Multiscale dynamic immunomodulation by a nanoemulsified Trojan-TLR7/8 adjuvant for robust protection against heterologous pandemic and endemic viruses

Abstract

The demand for safe vaccines that ensure long-term and broad protection against multiple viral variants has dramatically increased after the emergence of catastrophic infectious diseases such as COVID-19. To ensure long-term and broad protection against heterologous virus variants, antigen-specific polyfunctional T cells should be orchestrated with the activation of follicular helper T (T_FH) cells and germinal center (GC) B cells. Herein, we suggest a novel engineered nanoadjuvant (SE(Trojan-TLR7/8a)) that enhances the migration of nonexhausted antigen-presenting cells (APCs) into lymph nodes and elicits the activation of T_FH cells, the generation of GC B cells, and polyfunctional T cells via multiscale dynamic immunomodulation through squalene nanoemulsion (SE)-mediated macroscopic control of vaccine delivery and Trojan-TLR7/8a-enabled dynamic and sustained activation of APCs at the cellular level. SE(Trojan-TLR7/8a) can be lyophilized, reduce systemic toxicity, and outperform current commercial vaccine adjuvants (Alum or AS03) and mRNA vaccines. SE(Trojan-TLR7/8a) ensures cross-protection against diverse influenza and SARS-CoV-2 variants, providing 100% protection while maintaining a healthy state. SE(Trojan-TLR7/8a) also sustains a potent T-cell response in an aged ferret model of SFTSV infection. SE(Trojan-TLR7/8a) suggested herein provides a novel vaccine design principle for dynamic modulation at the multiscale level and demonstrates long-term and broad protective immunity against emerging pandemic and endemic infectious viruses.

Keywords: Dynamic immunomodulation; Heterologous viruses; Nanoemulsion; TLR7/8 agonist; Vaccine adjuvant.

Fig. 1

Schematic of multiscale dynamic immunomodulation by a lyophilizable squalene-based nanoemulsion loaded with a Trojan-TLR7/8 agonist (SE(Trojan-TLR7/8a)) to enhance broad and sustained protection. A Synthesis process of SE(Trojan-TLR7/8a). B Mechanism of action in promoting sustained cellular and humoral responses

Fig. 2

Fabrication and characteristics of SE(Trojan-TLR7/8a). Hydrodynamic size distribution (A) and representative transmission electron microscopy image (B) of SE(Trojan-TLR7/8a), showing a uniform size and morphology (scale bar, 100 nm). C Zeta potentials of SE and SE(Trojan-TLR7/8a) were measured via dynamic light scattering. D Average size of SE(Trojan-TLR7/8a) stored at 4 °C for over one month, demonstrating storage stability (n = 2). E Morphology of SE(Trojan-TLR7/8a) as a liquid (left), lyophilized (middle), and reconstituted (right) form. F Hydrodynamic size distribution of SE(Trojan-TLR7/8a) before and after lyophilization. G Comparison of IL-12(p70) secretion by SE(Trojan-TLR7/8a) before and after lyophilization (n = 3). H Comparison of antibody titers in serum after intramuscular immunization with SE(Trojan-TLR7/8a) dispersed immediately after freeze-drying and after 30 days of storage (n = 3). I Representative confocal laser scanning microscopy (CLSM) image and the mean fluorescence intensity of SE(Trojan-TLR7/8a) and IL-12(p70) secretion in the absence or presence of dynasore (an endocytosis inhibitor, 40 μM) for 24 h (n = 3). Scale bar, 10 μm. J, K Bone marrow-derived dendritic cells (BMDCs) were treated with OVA or SE (the same volume used as SE(Trojan-TLR7/8a), SE + R848 (3.18 μM) or SE(Trojan-TLR7/8a) (3.18 μM) mixed with OVA (10 μg ml^-1). J IL-12(p70) and IL-6 concentrations were measured via ELISA (n = 3). K Naïve CD4⁺ T cells were cocultured with BMDCs treated for 12 h for 3 days. The ratio of the concentrations of IFN-γ/IL-4 secreted from the coculture supernatants was measured via ELISA, and the percentage of differentiated follicular helper T cells (T_FH cells, PD-1⁺ CXCR5⁺ in CD3⁺ CD4⁺) was measured via flow cytometry (n = 3). L C57BL/6 mice were immunized intramuscularly with SE + R848 (25 μg, 79.5 nmol) or SE(Trojan-TLR7/8a) (72.1 μg, 79.5 nmol), each mixed with OVA (20 μg). Blood was collected at 1, 2, 4, 6, 8, 12, and 24 h following immunization. The concentrations of IL-6 (n = 3 mice per group) and ALT (n = 2 mice per group) in the serum were quantified via ELISA. The data are presented as the mean ± standard deviation (s.d.). In J, K, analysis was performed via one-way ANOVA with Tukey’s multiple comparison test; in H, I, analysis was performed via two-way ANOVA with Tukey’s multiple comparison test; and in C, G, I, analysis was performed via a two-tailed unpaired t test. P values are indicated (n.s. not significant; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001)

Fig. 3

Multiscale dynamic immunomodulation by SE(Trojan-TLR7/8a). A C57BL/6 mice were immunized intramuscularly with soluble DiD, SE(DiD), or SE(Trojan-TLR7/8a)(DiD) mixed with OVA on day 0. Representative whole-body fluorescence in vivo imaging system (IVIS) images and normalized average radiant efficiency at the injection site over time (n = 2 mice per group). B Representative images of H&E-stained muscle tissue showing cell recruitment by SE(Trojan-TLR7/8a) at 24 h. Scale bar: 200 μm. C Proportions of monocytes (Ly6G⁻ Ly6C⁺), neutrophils (Ly6G⁺ Ly6C⁻), CD11c⁺, CD11b⁺, NK cells (CD3⁻ NK1.1⁺), CD8⁺ T cells (CD3⁺ CD8⁺), and CD4⁺ T cells (CD3⁺ CD4⁺) recruited into the muscle at 24 h (n = 3 mice per group). D, E Bone marrow-derived dendritic cells (BMDCs) were treated with OVA, SE (the same volume used as SE(Trojan-TLR7/8a)), SE + R848 (3.18 μM) or SE(Trojan-TLR7/8a) (3.18 μM) mixed with OVA (10 μg ml^-1). D Supernatants were harvested at 6, 12, 24, 36, and 48 h, and the kinetics of IL-12(p70) secretion were measured via ELISA (n = 3). E Secretion of IL-12(p70) by BMDCs was measured after treatment with SE + R848 or SE(Trojan-TLR7/8a) mixed with OVA for 12 h (before washing) and then changed to fresh medium (after washing) (n = 3). Expression of CCR7 in BMDCs treated with the indicated samples for 12 h (F) and assessment of the cell migration capacity compared with that of the control group via a Transwell assay (G) (n = 3). H, L C57BL/6 mice were immunized intramuscularly with OVA (FITC) or OVA alone or in combination with SE (the same volume used as SE(Trojan-TLR7/8a)), SE + R848 (79.5 nmol), or SE(Trojan-TLR7/8a) (79.5 nmol) (n = 3 mice per group). I The total number of migratory DCs (CD103⁺) in the inguinal LNs (iLNs) on days 0.5, 1, and 3 (n = 3 mice per group). J OVA(FITC) fluorescence images and total radiant efficiency of excised iLNs at 0.5, 1, 3, 5, and 7 days postadministration were obtained via IVIS (n = 2). K Synergistic intracellular signaling of TLR7/8a and SE induced by SE(Trojan-TLR7/8a). M The expression levels of 17 genes related to humoral and cellular immunity in the iLNs at 7 days postadministration (n = 3 mice per group). N Immunofluorescence images of sectioned iLNs on day 7 showing the distribution of T cells (CD3, green) and germinal center (GC) B cells (GL7, red) via CLSM observation. Scale bar: 500 μm. Total numbers of CD11b⁺, CD11c⁺ (O), T_FH, GC B (Fas⁺ CD38⁻ in B220⁺), and CD8⁺ T cells (P) in the iLNs measured on days 0.5, 1, 3, 5, 7 and 14 (n = 3 mice per group). Q The total number of antigen-specific CD8⁺ T cells secreting IFN-γ or TNF-α was evaluated on day 7 (n = 3 mice per group). The data are presented as the means ± s.d. In E–G, Q, analysis was performed via one-way ANOVA with Tukey’s multiple comparison test; in C, D, I, J, O, P, analysis was performed via two-way ANOVA with Tukey’s multiple comparison test; P values are indicated (n.s. not significant; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001)

Fig. 4

Compared with conventional vaccines, SE(Trojan-TLR7/8a) elicits a high-magnitude immune response. A–I C57BL/6 mice were immunized intramuscularly with OVA (20 μg) alone or in combination with SE (with the same volume used for SE(Trojan-TLR7/8a)) or SE(Trojan-TLR7/8a) (79.5 nmol) twice at 3-week intervals, after which the iLNs, spleens, and lungs were analyzed according to the same schedule (3‒4 mice per group). B Representative flow cytometry plots and analysis of the percentages of GC B cells, T_FH cells, and memory B cells in the iLNs (n = 4). C Percentages of antigen-specific CD8⁺ T cells secreting IFN-γ or granzyme B (GzB) and CD4⁺ T cells secreting IFN-γ or GzB (n = 3). D Schematic of the systemic responses of the spleen, lung, and blood to SE(Trojan-TLR7/8a). Percentages of GC B cells and T_FH cells (E) (n = 4) and antigen-specific CD8⁺ T cells secreting IFN-γ (F) in splenocytes (n = 3). Percentages of effector memory (CD44⁺ CD62L⁻) CD8⁺ T cells (G) and antigen-specific CD8⁺ T cells secreting IFN-γ (H) in the lungs (n = 3). I Serum samples were collected at 4 weeks, and the titer of OVA-specific IgG was assessed via ELISA (n = 4). J–N C57BL/6 mice were intramuscularly immunized with OVA (20 μg) alone or in combination with alum (200 μg), AS03 (with the same volume used for SE(Trojan-TLR7/8a)), or SE(Trojan-TLR7/8a) (79.5 nmol) twice at 3-week intervals, and then the iLNs and serum were analyzed according to schedule (n = 5 mice per group). J A diagram comparing the effectiveness of SE(Trojan-TLR7/8a) with that of other squalene-based nanoemulsions (SE and AS03) and alum. K, L Representative flow cytometry plots and percentages of GC B and T_FH cells (K), as well as the percentage of memory B cells (L) in the iLNs. M Total numbers of antigen-specific CD8⁺ T cells secreting IFN-γ or TNF-α and CD4⁺ T cells secreting IL-4 in the iLNs. N Serum samples were collected at 4 weeks, and the titer of OVA-specific IgG1 or IgG2c was assessed via ELISA. The data are presented as the means ± s.d. In B, C, E–I, K–N, analysis was performed via one-way ANOVA with Tukey’s multiple comparison test; P values are indicated (n.s. not significant; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001)

Fig. 5

Compared with the mRNA vaccine, SE(Trojan-TLR7/8a) elicits a high-magnitude immune response. A C57BL/6 mice were intramuscularly immunized twice at 3-week intervals, after which the iLNs and serum were analyzed according to the schedule. The experimental groups included a nontreated control group and mice receiving LNP (OVA-mRNA) (OVA-mRNA 5 μg) or SE(Trojan-TLR7/8a) (79.5 nmol) with OVA (20 μg). B Representative flow cytometry plots and percentages of T_FH cells and memory B cells in the iLNs (n = 5 mice per group). C Percentages of GC B cells, B cells (CD19⁺ B220⁺), and FDCs (CD21/35⁺) in iLNs (n = 5 mice per group). D Serum OVA-specific IgG, IgG2c, or IgG1 titers (n = 3 mice per group). Total numbers of antigen-specific CD8⁺ (E) and CD4⁺ (F) T cells secreting IFN-γ, GzB, TNF-α, or IL-4 in the iLNs (n = 5 mice per group). G–I C57BL/6 mice were intramuscularly immunized with a homologous (LNP (OVA-mRNA)) or heterologous (SE(Trojan-TLR7/8a)) booster, and the iLNs were analyzed according to the schedule (n = 3 mice per group). H Kinetics of GC B cells, T_FH cells_, FDCs, and memory B cells in the iLNs. I Total number of antigen-specific CD8⁺ T cells secreting IFN-γ or GzB in the iLNs at 6 weeks. J–N C57BL/6 mice were intramuscularly immunized three times, and the iLNs were analyzed according to schedule (n = 3 mice per group). Representative flow cytometry plots and percentages of GC B cells (K) and percentages of T_FH cells and FDCs (L) in the iLNs at 12 weeks. M Representative flow cytometry plots and percentages of memory B cells in iLNs. N Percentage of antigen-specific CD8⁺ T cells secreting IFN-γ or GzB. The data are presented as the means ± s.d. In B–F, I, K–N, analysis was performed via one-way ANOVA with Tukey’s multiple comparison test; in H, analysis was performed via two-way ANOVA with Tukey’s multiple comparison test; P values are indicated (n.s., not significant; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001)

Fig. 6

SE(Trojan-TLR7/8a) ensures cross-protection against SARS-CoV-2. A C57BL/6 mice were intramuscularly immunized with HexaPro (prime, 1 μg; boost, 5 μg) alone or combined with SE (with the same volume used for SE(Trojan-TLR7/8a)) or SE(Trojan-TLR7/8a) (79.5 nmol) twice at 3-week intervals, after which the serum was collected according to the same schedule (n = 4 mice per group). Neutralizing activity against SARS-CoV-2 variants was assessed via a microneutralization assay, which involved infecting HEK293T-ACE2 cells with pseudoviruses and analyzing serum samples collected 21 days after the initial immunization. B–H BALB/c mice were administered SARS-CoV-2 spike-stabilized trimer protein (spike protein, 0.5 μg) alone or in combination with SE or SE(Trojan-TLR7/8a) twice at 3-week intervals according to the schedule. C Mice were infected with 100 LD₅₀ of the mouse-adapted SARS-CoV-2 virus, and infectious virus titration was confirmed from lung homogenates of the mice (n = 9 mice per group) at 3, 5, and 7 dpi. D Percentages of T_FH cells (PD-1⁺ IL-21⁺ in CD4⁺ T cells) and GC B cells (GL7⁺ AID⁺ in CD19⁺ cells) among splenocytes at 5 weeks (n = 5 mice per group). E, F A serum neutralization assay was performed against the Omicron BA.2, Wuhan, Alpha, and Delta strains, and the results are presented as the geometric mean titers (n = 15 mice per group). G IFN-γ spot-forming units (SFUs) were enumerated via ELISPOT assays after stimulating the splenocytes with the S peptide pool or inactivated with Omicron BA.2, Wuhan, Alpha, or Delta variant for 72 h. H Polyfunctional antigen-specific CD8⁺ T cells were analyzed by flow cytometry after stimulating the splenocytes with the S peptide pool or inactivated with Omicron BA.2, Alpha, or Delta variant in the presence of monensin for 12 h (n = 5 mice per group). The data are presented as the means ± s.d. In A, D–G, analysis was performed via one-way ANOVA with Tukey’s multiple comparison test; in C, H, analysis was performed via two-way ANOVA with Tukey’s multiple comparison test. P values are indicated (n.s., not significant; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001)

Fig. 7

SE(Trojan-TLR7/8a) ensures cross-protection against influenza. A BALB/c mice were intramuscularly immunized with sM2HA2 (15 μg) alone or in combination with SE (with the same volume used for SE(Trojan-TLR7/8a)) or SE(Trojan-TLR7/8a) (79.5 nmol) twice in weeks 0 and 3 and then challenged with 10 LD₅₀ of the mouse-adapted influenza subtypes one week after the second immunization according to the schedule. The survival rate (upper panel) and changes in body weight (bottom panel) after lethal infection with H1N1 (B), H5N2 (C), H7N3 (D), H9N2 (E), and H3N2 (F) were monitored for 13 days (n = 7 mice per group). G Lung virus titers were determined according to the TCID₅₀ in MDCK cells at 3 and 5 days after H1N1 and H5N2 infection (n = 3 mice per group). H H&E-stained images of lung sections collected 5 days after H1N1 and H5N2 infection. The arrows indicate inflammatory cell infiltration. The data are presented as the means ± s.d. In G, analysis was performed via two-way ANOVA with Tukey’s multiple comparison test. P values are indicated (n.s. not significant; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001)

Fig. 8

SE(Trojan-TLR7/8a) promotes long-term protective immune responses against influenza and SFTSV. A C57BL/6 mice were immunized intramuscularly with OVA (20 μg) alone or in combination with SE(Trojan-TLR7/8a) (79.5 nmol), and then the iLNs and bone marrow were analyzed according to the schedule. B Heatmap comparing the expression levels of genes related to the long-term immune response in independent isolations of LNs (n = 3 mice per group). C Representative flow cytometry plots and percentages demonstrating that SE(Trojan-TLR7/8a) increased the number of plasma cells (CD138⁺ B220⁻) in the bone marrow at 6 weeks postimmunization (n = 4 mice per group). D–G Long-lasting immune response and protective efficacy of SE(Trojan-TLR7/8a)-adjuvanted sM2HA2. BALB/c mice were immunized intramuscularly twice at 0 and 3 weeks and challenged with 10 LD₅₀ of mouse-adapted H1N1 21 weeks after the second immunization according to the schedule. E Serum IgG, IgG1 and IgG2a antibody titers specific to sM2 or HA2 at a 1:100 serum dilution ratio 24 weeks after the first immunization (n = 7 mice per group). F The numbers of sM2HA2-specific IFN-γ SFU and IL-4 SFU were measured via an ELISPOT assay (n = 6 mice per group). G The survival rate (left panel) and percentage changes in initial body weight (right panel) after infection with the H1N1 virus were monitored for 13 days (n = 7 mice per group). H–N Aged ferrets were immunized intramuscularly with inactivated SFTSV alone or combined with SE(Trojan-TLR7/8a) (79.5 nmol) twice at 0 and 2 weeks and challenged with SFTSV 30 weeks after the initial immunization according to the schedule. I Serum neutralizing titers were analyzed at 2 and 4 weeks (n = 6 ferrets per group). J Representative image and numerical value of IFN-γ SFU in PBMCs, analyzed at 30 weeks, confirmed by an ELISPOT assay (n = 2 ferrets per group). K On days 4 and 6 after SFTSV infection, the spleen and PBMCs were collected, and the viral load was quantified via qRT‒PCR (n = 3 ferrets per day per group). L–N Changes in ferret body weight (L), body temperature (M), and platelet count (N) were measured until 6 dpi (n = 6 for days 1–4 and n = 3 for days 5–6). The data are presented as the mean ± s.d. In E, analysis was performed via one-way ANOVA with Tukey’s multiple comparison test; in F, I–K, analysis was performed via two-way ANOVA with Tukey’s multiple comparison test; in C, analysis was performed via a two-tailed unpaired t test. P values are indicated (n.s., not significant; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001)