A Nanovaccine for Enhancing Cellular Immunity via Cytosolic Co-Delivery of Antigen and PolyIC RNA

Abstract

Traditional approaches to cancer vaccines elicit weak CD8⁺ T cell responses and have largely failed to meet clinical expectations. This is in part due to inefficient antigen cross-presentation, inappropriate selection of adjuvant and its formulation, poor vaccine pharmacokinetics, and/or suboptimal coordination of antigen and adjuvant delivery. Here, we describe a nanoparticle vaccine platform for facile co-loading and dual-delivery of antigens and nucleic acid adjuvants that elicits robust antigen-specific cellular immune responses. The nanovaccine design is based on diblock copolymers comprising a poly(ethylene glycol)-rich first block that is functionalized with reactive moieties for covalent conjugation of antigen via disulfide linkages, and a pH-responsive second block for electrostatic packaging of nucleic acids that also facilitates endosomal escape of associated vaccine cargo to the cytosol. Using polyIC, a clinically-advanced nucleic acid adjuvant, we demonstrated that endosomolytic nanoparticles promoted the cytosolic co-delivery of polyIC and protein antigen, which acted synergistically to enhance antigen cross-presentation, co-stimulatory molecule expression, and cytokine production by dendritic cells. We also found that the vaccine platform increased the accumulation of antigen and polyIC in the local draining lymph nodes. Consequently, dual-delivery of antigen and polyIC with endsomolytic nanoparticles significantly enhanced the magnitude and functionality of CD8⁺ T cell responses relative to a mixture of antigen and polyIC, resulting in inhibition of tumor growth in a mouse tumor model. Collectively, this work provides a proof-of-principle for a new cancer vaccine platform that strongly augments anti-tumor cellular immunity via cytosolic co-delivery of antigen and nucleic acid adjuvant.

Graphical Abstract

1. Introduction

Cancer vaccines have recently re-emerged as a potentially powerful component of an expanding immunotherapy arsenal. Cancer vaccines aim to amplify preexisting antitumor CD8⁺ T cells, promote the generation of new tumor antigen-specific T cells, and/or promote CD8⁺ T cell infiltration into solid tumors.^1–4 However, most cancer vaccines have demonstrated only a modest capacity to elicit a tumor-specific CD8⁺ T cell response in patients with limited therapeutic efficacy and disappointing clinical outcomes.^4,5 These poor responses can be attributed, at least in part, to the relatively low immunogenicity of peptide and protein antigens, which suffer from several intertwined pharmacological shortcomings, including rapid degradation and/or clearance, low accumulation in secondary lymphoid organs (e.g., lymph nodes), and inefficient delivery of antigen to the major histocompatibility complex (MHC) class I (MHC-I) antigen processing pathway, which is essential for generating a CD8⁺ T cell response.^6–10 To address these challenges, various approaches have been devised to increase the immunogenicity of cancer vaccine antigens, including the use of more potent adjuvants and/or particle-based delivery systems that promote uptake and cross-presentation of antigen on MHC-I by specific dendritic cell (DC) subsets.^11–15 However, inefficient antigen cross-presentation, improper choice or delivery of adjuvant, poor vaccine pharmacokinetics, and/or suboptimal coordination of antigen and adjuvant delivery continue to limit the efficacy of many promising strategies.

The cytosol is rich in targets for augmenting cellular immunity. A number of pattern recognition receptors (PRRs) that are important in the sensing of foreign nucleic acids, including MDA-5, RIG-I, and cGAS, are located within the cytosol for immune surveillance against many intracellular pathogens.^16–19 Accordingly, nucleic acid agonists of cytosolic PRRs are promising candidates as adjuvants for cancer vaccines due to their capacity to induce type-I interferons (IFN-I) and other pro-inflammatory cytokines essential to generating a robust CD8⁺ T cell response.^20–22 However, major delivery barriers have limited the adjuvant activity of nucleic acid agonists of cytosolic PRRs, including degradation by nucleases, rapid systemic distribution that can cause undesirable inflammatory side effects, low delivery to and uptake by antigen presenting cells (APCs), and critically, negligible delivery to the cytosol where these PRRs are localized.^17,23,24 Hence, the development of delivery platforms that address these barriers will be critical to realizing the potential of this promising and expanding class of vaccine adjuvants. Several groups, including ours, have devised ways to deliver antigens and/or adjuvants to the cytosol of professional APCs such as DCs to promote MHC-I presentation via the cytosolic antigen processing pathway, resulting in enhanced CD8⁺ T cell responses.^25–31

The double-stranded RNA (dsRNA) analogue polyinosinic:polycytidylic acid (polyIC) is one of the most clinically advanced nucleic acid adjuvants.^4,32 PolyIC can act through two distinct dsRNA sensing pathways: Toll-like receptor 3 (TLR-3), which resides in the endosomal membrane, and melanoma differentiation-associated protein 5 (MDA5) in the cytosol.^32–35 To protect polyIC from RNAse degradation, several formulations have been developed based on electrostatic complexation with cationic polymers, such as polyethylenimine (PEI) or poly-L-lysine (PLL).^36,37 Most notably, polyIC complexed with PLL and further stabilized with carboxymethylcellulose (polyICLC; also known as Hiltonol^®) was evaluated in multiple clinical trials and has been shown to stimulate IFN-I production and/or enhance T cell responses in humans and non-human primates.^36,38–42 Interestingly, polyICLC activates MDA-5 more effectively than polyIC and, thereby, enhances IFN-I and other proinflammatory cytokine responses to increase adjuvant potency.³⁴ This has been attributed to the ability of PLL to promote endosomal escape of polyIC via the “proton sponge effect.” Hence, cytosolic delivery of polyIC is an important design consideration for vaccine delivery technologies that aim to maximize its adjuvant properties.^43,44

PolyICLC has recently been investigated as a promising adjuvant for enhancing responses to peptide-based cancer vaccines in patients. Results from clinical trials demonstrated the ability of polyICLC to promote antigen-specific T cell responses to co-administered tumor antigens.^{4,38,41,45,46} In these clinical studies, however, antigens were mixed with polyICLC prior to vaccine administration. While translationally appealing, ample evidence show that simple mixing of antigen and adjuvant is less effective than co-delivering antigen and adjuvant on a common carrier.^{25,26,47–51} Co-delivery increases the probability that both antigen and adjuvant are internalized by the same APC, allowing antigen presentation to occur in an appropriate pro-inflammatory context and avoiding induction of T cell tolerance when only antigen is presented in the absence of signals required for APC maturation and T cell activation.⁵² Indeed, in a recent study, Seder and colleagues demonstrated that mixing tumor peptide antigens with polyICLC yielded a poor CD8⁺ T cell response when compared to a nanovaccine platform designed for co-delivery of antigen and TLR7/8 agonists.⁵³ Therefore, packaged co-delivery of antigen and polyIC may also be an important design criterion for maximizing cellular immunity in response to vaccination.

Here, we describe a versatile nanoparticle (NP) vaccine platform for facile co-loading and cytosolic co-delivery of antigens and polyIC as the adjuvant. This NP is designed with a PEG-rich corona that displays pyridyl disulfide (PDS) groups for covalent conjugation of thiol-containing antigens via thiol-disulfide exchange reactions, and a pH-responsive, endosomolytic core for electrostatic loading of immunostimulatory nucleic acids that also facilitates cytosolic delivery of both antigen and adjuvants. Upon intracellular uptake and in response to the decrease in pH within endosomal compartments, the NP destabilizes to trigger endosomal escape of vaccine cargo to the cytosol, allowing antigen to be processed and presented on MHC-I molecules and polyIC to activate the MDA-5 pathway. We demonstrate that the NP platform can efficiently co-load a model antigen, chicken ovalbumin (OVA), and polyIC to promote dual-delivery of OVA and polyIC to the cytosol. This enhances the immunostimulatory activity of polyIC via the MDA-5 pathway and promotes MHC-I restricted antigen presentation by APCs. Consequently, this coordinated innate immune activation via MDA-5 and MHC-I antigen presentation enhances the magnitude and functionality of the CD8⁺ T cell response, resulting in significant inhibition of tumor growth.

2. Materials and Methods

2.1. Polymer synthesis and characterization

Reversible addition-fragmentation chain transfer (RAFT) was used to synthesize the amphiphilic diblock copolymer poly[(polyethylene glycol) methacrylate)_0.9-co-(pyridyl disulfide ethyl methacrylate)_0.1]_13.7kDa-block-[(dimethylamino ethyl methacrylate)_0.5-co-(butyl methacrylate)_0.5]_15.3kDa (p[(PEGMA_0.9-co-PDSMA_0.1)]_13.7kDa-b-[(DMAEMA_0.5-co-BMA_0.5)]_15.3kDa (Supplementary Scheme S1). The chain transfer agent (CTA) used was 4-cyano-4-(phenyl-carbonothioylthio) pentanoic acid (Sigma-Aldrich) and the initiator used was 2,2’-azobis(4-methoxy-2,4 dimethylvaleronitrile) (V-70) (Wako Chemicals, Richmond, VA). Inhibitors were removed from monomer stocks by gravity filtration through an aluminum oxide column (activated, basic, Brockmann I; Sigma Aldrich).

For synthesis of the first block, filtered PEGMA (Mw = 300 Da, Sigma-Aldrich) and PDSMA (synthesized as previously described with minor modification)⁵⁴ was allowed to react under a nitrogen atmosphere in dioxane (40 wt % monomer) at 40°C for 24 h in an oil bath. The initial molar ratio of DMAEMA to PDSMA was 90:10, and the initial monomer ([M]₀) to CTA ([CTA]₀) to initiator ([I]₀) ratio was 55:1:0.2. The resultant poly(PEGMA-co-PDSMA) macro-chain transfer agent (mCTA) was isolated by dialysis against pure acetone (3x) using a 3.5 kDa MWCO SnakeSkin^™ membrane (Thermo Fisher Scientific), followed by half-acetone and half-deionized water (2x), and pure deionized water (1x). Following dialysis, the purified mCTA was frozen at −80°C for 5 h and then lyophilized for 3 days.

For the second block, purified mCTA was used for block copolymerization of filtered DMAEMA (Sigma Aldrich) and BMA (Sigma Aldrich) to create a pH-responsive polymer. DMAEMA (50%) and BMA (50%) ([M]₀/[mCTA]₀ = 143) were added to the mCTA dissolved in dioxane (40 wt % monomer and mCTA) along with V-70 initiator ([mCTA]₀/[I]₀ = 5). The solution was sealed and purged with nitrogen gas for 30 minutes and then allowed to react for 18 h at 30°C in an oil bath. The resulting amphiphilic diblock copolymer was isolated by precipitation into 80:20 pentane/ether and dried under vacuum for 48 h and stored in solid form at 4°C.

¹H NMR (CDCl₃ with TMS; Bruker AV400 spectrometer) was used to calculate polymer composition, degree of polymerization, and theoretical molecular weight of both the mCTA and diblock copolymer (Supplementary Figure 1A,B).

2.2. Preparation and characterization of polymer nanoparticles

Self-assembled micellar nanoparticles (NPs) were formulated by dissolving lyophilized polymer in ethanol to 50 mg/mL, followed by rapidly pipetting aliquots into phosphate buffer (100 mM, pH 7) to a final concentration of 10 mg/mL. For in vivo studies, ethanol was removed by buffer exchange into PBS (pH 7.4) via three cycles of centrifugal dialysis (Amicon, 3 kDa MWCO, Millipore), and NP solutions were then passed through a 0.2 μm Whatman® Puradisc polyethersulfone sterile filter (MilliporeSigma). Final polymer concentration was determined by measuring absorbance at 300 nm using a conventional UV–vis spectrophotometer (Synergy H1 Multi-Mode Microplate Reader, BioTek). The hydrodynamic size of the NPs was measured by dynamic light scattering (DLS) using a Malvern Zetasizer (Malvern, USA).

2.3. Preparation of antigen–nanoparticle conjugates

To covalently conjugate the model antigen, ovalbumin (OVA), to the pendant PDSMA groups on the NP via thiol–disulfide exchange reaction, free amines on the OVA protein were thiolated by incubation with ~24 molar excess of 2-iminothiolane (Traut’s Reagent, Thermo Fisher Scientific) in reaction buffer (100 mM phosphate buffer, pH 8, supplemented with 1 mM EDTA), as previously described.⁵⁵ The unreacted 2-iminothiolane was removed by buffer exchanging thiolated OVA into 1x PBS (pH 7.4) using Zeba Spin desalting columns (0.5 mL, 7 kDa MWCO, Thermo Fisher Scientific). For in vivo studies, thiolated OVA was sterilized via syringe filtration (0.2 μm, Millipore). The molar ratio of thiol groups to OVA protein was determined using Ellman’s reagent (Thermo Fisher Scientific) according to the manufacturer’s instructions and ~3–5 thiols per OVA were incorporated. Thiolated OVA was then reacted with polymer NPs at various molar ratios of NP:OVA (4:1, 8:1, 10:1) in PBS to make OVA–NP conjugates. Conjugation was done overnight, in the dark, at room temperature, and under sterile conditions, as previously described.^55,56 Antigen conjugation was verified via nonreducing SDS-polyacrylamide gel electrophoresis (SDS-PAGE) using 4–20% Mini-Protean TGX Precast protein gels (Bio-Rad). Gels were run at 130 V for 1 h and imaged with a Gel Doc EZ System (Bio-Rad). A conjugation ratio of 8:1 was used for all in vitro and in vivo formulations in order to maximize the amount of antigen delivered while maintaining particle stability. OVA from chicken egg white (MilliporeSigma) was used for conjugation characterization, and Endotoxin-free (<1 EU/mg) EndoFit OVA (Invivogen) was used for in vivo studies. OVA labeled with Alexa Fluor 647-NHS ester (AF647; Thermo Fisher Scientific) at a 2:1 AF647:OVA molar ratio was used to evaluate conjugation efficiency via fluorescent imaging of SDS-PAGE gels and for tracking conjugates after in vivo administration. DLS was used to measure the size of OVA–NP conjugates, as described above.

2.4. Formulation of nanoparticle/polyIC complexes

NP/polyIC complexes were formed by combining low molecular weight polyIC (Invivogen) with NP and OVA-NP in citric acid buffer (pH 4, 100 mM) at different theoretical charge ratios (+/−). After incubating at room temperature for 30 min, 2x volume 100 mM phosphate buffer (pH 8) supplemented with 2 mM NaOH was added and mixed rapidly to form NP/polyIC complexes. The charge ratio was defined as the molar ratio between protonated DMAEMA tertiary amines in the second block of the copolymer (positive charge; assuming 100% protonation at pH 4) and phosphate groups on polyIC backbone (negative charge). The charge ratios at which complete complexation of polyIC to the polymer occurred was determined via an agarose gel retardation assay. Free polyIC and complexes prepared at various charge ratios were loaded into lanes of a 2% agarose gel and run at 100 V for 1 h. Gels were stained with SYBR Safe (Invitrogen) for 1 h while protected from light and then imaged with a Gel Doc EZ system (Bio-Rad). A charge ratio of 6:1 was used for all in vitro and in vivo formulations in order to maximize the amount of polyIC delivered while maintaining stability of formulation. DLS was used to measure the size of the polyIC complexed with OVA-NP conjugates (OVA-NP/polyIC) formulation, as described above.

2.5. Erythrocyte lysis assay

The capacity of free polymer, ova polymer conjugates, and conjugate/polyIC complexes to induce pH-dependent disruption of lipid bilayer membranes was assessed via a red blood cell hemolysis assay as previously described.⁵⁷ Briefly, whole blood from de-identified patients was acquired from Vanderbilt Technologies for Advanced Genomics (VANTAGE) core. Blood was centrifuged to pellet erythrocytes, plasma was aspirated, and erythrocytes were resuspended in pH 7.4 PBS, and washed three times with PBS. After the final wash, erythrocytes were resuspended in 100 mM sodium phosphate buffer (supplemented with 150 mM NaCl) in the pH range of the endosomal processing pathway (pH 7.4, 7.0, 6.6, 6.2 and 5.8) and incubated with 10, 5, or 1 μg/mL polymer formulations (NP, OVA-NP, NP/polyIC, and OVA-NP/polyIC in a 96-well U-bottom plate for 1 h at 37°C. The extent of red blood cell lysis was determined using a UV–vis spectrophotometer (Synergy H1 Multi-Mode Microplate Reader, BioTek) by measuring the amount of hemoglobin released (abs = 541 nm) and normalized to a 100% lysis control (1%Triton X-100). Samples were run in quadruplicate.

2.6. Gal8 recruitment assay

Gal8 recruitment assay was performed as previously described with minor modifications.⁵⁸ Gal8-MDA-MB-231 cells were cultured and maintained in DMEM containing 4.5 g/L d-glucose and supplemented with 25 mM HEPES, 10% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin. Gal8-MDA-MB-231 cells were seeded in a 96 well cell culture microplate (Grenier) at a density of 5000 cells per well and allowed to adhere overnight. The following day, cells were treated with indicated nanoparticle formulations or PBS (Gibco). After a 16 h treatment, the media was aspirated and replaced with 100 μL of imaging media (FluoroBrite DMEM supplemented with 25 mM HEPES, 10% FBS, 100 U/mL penicillin and 100 μg/mL streptomycin, and 4 μM Hoechst 33342 (ThermoFisher Scientific). Cells were imaged using a 20x objective on an ImageXpress Micro XLS Widefield High-Content Analysis System. Four images were taken per well, and four wells were replicated per treatment, for a total of sixteen images per treatment (for PBS, twelve wells were replicated for a total of forty-eight images). Images were analyzed using the MetaXpress software Transfluor Application module to quantify the integrated YFP intensity and fluorescent nuclei per image, and integrated YFP intensity was normalized to the number of nuclei in each image. Statistics and graphing were then performed treating each image as an independent replicate.

2.7. Cell culture

The IRF and NF-κB reporter cell lines, human lung carcinoma A549-Dual^™ (Invivogen) and the murine macrophage cell line RAW-Dual^™ (Invivogen), were cultured in DMEM (Gibco) supplemented with 2 mM l-glutamine, 4.5 g/L d-glucose, 10% heat inactivated fetal bovine serum (HI FBS, Gibco), and 100 U/mL penicillin/100 μg/mL streptomycin (Gibco). The human monocyte cell line THP1-Dual^™ (Invivogen) was cultured in RPMI 1640 (Gibco) supplemented with 2 mM l-glutamine, 10% fetal bovine serum (FBS, Gibco), and 100 U/mL penicillin/100 μg/mL streptomycin (Gibco). The murine dendritic cell line DC2.4 was kindly provided by K. Rock (University of Massachusetts Medical School) and cultured in RPMI 1640 (Gibco) supplemented with 10% fetal bovine serum (HI FBS; Gibco), 2 mM l-glutamine, 100 U/mL penicillin/100 μg/mL streptomycin (Gibco), 50 μM 2-mercaptoethanol (Gibco), 1x nonessential amino acids (Cellgro), and 10mM HEPES (Invitrogen). Gal8-MDA-MB-231 cells were cultured and maintained in DMEM containing 4.5 g/L d-glucose and supplemented with 25 mM HEPES, 10% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin. All cell types were grown in a humidified atmosphere at 37 °C in 5% CO2.

2.8. In vitro evaluation of polyIC activity

Reporter cell lines that allow the simultaneous detection of activated NF-κB pathways, quantified by monitoring SEAP activity, and the IRF pathway, assessed by secreted luciferase activity, were used to evaluate the biological activity of NP/polyIC. Briefly, murine RAW-Dual^™ cells (macrophages; Invivogen) and human A549-Dual^™ cells (lung epithelial; Invivogen) were plated at 5 × 10⁴ cells/well in a 96-well plate, and allowed to adhere overnight. Human THP1-Dual^™ cells (monocytes; Invivogen) are a suspension cell line so they were plated the same day as dosing at 1 × 10⁵ cells/well. NP/polyIC was formulated as detailed above. A dose range was used to stimulate cells in the following groups: OVA-NP/polyIC, NP/polyIC, empty NP, free polyIC, or PBS for 24 h at 37°C with 5% CO₂. After incubation, cells were centrifuged at 1,500 rpm for 5 min; supernatant was collected and reporter proteins were measured using QUANTI-Blue^™ (Invivogen), a SEAP detection reagent, and QUANT-Luc^™ (Invivogen), a luciferase detection reagent by following the manufacturer’s instructions. Luminescence from luciferase activity and absorbance of SEAP were measured using a spectrophotometer (Synergy H1 Multi-Mode Microplate Reader, BioTek). All measurements were normalized after baselining to the average value of the PBS-treated negative control group. Values for EC₅₀ were extrapolated from dose-response curve fits using GraphPad Prism software.

2.9. In vitro evaluation of polyIC activity in MDA-5 deficient cells

RAW-Lucia^™ ISG cells and RAW-Lucia^™ ISG KO-MDA-5 cells were used to evaluate the role in MDA-5 in NP/polyIC function by monitoring interferon regulatory factor (IRF)-induced Lucia luciferase activity. Briefly, 5 × 10⁴ cells/well in a 96-well plate were allowed to adhere overnight. A dose range of NP/polyIC formulated as above were used to treat cells in the following groups: OVA-NP/polyIC, NP/polyIC, empty NP, free polyIC, a physical mixture of Lipofectamine 2000 and polyIC, or PBS for 24 h at 37°C with 5% CO₂. Following incubation, cells were centrifuged for 5 min at 1,500 rpm and the levels of IRF-induced Lucia in the supernatant were measured using QUANTI-Luc^™ (Invivogen), a Lucia luciferase detection reagent as described above. All measurements were normalized to the average value of the PBS-treated negative control group.

2.10. In Vitro evaluation of BMDC activation and maturation

Bone marrow cells were harvested from femurs and tibias of 6–8 week-old female C57Bl/6J mice by flushing them with cold PBS. Cells were centrifuged for 5 minutes at 450 × g and resuspended in complete BMDC culture media (RPMI 1640 medium supplemented with 10% heat-inactivated FBS, 100 U/mL penicillin, 100 μg/mL streptomycin, 2 mM l-glutamine, 10 mM HEPES, 1 mM sodium pyruvate, 1x non-essential amino acids, 50 μM β-mercaptoethanol, and 20 ng/mL GM-CSF). The cell suspension was passed through a 70 μM cell strainer (FisherbrandTM; Thermo Fisher Scientific), and seeded in 100 × 15 mm non-tissue-culture-treated Petri dishes (Corning Inc.) in complete medium containing 20 ng/mL granulocyte-macrophage colony-stimulating factor (GM-CSF) to induce differentiation into BMDCs. Cell were cultivated in a humidified chamber maintained at 37°C with 5% CO₂. Fresh complete BMDC culture medium was added on days 3, 5, and 7. On day 8, the percentage of CD11c+ cells (i.e., BMDCs) was confirmed to be greater than 80% as measured by flow cytometry using anti-CD11c-FITC (clone N418; BioLegend) (Supplementary Figure 2). BMDCs were seeded into wells of 12-well plates (Greiner Bio-One) at 6 × 10⁵ cells/well. BMDCs were treated with OVA-NP/polyIC, OVA + NP/polyIC, OVA-NP, NP/polyIC, OVA + polyIC, free polyIC, MPLA (TLR-4 agonist; positive control), or PBS for 24 h at 37°C with 5% CO₂. Following incubation, cells were scraped, washed with FACS buffer (PBS supplemented with 2% FBS), incubated with Fc-block (anti-CD16/CD32; clone 2.4G2; Tonbo) for 15 min at 4°C, and then stained with antibodies against the markers of DC activation: anti-CD80-PE/Cy5 (clone 16–10A1; BioLegend), anti-CD86-PE/Cy7 (clone GL-1; BioLegend) and anti-MHC-II-APC/Cy7 (clone M5.114.15.2; BioLegend) for 1 h at 4°C. Cells were then washed twice in FACS buffer, resuspended using FACS buffer supplemented with 1 μg/mL DAPI, and analyzed using an Luminex CellStream flow cytometer. Each treatment was performed with 3 technical replicates, and the experiment was conducted 3 times with similar results.

For qRT-PCR analysis of gene expression, BMDCs were seeded in 12-well plates at 6 × 10⁵ cells/well. BMDCs were treated as above for 6 h. Following incubation, cells were washed with PBS and 700 μL of RLT lysis buffer (Qiagen) was added to each well. Lysates were stored at −80°C until use. Messenger RNA (mRNA) was extracted from cell lysates using an RNA isolation kit (RNeasy mini kit, Qiagen). Complementary DNA (cDNA) was synthesized for each sample using a cDNA synthesis kit (iScript, Bio-Rad) and analyzed using qRT-PCR using Taqman kits (Thermo Fischer Scientific) and a CFX real-time PCR detection system (Bio-Rad) following the manufacturer’s instructions. Taqman probes for mouse Ifnb1 (Mm00439552_s1), Tnf (Mm00443258_m1), and Hmbs (Mm01143545_m1) were purchased from Thermo Fischer Scientific. Fold change was calculated using the ΔΔC_t method.

For analysis of secreted cytokines, supernatants were collected from BMDCs 24 h after treatment. A LEGENDplex Multi-Analyte Flow Assay Kit was used to measure secreted IFNβ, CXCL10, TNF-α, and IL-6 following the manufacturer’s instructions using a V-bottom plate. Data were collected on an Amnis CellStream Luminex Flow Cytometer equipped with 405, 488, 561, and 642 nm lasers and analyzed with LEGENDplex Data Analysis software v8.0 (VigeneTech).

2.11. In Vitro dendritic cell antigen presentation assay

An antibody that recognizes the mouse MHC-I (H-2K^b)-bound SIINFEKL was used to determine the effect of pH-responsive, endosomolytic NPs on protein antigen cross-presentation. Briefly, DC2.4 cells were plated at 2 × 10⁵ cells/well in a 12-well plate and allowed to adhere overnight. DC2.4 cells were treated with either OVA-NP/polyIC, OVA + NP/polyIC, OVA + polyIC, free OVA, PBS (negative control), or SIINFEKL peptide (positive control) for 24 h as described above. Following incubation, cells were treated with trypsin, washed, and resuspended with FACs buffer, incubated with Fc-block (anti-CD16/CD32; clone 2.4G2; Tonbo) for 15 minutes at 4°C, and then stained with PE/Dazzle 594-conjugated SIINFEKL/H-2K^b-reactive monoclonal antibody (clone 25.D1.16; Biolegend) for 1 h at 4°C. Cells were then washed 3x in FACS buffer, resuspended using FACS buffer supplemented with 1 μg/mL DAPI, and the relative levels of SIINFEKL/H-2K^b presentation was analyzed using an Amnis CellStream Luminex flow cytometer. Each treatment was performed with 3 technical replicates, and the experiment was conducted 3 times with similar results.

2.12. Animal care and experimentation

Female C57BL/6 mice (6–8 weeks old) were purchased from the Jackson Laboratory (Bar Harbor, ME). All animal experiments were reviewed and approved by the Vanderbilt University Institutional Animal Care and Use Committee (IACUC), and all surgical and experimental procedures were performed in accordance with the regulations and guidelines of the Vanderbilt University IACUC.

2.13. Intravital fluorescent imaging of antigen and polyIC at injection site

Female C57BL/6 mice (6–8 weeks old) were injected subcutaneously in the right lower flank with formulations containing 58 μg Alexa Fluor 647-labeled OVA (labeling detailed in Section 2.3) and 32 μg rhodamine-labeled LMW polyIC (Invivogen). Experimental groups were as follows: PBS, polyIC complexed OVA-NP conjugates (OVA-NP/polyIC), or a soluble mixture of OVA and polyIC (OVA + polyIC). Injection site was longitudinally monitored for one week via intravital imaging using the IVIS Lumina III (PerkinElmer). The levels of Alexa Fluor 647 and rhodamine were determined using an Ex 650 /Em 668 nm and Ex 571/ Em 591 filter set, respectively. Spectral unmixing, which is a tool provided by the IVIS Lumina III, was used to isolate and quantify each individual fluorescence source.

2.14. In vivo analysis of lymph node accumulation

Female C57BL/6 mice (6–8 weeks old) were injected subcutaneously at the base of tail with formulations containing 58 μg Alexa Fluor 647-labeled OVA and/or 32 μg rhodamine-labeled LMW polyIC (Invivogen). Experimental groups were as follows: PBS, OVA-NP/polyIC, or OVA + polyIC). After 48 h, inguinal (draining) LNs were harvested, and the AF647 and rhodamine fluorescence signal was imaged and measured with the IVIS Lumina III (PerkinElmer). The injection site retention half-life of fluorescently-labeled OVA and polyIC was estimated by fitting fluorescence intensity as a function of to an exponential decay model using non-linear regression.

2.15. qRT-PCR analysis of gene expression in lymph nodes

Female C57BL/6 mice (6–8 weeks old) were injected subcutaneously at the base of tail with formulations containing 58 μg Alexa Fluor 647-labeled OVA and/or 32 μg rhodamine-labeled LMW polyIC (Invivogen). Experimental groups were as follows: PBS, OVA-NP/polyIC, or OVA + polyIC). After 48 h, inguinal LNs were harvested and placed in RLT lysis buffer (Qiagen) supplemented with 2% β-mercaptoethanol (Sigma) in a gentleMACS M tube with mechanical disruption using an OctoMACS tissue dissociator (Miltenyi). LN RNA was isolated with a RNeasy RNA isolation kit (Qiagen) with the RNase-free DNase Set (Qiagen), used according to manufacturer’s specifications. Complementary DNA (cDNA) was synthesized with the Bio-Rad iScript cDNA kit and analyzed via qPCR using the appropriate TaqMan kits (Thermo Fisher Scientific). The TaqMan gene expression kits were: Ifnb1 (Mm00439552_s1), Cxcl10 (Mm00445235_m1), Tnf (Mm00443258_m1), Il6 (Mm00445235_m1), and Hmbs (Mm01143545_m1).

2.16. In vivo immunization

C57BL/6 mice (6–8 weeks old) were immunized via subcutaneous injection at the base of the tail on days 0, 7, and 14 with formulations containing 58 μg of OVA and/or 32 μg of polyIC with or without 400 μg of polymer in PBS. The groups were as follows: PBS, OVA-NP/polyIC, OVA + NP/polyIC, OVA-NP, OVA + polyIC, or OVA. On day 20, whole blood was collected for SIINFEKL (pOVA/H-2K^b) tetramer staining. On day 21, mice were euthanized to evaluate antibody titer and track CD8⁺ T cell response by tetramer staining, as well as to determine T cell function by intracellular cytokine staining and ELISpot assay.

2.17. Analysis of OVA-specific CD8⁺ T cell responses in whole blood

On day 20 after immunizations, whole blood was collected in K₂EDTA treated tubes (BD Biosciences), treated with ACK lysis buffer (KD Medical), washed, resuspended in cold FACS buffer (PBS supplemented with 2% FBS and 50 μM dasatinib), and plated in a 96-well U-bottom plate. Next, the cells were centrifuged for 5 min at 1,500 rpm and resuspended in FACS buffer and incubated with Fc-block (anti-CD16/CD32, clone 2.4G2; Tonbo) for 15 min at 4°C, and stained with antibodies CD45.2 (APC; clone 104; BioLegend), CD3ε (PE/Cy7; clone 145.2C11; BioLegend), and CD8α (APC/Cy7; clone 53–6.7; Tonbo) for 1 h at 4°C. Cells were then washed 3x in FACS buffer and then stained for 2 h with 1.5 μg/mL of PE-labeled OVA _257–264 pOVA/H-2K^b tetramer prepared according to a previously reported procedure.⁵⁹ Cells were then washed 3x in FACS buffer, resuspended using FACS buffer supplemented with 1 μg/mL DAPI, and analyzed using an Amnis CellStream Luminex flow cytometer. Representative flow cytometry data and gating strategies for determining the frequency of SIINFEKL-specific CD8⁺ T cells are shown in Supplementary Figure 3.

2.18. Preparation of splenocytes

On day 21 after immunization, mice were euthanized and spleens harvested and mechanically disrupted into single-cell suspensions in complete RPMI 1640 (cRPMI; 10% FBS, 100 U/mL penicillin, 100 μg/mL streptomycin, 50 μM 2-mercaptoethanol, and 2 mM l-glutamine) by forcing them through a 70 μm cell strainer (FisherbrandTM; Thermo Fisher Scientific) using a sterile syringe plunger. Cells were passed through the strainer two more times to remove any residual tissue fragments. The cells were centrifuged for 5 min at 1,500 rpm and resuspended in ACK lysis buffer (KD Medical) to remove erythrocytes. After 5 min incubation, cRPMI was added to deactivate ACK lysis buffer. Finally, cells were centrifuged and resuspended in cRPMI.

2.19. Analysis of OVA-specific CD8⁺ T cell responses in the spleen

Splenocytes were plated at 3 × 10⁶ cells/well in a 96-well U-bottom plate. Next, the cells were centrifuged for 5 min at 1,500 rpm and resuspended in FACS buffer and incubated with Fc-block (anti-CD16/CD32, clone 2.4G2; Tonbo) for 15 min at 4°C, and then stained with antibodies CD45.2 (APC; clone 104; BioLegend), CD3ε (PE/Cy7; clone 145.2C11; BioLegend), and CD8α (APC/Cy7; clone 53–6.7; Tonbo) for 1 h at 4°C. Cells were then washed 3x in FACS buffer and then stained for 2 h with 1.5 μg/mL of PE-labeled pOVA/H-2K^b tetramer. Cells were then washed 3x in FACS buffer and analyzed as described above. Representative flow cytometry data and gating strategies for determining the frequency of SIINFEKL-specific CD8⁺ T cells are shown in Supplementary Figure 3.

2.20. Intracellular cytokine staining of OVA-specific CD8⁺ and CD4⁺ T Cells

Splenocytes were plated in 96-well U-bottom plates at 2 × 10⁶ cells/well in cRPMI and the appropriate stimulant or control was added: 10 μM of MHC-I H-2K^b epitope SIINFEKL (OVA_257–264; Invivogen), 10 μM of MHC-II H-2A^b epitope ISQAVHAAHAEINEAGR (OVA_323–339; Invivogen), 1x cell stimulation cocktail (PMA and ionomycin; Thermo Fischer Scientific) as the positive control, and cRPMI as the negative control. Cells were incubated at 37°C in an atmosphere of 5% CO₂ for 1 h 30 min. BD GolgiPlug protein transport inhibitor (BD Biosciences) was then added to each well, and cells were incubated for an additional 5 h 30 min. Following incubation, cells were washed with PBS and stained with eFluor 450 fixable viability dye (eBioscience) for 30 min at 4°C. Cells were next washed with FACS buffer (PBS supplemented with 2% FBS) and incubated with Fc-block (anti-CD16/CD32, clone 2.4G2; Tonbo) for 15 min at 4°C, and then stained with antibodies for CD3ε (PE/Cy7; clone 145.2C11; Biolegend), CD8α (APC/Cy7; clone 53–6.7; Tonbo) and CD4 (AF488; clone RM4–5; Biolegend) for 1 h at 4°C. Cells were washed 2x in FACS buffer, then fixed and permeabilized by incubating for 10 min at 4°C with BD Cytofix/Cytoperm (BD Biosciences), according to manufacturer instructions. Cells were then washed 2x with 1x BD perm/wash buffer (BD Biosciences) and incubated for 1 h at 4°C with antibodies against intracellular cytokines: anti-IFNγ-APC (clone XMG1.2; BD Biosciences) and anti-TNFα-PE (clone MP6-XT22; BD Biosciences). Finally, cells were washed once with 1x perm/wash buffer, resuspended in FACS buffer supplemented with 50 nM dasatinib, and analyzed as described above. Data are reported as the percentage of CD8α⁺ cells that are IFNγ⁺ and/or TNFα⁺ after subtraction of background values from unstimulated negative controls. Representative gating for ICCS analysis of splenocytes is presented in Supplementary Figure 4.

2.21. Enzyme-Linked Immunosorbent Spot assay (ELISpot)

Splenocytes from each vaccinated mouse were evaluated for antigen-specific IFN-γ production by ELISpot assay (Mouse IFN-gamma Single-Color ELISpot; ImmunoSpot) according to manufacturer’s instructions with minor modifications. Microtiter 96-well plates pre-coated with anti-mouse IFN-γ monoclonal antibody (capture antibody) were washed 3x with sterile PBS and blocked with 200 μL of complete RPMI 1640 for 2 h at 37°C in an atmosphere of 5% CO₂. Medium was aspirated and the appropriate stimulant or control added: 10 μg/mL SIINFEKL peptide, 10 μM ISQAVHAAHAEINEAGR peptide, 10 μg/mL Concanavalin A (Invivogen) as the positive control, and 10 μg/mL Influenza A NP (366–374) (GenScript) as the negative control. Immediately thereafter, splenocytes were plated in quadruplicate at 2.5 × 10⁵ cells/well and incubated for 48 h at 37°C in an atmosphere of 5% CO₂. Plates were washed 3x with wash buffer (PBS supplemented with 1% v/v FBS and 0.05% v/v Tween 20) and incubated for 2 h at room temperature with biotin-conjugated rat anti-mouse IFN-γ detection antibody (BDbiosciences). Plates were washed four times with wash buffer and incubated with 1 μg/mL of avidin-HRP for 45 min at room temperature. After three washes with wash buffer and two washes with PBS, 100 μL of Blue Developer Solution prepared according to (ImmunoSpot) was added to each well and left to develop for ~4 min at room temperature or when solution turns blue in the darkest wells. Plates were then washed 5x with water, dried overnight, and the number of spots was counted using an ImmunoSpot ELISpot reader and analysis software package (Cellular Technology Limited). The average number of spots counted upon incubation with cRPMI (i.e., background) was subtracted from the number of spots counted upon peptide stimulation, and data are reported as the number of IFN-γ spot forming cells (SFCs) normalized to 2 × 10⁵ cells.

2.22. Antibody titer

Approximately 100 μL of blood was collected from each mouse via cardiac puncture, and sera were tested for OVA-specific IgG. Nunc MaxiSorp plates (high protein binding plates; Thermo Fisher Scientific) were coated with 10 μg/mL OVA in 1x PBS overnight at 4°C. Plates were then washed two times with PBS and blocked with PBS/0.01% tween 20 for 1 h at room temperature. Sera were added at a 1/100 dilution and subsequent 10-fold serial dilutions in PBS/0.01% tween 20 and incubated for 2 h at room temperature. Sera from one naïve mouse (negative control) and monoclonal anti-chicken OVA antibody (positive control; Sigma-Aldrich) were included in each plate to determine cutoff values. Following incubation, plates were washed 3x with PBS and incubated with secondary antibody (anti-IgG-HRP; EMD Millipore) at a 1:5000 dilution in PBS for 1 h at room temperature. Plates were again washed 3x with PBS and incubated with 100 μl of developing agent (1-step Ultra-TMB ELISA; Thermo Fisher Scientific). After 1 min, the enzymatic reaction was quenched with 100 μl of 0.18 M sulfuric acid and absorption of the colormetric reaction measured within 30 min at 450 nm using a plate reader (Synergy HTX). End point titers were determined from reciprocal dilutions using a sigmoidal fit (GraphPad Prism 5; GraphPad Software Inc.) to determine the dilution at which the A_{450 nm} value was equal to the mean + two standard deviations of that of naïve serum.

2.23. Tumor studies

For the prophylactic tumor challenge studies, C57BL/6 mice (6–8 weeks old) were immunized via the subcutaneous route at the base of the tail on days 0, 7, and 14 with formulations containing 58 μg of OVA and/or 32 μg of polyIC with or without 400 μg of polymer in PBS. The groups were as follows: PBS, OVA-NP/polyIC, OVA + NP/polyIC, OVA-NP, OVA + polyIC, or OVA. Mice were challenged 7 days following the final vaccination by subcutaneous flank injection of 3 × 10⁵ EG7.OVA cells (C57BL/6 mouse derived EL-4 thymoma line expressing OVA cDNA). Tumor volume was measured three times per week via caliper measurements using the formula V = (L × W × W)/2. Mice were euthanized at a tumor burden end point of 1500 mm³.

For the EG7.OVA therapeutic vaccination model, C57BL/6 mice (6−8 weeks old) were inoculated via subcutaneous flank injection with 3 × 10⁵ EG7.OVA cells. Mice were then vaccinated as described above on days 5, 12, and 19 with formulations containing 58 μg of OVA and/or 32 μg of polyIC with or without 400 μg of polymer in PBS. The groups were as follows: PBS, OVA-NP/polyIC, OVA + NP/polyIC, OVA-NP, OVA + polyIC, or OVA. Tumor growth was monitored as indicated above. Mice were euthanized at a tumor burden end point of 2000 mm³.

2.24. Statistical analysis

Significance for each experiment was determined as indicated in the corresponding figure captions. All analyses were done using GraphPad Prism software, version 7.0c. Plotted values represent experimental means, and error bars represent SD unless otherwise noted in the figure captions. **** P < 0.0001, *** P < 0.005, **P < 0.01, * P < 0.05.

3. Results and Discussion

3.1. Synthesis and characterization of polymeric carriers for dual-delivery of antigen and nucleic acid adjuvants

We synthesized a pH-responsive diblock copolymer designed to facilitate cytosolic delivery of vaccine cargo, which we expected would increase the immunostimulatory activity of polyIC by enhancing access to MDA-5 while also promoting antigen presentation on MHC-I, resulting in increased CD8⁺ T responses. The polymer is composed of two blocks synthesized by reversible addition–fragmentation chain transfer (RAFT) polymerization (Figure 1A and Supplementary Scheme 1). The first block is primarily composed of the hydrophilic monomer poly(ethylene glycol) methacrylate (PEGMA) (MW 300 Da), to confer colloidal stability to the NP, and is copolymerized with a small percentage (10%) of pyridyl disulfide ethyl methacrylate (PDSMA) for conjugation of thiol-bearing protein or peptide antigens via thiol-disulfide exchange reaction. The first block is then chain extended with the second block which is composed of cationic, pH-responsive dimethylamino ethyl methacrylate (DMAEMA) and hydrophobic butyl methacrylate (BMA) at a 50:50 molar ratio, which we and others have previously demonstrated exhibits potent endosomal escape activity.^60–62 In aqueous solution, this amphiphilic diblock copolymer self-assembles into micellar nanoparticles of ~30 nm in diameter. After cellular uptake, the acidic endosomal environment protonates DMAEMA residues, triggering a micelle-to-unimer conversion that exposes hydrophobic BMA groups, which act cooperatively with DMAEMA residues to destabilize the endosomal membrane, allowing release of cargo into the cytosol (Figure 1B). This is distinct from the inefficient and still poorly understood and debated proton sponge mechanism harnessed by PLL, PEI, and several other cationic polymers, which relies on the buffering capacity of cationic polymers with ionizable amino groups to generate osmotic stress within endo/lysosomes.⁶³

Figure 1: Fabrication and characterization of nanovaccine platform for cytosolic dual-delivery of antigen and nucleic acid adjuvants.

(A) Chemical structure and composition of pH-responsive endosome-destabilizing poly[(PEGMA-co-PDSMA)-block-(DMAEMA-co-BMA)] diblock copolymer. (B) Schematic representation of polymeric nanoparticle promoting antigen and adjuvant (polyIC) delivery in the cytosol via endosomal escape, resulting in MHC-I antigen presentation and activation of innate immunity via MDA-5 signaling, which act synergistically to enhance antigen-specific CD8⁺ T cell responses. (C) Schematic representation of a rapid, facile, and scalable process for co-loading antigen and nucleic acid adjuvants via simple mixing of thiol containing antigen (OVA protein) and micellar nanoparticle, resulting in covalent linkage via disulfide bridge bond, followed by electrostatic complexation of polyIC to the cationic DMAEMA groups in the second block (core) of carrier. (D) Schematic of the experimental and control formulations evaluated. (E) Thiolated OVA protein labeled with Alexa Fluor 647 was reacted with NPs to form conjugates at a 6:1 molar ratio. SDS-PAGE was used to confirm antigen conjugation. Lane (1) free OVA protein; (2) OVA + NP/polyIC; (3) OVA-NP; (4) OVA-NP/polyIC; (5) NP alone (no OVA). Material loaded into each lane was normalized based on 2 μg OVA. (F) PolyIC was complexed with NP and conjugate (OVA-NP at 6:1 ratio) at a N:P (+/−) charge ratio of 4:1. Agarose gel electrophoresis was used to confirm polyIC complexation. Lane (1) free polyIC; (2) NP alone (no polyIC); (3) OVA-NP (no polyIC); (4) NP/polyIC; (5) OVA-NP/polyIC. Material loaded into each lane was normalized to 1 μg polyIC. (G) Representative number-average size distributions of NP alone, OVA-NP (6:1 molar ratio), and OVA-NP/polyIC (6:1 molar ratio, 4:1 charge ratio) as measured by dynamic light scattering. (H) Erythrocyte hemolysis assay demonstrating pH-dependent membrane destabilizing activity of the empty NP, nanoparticle OVA conjugate (OVA-NP), and conjugate complexed with polyIC (OVA-NP/polyIC). Concentrations are normalized to 5 μg/ml of polymer. (mean ± SD; n = 4; statistical significance between OVA-NP/polyIC and all other formulations at each pH are shown; *P <0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; one-way ANOVA with Dunnett’s multiple comparisons test). (I) Schematic of Galectin 8 (Gal8) recruitment assay used to investigate endosomal escape of selected formulations. (J) Representative fluorescent images of MDA-MB-231 cells expressing Gal8-YFP fusion protein upon treatment with indicated nanovaccine formulations or controls. The fluorescent intensity of puncta increases significantly when YFP is recruited to the disrupted endosome, which is quantified by integrating YFP fluorescent intensity of the puncta per cell. (K) Integrated Gal8-YFP intensity per cell for indicated formulations (mean ± SD; n = 6; statistical significance between all formulations vs. PBS are shown; *P <0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; one-way ANOVA with Dunnett’s multiple comparisons test).

The ability of the NP to co-load the model protein antigen ovalbumin (OVA) and the nucleic acid adjuvant polyIC was next assessed. To achieve this, and with an eye towards future translation, we devised a simple and scalable two-step formulation process that required no intermediate purification steps prior to administration. As shown in Figure 1C, micellar NPs in aqueous solution are first mixed with thiol-containing antigen (e.g., OVA) to covalently link antigen to the hydrophilic NP corona. The antigen-NP conjugates are then electrostatically complexed with polyIC at pH 4, followed by neutralization of the solution to pH 7.4 to yield NPs that are dual-loaded with antigen and nucleic acid adjuvant. To first investigate antigen conjugation, we thiolated AlexaFluor647-labeled OVA using Traut’s reagent (3–5 thiols/OVA) and subsequently reacted it with PDSMA groups presented on the corona of the NP at various molar ratios of NP:OVA (2:1, 4:1, 6:1, 8:1, 10:1, 20:1) to form OVA-NP conjugates. To assess conjugation efficiency, we used SDS-polyacrylamide gel electrophoresis (SDS-PAGE) to monitor the band shift and disappearance of AF647-labeled OVA due to conjugation to NP. Complete conjugation was achieved at all tested ratios (Supplementary Figure 5A) demonstrating the potential to stoichiometrically tailor the antigen loading capacity. Following conjugation, negatively charged polyIC was electrostatically complexed to the tertiary amine groups on DMAEMA in the second block at various N:P charge ratios (i.e., the molar amount of protonatable amines (N = nitrogen) groups to the molar amount of phosphate (P) groups in the nucleic acid backbone). To achieve this, complexes were prepared at pH = 4, which is below the acid dissociation constant (pKa ~ 7) of DMAEMA, ensuring that the DMAEMA groups are predominantly protonated and positively charged, thus allowing for electrostatic complexation with negatively charged polyIC. Following complexation, the solution was further diluted in 100 mM phosphate buffer (pH 8) to a final pH of 7.4 before use. N:P ratios of 1:1, 2:1, 4:1, 6:1, 8:1, 12:1, 20:1 were tested and agarose gel electrophoresis was used to evaluate polyIC complexation efficiency (Supplementary Figure 5B). It was determined that N:P ratios of 6:1 and greater enabled nearly complete complexation of polyIC, as demonstrated by the lack of migration of NP/polyIC complexes. A schematic representation of the formulations used in subsequent experiments is shown in Figure 1D.

We next assessed whether OVA and polyIC could be co-loaded on the same NP using SDS-PAGE and agarose gel electrophoresis. Using an NP:OVA molar ratio of 8:1 and charge ratio of 6:1 for polyIC complexation, it was determined that both species could be loaded onto NPs (Figure 1E,F). We further confirmed the assembly and evaluated the particle size of NP micelles, OVA-NP conjugates (8:1), NP/polyIC complexes (6:1), and NPs co-loaded with both OVA and polyIC (8:1 OVA-NP/polyIC 6:1) via dynamic light scattering (DLS) (Figure 1G). DLS analysis demonstrated that free NP micelles are ~30 nm in diameter and after co-loading OVA and polyIC the diameter increased to ~60 nm. A conjugation ratio of 8:1 and a complexation ratio of 6:1 were selected for all subsequent investigations in order to maximize the amount of antigen and adjuvant delivered per particle, while maintaining particle size and colloidal stability.

Next, we assessed the capacity of the NPs alone and formulated with OVA and polyIC to induce pH-dependent disruption of lipid bilayer membranes using an erythrocyte hemolysis assay (Figure 1H), which is commonly used to predict the membrane-destabilizing activity of polymers.⁵⁷ Significant pH-dependent hemolytic activity was evident with the all of the formulations tested, with a slight decrease observed following complexation of polyIC, which may reflect reduced access of endosomolytic segments of the NP to erythrocyte membranes when complexed with the nucleic acid. To further validate the endosomolytic activity of the NP, we employed a Gal8-YFP reporter assay that directly measures endosomal disruption (Figure 1I).⁵⁸ This assay utilizes the fusion protein Gal8-YFP, which is a fusion between Galectin 8 (Gal8), an endogenous cytosolic protein that binds glycans, such as those found on the intraluminal membrane of endosomes, and yellow fluorescent protein (YFP). Following treatment with an endosomolytic agent, Gal8-YFP redistributes from the cytosol to the site of the ruptured endosomes to bind the newly exposed glycan. The degree of Gal8-YFP recruitment directly measures endosomal disruption, and has been shown to correlate with increased activity for cytosolic-acting nucleic acid drugs. In this study, we used a previously validated Gal8-YFP-expressing MDA-MB-231 breast cancer cell line (Gal8-MDA-MB-231).^58,64 Following treatment with 100 μg/mL of OVA-NP/polyIC, OVA-NP, NP/polyIC, empty NP, or PBS as a negative control, we quantified the integrated YFP fluorescence intensity of the puncta per cell, reflective of the quantity of bound Gal8-YFP molecules at the disrupted endosomes. Consistent with the hemolysis assay, all treatments significantly increased Gal8 recruitment compared to PBS, with empty NP inducing the most endosomal disruption, followed by NP/polyIC, OVA-NP/polyIC, and OVA-NP, though differences between the groups were not statistically significant (Figure 1J,K). Collectively, these data demonstrate that our nanovaccine design strategy enables rapid, facile, and efficient co-loading of antigen and polyIC into sub-100 nm NPs that display potent endosomolytic activity.

3.2. Endosomolytic nanoparticles enhance co-delivery of protein antigens and polyIC to cytosolic targets

Next, we evaluated the ability of the NP platform to enhance the immunostimulatory activity of polyIC using reporter cells that stably express an inducible reporter (Lucia Luciferase), which facilitates quantification of interferon-regulatory factor (IRF) pathway activation. THP1-Dual^™ cells (human monocyte-like cell line), RAW-Dual^™ cells (murine macrophage-like cell line), and A549-Dual^™ cells (human lung epithelial cell line) were treated with OVA-NP/polyIC, NP/polyIC, empty NP, and free polyIC over a range of polyIC doses; PBS (vehicle) served as the negative control. After 24 h, cell culture supernatant was collected and the relative level of IFN-I production was quantified by luciferase luminescence assay. In all cell types, NPs complexed with polyIC led to a substantial increase in IFN-I production compared to free polyIC alone (Figure 2A), which exerted activity only at relatively high concentrations in vitro (>100 μg/mL).

Figure 2: pH-responsive nanoparticles enhance dual-delivery of polyIC and antigens to the cytosol to enhance class I antigen presentation and dendritic cell activation in vitro.

(A) In vitro evaluation of TLR3 and MDA-5 activation in reporter cell lines in response to stimulation with indicated formulations for 24 h (Data were collected in technical triplicates). (B) In vitro evaluation of the role of MDA-5 activation using RAW-ISG Lucia vs. RAW-ISG Lucia KO MDA-5 reporter cells to monitor the type-I interferon production in response to 24 h incubation with indicated formulations (Data were collected in technical triplicates). (C) Flow cytometric quantification of median fluorescence intensity (MFI) of CD80, CD86, MHC-II expression by BMDCs treated with indicated formulations. (D) Analysis of Ifnb1 and Tnf gene-expression by qRT-PCR in BMDCs treated for 24 h as indicated. (E) Concentration of secreted cytokines by BMDCs treated with indicated formulations for 24 h. (F) Flow cytometric analysis of the median fluorescent intensity (MFI) of DC2.4 cells treated with indicated formulations and stained with PE-labeled anti SIINFEKL/H-2k^b antibody. Statistical data are presented as mean ± SD; n = 4; statistical significance between OVA-NP/polyIC vs other indicated formulations are shown; *P <0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; one-way ANOVA with Tukey’s multiple comparisons test.

Since polyIC is an agonist for both endosomal TLR3 and cytosolic MDA-5 signaling pathways, we directly assessed the contribution of MDA-5 in mediating this response by treating RAW-Lucia^™ ISG MDA-5-deficient reporter cell line (RAW-Lucia^™ ISG KO MDA-5) and monitoring IRF-induced Lucia luciferase activity compared to RAW-Lucia^™ ISG cells that express endogenous MDA-5 (Figure 2B). IFN-I response was significantly diminished, though not fully abrogated, in the MDA-5 knockout cells, indicating that NPs facilitate delivery of polyIC to the cytosol and activate an MDA-5-dependent IFN-I response. Interestingly, the activity of polyIC delivered using the commercial transfection agent Lipofectamine 2000 was only slightly inhibited in the MDA-5 knockout cells, implicating the activation of other PRRs, likely TLR3, by polyIC, and further demonstrating the relative potency of the NPs to enhance cytosolic delivery of the adjuvant cargo. In this work we selected polyIC as a nucleic acid adjuvant due to its high translational relevance in immuno-oncology, but in principle the NP platform could be used to co-deliver antigen with virtually any nucleic acid adjuvant, including agonists of other cytosolic PRRs. For example, we have previously demonstrated that the immunostimulatory activity of 5’triphosphate RNA (3pRNA), an agonist of RIG-I, and interferon stimulatory DNA (ISD), a ligand for cGAS, can be dramatically enhanced when delivered using other endosomolytic carriers.^65,66 Hence, this nanovaccine platform may also open new opportunities for exploring other emerging adjuvants to maximize their in vivo activity.

Consistent with their capacity to enhance polyIC activity, we also found that NP-mediated delivery of polyIC increased expression of the DC maturation markers MHC-II and co-stimulatory molecules CD80 and CD86 by murine bone marrow-derived dendritic cells (BMDCs) to a greater extent than free polyIC or NP formulations lacking polyIC (Figure 2C). Additionally, NPs complexed with polyIC significantly increased gene expression of Ifnb1 and Tnf by BMDCs as measured by quantitative RT-PCR 6 h after treatment (Figure 2D). To further evaluate DC activation, we quantified cytokine and chemokine levels in the supernatant of BMDCs treated with NPs formulated with and without polyIC (Figure 2E). We observed that complexation of polyIC with NPs led to higher concentrations of secreted IFN-β1, an important cytokine for CD8⁺ T cell activation.⁶⁷ Interestingly, we also found that free polyIC induced a higher level of CXCL10, TNFα, and IL-6 production than when complexed with the NP, potentially reflecting a differential cytokine response elicited by TLR3 and MDA5.^34,43,68 Additional studies, including a deeper analysis of the cytokine profiles induced by both free and NP-complexed polyIC, are necessary to further interrogate this possibility. The importance of IFN-I in generating CD8⁺ T cell responses is notable and should be weighed against the potentially deleterious inflammatory side effects of IL-6 and TNFα that may limit vaccine tolerability, safety, and/or efficacy.^69,70 It should also be noted that endosomal escape is not a fully efficient event^71,72 and, therefore, it is likely that a fraction of internalized polyIC is retained within the endosome for activation of TLR3. This raises the possibility that the cytokine profile can be tailored by controlling the relative activation of TLR3 and MDA5 pathways, which have been shown to act synergistically in some settings,^17,35,36 via delivery of polyIC with NPs of varying endosomolytic activity, a property we have demonstrated can be precisely tuned via control of polymer composition.⁶⁰ If true, this may also afford an opportunity to design NPs capable of leveraging, and optimizing, synergy between multiple nucleic acid adjuvants that engage both cytosolic (e.g., RIG-I, cGAS) and endosomal (e.g., TLR7, 9) PRRs.

Finally, we evaluated the capacity of endosomolytic NPs to enhance MHC-I antigen presentation, using OVA as a model antigen. We treated DC2.4 dendritic cells with free OVA, OVA-NP, OVA-NP/polyIC, OVA + NP/polyIC, and OVA + polyIC, and OVA_257–264 (SIINFEKL) peptide. In vitro presentation of the immunodominant MHC-I restricted OVA epitope SIINFEKL was assessed by flow cytometry using a fluorescently-labeled antibody that recognizes H-2K^b-bound SIINFEKL (Figure 2F). First, we found that covalent conjugation of OVA to NPs via a disulfide linker significantly enhanced SIINFEKL presentation on MHC-I compared to free OVA and a mixture of OVA and polyIC. This is consistent with our previous findings using other pH-responsive endosomolytic polymers that promote cytosolic antigen delivery.^25–27,73 MHC-I presentation of SIINFEKL was further augmented upon complexation of OVA-NP conjugates with polyIC. Collectively, these results demonstrate that pH-responsive, endosomolytic NPs facilitate the cytosolic co-delivery of OVA and polyIC, resulting in DC activation, production of IFN-I, and enhanced class I antigen presentation, with the potential to enhance CD8⁺ T cell activation.

3.3. NP delivery improves the pharmacological properties of antigen and polyIC in vivo

Rapid clearance, poor cellular uptake, and inefficient lymph node (LN) accumulation are major pharmacological barriers that limit the immunogenicity of protein and peptide antigens as well as the potency, and potentially tolerability and safety, of many promising nucleic acid adjuvants.^8,74 It is well-established that nanoparticles less than ~100 nm in diameter can exploit lymphatic drainage to enhance cargo accumulation in vaccine site draining lymph nodes.^8,75 Hence, we postulated that delivery of OVA and polyIC with NPs could address these barriers. To evaluate this, we first used intravital imaging to quantify injection site retention kinetics of AlexaFluor647-labeled OVA and rhodamine-labeled polyIC, formulated with or without loading onto NPs, following subcutaneous administration (Figure 3A). As anticipated, free OVA and polyIC very rapidly cleared the injection site (i.e., half-life ~1.5 h); polyIC is polydisperse (0.2–1 kbp; ~100–500 kDa) and is highly susceptible to nuclease degradation,³⁶ and though OVA (43 kDa) is of sufficient molecular weight to be absorbed via the lymphatics⁷⁶ it accumulates only minimally within lymph nodes. By contrast, OVA and polyIC loaded onto NPs cleared the injection site slowly, with injection site half-lives of ~44 h and ~28 h, respectively. This can likely be attributed to the larger size of the OVA-NP/polyIC formulation as well as non-specific interactions with surrounding tissue at the injection site. Additionally, the retention profiles of OVA and polyIC were closely matched when integrated into NPs, an indication of co-transport on common nanocarrier.

Figure 3: NP vaccines modulate antigen and polyIC clearance kinetics to enhance and coordinate delivery to draining lymph nodes.

(A) Representative fluorescence IVIS images (left) of mice following subcutaneous administration of NP vaccine formulated with AF647-labeled OVA and rhodamine-polyIC or a soluble mixture of AF647-OVA and rhodamine-polyIC and (right) quantification of relative amounts of OVA and polyIC at the injection site as a function of time post-injection (n = 2 mice/PBS group and n = 5 mice for all other groups tested). (B) Representative images and IVIS quantification of fluorescence intensity of the vaccine site draining inguinal LN 48 h following subcutaneous administration of NP vaccine formulated with AF647-labeled OVA and rhodamine-polyIC or a soluble mixture of AF647-OVA and rhodamine-polyIC (mean ± SD; n = 5; statistical significance between OVA-NP/polyIC vs other indicated formulations are shown; *P <0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; one-way ANOVA with Dunnett’s multiple comparisons test). (C) Ifnb1, Cxcl10, Tnf, and Il6 expression in the inguinal LN 6 h following administration of indicated vaccine formulation (mean ± SD; n = 5; statistical significance between formulations vs PBS are shown; *P <0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; one-way ANOVA with Tukey’s multiple comparisons test).

To evaluate the distribution of the NP vaccine components to the injection site LN, vaccine formulations containing fluorescently labeled OVA and polyIC were administered subcutaneously to allow for monitoring of carrier and cargo distribution to the draining LN – the inguinal LN (Figure 3B). Intravital imaging of inguinal LNs isolated 48 h after injection demonstrated that loading of antigen and polyIC onto NPs significantly increased their accumulation within the inguinal LN compared to soluble antigen and adjuvant. A slight increase in OVA accumulation was observed when mixed with NP/polyIC, potentially a result of weak non-covalent associations with OVA that affected its distribution behavior. Additionally, a significant increase in the expression of Ifnb1, Cxcl10, and Tnfa in the inguinal LN was observed 6 hours after administration, further demonstrating the ability of NP vaccine to enhance polyIC activity and delivery to the draining LN (Figure 3C). Interestingly, and consistent with in vitro data, free polyIC stimulated significantly more Il6 expression than when complexed with the NP, further highlighting the potential to tune the cytokine profile elicited by polyIC via control of nanocarrier properties, with implications for improving vaccine efficacy as well as safety.⁶⁹ Collectively, these data demonstrate the ability of the NP vaccine platform to modulate the local pharmacokinetics of a protein antigen and polyIC and thereby enhance their delivery to vaccine site draining LNs.

3.4. Co-delivery of antigen and polyIC with endosomolytic NPs enhances the magnitude and functionality of the CD8⁺ T cell response

We next evaluated the capacity of nanovaccines to enhance CD8⁺ T cell responses to the H-2K^b-restricted OVA epitope, SIINFEKL (pOVA), after vaccination. Mice were administered OVA-NP/polyIC, OVA + NP/polyIC, OVA-NP, OVA + polyIC, OVA, or PBS (vehicle), and boosted on days 7 and 14 (Figure 4A). On day 20, pOVA/H-2K^b tetramer staining was used to monitor the magnitude of the SIINFEKL-specific CD8⁺ T cell response in peripheral blood (Figure 4B). OVA-NP/polyIC generated the highest antigen-specific CD8⁺ T cell response of all of formulations tested, resulting in ~8% SIINFEKL-specific CD8⁺ T cells in the blood. By contrast, free OVA and a soluble mixture of OVA and polyIC elicited responses undetectable beyond background. Similar to in vitro findings, a modest increase in the percentage of pOVA/H-2K^b tetramer-positive CD8⁺ T cells was observed for free OVA mixed with NPs complexed with polyIC. This likely reflects both the enhanced activity of polyIC when delivered using NPs as well as the slight increase in OVA accumulation associated with this formulation.

Figure 4: Dual-delivery of antigen and polyIC with NPs enhances the magnitude and functionality of CD8⁺ T cell response.

(A) Administration and analysis scheme for mice vaccinated with OVA-NP/polyIC or indicated formulations containing OVA protein. (B) Quantification of the frequency of SIINFEKL-specific CD8⁺ T cells in peripheral blood via peptide/MHC tetramer staining (mean ± SEM; n = 8 mice/group; statistical significance between OVA-NP/polyIC and all other formulations are shown; ***P < 0.001, ****P < 0.0001; two-way ANOVA with Dunnett’s multiple comparison test). (C) Quantification of the frequency of SIINFEKL-specific CD8⁺ T cells in the spleen via peptide/MHC tetramer staining (mean ± SEM; n = 8 mice/group; statistical significance between OVA-NP/polyIC and all other formulations are shown; *P <0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; two-way ANOVA with Dunnett’s multiple comparison test). (D) ICCS was used to determine the percentage of CD8⁺ T cells positive for IFNγ and/or TNFα after ex vivo restimulation of splenocytes with SIINFEKL peptide (mean ± SD; n = 8 mice/group; statistical significance between OVA-NP/polyIC and all other formulations are shown; ****P < 0.0001; two-way ANOVA with Tukey’s multiple comparison test). (E) Representative images of ELISPOT and quantification of CD8⁺ IFN-γ⁺ T cell response after ex vivo restimulation of splenocytes with SIINFEKL peptide (mean ± SEM; n = 8 mice/group; statistical significance between OVA-NP/polyIC and all other formulations are shown; *P <0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; two-way ANOVA with Tukey’s multiple comparison test). (F) ICCS was used to determine the percentage of CD4⁺ T cells positive for IFNγ and/or TNFα after ex vivo restimulation of splenocytes with ISQAVHAAHAEINEAGR peptide (mean ± SD; n = 8 mice/group; statistical significance between OVA-NP/polyIC and all other formulations are shown; ****P < 0.0001; two-way ANOVA with Tukey’s multiple comparison test). (G) Serum IgG antibody titer at day 21 as measured by ELISA (mean ± SEM; n = 8 mice/group; statistical significance between OVA-NP/polyIC and all other formulations are shown; *P <0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; one-way ANOVA with Tukey’s multiple comparison test).

On day 21, mice were euthanized and spleens were harvested to track T cell responses with pOVA-H-2K^b tetramer, ICCS, and ELISpot assay. In evaluating the magnitude of SIINFEKL-specific CD8⁺ T cell response in the spleen by pOVA/H-2K^b tetramer staining, we saw similar results as in the peripheral blood, where OVA-NP/polyIC generated the highest antigen-specific CD8⁺ T cell response of all the formulations tested, resulting in ~6% SIINFEKL-specific CD8⁺ T cells (Figure 4C). The functionality of the OVA-specific T cell response was also evaluated via peptide restimulation of splenocytes followed by ICCS for TNFα and IFNγ (Figure 4D). Vaccination with OVA-NP/polyIC increased the frequency of TNFα⁺IFNγ⁺ polyfunctional antigen-specific CD8⁺ T cells to a greater degree relative to all other formulations tested. A mixture of free OVA and NP complexed with polyIC (OVA+NP/polyIC) also resulted in an increase in the percentage of polyfunctional OVA-specific CD8⁺ T cells compared OVA + polyIC (P=0.055), consistent with the strong enhancement in immunostimulatory potency achieved via delivery of polyIC with endosomolytic NPs. We further evaluated the frequency of cytokine-secreting CD8⁺ T cells by IFNγ ELISpot assay after restimulation of splenocytes with the peptide SIINFEKL. Again, we found that OVA-NP/polyIC enhanced the number of IFNγ+ secreting antigen-specific CD8⁺ T cells relative to other groups (Figure 4E). The improved adjuvant effects of NP/polyIC was further demonstrated via restimulation of splenocytes with the MHC-II H-2A^b-restricted OVA epitope ISQAVHAAHAEINEAGR followed by ICCS analysis, which demonstrated that OVA-NP/polyIC and OVA + NP/polyIC similarly enhanced the frequency of IFNγ⁺TNFα⁺ CD4⁺ T cells (i.e., helper type 1 CD4⁺ T cell; Th1) compared to all other formulations (Figure 4F). This is also consistent with the ability of NP/polyIC complexes to increase total IgG endpoint antibody titer against OVA (Figure 4G). Hence, while a distinctive feature of endosomolytic NPs is their capacity to promote MHC-I presentation to enhance CD8⁺ T cell responses, these data also corroborate our previous findings that they are also able to augment CD4⁺ T cell responses, particularly when co-loaded with a Th1-directing adjuvant.^25,56 The capacity of the platform to promote a balanced CD8⁺/CD4⁺ Th1 response may also further enhance cancer vaccine efficacy given the important role that CD4⁺ T cells play in supporting CD8⁺ T cell effector function as well as the intrinsic roles of CD4⁺ T cells in antitumor immunity.^77,78 Together, these data demonstrate that the adjuvant effects of polyIC, which is a relatively weak adjuvant when administered alone, are strongly augmented when delivered into the cytosol using endosomolytic polymer NPs, and that the CD8⁺ T cell response can be further enhanced by dual-delivery of antigen and polyIC formulated together using the NP platform.

3.5. Nanovaccine protects from tumor formation and inhibits tumor growth in mice

As a functional validation of the T cell response elicited by vaccination, we next evaluated the ability of NP vaccines to protect against tumor growth following challenge with a murine thymoma EL-4 cell line that expresses OVA as a model antigen (EG7.OVA). Mice were vaccinated subcutaneously and boosted on days 7 and 14 (Figure 5A). On day 23, mice were challenged with a contralateral subcutaneous inoculation of EG7.OVA cells, and tumor growth and survival were measured (Figure 5B–D). Consistent with the increased magnitude and polyfunctionality of the CD8⁺ T cell response, vaccination with OVA-NP/polyIC conferred the greatest degree of protection from tumor growth. The OVA + NP/polyIC formulation also afforded some protection from tumor formation. It is also notable that OVA-NP conjugates (i.e., in the absence of polyIC) also inhibited tumor growth to a limited degree but to a level similar to a mixture of OVA + polyIC, a finding consistent with our previous work demonstrating that conjugation of antigens to endosomolytic NPs can enhance cellular immunity even in the absence of an additional adjuvant.^27,73

Figure 5: Dual-delivery of antigen and polyIC protects from tumor formation and inhibits tumor growth in a mouse tumor model.

(A) Administration, analysis, and tumor challenge scheme for mice immunized with OVA-NP/polyIC or indicated formulations containing OVA protein. (B) Spider plots of individual tumor growth curves, with the numbers of mice exhibiting complete responses (CR) denoted. (C) Average tumor volume following challenge with EG7.OVA cells of mice immunized with indicated vaccine formulations (mean ± SEM; n = 8 mice/group; P=0.052 for OVA-NP/polyIC vs. OVA + NP/polyIC on day 31 via unpaired t test). (D) Kaplan-Meier survival curves of mice growing EG7.OVA tumors treated with indicated formulations using 1500 mm³ tumor volume as the end point (n = 8 mice/group; statistical significance between OVA-NP/polyIC and all other formulations are shown; *P <0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; Mantel-Cox log-rank test; statistical significance between PBS and all other formulations are shown; +P <0.05, ++P < 0.01, +++P < 0.001, ++++P < 0.0001; Mantel-Cox log-rank test). (E) Tumor inoculation and therapeutic vaccination scheme for mice immunized with OVA-NP/polyIC or indicated formulations containing OVA protein. (F) Spider plots of individual tumor growth curves, with the numbers of complete responders (CR) denoted. (G) Average EG7.OVA tumor volume in response to indicated treatments (mean ± SEM; n = 8–10 mice/group; *P <0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; unpaired t test of OVA-NP/polyIC vs. all other formulations on day on day 17). (H) Kaplan-Meier survival curves of mice growing EG7.OVA tumors treated with indicated formulations using 2000 mm³ tumor volume as the end point (n=8–10 mice/group; statistical significance between OVA-NP/polyIC and all other formulations are shown; *P <0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; Mantel-Cox log-rank test; statistical significance between PBS and all other formulations are shown; +P <0.05, ++P < 0.01, +++P < 0.001, ++++P < 0.0001; Mantel-Cox log-rank test).

Finally, we determined whether endosomolytic NPs improve the efficacy of therapeutic cancer vaccines in proof-of-concept experiments. Mice were treated beginning 5 days after subcutaneous inoculation of EG7.OVA cells and were boosted on days 12 and 19, and tumor growth was monitored (Figure 5E). Therapeutic vaccination with the OVA-NP/polyIC formulation significantly inhibited tumor growth and extended mean survival time relative to mice vaccinated with OVA only, a mixture of polyIC and OVA, or OVA-NP conjugates, which had an insignificant effect on tumor growth (Figure 5F–H). In accord with our other studies, immunization with OVA + NP/polyIC also inhibited tumor growth, though to a slightly lesser degree than when OVA was also covalently bound to the carrier.

While these results are promising, future studies are necessary to validate the capacity of this nanovaccine platform to enhance the immunogenicity of bone fide tumor antigens, many of which may be less immunogenic than OVA. We anticipate that the platform would be directly amenable to covalent loading of tumor peptide antigens containing cysteine residues, though antigenic sequences can also be cloned into OVA or similar proteins (e.g., albumin),^79,80 a strategy that would allow for direct integration of diverse epitopes using the same fabrication and assembly process used here. Additionally, this cancer vaccine technology also remains to be tested in more aggressive and poorly immunogenic tumor models (e.g., B16 melanoma or MOC2 oral squamous cell carcinoma) and evaluated in combination with immune checkpoint blockade or other adjunctive therapies that address established barriers to T cell function and tumor infiltration.^81–83 Nonetheless, these studies validate proof-of-principle for a new cancer vaccine platform that significantly augments antitumor cellular immunity via cytosolic dual-delivery of antigen and adjuvant using the clinically advanced nucleic acid adjuvant, polyIC.

4. Conclusion

Cancer vaccines offer a promising strategy for bolstering the magnitude, breadth, and quality of the tumor antigen-specific T cell response and have emerged as important components of an expanding immunotherapeutic armamentarium. While there is clinical evidence that protein- and peptide-based cancer vaccines generate immune responses in patients, their clinical efficacy remains limited by an insufficient capacity to generate robust cytotoxic CD8⁺ T cell responses, the primary mediator of antitumor immunity in most cancers. To meet this challenge, we describe herein a new nanoparticle cancer vaccine platform for enhancing antitumor CD8⁺ T cell responses via cytosolic dual-delivery of antigen and polyIC – a clinically advanced nucleic acid vaccine adjuvant. We utilized RAFT polymerization to synthesize a well-defined, multifunctional NP platform that enables co-loading of antigens and nucleic acid adjuvants and also harnesses a potent endosomal escape mechanism to facilitate cargo delivery into the cytosol. Cytosolic delivery of antigen resulted in increased antigen presentation on MHC-I molecules, while endosomal escape of polyIC strongly increased its immunostimulatory potency via the MDA-5 pathway, resulting in proinflammatory cytokine production, costimulatory molecule upregulation, and IFN-I secretion by dendritic cells. Additionally, owing to its nanoscale dimensions, the NP vaccine platform extended the injection site half-life of protein antigen and polyIC and increased their accumulation in vaccine site draining LNs. By overcoming these intracellular and physiological delivery barriers, NPs co-loaded with antigen and polyIC stimulated a strong, multifunctional CD8⁺ T cell response that conferred protection from tumor formation and inhibited the growth of established tumors in a mouse tumor model. In summary, the NP vaccine described herein is a promising and versatile platform for cytosolic dual-delivery of antigen and nucleic acid adjuvants that potently augments antitumor cellular immunity.