Responsive Multivesicular Polymeric Nanovaccines that Codeliver STING Agonists and Neoantigens for Combination Tumor Immunotherapy

Abstract

Immune checkpoint blockade (ICB) has significantly advanced cancer immunotherapy, yet its patient response rates are generally low. Vaccines, including immunostimulant-adjuvanted peptide antigens, can improve ICB. The emerging neoantigens generated by cancer somatic mutations elicit cancer-specific immunity for personalized immunotherapy; the novel cyclic dinucleotide (CDN) adjuvants activate stimulator of interferon genes (STING) for antitumor type I interferon (IFN-I) responses. However, CDN/neoantigen vaccine development has been limited by the poor antigen/adjuvant codelivery. Here, pH-responsive CDN/neoantigen codelivering nanovaccines (NVs) for ICB combination tumor immunotherapy are reported. pH-responsive polymers are synthesized to be self-assembled into multivesicular nanoparticles (NPs) at physiological pH and disassembled at acidic conditions. NPs with high CDN/antigen coloading are selected as NVs for CDN/antigen codelivery to antigen presenting cells (APCs) in immunomodulatory lymph nodes (LNs). In the acidic endosome of APCs, pH-responsive NVs facilitate the vaccine release and escape into cytosol, where CDNs activate STING for IFN-I responses and antigens are presented by major histocompatibility complex (MHC) for T-cell priming. In mice, NVs elicit potent antigen-specific CD8⁺ T-cell responses with immune memory, and reduce multifaceted tumor immunosuppression. In syngeneic murine tumors, NVs show robust ICB combination therapeutic efficacy. Overall, these CDN/neoantigen-codelivering NVs hold the potential for ICB combination tumor immunotherapy.

Keywords: cGAS-STING; cancer immunotherapy; nanovaccine; neoantigen; pH responsiveness; polymeric nanocarrier; vaccine codelivery.

Scheme 1

pH‐responsive multivesicular polymeric nanovaccines (NVs) for the codelivery of STING agonists and neoantigens in combination tumor immunotherapy. The star‐shaped polymers were self‐assembled into pH‐responsive nanoparticles (NPs) that coloaded cGAMP and neoantigen peptides through electrostatic and hydrophobic interactions, respectively. The NVs efficiently codelivered cGAMP and neoantigens to the draining lymph nodes (LNs) and intranodal antigen presenting cells (APCs). Upon cell internalization by endocytosis, NVs were disassembled in response to the acidity in the endosome and escaped to the cytosol for STING activation by cGAMP and neoantigen presentation by major histocompatibility complex (MHC). NVs elicited sustained antigen presentation, potentiated and prolonged the neoantigen‐specific T‐cell responses, and ameliorated the tumor immunosuppression in the tumor microenvironment (TME). As a result, the combination of these NVs with immune checkpoint blockade (ICB) showed robust tumor therapeutic efficacy in multiple tumor models.

Figure 1

Characterization and screening of pH‐responsive star‐shaped polymer nanoparticles (NPs). A) Schematic illustration of the self‐assembly of star‐shaped multifunctional polymers into pH‐responsive NPs. B) Titration plots for the critical micelle concentration (CMC) measurement for S20, S40, and S60 NPs via the ratios of pyrene fluorescence intensities at 334.2 nm over 331.5 nm (I334.2/I331.5). C) pH titration of NPs (1 mg mL⁻¹) using NaOH (0.01 m). D) DLS results showing the hydrodynamic sizes of S40 NPs (1 mg mL⁻¹) at different pH conditions. E,F) The hydrodynamic sizes (E) and zeta potential (F) of NPs at the range of pH 5.5 – pH 7.4 indicate the NPs disassembly and the increased electrostatic charge on NPs in response to acidity. G) TEM images of S40 NPs at pH 7.0 and pH 5. The large NP sizes at pH 7.4 and the smaller NP sizes at pH 5 showed NP disassembly at pH5. H) pH‐dependent erythrocyte membrane destabilization by NPs, as measured using an erythrocyte hemolysis assay. Percent hemolysis was calculated relative to H₂O. I) MTT assay results showed the cell viability of DC2.4 cells treated with different NPs for 24 h. Polyethylenimine (PEI) was used as a positive control. In H and I, data represent mean ± SD (n = 3); ns: nonsignificant, *p < 0.05, and ****p < 0.0001 (Student's t‐test).

Figure 2

Characterization of nanoparticle (NP)/cGAMP. A) pH‐responsive cGAMP release kinetics from S40 polymeric NPs. B) Confocal microscopy images showed the uptake of free Fluo‐CDG or NP/Fluo‐CDG (1 µg mL⁻¹) into DC2.4 cells after a 5‐h incubation. DIC: differential interference contrast. C) Colocalization ratios of Fluo‐CDG with endolysosome as quantified from 20 random cells in confocal microscopy results. D) Flow cytometry results showing the time‐dependent intracellular delivery of free Fluo‐CDG or/Fluo‐CDG in DC2.4 cells (Fluo‐CDG: 1 µg mL⁻¹). E) Flow cytometry results showing that, relative to controls, NP/cGAMP enhanced the upregulation of MHC‐II and costimulatory factors CD40, CD80, and CD86 in DC2.4 cells. F) ELISA results showing that NPs significantly promoted the ability of cGAMP to induce murine IFN‐β (mIFN‐β) in DC2.4 cells after a 24‐h treatment in a dose‐dependent manner. G) NP/cGAMP outperformed controls to induce mIFN‐β) in DC2.4 cells (cGAMP: 1 µg mL⁻¹, 24‐h treatment). H,I) Fold changes of interferon (IFN) dependent reporter signals in RAW‐ISG cells treated with cGAMP formulations demonstrated the dose‐dependent cGAMP‐selective IFN response (H) (24‐h treatment) and the superior INF induction ability of NP/cGAMP than controls (I) (cGAMP: 1 µg/mL, 24‐h treatment). Lipo/cGAMP: Lipofectamine2000‐transfected cGAMP. PEI/cGAMP: PEI‐transfected cGAMP. NP/cGAMP: cGAMP‐loaded S40 NPs. Data: mean ± SD (n = 3); ns: nonsignificant, *p < 0.05, **p < 0.01, and ****p < 0.0001 (Student's t‐test).

Figure 3

cGAMP/antigen‐codelivering nanovaccines (NVs) sustained antigen presentation in dendritic cells (DCs). A) Study scheme of the antigen presentation sustainability in DC2.4 cells. B) Flow cytometry results showing the levels of SIINFEKL presented on DCs after incubation with NVs [(NP/(cGAMP/SIINFEKL)] and controls for 5 h, washing off extracellular vaccines, and further incubation for a series of durations, prior to antibody staining of cell surface H‐2K^b‐SIINFEKL complexes. C) Confocal microscopy images showing the uptake and presentation of SIINFEK_(FITC)L on DCs treated with free vaccines or NVs for 5 h, followed by washing off extracellular vaccines and 24‐h incubation. D) The assay absorption at 570 nm (A₅₇₀) indicated the activity of SIINFEKL‐specific B3Z CD8⁺ T cells after coculture with vaccine‐treated DCs. Relative to controls, NVs enabled DCs to promote antigen‐specific T‐cell activation. Data: mean ± SD (n = 3; **p < 0.01, and ****p < 0.0001 (Student's t‐test).

Figure 4

Nanovaccines (NVs) improved the codelivery of DY547‐CDG and SIINFEK_(FITC)L to draining lymph nodes (LNs) and intranodal antigen presenting cells (APCs). A) Signal quantification (left) and representative photos (right) of draining inguinal LNs 18 h after s.c. administration of NVs or a soluble mixture of SIINFEKL and CDG at tail base in C57BL/6 mice (n = 3). B,C) Flow cytometry data quantification showing the NV codelivery of CDG and SIINFEKL into LN‐residing dendritic cells (DCs) (B) and macrophages (C), two primary intranodal APC subsets. Data: mean ± SEM (n = 3); *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 (Student's t‐test).

Figure 5

cGAMP/SIINFEKL‐codelivering nanovaccines (NVs) elicited potent and durable T‐cell responses in mice. A) Study design for T‐cell responses in C57Bl/6 mice (n = 4). B,C) Representative flow cytometry plots (B) and quantification (C) of a H‐2K^b‐SIINFEKL tetramer staining assay showing that NVs (s.c. injected at tail base; days 0, 14) augmented the peripheral SIINFEKL‐specific CD8⁺ T cells (day 21) in mice. D) Quantification of CD8⁺ memory T cells in the above immunized mice (day 21). E) Representative flow cytometry plots of memory T cells in total CD8⁺ and SIINFEKL⁺ CD8⁺ T cells in NV‐treated mice (day 21). F) PD‐1 median fluorescence intensity (MFI) on total live PBMC CD8⁺ T cells. G) PD‐1 MFI on total live PBMC CD8⁺ T cells and SIINFEKL⁺ CD8⁺ T cells from NV‐immunized mice, indicating elevated PD‐1 levels specifically on SIINFEKL⁺ CD8⁺ T cells relative to total CD8⁺ T cells. H,I) EG7.OVA (H) and EL4 (I) tumor growth curves and tumor weights at day 20 post tumor challenge in immunized mice challenged with EG7.OVA (right flank) and EL4 (left flank) tumor cells on day 34 post priming vaccination. ND: nondetectable. Statistics are indicated in comparison with NVs. Data represent mean ± SEM; ns: not significant, *p < 0.05, **p < 0.01; ***p < 0.001, and ****p < 0.0001 (one‐way ANOVA with Dunnett test).

Figure 6

cGAMP/E7‐loaded nanovaccines (NVs) mediated robust immunotherapy in combination with αPD‐1 in TC‐1 tumor. A) Study design of TC‐1 combination immunotherapy in syngeneic C57Bl/6 mice. Vaccine: 10 nmole cGAMP and 20 µg antigen, s.c. administration at mouse tail base on days 7, 13, 19; αPD‐1: i.p. administration, 200 µg, on days 7, 10, 13, 16, 19. B,C) TC‐1 tumor growth curves (B) and the tumor weights at the end of study on day 28 (C). D) Kaplan–Meier mouse survival curves. E) Spleen/body weight ratios of mice at day 28 after treatment. F) Mouse body weights during the course of treatment. Data represent mean ± SEM; ns: not significant, *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 (one‐way ANOVA with Dunnett test).

Figure 7

Immune milium analysis in TC‐1 tumor microenvironment after nanovaccines (NVs) and immune checkpoint blockade (ICB) combination immunotherapy. A,B) Photographs (A) and quantification (B) of ELISPOT results showing the INF‐γ spots in the as‐treated tumors. C) RT‐PCR results of the mRNA levels of immunostimulatory cytokines and chemokines in as‐treated tumors. D) RT‐PCR results of the mRNA levels of M1‐like macrophage marker Nos2 and M2‐like macrophage markers Mrc1, Ym1, and Arg1 in as‐treated tumors. Data were relative to the expression of house‐keeping gene Gapdh. E) CD8⁺/CD4⁺ T‐cell ratio in TC‐1 tumor after combination immunotherapy. F) Tumor‐infiltrating dendritic cells (DCs) in TC‐1 tumor after combination immunotherapy. G) DC levels in the nontumor draining inguinal lymph nodes (LNs) and spleens of TC‐1‐tumor‐bearing mice after combination immunotherapy. Data represent mean ± SEM; ns: not significant, *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 (one‐way ANOVA with Dunnett test).