Manipulating the delivery and immunogenicity of DNA vaccines through the addition of CB[8] to cationic polymers

Abstract

Challenges with vaccine reactogenicity, stability, and access have highlighted the need to develop alternative strategies for formulation and delivery. We explored the incorporation of cucurbit[n]urils (CBs), as supramolecular "hosts," into nucleic acid-polymer polyplexes. CBs are small, non-toxic, barrel-shaped molecules that transiently crosslink polymers containing supramolecular "guests," thereby increasing molecular weight (MW) of the complex, a correlate of transfection efficiency. We tested whether the supramolecular interactions of CB[8] impact polyplex function. We generated a library of different CB[8] polyplexes using plasmid DNA (pDNA), varying N/P (the ratio of polymer to plasmid), the length, and guest (phenylalanine [Phe]) group frequency of the polyethylenimine (PEI) polymer backbone. We found that N/P 32 and the 20Phe1 (20kDa PEI with 1 mol% Phe) gave optimal gene expression and that incorporating CB[8] in polyplex formulations improved gene expression, both in vitro and in vivo. Despite increases in gene expression, inclusion of CB[8] in formulations with higher guest-binding capacity led to decreased immunogenicity, possibly as a result of dampened innate immune responses. Our data show that CB[8] polyplexes increase gene delivery and expression but alter inflammatory responses. These findings highlight that rational design of the CB[8] polymer system can enable nucleic acid delivery for both vaccine and therapeutic applications.

Keywords: DNA vaccine; MT: Delivery Strategies; cucurbituril; nucleic acid delivery; polymeric nanoparticles; polyplex.

Graphical abstract

Figure 1

Altering polymer size and guest group percentage affects expression level when combined with CB[8] CB repeat monomer (A), CB[8] structure indicating cavity (B), schematic of interaction between polymers and CB[8] (C). Table indicating variables for design of experiments study (D). Quantification of GFP expression by cell percentage (%) 24 h post-transfection with a library of polyplexes formulated with varying MW, guest-loading capability polymers, and N/P ratios (E). Particle diameter and PdI of polyplexes formed from different CB[8] polymer systems at N/P ratio 32:1 (F). Zeta potential of polyplexes (G). Correlation of characterization data measured against GFP expression (H). Bars represent mean ± SEM; statistical analysis performed by ANOVA with Tukey’s test. ∗∗p < 0.01; ∗∗∗p < 0.001.

Figure 2

Increasing N/P ratio of CB[8] polyplexes improves transfection efficiency of plasmid DNA in vitro and immunogenicity in vivo Particle diameter (A) and zeta potential (B) were assessed of polyplexes formed using the CB[8] polymer system and pDNA at N/P ratios ranging from 3:1 to 48:1. Diameter of 20Phe1 CB[8] polyplexes and different N:P ratios measured by DLS (C). GFP expression was assessed by flow cytometry, presented as % cell, 24 h after transfection (D). Toxicity of polyplexes on cells (E). BALB/c mice were immunized IM with 5 μg of HA pDNA in a prime-boost regime with a 3-week interval between immunizations. HA-specific IgG determined by ELISA from blood collected 3 weeks after both prime and boost (F). Three weeks after immunization, mice were intranasally infected with A/California/07/2009 (H1N1); weight loss (G); viral load (H); and cell recovery (I) after infection. (A, B, D, E) Error bars represent mean ± SEM for n = 3. (F–I) n = 5; symbols represent individual mice; error bars represent mean ± SEM. Statistical analysis was performed by ANOVA with Tukey’s test. ∗∗p < 0.01; ∗∗∗p < 0.001.

Figure 3

CB[8] increases GFP expression in vitro but decreases immunogenicity in vivo TEM of formulated pDNA with and without CB[8] (A). pDNA was formulated at N/P ratio of 32 with various combinations of PEI +/− CB[8], prior to transfection into cells or in vivo studies. Quantification of GFP expression by cell percentage (%) 24 h post-transfection of HEK293T (B). BALB/c mice were immunized IM with 5 μg of HA pDNA in a prime-boost regime with a 3-week interval between immunizations. HA-specific IgG determined by ELISA from blood collected 3 weeks after both prime and boost (C). Three weeks after immunization, mice were intranasally infected with A/California/07/2009 (H1N1); weight loss over time course (D) and at d5 (E); viral load (F) and HA-specific CD8 after infection (G). Error bars represent mean SEM for n = 3 (B); n = 5 (C–G); symbols represent individual mice; error bars represent mean ± SEM. Statistical analysis was performed by ANOVA with Tukey’s test. ∗∗p < 0.01; ∗∗∗p < 0.001.

Figure 4

CB[8] dampens innate immune responses in THP-1 cells Quantification of fLuc expression in relative light units (RLU) in THP1 cells 48 h after transfection with 1 μg of pDNA polyplexes +/− CB[8] (N/P 32) (A). n = 3, where symbols represent technical replicates; error bars represent SEM. Cytokine responses assessed by MSD from supernatant of THP1 transfection at 48 h; each point represents mean pg/mL (B). Statistical analysis was performed by ANOVA with Šídák test as a paired comparison between +/− CB[8]. ∗∗∗p < 0.001 between 10Phe2 +/− CB[8].

Figure 5

CB[8] increases gene expression in vivo but reduces innate cell recruitment Cytokine responses assessed by MSD from mouse sera 24 h after IM injection with 5 μg of pDNA polyplexes +/− CB[8] (N/P 32) (A); each point represents mean pg/mL. n = 5; statistical analysis was performed by ANOVA with Šídák test as a paired comparison between +/− CB[8]. Cell counts of muscle tissue collected 24 h after administration of 20 μg of polyplex formulated GFP pDNA (B). Live muscle cells expressing GFP (C), muscle neutrophils (D), and muscle macrophages (E). Cell counts of lymph nodes collected 24 h after administration of 20 μg of polyplex-formulated GFP pDNA (F); lymph node neutrophils (G); lymph node macrophages (H); and lymph node cDC2 (I). n = 5, where symbols represent individual mice; error bars represent SEM. ∗p < 0.05 ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 as indicated. Statistical analysis was performed by ANOVA with Šídák test as a paired comparison between +/− CB[8].