブランク) ) / (OD450 (対照) - OD450 (ブランク) )] × 100%

[0097] Immunofluorescence imaging CA1a cells were pre-seeded on 12 mm coverslips (Citoglass, China) coated with poly-D-lysine (Gibco, USA) 24 h prior to treatment. After treatment with FAM-NC-ASO-loaded RBCEVs, coverslips were rinsed with fresh medium and stained with CellMask™ Deep Red Plasma Membrane Stains (Thermo Fisher Scientific) for 10 min at 37°C. Cells were rinsed twice in PBS and stained with Hoechst 33342 (Abcam) for 5 min at room temperature. Coverslips were rinsed three times with PBS and then fixed with 4% paraformaldehyde in PBS (Alfa Aesar, USA) for 12 min. Coverslips were then washed three times with PBS followed by a final wash with distilled water before mounting on slides with antifade fluorescent mounting medium (Abcam). Images were acquired using an Olympus FV3000 confocal microscope (Olympus, Japan). Image acquisition was performed using FluoView software, and further analysis and quantification were performed using ImageJ v1.8.0. Cellular areas were selected as regions of interest (ROI) based on the extended mask of the Hoechst signal. The FAM signal was measured as the mean pixel intensity of the ROI. The total measurement area covered 1200-1600 cells in each condition.

[0098] IFN reporter assay A549-Dual™ and A549-Dual™ RIG-I - / - Cells were seeded in 96-well plates at 10,000 cells / well in medium for 18–24 h, followed by incubation with RBCEVs. 0.05 μg / μL immRNA-, 3p-125b-ASO-, or NC-RNA-loaded RBCEVs were incubated with cells for 24 and 48 h. Lucia luciferase in the supernatant was detected with QUANTI-Luc reagent (InvivoGen) on a Synergy H1 microplate reader (Biotek, USA).

[0099] Activation of immune cells in co-culture systems AsPC-1 cells were seeded at a density of 50,000 cells / well in 12-well plates and cultured overnight in a humidified incubator at 37°C and 5% CO2. The cells were then treated with RBCEVs that delivered either 1 μg / ml of NC ASO, KRAS G12D ASO, immRNA, or a combination of KRAS G12D ASO and immRNA. After 48 hours, the cell medium was harvested and centrifuged at 21,000 × g for 20 minutes to completely remove free RBCEVs. The supernatant was collected and added to THP-1 cells pre-seeded at a density of 100,000 cells / well in 12-well plates. After 24 hours, the treated THP-1 cells were harvested, blocked with anti-human CD16 / CD32 antibodies, and stained with anti-human CD86-APC (Biolegend). The stained cells were then analyzed by a FACS LSRII cytometer (BD BioSciences, USA).

[0100] Monocytes from Ficoll-isolated human peripheral blood mononuclear cells (PBMCs) were separated from lymphocytes by using human CD14 microbeads according to the manufacturer's protocol (Miltenyi Biotec, USA). Monocytes were differentiated into macrophages by culturing in RPMI medium supplemented with 10% fetal bovine serum, 1x penicillin-streptomycin, and 10 ng / ml M-CSF for 7 days before incubation with the supernatants of treated PDOs. PCA067 PDOs were seeded in 24-well plates until they reached a diameter of 50–200 nm. The PDOs were then treated with RBCEVs that delivered either 1 μg / ml NC ASO, KRAS G12D ASO, immRNA, or a combination of KRAS G12D ASO and immRNA. After 48 h, cell media was collected and centrifuged at 21,000 × g for 20 min to completely remove free RBCEVs. The collected supernatants were added to PBMC-derived macrophages. After 24 hours, the macrophages were harvested for RNA extraction and qPCR analysis.

[0101] Generation of in vivo cancer models and treatment with RBCEVs All mouse experiments were performed in accordance with experimental protocols approved by the Institutional Animal Care and Use Committee of the National University of Singapore. Age-matched mice were tagged and randomly divided into control and test treatment groups. Experiments were performed in a blinded manner. Mice that died accidentally due to anesthesia were excluded. BALB / c mice were purchased from InVivos Pte Ltd, Singapore. NSG-SGM3 mice (NOD.CgPrkdc <scid>Il2rg <tm1wjl> / Tg(CMV-IL3,CSF2,KITLG)1Eav / MloySzJ) and BALB / c nude mice (NU / JInv strain) were purchased from Jackson Laboratory, USA.

[0102] Intratumoral administration of RBCEVs 1.25 × 10 5 4T1 cells were injected into the 4T1 tumor-bearing mice. Ten days later, mice were randomly divided into groups and intratumorally injected with 2.5 mg / kg immRNA-loaded RBCEVs, 5 mg / kg 3p-125b-ASO-loaded RBCEVs, or the same amount of NC-RNA-loaded RBCEVs as a control. Intratumoral injections were repeated every 3 days for a total of five times. For PD-L1 blockade treatment, 1 day after intratumoral injection of immRNA-loaded RBCEVs, mice bearing 4T1 tumors were intraperitoneally injected with 2 mg / kg anti-mouse PD-L1 monoclonal antibody (BioXCell, USA). Treatments were repeated five times at 3-day intervals. Tumor size was measured every 2 days using digital calipers. Mice were sacrificed when untreated tumors approached approximately 15 mm in diameter. Tumors were collected for RNA extraction, flow cytometry analysis, or imaging.

[0103] 1 × 10 6 CA1a cells were injected into the mice. After 14 days, the mice were randomized and intratumorally injected with 2.5 mg / kg immRNA- or NC-RNA-loaded RBCEVs. Intratumoral injections were repeated every 3 days for a total of 5 times. Tumor size was measured every 2 days using a digital caliper. Mice were sacrificed when untreated tumors approached approximately 15 mm in diameter. Tumors were collected for RNA extraction, flow cytometry analysis, or imaging.

[0104] 5 × 10 5 CT26 cells were injected subcutaneously. 3 At the time of reaching the peri-peritoneal stage, mice with comparable tumor volumes and body weights were selected and randomly divided into four groups (designated as day 0) and injected intratumorally with 2.5 mg / kg RBCEV, NC ASO, KRAS G12D ASO, or 5 mg / kg KRAS G12D ASO and immRNA-loaded RBCEV at 2-day intervals. Tumor size was measured every 2 days using digital calipers. Mice were sacrificed when untreated tumors approached approximately 15 mm in diameter. Isolated tumors were measured for absolute wet weight on an automated electronic balance. Tumors were then subjected to RNA extraction or imaging.

[0105] Intrapulmonary targeted delivery of RBCEVs To generate a breast cancer metastasis model, 4T1 cells were transfected with a lentiviral vector (pHAGE-EGFR, Addgene, USA), selected with puromycin (Santa Cruz), and sorted using an Aria II sorter (BD Biosciences) to generate stable cell lines with high expression levels of human EGFR. A total of 2.5 × 10 5 4T1-hEGFR cells were intravenously injected into 7-week-old female BALB / c mice. One day after inoculation, pulmonary delivery of 25 mg / kg of uncoated or VHH-coated immRNA-RBCEVs was performed in each mouse every 2 days using a Microspray Aerosolizer for Mice (Yuyan, China). After five treatments, the mice were sacrificed and the lungs were harvested for RNA extraction, flow cytometry analysis, or imaging. To obtain the pulmonary biodistribution of RBCEVs, 5 × 10 5 4T1-hEGFR cells were injected intravenously into 7-week-old female BALB / c mice. Seven days later, intrapulmonary delivery of 25 mg / kg of uncoated or VHH-coated CFSE-RBCEVs was performed. After 24 hours, mice were sacrificed and lungs were removed for flow cytometry analysis.

[0106] To generate an orthotopic lung cancer model, H1975 cells were transduced with a lentiviral vector (pLenti-mCherry-luc, Addgene, USA), selected with puromycin (Santa Cruz), and sorted using an Aria II sorter (BD Biosciences) to generate a stable luciferase- and mCherry-expressing H1975 cell line (Luc-mCherry-H1975). A total of 1 × 10 6 Luc-mCherry-H1975 cells were injected intravenously into 8-week-old NSGS mice. After 5 weeks, mice were subjected to IVIS imaging to confirm tumor growth in the lungs. Mice with comparable bioluminescence intensity were divided into two groups (referred to as day 0) and intratracheally administered 5 mg / kg NC ASO or EGFR L858R ASO-loaded RBCEVs at 3-day intervals. On day 14, mice were sacrificed and lung tissues were collected for H&E and immunohistochemical staining.

[0107] Flow cytometry analysis RBCEV analysis To quantify FAM-NC-ASO loading efficiency, 50 μg of RBCEVs were incubated with 2.5 μg of latex beads (Thermo Fisher Scientific) overnight at 4°C on a shaker, washed three times with PBS, and then resuspended in 100 μL of FACS buffer. FACS analysis of the latex beads was performed using a CytoFlex LX hemocytometer (Beckman Coulter). FACS plots were generated using FlowJo v10.0.7.

[0108] Apoptosis analysis After treatment with RBCEVs, cells were collected and washed with PBS. Apoptosis was determined by Annexin V (ANXV) / propidium iodide (PI) staining using an Apoptosis Detection Kit (Life Technologies). Briefly, 50,000 treated cells were incubated with ANXV and PI in binding buffer for 15 min at 4°C. Cells were then analyzed by a FACS LSRII cytometer (BD BioSciences, USA) or a CytoFlexS cytometer (Beckman Coulter). FACS plots were generated using FlowJo v10.0.7.

[0109] Immune cell analysis Tumors were excised, washed with PBS, and dissociated in DMEM medium containing 10% FBS and 5 mg / mL collagenase IV (Thermo Fisher Scientific) using a GentleMACS dissociator (Miltenyi Biotech, Germany). Cells were filtered through a 70 μm strainer and blocked with anti-mouse CD16 / CD32 antibody (BioLegend), followed by incubation with anti-mouse CD45-PECy7 (BioLegend), anti-mouse CD11b-FITC (BioLegend), anti-mouse F4 / 80-APC (BioLegend), anti-mouse Ly6G / C-APC (BioLegend), anti-mouse CD3ε-APC (BioLegend), anti-mouse CD11c-PB (BioLegend), anti-mouse CD49b-APC (BioLegend), anti-mouse SiglecF-PE (BioLegend), anti-mouse CD103-APC (BioLegend), anti-mouse CD4-BV421 (BioLegend), anti-mouse CD8-APC (BioLegend), and anti-mouse MHCII-PE-Cy7. (BioLegend), anti-mouse CD206-APC (BioLegend), anti-mouse CD86-APC-Cy7 (BioLegend), anti-FLAG-APC antibody (BioLegend), or anti-human EGFR nanobody for 30 min at 4 °C. Cells were then washed three times in FACS buffer. Cells were analyzed using a FACS LSRII cytometer (BD BioSciences) or a CytoFLEX S cytometer (Beckman Coulter) or a CytoFLEX LX cytometer (Beckman Coulter). Dead cells were identified by staining with SYTOX™ dye (Thermo Fisher Scientific) and excluded from the analysis. FCS files were analyzed using FlowJo v10.0.7, Kaluza (Beckman Coulter) or Cytobank viSNE (Beckman Coulter). Briefly, cells were first gated using an FSC-A vs. SSC-A plot to exclude debris and dead cells.Single cells were then gated via FSC-W vs. FSC-H plots to exclude doublets and aggregated cells. Fluorescent positive cell populations were then gated by the target fluorescent channel and then subjected to viSNE analysis. Equal event sampling was selected. viSNE plots of each individual parameter were downloaded from Cytobank. Cell phenotypes were assigned in the viSNE plots based on distribution and expression characteristics using phenotypic markers.

[0110] Ex vivo analysis of tumor-specific T cell activity For detection of CD69 and granzyme B expression on tumor-infiltrating CD8+ T cells, CD8+ T cells were selected from dissociated tumor cells using a CD8+ tumor-infiltrating lymphocyte isolation kit (Miltenyi Biotec). The isolated CD8+ T cells were blocked with anti-mouse CD16 / CD32 antibodies and stained with anti-mouse CD69-PE antibody (BioLegend) or anti-mouse granzyme B-FITC antibody (BioLegend). Data were acquired using a FACS LSRII hemocytometer and analyzed using FlowJo. For detection of tumor-specific T cell activity, the isolated CD8+ T cells were incubated with 4T1-hEGFR-luciferase cells at a ratio of 50:1 for 48 hours. The supernatants were harvested after removing floating cells. The amount of IFNγ in each supernatant was analyzed using a mouse IFNγ ELISA kit (BioLegend) according to the manufacturer's instructions. Tumor cell viability was assessed using a luciferase assay kit (Promega, USA) according to the manufacturer's instructions, and data were obtained using a microplate reader (Tecan).

[0111] Mouse cytokine immunoassay Mouse blood was collected by cardiac puncture at the end of in vivo treatment. Whole blood was allowed to clot for 30 min at 37°C, and serum was separated after centrifugation at 5,000 rpm for 20 min. Separated serum was collected and stored at -80°C. IFNβ concentration in serum was measured using a Recombinant Mouse IFNβ ELISA kit (Biolegend) according to the manufacturer's instructions. For multiplex immunoassays, cell culture supernatants were collected after centrifugation at 300 × g for 5 min. Tumors were excised and homogenized in cold RIPA buffer supplemented with protease inhibitors. Tumor lysates were collected using centrifugation at 12,000 × g for 20 min. Cytokine concentrations in collected supernatants or tumor lysates were measured using a ProcartaPlex™ Multiplex Immunoassay kit (Invitrogen, USA) according to the manufacturer's instructions. Cytokines assayed include IFNα, IFNβ, IL-6, IL-10, IL-12p40, and TNFα.

[0112] RNA extraction and RT-qPCR Total RNA was extracted from cells or tissues using TRIzol (Thermo Fisher Scientific) according to the manufacturer's manual. RNA was converted to cDNA using a High Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific) according to the manufacturer's protocol. mRNA levels were quantified using the Ssofast® Green qPCR kit (Bio-Rad) and normalized to GAPDH (see Supplementary Table S1 for primer sequences). miRNA levels were quantified using the Taqman® miRNA qPCR kit (Thermo Fisher Scientific) and normalized to snoRNA234 (mouse) or U6B snRNA (human). All qPCR reactions were performed using a CFX96 Touch™ The PCR was performed using a Real-Time PCR Detection System (Bio-Rad) or a QuantStudio 6 Flex Real-Time PCR System (Life Technologies).

[0113] Hematoxylin and eosin (H&E) staining and TUNEL assay After overnight fixation in 10% neutral buffered formalin (Sigma), tumor or lung tissues were dehydrated sequentially in 70%, 80%, 90%, and 100% ethanol at 37°C using a Leica TP1020 tissue processor (Leica, Germany). Samples were cleared in three baths of Histo-Clear (National Diagnostics, USA) for 1.5 h each at 37°C and impregnated in three baths of paraffin wax (Thermo Fisher Scientific) for 1 h each at 62°C, respectively. Paraffin blocks were sectioned at 5 μm using a Leica RM2255 rotary microtome. Sections were dried at 37°C and degreased sequentially in three baths of Histo-Clear, followed by immersion in two baths of absolute ethanol and one bath of 70% ethanol for 10 min each. Sections were rehydrated in 90%, 75%, and 50% ethanol for 5 min each and in distilled water for 10 min. Sections were then stained with hematoxylin (Abcam) for 5 min. After washing in water, sections were treated with 0.3% acid alcohol, washed, and stained blue with Bluing Reagent (Abcam). Sections were then stained with 0.5% eosin (Abcam) for 1 min. After washing in water, sections were dehydrated in absolute ethanol, cleared in Histo-Clear, and then mounted with Histomount mounting solution (National Diagnostics). Images were acquired with a TissueFAXS PLUS slide scanner (TissueGnostics, Austria). Image acquisition was performed with TissueFAXS viewer software, and further analysis and quantification were performed with ImageJ v1.8.0.

[0114] Apoptosis was assessed using the TUNEL assay BrdU-Red kit (Abcam) in combination with immunofluorescence staining. 4T1 tumor sections were washed twice with PBS. Antigen retrieval was performed by heating the sections in a microwave oven for 15 minutes until boiling of antigen retrieval solution (Tris-EDTA, pH 9.0) was achieved. Sections were allowed to cool in the retrieval solution for 30 minutes. Blocking buffer (5% normal donkey serum and 0.3% TritonX-100 in PBS) was applied for 1 hour at room temperature. After overnight incubation at 4°C with rabbit anti-mouse α-smooth muscle actin antibody (Abcam, 1:250 dilution), sections were incubated with Alexa Fluor® 647 fluorophore-conjugated donkey anti-rabbit secondary antibody (Jackson ImmunoResearch, 1:200 dilution) for 1 hour at room temperature. Sections were then stained using the TUNEL assay kit according to the manufacturer's protocol. After staining, images of TUNEL staining were acquired using an LSM-710 NLO confocal microscope (Zeiss, Germany) or an Olympus FV3000 confocal microscope (Olympus). Image acquisition was performed using Zeiss Zen software or FluoView software, and further analysis and quantification were performed using ImageJ v1.8.0. Nuclear areas were selected as regions of interest (ROI) based on Hoechst signal. BrdU-Red signal was measured as the average pixel intensity of the selected ROI.

[0115] statistical analysis All statistical analyses were performed using Student's two-tailed t-test in GraphPad Prism 8 (GraphPad Software, CA) to determine significant differences between treated samples and controls. A p-value of less than 0.05 was considered significant based on at least three independent repeats. In all graphs, data are presented as median or mean and standard error of the mean (SEM). Animal experiments were repeated in groups of 5–6 mice. A minimum sample size of 3 was used to compare mean differences between two independent groups with an α error probability = 0.05, effect size d = 5, and power = 0.95 for G-tests. * The results were determined using power analysis.

[0116] result RBCEVs can be loaded with small RNAs using multiple methods for RNA delivery to cancer cells RBCEVs were purified according to a previous protocol (Usman et al., 2018). Purified RBCEVs were enriched in common EV markers, such as ALIX and TSG101, as well as the major RBC protein, hemoglobin A (HBA) (Figure 11A). RBCEVs were also found to express glycophorin A (GPA) on their surface as an exclusive marker for EVs derived from human RBCs (Figure 11A-B). The cytoskeletal protein β-actin was nearly absent in RBCEVs, suggesting that purified RBCEVs were free of cellular debris (Figure 11A). Furthermore, purified RBCEVs were uniform in size, ranging from 120 to 200 nm in diameter, as measured using a nanoparticle analyzer (Figure 1D).

[0117] To deliver therapeutic RNA to cancer cells, RBCEVs were loaded with small RNAs using multiple methods, including electroporation, Exo-Fect-mediated transfection, and REG1-mediated transfection. Three rounds of centrifugation were used to separate RBCEVs from the RNA-transfectant complexes. Bead-assisted flow cytometry analysis of FAM-ASO-loaded RBCEVs revealed that FAM-ASOs were efficiently loaded into RBCEVs using the three different methods (Figure 1A-B). The loading efficiency by Exo-Fect and REG1 was relatively higher than that by electroporation (Figure 1A-B). We also observed that loading with REG1 resulted in a homogenous population of RBCEVs with strong FAM fluorescence (Figure 1A-B). In contrast, simple incubation did not promote the entry of FAM-ASOs into RBCEVs (Figure 1A). Transmission electron microscopy revealed that after REG1 transfection, RBCEVs retained a similar morphology to untreated controls, whereas electroporated and Exo-Fect transfected RBCEVs showed signs of aggregation, as indicated by a greater extent of pleated membranes (Figure 1C). Nanoparticle tracking analysis also revealed that REG1 maintained the size of RBCEVs (Figure 1D). Electroporation and Exo-Fect transfection made RBCEVs more prone to aggregation, as evidenced by an increase in the proportion of EVs with large particle sizes, approximately 400–600 nm in diameter (Figure 1D).

[0118] The uptake of FAM-ASO-loaded RBCEVs by human breast cancer MCF10aCA1a (CA1a) cells was further investigated using the three loading methods. FACS and immunofluorescence analysis demonstrated that CA1a cells readily internalized RBCEVs containing FAM-ASO (Figure 1E-G). Confocal imaging revealed that FAM signals were present intracellularly (Figure 1E). Exo-Fect-transfected RBCEVs showed the highest uptake by CA1a cells as shown by FACS (Figure 1G), whereas REG1-transfected RBCEVs were most highly internalized by CA1a cells as shown by immunofluorescence analysis (Figure 1E-F). To evaluate the potential toxic effects of the three loading methods, CA1a cells were incubated with electroporated, Exo-Fect-transfected, and REG1-transfected RBCEVs for 3 days. CCK-8 assay analysis revealed that electroporation and REG1 were non-toxic to CA1a cells, whereas Exo-Fect showed mild toxicity to cells after 24 h incubation but no toxicity at later time points (Figure 1H). Overall, RBCEVs treated with REG1 and Exo-Fect are comparable in delivery. Both REG1 and Exo-Fect perform better than electroporation for loading RNA into RBCEVs. Therefore, Exo-Fect and REG1 were used as transfectants for RBCEVs in subsequent experiments. To estimate the loading efficiency of the transfectants, RBCEVs were transfected with miR-125b ASO using Exo-Fect and REG1, and then dissolved by detergent. The miR-125b ASO was separated by gel electrophoresis (Figure 1I). Based on the fluorescence intensity of the bands, it was observed that approximately 94.2% and approximately 86% of the 125b-ASO was loaded into RBCEVs by Exo-Fect and REG1, respectively (Figure 1I).Of the two transfection methods, REG1 resulted in stronger inhibition of miR-125b in CA1a cells with similar transfection efficiency, as shown by dose-response assay (Figure 1J).

[0119] RBCEVs deliver immunomodulatory RNA to activate the RIG-I pathway and induce immunogenic cell death of cancer cells A series of immRNA species are known in the art as potent RIG-I agonists in human skin dendritic cells and macrophages (Yong et al., 2019; Ho et al., 2019). Among the immRNA variants, 3p10LA9 showed the strongest cellular IFN production activity. To examine the effect of immRNA 3p10LA9 (Figure 2A) on breast cancer cells, we transfected immRNA into RBCEVs using REG1 and incubated the immRNA-loaded RBCEVs with highly metastatic mouse breast cancer 4T1 cells for 24 hours. Scrambled RNA was used as a negative control (NC). qPCR analysis revealed that uptake of immRNA-loaded RBCEVs by 4T1 cells resulted in significant upregulation of cellular RLRs, Ddx58 (the gene encoding RIG-I) and Mda5, as well as RLR downstream effectors including Irf3, Irf7, Ifnb, Rsad2 (viperin) and Isg56, compared with NC-RNA-loaded RBCEVs (Figure 2B). The effect of immRNA was also evaluated in human triple-negative basal-like breast cancer (TNBC) CA1a cells, an advanced aggressive and metastatic derivative of MCF10A cells, and similar upregulation of RIG-I-associated genes was obtained with the same dose of immRNA-loaded RBCEVs. Uptake of immRNA-loaded RBCEVs by CA1a cells significantly upregulated intracellular levels of DDX58, MDA5, IRF3, IRF7, IFNB, RSAD2, and ISG56 (Figure 2C). Similarly, human TNBC MDA-MB-468 cells were highly responsive to immRNA-loaded RBCEVs, as shown by substantial upregulation of DDX58, MDA5, IRF7, IFNB, RSAD2, and ISG56 (Figure 12A). However, immRNA-loaded RBCEVs did not induce RIG-I activation in human TNBC MDA-MB-231 cells (Figure 12B).

[0120] We also tested the delivery of im mRNA to lung cancer cells. Human lung cancer NCI- H358 (H358) cells showed significant upregulation of DDX58, MDA5, MAVS, IRF7, IFNB, RSAD2, and ISG56 after incubation with im mRNA-loaded RBCEVs (Figure 2D). RSAD2, a known IFN-stimulated gene, was increased 305-, 518-, and 143-fold in all three cell lines, respectively (Figures 2B-D and 12A).

[0121] To evaluate the effect of immRNA on normal cells, non-transformed mammary epithelial MCF10A cells were incubated with immRNA-loaded RBCEVs, which resulted in a slight increase in intracellular DDX58 and RSAD2 levels, but not other RIG-I-related genes (Figure 12C). Mouse lung epithelial (mLE) cells were also generated using a previously established protocol (Kasinski & Slack, 2013). The cell surface marker EpCAM was measured by flow cytometry (Figure 11C). Uptake of immRNA-loaded RBCEVs by lung epithelial cells resulted in significant upregulation of Ddx58, Mda5, Irf7, Rsad2, and Isg56. Ifnb was not detected in the cells, whereas Ifna4, Ifna11, and Ifna12 were significantly upregulated (Figure 12D).

[0122] The type I IFN production activity induced by immRNA was evaluated. immRNA-loaded RBCEVs were cultured in human lung epithelial carcinoma A549-Dual™ and A549-Dual™ RIG-I - / - These two cell lines secrete Lucia luciferase via the interferon regulatory factor (IRF) pathway under the control of the ISG54 minimal promoter coupled with five IFN-stimulated response elements. As a result, wild-type (WT) A549-Dual™ cells incubated with immRNA-RBCEVs were responsive to this RIG-I activator, as shown by increased luciferase activity after 24 and 48 hours (Figure 2E). Removal of the RIG-I gene from the cells resulted in a complete loss of IFN signaling in response to immRNA (Figure 2E), confirming that RIG-I is the main immRNA sensor in cells. To evaluate the effect of immRNA on the inflammatory response, the concentrations of key inflammatory mediators were measured in 4T1 conditioned medium. 4T1 cells treated with immRNA-loaded RBCEVs secreted significantly higher levels of IFNα (approximately 55.95 pg / mL) and IFNβ (approximately 1844.33 pg / mL) (Figure 2F). Furthermore, immRNA-RBCEVs taken up by 4T1 cells also induced the secretion of pro-inflammatory cytokines TNFα (approximately 10.91 pg / mL) and IL-6 (approximately 793.22 pg / mL) (Figure 2F).

[0123] The induction of immunogenic cell death in cancer cells was then quantified using Annexin V (ANXV) and propidium iodide (PI) staining for each EV treatment. After 72 h of incubation, immRNA-loaded RBCEVs induced apoptosis in 4T1 cells (approximately 15.9%) (Figures 2G and 12E), CA1a cells (approximately 24.6%) (Figures 2H and 12E), H358 cells (approximately 17.6%) (Figures 2I and 12E), and MDA-MB-468 cells (approximately 16.0%) (Figures 12E-F), as indicated by the average percentage of the ANXV+PI+ population. immRNA-loaded RBCEVs failed to induce cell death in RIG-I-low MDA-MB-231 cells (Figures 12E-F), indicating that RIG-I is an essential receptor responding to immRNA. MCF10A and mLE cells did not undergo apoptosis after incubation with immRNA-loaded RBCEVs (Figure S12E-F). These data suggest that immRNA-loaded RBCEVs activate the cellular RIG-I pathway, induce type I IFN secretion, and induce cell death in cancer cells while sparing non-malignant cells at the same time.

[0124] RBCEVs deliver bifunctional ASOs to inhibit oncogenic miR-125b while simultaneously activating the RIG-I pathway leading to cancer cell death After confirming that 5' triphosphate immRNA could effectively trigger RIG-I activation, we tested a combination approach that simultaneously incorporated ASO-mediated oncogene silencing and RIG-I-mediated immune activation. We designed a miR-125b ASO with a triphosphate group attached to its 5' end (3p-125b-ASO) (Figure 3A). To improve the yield of 3p-125b-ASO, the first four nucleotides of the original 125b-ASO (full-length complementary to miR-125b) were removed and replaced with two guanine nucleotides. A pair of 3'-end truncations was also tested. The dsDNA template of the modified 125b-ASO sequence was inserted behind a T7 promoter for efficient in vitro transcription (IVT) following established protocols (Luo et al., 2012; Yong et al., 2019). This allowed the product RNA to rebind to T7 RNA polymerase, creating a self-priming (in cis) hairpin duplex, resulting in longer products (Figure S13A). The sequence of the 3p-125b-ASO produced by IVT was confirmed by sequencing.

[0125] To evaluate the functional impact of 3p-125b-ASO in cancer cells, we delivered 3p-125b-ASO using RBCEVs. After 24 h of incubation of ASO-loaded RBCEVs with cancer cells, we compared the miR-125b silencing activity of 3p-125b-ASO produced in IVT with that of unmodified miR-125 ASO (125b-ASO), which possesses a free 5'-OH group, using qPCR. As a result, unmodified 125b-ASO and modified 3p-125b-ASO inhibited endogenous miR-125b levels to a similar extent in 4T1, CA1a, and H358 cells (Figure 3B, D, F). Thus, silencing activity was not hindered by the presence of a 5'-terminal triphosphate group. Next, RIG-I signaling in cancer cells was evaluated in response to treatment with 3p-125b-ASO-loaded RBCEVs. Uptake of 3p-125b-ASO-loaded RBCEVs induced significant upregulation of Ddx58, Mda5, Irf7, Ifnb, Rsad2, and Isg56 in 4T1 cells (Figure 3C), DDX58, MDA5, MAVS, IRF3, IRF7, IFNB, RSAD2, and ISG56 in CA1a cells (Figure 3E), and DDX58, MDA5, MAVS, IRF7, IFNB, RSAD2, and ISG56 in H358 cells (Figure 3G). Silencing of miR-125b by unmodified 125b-ASO did not affect RIG-I activation in the three cell lines (Figure 3C,E,G). Similar to immRNA-loaded RBCEVs, 3p-125b-ASO-loaded RBCEVs showed no effect on MDA-MB-231 cells, a moderate effect on MCF10A cells, and upregulation of Ifna but not Ifnb via RIG-I activation in mLE cells (Figure 13B–D).

[0126] To verify the activity of 3p-125b-ASO-induced type I IFN secretion, A549-Dual™ and A549-Dual™ RIG-I - / - The reporter assay was repeated using A549-Dual™ RIG-I cells. RIG-I stimulation by 3p-125b-ASO-loaded RBCEVs suppressed RIG-I expression in A549-Dual™ wild-type (WT) cells. - / - Compared with 3p-125b-ASO-loaded RBCEVs, 4T1 cells significantly induced IFN secretion after 24 and 48 h (Figure 3H). RIG-I knockout confirmed the critical role of RIG-I in inducing IFN secretion. Similarly, cytokine immunoassay analysis revealed that 4T1 cells loaded with 3p-125b-ASO-loaded RBCEVs released significantly higher levels of IFNα (approximately 25.42 pg / mL), IFNβ (approximately 815.41 pg / mL), and TNFα (approximately 8.23 ​​pg / mL) (Figure 3I).

[0127] To examine the functional impact of 3p-125b-ASO on intrinsic apoptosis in cancer cells, cancer cells were incubated with 3p-125b-ASO-RBCEV for 72 h. Treatment with 3p-125b-ASO-RBCEV strongly induced apoptosis in 4T1 (approximately 15.5%) (Figures 3J and 13E), CA1a (approximately 21.7%) (Figures 3K and 13E), and H358 cells (approximately 14.6%) (Figures 3L and 13E), similar to the effect of the original 125b-ASO, as determined by ANXV and PI binding. 3p-125b-ASO-RBCEV also induced cell death in MDA-MB-231 cells and exerted moderate cytotoxicity against non-malignant cells, an effect attributed to the inhibition of miR-125b in cells (Figures 13E-F). Of note, combinatorial treatment of immRNA-RBCEV and 3p-125b-ASORBCEV in 4T1 and CA1a cells did not enhance the effect of each agent on RIG-I activation and apoptosis (Figure S13G-H). In contrast, the combination of immRNA-RBCEV and 3p-125b-ASO-RBCEV activated the RIG-I response pathway less and had less effect on apoptosis than single-agent treatment (Figure S13G-H). Together, these data suggest that the bifunctional 3p-125b-ASO contains two distinct and independent functions, which can simultaneously inhibit oncogenic miR-125b and activate RIG-I signaling, leading to immunogenic death of cancer cells.

[0128] Intratumoral delivery of immRNA suppresses breast cancer growth by inducing RIG-I-mediated immune responses To examine the antitumor activity of immRNA in vivo, 4T1 cells were implanted into the fourth mammary fat pad (MFP) of BALB / c mice (Figure 4A). Tumors were treated intratumorally with RBCEVs containing immRNA or NC RNA every 3 days, and tumor growth was monitored (Figure 4A). Treatment with immRNA-loaded RBCEVs was observed to significantly suppress 4T1 tumor growth compared to NC-RNA-loaded RBCEVs (Figure 4B). Mice were sacrificed on day 18, when untreated tumors reached approximately 15 mm in diameter, and tumors were excised. qPCR analysis of dissociated tumor cells revealed that delivery of immRNA-loaded RBCEVs led to the upregulation of Ddx58, Mda5, Mavs, Irf3, Irf7, Ifnb, Rsad2, and Isg56 in tumors (Figure 4C). To confirm the cytotoxic effect of immRNA-loaded RBCEVs on tumors, TUNEL staining was performed and tumor sections were co-stained with an antibody against α-smooth muscle actin (αSMA) to mark cancer-associated fibroblasts, the most abundant cell type in the tumor stroma. Increased apoptosis was observed in tumors treated with immRNA-loaded RBCEVs compared to treatment with NC-RNA-loaded RBCEVs (Figure 4G-H). Furthermore, TUNEL-positive cells were clustered in the tumor bulk rather than within the stromal compartment (Figure 4G). Interestingly, robust immune activation was observed in tumors treated with immRNA-RBCEVs compared to NC-RNA-loaded RBCEVs, as evidenced by increased infiltration of neutrophils (CD45+CD11b+Ly6G / C+), natural killer (NK) cells (CD45+CD11b+CD49b+), macrophages (CD45+CD11b+F4 / 80+), dendritic cells (DCs) (CD45+CD11c+), total T cells (CD45+CD3ε+), and CD8+ T cells (CD45+CD3ε+CD8+) (Figures 4D and 14A). Furthermore, serum IFN-β levels were measured at the end point. Intratumoral administration of immRNA-loaded RBCEVs slightly elevated serum IFN-β levels in mice (approximately 6.99 pg / mL) (Figure 4E). To further characterize the cytokine milieu within the tumor, qPCR of cytokine-encoding genes was performed in dissociated tumor cells.The results revealed that immRNA-loaded RBCEVs significantly increased the mRNA levels of several pro-inflammatory cytokine genes, including Ifna4, Ifna11, and Ifna12, as well as Tnf, Il6, Il12b, Il2, Ifng, Il1b, and Il18, while decreasing the levels of anti-inflammatory cytokine genes, including Il4, Il5, and Tgfb1, indicating a shift from Th2 to Th1 immune responses ( Figure 4F ).

[0129] 4T1 tumors were treated with 2.5 mg / kg immRNA-loaded RBCEVs (intratumoral (it)) and 2 mg / kg anti-PD-L1 monoclonal antibody (intraperitoneal (ip)) on different days. ImmRNA-loaded RBCEVs or anti-PD-L1 antibody monotherapy were used as controls. The dose and frequency of anti-PD-L1 antibody treatment were applied based on previous studies (Hong et al., 2019). 2 mg / kg anti-PD-L1 is 34-fold lower than the clinical equivalent dose (Herbst et al., 2014). Surprisingly, we observed that 2 out of 4 mice treated with anti-PD-L1 alone and 2 out of 4 mice treated with immRNA RBCEVs in combination with anti-PD-L1 died after 3 or 5 doses (Figure 14B). In contrast, all mice survived after 5 doses of immRNA-RBCEV monotherapy treatment (Figure 14B). Serum IFNβ levels in surviving mice at the end point indicated that anti-PD-L1 alone induced cytokine release syndrome, which appeared to be alleviated when anti-PD-L1 was combined with immRNA-RBCEVs, even though the mortality associated with the end point was the same as that caused by anti-PD-L1 monotherapy (Figure 14D). Combination treatment and anti-PD-L1 treatment alone did not have a further favorable effect on tumor suppression compared with immRNA-RBCEV monotherapy (Figure 14C). Therefore, these data suggest that immRNA-loaded RBCEVs are safe and well-effective as a single-agent immunotherapy for breast cancer treatment.

[0130] As a model of human breast cancer, CA1a cells were implanted into the fourth MFP of immunodeficient NSG-SGM3 mice (Figure 4I). Similarly, tumors were treated intratumorally with immRNA-loaded RBCEVs every 3 days, and tumor size was measured (Figure 4I). Treatment with immRNA-loaded RBCEVs significantly inhibited CA1a tumor growth (Figure 4J). qPCR results showed that the RIG-I cascade, including DDX58, MDA5, MAVS, IRF3, IRF7, IFNB, RSAD2, and ISG56, was upregulated in tumors by immRNA-RBCEV treatment (Figure 4K). Furthermore, treatment with immRNA-RBCEVs was associated with enhanced infiltration of neutrophils, macrophages, and dendritic cells in tumors (Figures 4L and 14E). The same series of experiments was performed in nude mice in the MDA-MB-468 tumor model, with four mice per group. ImmRNA-loaded RBCEV treatment led to rapid and extensive shrinkage of MDA-MB-468 tumors (Figure 14F), one of which disappeared. RIG-I activation was addressed in a limited number of tumor cells by qPCR. The results revealed that immRNA-loaded RBCEV treatment activated the RIG-I pathway by upregulating DDX58, MDA5, RSAD2, MAVS, IRF3, IRF7, IFNB, and ISG56 in tumor cells (Figure 14G). Taken together, these data suggest that intratumoral administration of immRNA-RBCEVs activates the RIG-I pathway and induces antitumor responses by recruiting immune cells for the suppression of orthotopic and xenograft breast cancer.

[0131] Intratumoral administration of a dual-function ASO suppresses breast cancer growth by inducing apoptosis and RIG-I-mediated immune responses Next, the antitumor efficacy of 3p-125b-ASO was tested in vivo in a 4T1 cancer model. 4T1 cells were implanted into the 4th MFP of BALB / c mice (Figure 5A). 4T1 tumors were treated with 5 mg / kg of 3p-125b-ASO-loaded RBCEVs every 3 days and monitored every 2 days (Figure 5A). Mice were sacrificed and tumors were collected when untreated tumors reached a diameter of approximately 15 mm. Similar to immRNA treatment, 3p-125b-ASO delivered using RBCEVs was observed to significantly suppress the growth of 4T1 tumors (Figure 5B). qPCR analysis of dissociated tumor cells revealed decreased levels of miR-125b and increased levels of Ddx58, Mda5, Mavs, Irf3, Irf7, Ifnb, Rsad2, and Isg56 in tumors treated with 3p-125b-ASO-loaded RBCEVs compared to NC-RNA-loaded RBCEVs (Figure 5C-D). Furthermore, FACS analysis revealed increased numbers of tumor-infiltrating neutrophils, NK cells, macrophages, dendritic cells, total T cells, and CD8+ T cells in tumors treated with 3p-125b-ASO-loaded RBCEVs compared to NC-RNA-loaded RBCEVs (Figures 5E and 15). Because low systemic IFNβ levels were detected by ELISA in tumor-bearing mice that received intratumoral administration of immRNA-RBCEVs, we performed cytokine immunoassays on tumor lysates to improve the sensitivity of detecting the lowest concentrations of cytokines. The data revealed that elevated levels of IFNα (approximately 25.62 pg / mL), IFNβ (approximately 38.71 pg / mL), TNFα (approximately 72.64 pg / mL), IL-6 (approximately 139.26 pg / mL), IL-12p40 (approximately 100.78 pg / mL), and IL-10 (approximately 23.06 pg / mL) were detected in tumors treated with 3p-125b-ASO-loaded RBCEVs compared to controls (Figure 5F). The IL-12 / IL-10 ratio was significantly increased by 3p-125b-ASO-loaded RBCEV treatment (Figure 5F), suggesting a proinflammatory tumor microenvironment.Furthermore, qPCR analysis of tumor cells showed a significant increase in Il2, Ifng, Il1b, and Il18, as well as a decrease in Il4, Il5, and Tgfb1, indicative of a Th1-dominated immune response (Figure 5G). Indeed, 3p-125b-ASO-loaded RBCEVs substantially induced apoptosis, with TUNEL-positive cells concentrating in the tumor mass rather than in the stroma, as determined by TUNEL staining of tumor sections, suggesting that miR-125b knockdown and RIG-I activation contributed to the tumoricidal activity mediated by 3p-125b-ASO (Figure 5H-I). Taken together, these data provide evidence that both miR-125b inhibition and RIG-I-mediated activation enhanced the antitumor efficacy of 3p-125b-ASO in an orthotopic breast cancer model.

[0132] Conjugation of RBCEVs with EGFR-binding nanobodies promotes specific delivery of iM mRNA to metastatic EGFR-positive breast cancer cells Specific targeting of cancer cells is a key feature of EV-based drug delivery, as it enhances therapeutic efficacy while protecting normal cells from toxicity. EGFR is a known proto-oncogene in many types of cancer and is considered a common cancer biomarker. Mouse breast cancer 4T1 cells were transduced with lentivirus carrying a human EGFR expression vector. A RBCEV surface functionalization method using OaAEP1 ligase was previously developed to ligate peptides with a ligase-binding motif (NGL) at the C-terminus to the RBCEV surface (Pham et al., 2021). In this study, the enzymatic ligation method was combined with the streptavidin-biotin conjugation method to conjugate RBCEVs with nanobodies via a linker peptide (Figure 6A). Specifically, RBCEVs were enzymatically ligated with a biotinylated linker peptide (biotin-TRNGL) and sequentially conjugated with tetrameric streptavidin and biotinylated anti-EGFR nanobody (α-EGFR-VHH) (Figure 6A). Flow cytometry analysis showed that α-EGFR-VHH specifically bound to human EGFR-expressing cells (CA1a, H358 and 4T1-hEGFR cells) but not to mouse EGFR-expressing 4T1 cells (Figure 11C). To evaluate the specificity of EGFR-targeted RBCEVs for EGFR-expressing cancer cells, EGFR-VHH-coated RBCEVs were labeled with CFSE and incubated with human EGFR-expressing 4T1 cells (4T1-hEGFR) and parental 4T1 cells at suboptimal doses for 2 h. Anti-mCherry nanobody was used as a negative control (Ctrl-VHH). The CFSE fluorescence intensity of 4T1-hEGFR cells treated with EGFR-targeted CFSE-labeled RBCEVs was approximately 28.6-fold higher than that in 4T1-hEGFR cells treated with non-targeted CFSE-labeled RBCEVs (Figure 6B). Non-targeted CFSE-labeled RBCEVs were taken up by 4T1-hEGFR cells at similar levels as non-targeted and EGFR-targeted CFSE-labeled RBCEVs by 4T1 cells (Figure 6B). Specific uptake of EGFR-targeted RBCEVs carrying fluorescent RNA was observed in 4T1-hEGFR cells.Consistently, FAM fluorescence intensity was significantly higher in hEGFR-positive 4T1 cells treated with EGFR-VHH-coated FAM-ASO-loaded RBCEVs compared to parental 4T1 cells treated with EGFR-VHH-coated FAM-ASO-loaded RBCEVs and 4T1-hEGFR cells treated with non-targeting FAM-ASO-loaded RBCEVs (Figure 6C). To verify whether EGFR-targeting RBCEVs specifically deliver functional RNA to 4T1-hEGFR cells, EGFR-VHH-coated RBCEVs were loaded with immRNA. EGFR-VHH-coated immRNA-loaded RBCEVs showed higher functional uptake by 4T1-hEGFR cells compared to 4T1 cells, as evidenced by substantial increases in cellular Ddx58, Rsad2, and Ifnb expression (Figures 6D and 16A).

[0133] To generate a lung metastatic breast cancer model, 4T1-hEGFR cells were injected intravenously into the tail vein of BALB / c mice (Figure 6E). Seven days later, non-targeted and EGFR-targeted CFSE-labeled RBCEVs were delivered via intrapulmonary administration to mice bearing metastatic breast cancer. CFSE-labeled RBCEVs were washed extensively using size-exclusion chromatography and centrifugation. The flow-through of the final wash was used as a negative control. 24 hours after intrapulmonary delivery, the biodistribution of non-targeted and EGFR-targeted CFSE-labeled RBCEVs in the lungs was analyzed using flow cytometry. Pulmonary delivery of EGFR-targeted and non-targeted RBCEVs was consistent, as evidenced by similar percentages of CFSE-positive cells (approximately 16%) in homogenized total lung cells (Figures 6F and 16B). Conjugation of RBCEVs with EGFR nanobodies increased the percentage of CFSE-positive tumor cells to approximately 68% (Figures 6F and 16B). No significant changes were observed in alveolar macrophages, which are located in the airway lumen and consistently take up EVs at a high rate (Figs. 6F and 16B). Furthermore, a significant decrease in the percentage of CFSE-positive neutrophils and a slight decrease in the percentage of CFSE-positive interstitial macrophages were observed in the lungs of mice treated with EGFR-targeting RBCEVs (Figs. 6F and 16B). This may be due to the more preferential uptake of hEGFR-specific RBCEVs into hEGFR-positive tumor cells than into hEGFR-negative cells in the lung parenchyma. These data suggest that conjugation of EGFR nanobodies can specifically target RBCEVs to EGFR-positive metastatic breast cancer cells in vivo.

[0134] Pulmonary delivery of EGFR-targeted immRNA using RBCEVs suppresses breast cancer metastasis in the lungs To treat metastatic breast cancer in the lungs, multiple doses of EGFR-targeting immRNA-loaded RBCEVs were delivered intrapulmonary every other day (Figure 7A). After five treatments, mice were sacrificed and lungs were excised for subsequent analysis. As expected, non-targeting and EGFR-targeting immRNA-loaded RBCEVs suppressed metastatic tumor cell engraftment, as indicated by a decrease in the percentage of hEGFR-positive cells in the lungs (Figures 7B, C and 17A). Notably, EGFR-targeting immRNA-loaded RBCEVs suppressed tumor growth in the lungs, an effect attributed to increased apoptosis in this treatment. H&E staining of lung sections showed that mice treated with EGFR-targeting immRNA-loaded RBCEVs had significantly lower metastatic areas in the lungs (Figures 7D-E). qPCR analysis of whole lung lysates revealed that immRNA-loaded RBCEVs induced activation of the RIG-I cascade by upregulating Ddx58, Mda5, Rsad2, Irf7, Ifnb, and Isg56. Although no substantial increase in RLR genes was observed, Ifnb and IFN-stimulated genes Isg56 and Rsad2 were significantly increased by EGFR-targeted RBCEV treatment compared to non-targeted RBCEV treatment (Figure 7F). By FACS analysis, increased infiltration of T cells, including neutrophils, NK cells, interstitial macrophages, cDCs, and CD8+ T cells but not CD4+ T cells, was observed in the lungs of mice treated with immRNA-loaded RBCEVs (Figures 7G and 17A). EGFR-targeted RBCEVs further increased the numbers of infiltrating neutrophils, interstitial macrophages, cDCs, T cells, and CD8+ T cells (Figures 7G and 17A). Alveolar macrophages were nearly absent in the airways of mice treated with EGFR-targeted RBCEVs containing immRNA and controls (Figures 7G and 17A). TUNEL staining of lung sections further confirmed that treatment with EGFR-targeted RBCEVs containing immRNA increased apoptosis in the lungs compared to treatment with non-targeted immRNA-loaded RBCEVs (Figures 7H-I). To examine whether the respiratory inflammatory state in mice affected metastasis, we detected cytokine concentrations in lung homogenates.All cytokines tested were significantly elevated in the lungs of mice treated with EGFR-targeted and non-targeted immRNA-loaded RBCEVs compared to control mice (Figure 7J). Mice receiving EGFR-targeted immRNA-loaded RBCEVs showed higher IFNα (approximately 17.97 pg / mL), IFNβ (approximately 22.47 pg / mL), TNFα (approximately 312.86 pg / mL), IL-12p40 (approximately 72.73 pg / mL), IL-10 (approximately 16. 91 pg / mL), and IL-12 / IL-10 ratios compared to mice receiving non-targeted immRNA-loaded RBCEVs, but there was no significant difference in IL-6 levels (Figure 7J). qPCR analysis of lung lysates revealed that immRNA-loaded RBCEVs increased mRNA levels of Il2, Ifng, Il1b, and Il18, while decreasing Il4, Il5, and Tgfb1 (Figure 17B), suggesting Th1-dominated immunity in the lung. EGFR-targeted RBCEVs further enhanced Il2 and Ifng expression and decreased Tgfb1 expression compared to non-targeted RBCEVs (Figure 17B). Together, these data suggest that EGFR-targeted RBCEVs enhance the specificity of delivery to EGFR-positive cancer cells, leading to improved therapeutic efficacy of immRNA in vivo.

[0135] EGFR-targeted immRNA-loaded RBCEVs induce DC activation, promote M1 macrophage polarization, and enhance tumor-specific CD8+ T cell activity To evaluate the ability of EGFR-targeted immRNA-loaded RBCEVs to induce DC activation and modulate macrophage polarization in the lung, flow cytometry was used to determine the expression of DC activation markers and macrophage M1 / M2 markers. The data revealed that non-targeted and EGFR-targeted immRNA-loaded RBCEV treatment increased the percentage of MHCII+CD11c+ DCs among CD45+ cells (Figure 8A-B). EGFR-targeted immRNA-loaded RBCEV treatment further increased the percentage of MHCII+CD11c+ DCs compared to non-targeted immRNA-loaded RBCEV treatment (Figure 8A-B). Furthermore, immRNA-loaded RBCEV treatment resulted in a lower percentage of CD206+F4 / 80+ M2-like macrophages and a higher percentage of CD86+F4 / 80+ M1-like macrophages among CD11b+CD45+ cells (Figure 8A-B). The M1 / M2 ratio was significantly increased by RBCEV treatment loaded with EGFR-targeting immRNA compared to RBCEV treatment loaded with non-targeting immRNA (Figure 8B). To quantify the activity of CD8+ tumor-infiltrating lymphocytes (TILs) against tumor-associated antigens, CD8+ TILs were isolated from the lungs for cytotoxicity analysis. As a result, the levels of CD69 and granzyme B were significantly increased in CD8+ TILs by RBCEV treatment loaded with immRNA (Figure 8C-D). Considering that tumor-specific immune responses are an important factor in cancer immunotherapy, tumor-specific T cell responses were evaluated. CD8+ TILs were incubated with 4T1-hEGFR-luciferase cells at a ratio of 50:1 for 48 h. Restimulation of CD8+ TILs isolated from RBCEV-treated mice loaded with immRNA with tumor antigen markedly enhanced IFNγ production and was associated with a significant increase in cytotoxic T lymphocyte (CTL) activity (Figure 8E-F). RBCEVs loaded with immRNA targeting EGFR further enhanced the cytotoxicity of CD8+ TILs, as indicated by increased cellular expression of CD69 and granzyme B, secretion of IFNγ, and CTL activity (Figure 8C-F).Collectively, these data correlated with increased expression of Th1 cytokines and decreased expression of Th2 cytokines in the lungs of EGFR-targeted immRNA-RBCEV-treated mice ( Figures 7 J and 17B ), suggesting a Th1-associated and CTL-mediated antitumor immune response.

[0136] Specific knockdown of mutant KRAS and EGFR in various cancers using RBCEV-delivered ASOs Cancer cells induce many genetic alterations to support tumor progression, not only by favoring survival and proliferation but also by forming a less immunogenic tumor microenvironment (TME). The most common oncogenic driver in cancer is KRAS activating mutation, which is found in approximately 30% of patients with lung and colorectal cancer, and nearly 100% of patients with pancreatic cancer. Most KRAS mutant cancer cells harbor mismatch mutations at the glycine 12 (G12) position. The small molecule drug ARS1620 was recently invented as an effective inhibitor of KRAS-G12C, but it is not applicable to other KRAS mutants, nor to other mismatches at G12. Herein, we used RBCEVs to deliver ASO drugs targeting oncogenic drivers, including KRAS G12D, KRAS G12V, EGFR L858R, and EGFR T790M, and achieved excellent anticancer activity by combining oncogenic gene-targeting ASO with immRNA.

[0137] ASOs were designed using the Gapmer method. ASOs were chemically modified with locked nucleic acid (LNA) at the two terminal positions and at the nucleotides matching the missense mutation to enhance specificity. In addition, phosphorothioate (PS) modifications were included at all nucleotides, and 2'-O-methoxyethyl (2'MOE) modifications were added to five nucleotides in each wing to enhance the stability of the ASO (Figures 9A and 18A). Specific knockdown by the corresponding ASO-loaded RBCEVs was examined using cancer cell lines A427 (KRAS G12D), AsPC-1 (KRAS G12D), and H441 (KRAS G12V), carrying KRAS mutations. H1975 and DU145 cells carrying the WT KRAS gene were used as control cell lines. Western blot and qPCR analyses demonstrated that KRAS G12D ASO 2 and KRAS G12V ASO 2 exhibited significant knockdown potency of the corresponding mutant genes in A427, AsPC-1, and H441 cells, respectively, resulting in a significant reduction in the viability of mutant-bearing cells as determined by CCK-8 assay, while showing minimal cross-reactivity in WT gene-bearing H1975 and DU145 cells (Figures 9B-I).

[0138] Similarly, we investigated the specific knockdown of EGFR L858R and T790M mutations by ASO delivered by RBCEV in H1975 cells carrying these two mutations. A549 cells carrying WT EGFR were used as a control cell line (Figure 18A). Figure 18B-G revealed that EGFR L858R ASO4 and T790M ASO4 specifically inhibited EGFR protein expression in target cells, while showing limited effect on EGFR WT in A549 control cells. Based on these data, KRAS G12D ASO 2, KRAS G12V ASO 2, EGFR L858R ASO 4, and EGFR T790M ASO 4 were selected for further testing.

[0139] Combination of ASO targeting oncogene addiction and immRNA synergistically enhances RIG-I pathway activation in cancer cells Using the RBCEV and LNP platforms, we investigated the anticancer efficacy of the combined therapy of oncogene addiction inhibition and RIG-I activation by co-delivering ASO and immRNA to various mutant-bearing cancer cell lines, including A427 (KRAS G12D) lung cancer cells, CT26 (Kras G12D) colon cancer cells, AsPC-1 (KRAS G12D) pancreatic cancer cells, H441 (KRAS G12V) lung cancer cells, and H1975 (EGFR L858R / T790M) lung cancer cells. First, we quantified the expression levels of DDX58 and IFN-β in treated cells using qPCR. Notably, cells treated with either KRAS ASO or EGFR ASO in combination with immRNA showed the highest expression of both DDX58 and IFN-β than cells that received single treatment (Figure 19A-D). We next measured the viability of cancer cells after various treatments with RBCEV formulations by CCK-8 assay (Figure S19E-H), and found that the combination exerted significantly enhanced cytotoxicity against KRAS G12D-expressing pancreatic AsPC-1 cells (Figure S19G). Similar observations were observed when AsPC-1 cells were treated with ASO and immRNA delivered by LNPs (Figure S19I-J).

[0140] We next examined the activation of THP-1 monocytic cells in coculture with pretreated AsPC-1 cells (Figure S1K).Flow cytometry analysis revealed that AsPC-1 cells pretreated with KRAS ASO and im mRNA expressed increased levels of CD86, an activation marker for monocytic THP-1 cells (Figure S1L).

[0141] RBCEVs delivered with EGFR ASO effectively inhibit lung tumor growth To investigate the therapeutic potential of mutation-specific ASOs in vivo, H1975 tumor cell xenograft mice were generated by intravenous injection. EGFR L858R ASO was loaded into EVs and delivered intratracheally every 3 days to tumor-bearing mice (Figure 20A). Notably, EGFR L858R ASO-loaded EVs exhibited a remarkable tumor-inhibitory effect, indicated by a significantly lower luciferase signal than that of mice administered NC ASO-loaded EVs (Figure 20B-C). By immunohistochemistry (IHC) staining, a significant decrease in EGFR levels was observed in the L858R ASO-EV-treated group compared with that of mice treated with NC ASO-loaded EVs (EV-NC ASO) (Figure 20D).

[0142] Combination of RBCEV-mediated immRNA and oncogene-targeting ASO shows potent anticancer effects in vivo A syngeneic CT26 colon cancer mouse model harboring the KRAS G12D mutation was generated by subcutaneous (sc) inoculation of CT26 colon cancer cells into BALB / c mice. RBCEVs loaded with KRAS ASO, immRNA, or RBCEVs loaded with a combination of KRAS ASO and immRNA were injected into the tumors every 2 days. It was observed that the combination of KRAS G12D ASO and immRNA exhibited superior tumor inhibition effects, as shown by the reduction in tumor volume (Figure 21A) and tumor weight (Figure 21B). In addition, it was also observed that the formation of necrotic areas and reduced tumor cell density were observed in tumors harvested from mice that received the combination treatment, as shown by H&E staining (Figure 21C) and TUNEL assay (Figure 21D). These results suggest the potential of the combination therapy of KRAS-inhibitory ASO and RIG-I activation using the RBCEV platform against KRAS-driven cancers.

[0143] In summary, a robust platform for local and targeted delivery of immunomodulatory RIG-I agonists using RBCEVs was demonstrated. Treatment of orthotopic and metastatic breast cancer with RBCEVs loaded with both immRNA and 3p-125b-ASO enabled activation of RIG-I signaling, resulting in a three-pronged attack: (i) potent tumor growth inhibition; (ii) increased levels of type I IFN and proinflammatory cytokines; and (iii) IFN-mediated recruitment of innate immune cells (macrophages, neutrophils, and NK cells), activation of dendritic cells and polarization of M1 macrophages, and cross-priming of adaptive immune effectors (CD8+ T cells) through antigen-presenting cells (dendritic cells and macrophages) (Figure 10), indicating that the immunosuppressive tumor microenvironment was reversed. Polarization of Th1 / Th2 immune factors promotes antitumor immunity. Th1 cytokines, such as IL-1β, IL-2, IL-12, IL-18, TNFα, and IFNγ, are associated with pro-inflammatory, whereas Th2 cytokines, such as IL-4, IL-5, and TGFβ, play suppressive roles in the tumor immune microenvironment. Data demonstrated that RBCEVs loaded with immRNA and 3p-125b-ASO could shift the Th1 / Th2 balance in the tumor microenvironment in favor of Th1, contributing to tumor suppression. Effective cancer immunotherapy requires T cell priming, activation, and tumor infiltration. Thus, intratumoral delivery of RBCEVs loaded with immRNA and 3p-125b-ASO induced massive cell apoptosis in the tumor bulk, but no signs of apoptosis were observed in the tumor stroma.

[0144] RBCEVs loaded with immRNA and anti-PD-L1 were combined for enhanced anti-tumor activity. However, approximately 50% mortality was observed in tumor-bearing mice receiving repeated doses of anti-PD-L1 systemic administration. The lethality of the mice may be due to a hypersensitivity reaction caused by anti-PD-L1. These findings highlight a previously uncharacterized adverse toxicity of anti-PD-L1 that has not been reported in previous studies of RIG-I agonist and anti-PD-L1 combinations. Indeed, various clinical studies have demonstrated that anti-PDL1 as a sole treatment greatly induces systemic IFNβ release and cytokine stormi. Although systemic administration of RIG-I agonists is likely to induce cytokine release syndrome as well, local administration of immRNA is safe and fully effective according to the data disclosed herein.

[0145] A combinatorial approach to simultaneously inhibit oncogenic miR-125b and activate the RIG-I pathway was demonstrated. Tumor cells tend to evade single-targeted therapies, such as specific kinase / oncogene inhibitors or immunotherapy, due to their genetic and epigenetic plasticity. Therefore, multitargeted therapies are needed. 3p-125b-ASO offers advantages over the combination of multiple single-targeted therapies. 3p-125b-ASO is small and easy to synthesize. 3p-125b-ASO contains two distinct and independent properties: RIG-I activator and oncogene suppressor. Furthermore, RIG-I-induced apoptosis of cancer cells was synergistic with apoptosis induced by ASO-mediated miR-125b inhibition. Such bifunctional ASOs can be adapted to various tumor entities by targeting key oncogenes that drive tumorigenicity. Furthermore, in pancreatic cancer, at least in an in vitro setting, synergistic anticancer effects were observed when targeting both the KRAS-G12D mutant and the RIG-I pathway. These results suggest the potential of immRNA to ignite the "chilled" state of poor immunogenicity in pancreatic tumors, to which KRAS mutations contribute greatly. Taken together, these data suggest the potential clinical applications of immRNA, 3p-125b-ASO, and KRAS-G12D ASO.

[0146] The biological safety and reproducibility of drug delivery systems are crucial for their clinical translation. In vivo-jetPEI is a widely used delivery system for RIG-I agonists and other types of RNA therapeutics in preclinical and clinical studies. However, it has been reported to be associated with fatal hepatotoxicity in mice. Due to their biocompatibility, stability, and limited immunogenicity, EVs offer multiple advantages over traditional synthetic delivery vehicles as a delivery system. RBCEVs, which have high availability and scalability and no risk of horizontal gene transfer, have been demonstrated to be biologically safe, efficient, and amendable for therapeutic delivery in cancer treatment. Effective immunomodulation of tumor microenvironment and potent antitumor activity using RBCEVs as a delivery vehicle for RIG-I agonists were achieved without observable side effects, suggesting the advantages of RBCEVs for clinical applications.

[0147] A method for RBCEV surface functionalization with EGFR-targeted nanobodies has also been described, which can enhance the delivery of RIG-I agonists to EGFR-positive cancer cells, thereby improving therapeutic efficacy while reducing side effects. In experiments, depletion of alveolar macrophages in the airways was observed following intrapulmonary administration of non-targeted and EGFR-targeted RBCEVs containing immRNA. Data on the biodistribution of intrapulmonary-delivered RBCEVs showed that alveolar macrophages were the primary recipients of both non-targeted and EGFR-targeted RBCEVs. It is clear that RBCEVs themselves have no effect on alveolar macrophages, as CFSE-labeled or NC-RNA-loaded RBCEVs did not cause rapid death of alveolar macrophages. Alveolar macrophages play an important role in removing antigens in the lungs by nonspecific phagocytosis. The depletion of alveolar macrophages could be an effect resulting from RIG-I-mediated immune responses from phagocytosed RBCEVs containing immRNA. Thus, described herein is the delivery of therapeutic RNA via EVs via pulmonary administration. Data show that RNA-loaded RBCEVs are stable and can penetrate deep into the lung parenchyma for delivery of immRNA. The compositions described herein can be used for therapeutic RNA delivery, which can be applied not only to cancer metastasis but also to other lung diseases such as influenza and asthma.

[0148] Table 1: Primer list TIFF2025510600000029.tif244141TIFF2025510600000030.tif244144TIFF2025510600000031.tif24495

[0149] Table 2: List of sequences and corresponding SEQ ID NOs. TIFF2025510600000032.tif249129TIFF2025510600000033.tif249143TIFF2025510600000034.tif249143TIFF2025510600000035.tif249129

[0150] References TIFF2025510600000036.tif143166 < / scid>

Claims

1. A composition comprising a vesicle for delivery of a retinoic acid-inducible gene I receptor (RIG-I) agonist to a cell.

2. The composition of claim 1, wherein the RIG-I agonist is selected from the group consisting of an immunomodulatory RNA (immRNA), an antisense oligonucleotide (ASO), a small interfering RNA (siRNA), and combinations thereof.

3. The composition of claim 2, wherein the RIG-I agonist comprises an immune modulatory RNA (immRNA) and / or an antisense oligonucleotide (ASO).

4. Immunomodulatory RNA (immRNA) sequences 4. The composition of claim 2 or 3, comprising:

5. The composition of any one of claims 2 to 4, wherein the immune modulatory RNA (immRNA) further comprises a modification at the 5' and / or 3' end, optionally wherein the modification is selected from the group consisting of a 5' triphosphate cap, phosphorylation, methylation, adenylation, and combinations thereof.

6. Antisense oligonucleotides (ASOs) are miRNA-125b-ASO, KRAS-ASO, or The composition of claim 2 or 3, comprising an ASO that targets EGFR.

7. miRNA-125b-ASO was synthesized using the sequence 7. The composition of claim 6, comprising:

8. The KRAS-ASO is selected from the group consisting of KRAS-G12D ASO, KRAS-G12V ASO, KRAS-G12C ASO, and KRAS-G12S ASO, and optionally the KRAS-G12D ASO has the sequence or the KRAS-G12V ASO comprises any one of the following sequences:

7. The composition of claim 6, comprising any one of:

9. The EGFR-targeting ASO is selected from the group consisting of an EGFR L858R-targeting ASO or an EGFR T790M-targeting ASO, and optionally the EGFR L858R-targeting ASO has the following sequence: or an ASO targeting EGFR T790M comprising the following sequence:

7. The composition of claim 6, comprising any one of:

10. The composition of any one of claims 6 to 9, wherein the antisense oligonucleotide (ASO) further comprises a modification at the 5' and / or 3' end.

11. 11. The composition of claim 10, wherein the modification is selected from the group consisting of 5' triphosphate cap, phosphorylation, methylation, adenylation, locked nucleic acid, phosphorothioate, 2-methoxyethyl, and combinations thereof.

12. The composition of any one of claims 1 to 11, wherein the vesicle for delivery of a retinoic acid-inducible gene I receptor (RIG-I) agonist to a cell comprises an extracellular vesicle or a lipid nanoparticle.

13. The composition of claim 12, wherein the extracellular vesicles are selected from the group consisting of red blood cell-derived extracellular vesicles (RBCEVs), milk-derived EVs, plasmid-derived EVs, and cancer cell-derived EVs.

14. The composition of claim 13, wherein the extracellular vesicle is further conjugated to an antibody, optionally wherein the antibody is selected from the group consisting of a nanobody, a monoclonal antibody, and a single chain antibody.

15. 15. The composition of claim 14, wherein the antibody is a nanobody that binds to a cancer marker, and optionally the cancer marker is selected from the group consisting of epidermal growth factor receptor (EGFR), HER1, HER2, HER3, human growth factor (HGF), CXCR4, PD-L1, CD49a, CD49b, CD49c, CD49d, CD49e, CD49f, ITGA7, ITGA8, ITGA9, ITGA10, ITGA11, CD11D, CD103, CD11a, CD11b, CD51, CD41, CD11c, CD29, CD18, CD61, CD104, ITGB5, ITGB6, ITGB, ITGB8, FOLR1, FOLR2, or FOLR3.

16. Red blood cell-derived extracellular vesicles (RBCEVs); and / or an immunomodulatory RNA (immRNA) comprising or consisting of the sequence: and further comprising a 5' triphosphate cap. A composition comprising:

17. Lipid nanoparticles; and / or an immunomodulatory RNA (immRNA) comprising: KRAS-ASO A composition comprising:

18. 20. A method for delivering a retinoic acid-inducible gene I receptor (RIG-I) agonist to a cell, comprising administering to the cell a pharma- ceutically effective amount of the composition of any one of claims 1 to 17.

19. 20. A method of treating a disease comprising administering to a subject in need thereof a pharma- ceutically effective amount of the composition of any one of claims 1 to 17.

20. 20. The method of claim 19, wherein the disease is cancer or a lung disease or disorder.

21. 21. The method of claim 20, wherein the cancer is selected from the group consisting of breast cancer, lung cancer, pancreatic cancer, brain cancer, head and neck cancer, liver cancer, stomach cancer, colon cancer, prostate cancer, cervical cancer, and bone cancer.

22. 22. The method of claim 21, further comprising administering a pharma- ceutically effective amount of an anti-cancer drug.

23. 23. The method of claim 22, wherein the anti-cancer drug is selected from the group consisting of an anti-PD1 antibody, one or more anti-cancer gene ASOs, and one or more anti-cancer cytokines.

24. 21. The method of claim 20, wherein the pulmonary disease or disorder is asthma or pneumonia.

Images