Adjuvant-carrying synthetic vaccine particles augment the immune response to encapsulated antigen and exhibit strong local immune activation without inducing systemic cytokine release

Abstract

Augmentation of immunogenicity can be achieved by particulate delivery of an antigen and by its co-administration with an adjuvant. However, many adjuvants initiate strong systemic inflammatory reactions in vivo, leading to potential adverse events and safety concerns. We have developed a synthetic vaccine particle (SVP) technology that enables co-encapsulation of antigen with potent adjuvants. We demonstrate that co-delivery of an antigen with a TLR7/8 or TLR9 agonist in synthetic polymer nanoparticles results in a strong augmentation of humoral and cellular immune responses with minimal systemic production of inflammatory cytokines. In contrast, antigen encapsulated into nanoparticles and admixed with free TLR7/8 agonist leads to lower immunogenicity and rapid induction of high levels of inflammatory cytokines in the serum (e.g., TNF-a and IL-6 levels are 50- to 200-fold higher upon injection of free resiquimod (R848) than of nanoparticle-encapsulated R848). Conversely, local immune stimulation as evidenced by cellular infiltration of draining lymph nodes and by intranodal cytokine production was more pronounced and persisted longer when SVP-encapsulated TLR agonists were used. The strong local immune activation achieved using a modular self-assembling nanoparticle platform markedly enhanced immunogenicity and was equally effective whether antigen and adjuvant were co-encapsulated in a single nanoparticle formulation or co-delivered in two separate nanoparticles. Moreover, particle encapsulation enabled the utilization of CpG oligonucleotides with the natural phosphodiester backbone, which are otherwise rapidly hydrolyzed by nucleases in vivo. The use of SVP may enable clinical use of potent TLR agonists as vaccine adjuvants for indications where cellular immunity or robust humoral responses are required.

Keywords: Adjuvant; CpG; R848; Resiquimod; Synthetic nanoparticle vaccine; TLR agonist.

Fig. 1

Adjuvant encapsulation in nanoparticles (SVP) results in suppressed TNFα induction in murine cells in vitro. SVP containing TLR7/8 agonist R848 (A, B) or TLR9 agonists PO-CpG 1826 (C) or PS-CpG1826 (D) were added to J774 cells (A, C) or fresh mouse splenocyte cultures (B, D) in parallel with free adjuvants in triplicates. The amount of TNF-α in culture supernatants was measured 6 h (splenocytes) or 16 h (J774) after incubation. Representative results out of two separate experiments with groups of three mice are shown.

Fig. 2

Nanoparticle encapsulation of both antigen and TLR7/8 agonist (R848), whether co-encapsulated or in separate particles, results in a higher humoral immune response than utilization of free TLR7/8 agonist and/or free antigen. Groups of 5–10 mice were immunized (3 times, 4-week intervals, s.c., hind limb) with various combinations of SVP-OVA or free OVA and/or SVP-R848 or free R848, as indicated. The same dose of OVA (5.6 μg) was used in all experimental groups and the same dose of R848 (6.8 μg) was used in all groups in which R848 was administered. Antibody generation was measured by ELISA at the indicated time-points and is presented as log₁₀ of EC₅₀ with standard deviations.

Fig. 3

Nanoparticle encapsulation of both antigen and TLR7/8 agonist (adjuvant) results in higher local and systemic cellular immune responses than utilization of free antigen or adjuvant. A – total draining lymph node (LN) cell count, B – SIINFEKL-positive CD8⁺ T cells in draining lymph node (absolute numbers – gray bars, left Y-axis; percentage of total CD8⁺ T cells – hatched bars, right Y-axis), C – flow cytometric analysis of SIINFEKL-positive CD8⁺ T cells ex vivo, D – flow cytometric analysis of SIINFEKL-positive CD8⁺ T cells after in vitro expansion, E – in vitro cytotoxic activity of LN-derived cells against OVA-expressing target cells, and F – in vivo cytotoxicity against SIINFEKL-pulsed syngeneic splenocytes. Groups of 3–6 mice were injected (s.c., hind limb) either with SVP-encapsulated or free OVA and/or R848 in combinations indicated. At 4 days after injection, total cells in the popliteal LNs were isolated and counted (A). LN cells were then stained for surface markers either immediately (B, C) or after 4 days of in vitro expansion on IL-2 without feeder/stimulator cells (D). The percentage of antigen-specific CTL stained with SIINFEKL-MHC class I pentamer and anti-CD8 antibody was assessed by flow cytometry (indicated in boxes in C and D; samples from two representative animals shown). In vitro cytotoxic activity was assessed after in vitro expansion of LN cells, as described. The difference between specific (E.G7-OVA transfected targets cells) and non-specific (EL-4 parental cells) cytotoxicity is shown (E). The amount of free or SVP-encapsulated OVA and R848 in μg/injection dose is indicated in parentheses. The effector-to-target ratio (E:T ratio) is depicted in the x-axis. In vivo CTL activity (F) was assessed in mice (n = 2–4/group) injected s.c. with SVP-encapsulated or free OVA either alone or with free or SVP-encapsulated R848, as indicated. The total dose of OVA was the same across all treatment groups, and the total dose of R848 was the same for all groups receiving R848. Specific in vivo cytotoxicity was determined as described in Section 2 at 6 days post-immunization. All experiments presented in this figure were run 2–3 times.

Fig. 4

Local cytokines are induced by SVP-encapsulated, but not free, TLR7/8 agonist R848. Mice were injected s.c. in both hind limbs with either SVP-OVA-R848, SVP-OVA, or SVP-OVA admixed with free R848. At the times indicated, popliteal LNs were taken and incubated in vitro overnight. Culture supernatants were collected and analyzed for the presence of specific cytokines by ELISA. A – IFN-γ, B – RANTES, C – IL-12(p40), D – IL-1β. Average of four LNs per time-point is shown.

Fig. 5

Focused local cytokine induction after a single-site injection with SVP-encapsulated, but not free, TLR7/8 agonist R848. Mice were injected s.c. in a single hind limb with either SVP-R848 or free R848. At the times indicated, popliteal LNs from the injection side (ipsilateral, I) and from the opposite side (contralateral, C) were collected and incubated in vitro overnight. Culture supernatants were analyzed for the presence of specific cytokines by ELISA. A – IFN-γ, B – IL-12(p40), C – IL-1β. Average of two mice per time-point is shown.

Fig. 6

Strong systemic cytokine induction after injection of free, but not SVP-encapsulated TLR7/8 agonist R848. Mice were injected s.c. with either SVP-OVA-R848, SVP-OVA or SVP-OVA admixed with free R848 and bled at times indicated. The same amount of free or SVP-encapsulated R848 was used for each group (6.8 μg). Serum samples were collected at the times indicated and analyzed for individual cytokines by ELISA. Average cytokine concentration is shown (three samples per group per each time-point). A – TNF-α, B – IL-6, C – RANTES, D – IP-10, E – IL-12(p40), F – MCP-1.

Fig. 7

Systemic cytokine induction after intranasal inoculation of free, but not SVP-encapsulated TLR7/8 agonist R848. Mice were inoculated intranasally with either SVPOVA-R848, SVP-OVA, or SVP-OVA admixed with free R848 and bled at times indicated. The same amount of free or SVP-encapsulated R848 was used for each group (3.8 μg). Serum samples were collected at the times indicated and analyzed for individual cytokines by ELISA. Average cytokine concentration is shown (three samples per group per each time-point). A – TNF-α, B – IL-6, C – IL-12(p40), D – MCP-1.

Fig. 8

Higher humoral immunogenicity of SVP-encapsulated antigen coupled with free TLR9 agonist PS-CpG or with SVP-encapsulated TLR9 agonist PO-CpG. Mice (5/group) were immunized (s.c., hind limb) with the indicated SVP or combinations of free antigen and PS-CpG 1826. Antibody titers were determined at times indicated. Mice were immunized either two (A, C) or three (B) times with 2-week intervals between immunizations. OVA (A, B) and prostatic acid phosphatase (PAP) (C) proteins were evaluated as test antigens, either in free form or encapsulated in SVP. In Panel B, the total dose of both free OVA and free PS-CpG 1826 was five times greater than that administered in the SVP-treated groups. In Panel C, the amount of free PAP used was 50 times higher than that of SVP-encapsulated PAP, and the amount of free PS-1826 was two times higher than used in SVP-encapsulated form (C). No adjuvant effect of free PO-1826 was detected (data not shown).

Fig. 9

Nanoparticle encapsulation of both antigen and TLR9 agonist PO-CpG (adjuvant) results in a higher local cellular immune response than utilization of free antigen and adjuvant. Groups of 2–3 mice were injected (s.c., hind limb, once or twice with 2-wk interval) with a combination of SVP-encapsulated OVA and SVP-encapsulated PO-CpG 1826, free OVA admixed with free PS-CpG 1826, or PBS as indicated. The amount of free OVA (50 μg) was ten times higher and the amount of free PS-CpG 1826 (20 μg) was five times higher than that present within SVP (5 and 4 μg, respectively). Tissues were taken at 5 days after the last immunization and restimulated in vitro for 7 days with mitomycin-treated E.G7-OVA cells. LN cells were stained with surface marker antibodies and analyzed by flow cytometry. Top row – popliteal LN cultures (initiated 5 days after a single immunization), bottom row – splenocyte cultures (initiated 5 days after prime-boost immunization). Representative flow cytometry dot plots from two separate experiments are shown. The numbers indicate the percentage of CD8⁺ T cells that are specific for the OVA SIINFEKL peptide (percent share of CD8⁺CD19⁻ cells).

Fig. 10

Focused local cytokine induction by SVP-encapsulated, but not free TLR9 agonist CpG. Mice were injected s.c. in a single hind limb with either SVP-PS-CpG 1826 or free PS-CpG 1826. At the times indicated, popliteal LNs from the injection side (ipsilateral, I) and from the opposite side (contralateral, C) were taken and incubated in vitro overnight. Culture supernatants were collected and analyzed for cytokine presence by ELISA. A – IFN-γ, B – IL-12(p40), C – IL-1β. Average of two mice per group per time-point is shown.