Reprogramming the adjuvant properties of aluminum oxyhydroxide with nanoparticle technology

Abstract

Aluminum salts, developed almost a century ago, remain the most commonly used adjuvant for licensed human vaccines. Compared to more recently developed vaccine adjuvants, aluminum adjuvants such as Alhydrogel are heterogeneous in nature, consisting of 1-10 micrometer-sized aggregates of nanoparticle aluminum oxyhydroxide fibers. To determine whether the particle size and aggregated state of aluminum oxyhydroxide affects its adjuvant activity, we developed a scalable, top-down process to produce stable nanoparticles (nanoalum) from the clinical adjuvant Alhydrogel by including poly(acrylic acid) (PAA) polymer as a stabilizing agent. Surprisingly, the PAA:nanoalum adjuvant elicited a robust TH1 immune response characterized by antigen-specific CD4⁺ T cells expressing IFN-γ and TNF, as well as high IgG2 titers, whereas the parent Alhydrogel and PAA elicited modest TH2 immunity characterized by IgG1 antibodies. ASC, NLRP3 and the IL-18R were all essential for TH1 induction, indicating an essential role of the inflammasome in this adjuvant's activity. Compared to microparticle Alhydrogel this nanoalum adjuvant provided superior immunogenicity and increased protective efficacy against lethal influenza challenge. Therefore PAA:nanoalum represents a new class of alum adjuvant that preferentially enhances TH1 immunity to vaccine antigens. This adjuvant may be widely beneficial to vaccines for which TH1 immunity is important, including tuberculosis, pertussis, and malaria.

Fig. 1

Top down schematic for producing stable nanoalum particles from Alhydrogel

Fig. 2

Addition of PAA as a stabilizer produces stable nanoalum particles. a The particle size of Alhydrogel, Alhydrogel mixed with PAA prior to sonication and microfluidization, and PAA:nanoalum were determined by dynamic light scattering (DLS) or laser diffraction. b Increasing the amount of PAA at the start of the process or increasing the number of passes through the microfluidizer results in smaller PAA:nanoalum particles. c PAA:nanoalum retains the original particle size for at least 3 months at 5, 25, or 37 °C as determined by DLS. d Cryo-TEM images of Alhydrogel on the left and PAA:nanoalum on the right reveal a highly aggregated and mono-dispersed state, respectively. Scale bars, 100 nm. c Cryo-TEM images of Alhydrogel on the left and PAA:nanoalum on the right reveal a highly aggregated and mono-dispersed state, respectively. d PAA:nanoalum retains the original particle size for at least 3 months at 5, 25, or 37 °C as determined by DLS. e Unlike Alhydrogel, PAA:nanoalum particle size is stable for at least three free-thaw cycles. Data are representative of three experiments with three replicates each. The means and standard deviations are shown

Fig. 3

PAA:nanoalum promotes TH1 immunity without exogenous TLR4 agonistsl. C57Bl/6 mice (five per group) were immunized twice with 0.5 µg of ID93 alone or adjuvanted as indicated. One week after a homologous booster immunization the frequency of antigen-specific CD4 T cells in the spleen was determined by intracellular cytokine staining after stimulation with the immunizing antigen. a A representative FACS plot is shown for antigen stimulated cells from the Alhydrogel and PAA:nanoalum animal (left and right, respectively). b Distribution of CD154, IFN-ƴ and TNF producing CD4 T cells. Representative FACS gating is shown in Supplementary Fig. 3A. N = 5 animals per group. Data are representative of three experiments with similar results. Bars are drawn to the means with whiskers to the standard deviations

Fig. 4

Augmentation of TH1 immunity is not a general property of nanoalum formulations or free PAA. a ID93 was mixed with Alhydrogel, PAA, or PAA:nanoalum for detected by silver stain before and after centrifugation to pellet the alum particles. Lane 1: ladder, 2: ID93, 3: ID93 + Alhydrogel, 4: ID93 + PAA:nanoalum 5: ID93 + PEG:nanoalum. The full gel is shown in Supplementary Fig. 2. b–d C57Bl/6 mice (five per group) were immunized twice with 0.5 µg of ID93 alone or adjuvanted with Alhydrogel, PAA, PAA:nanoalum, or PEG:nanoalum. b One week after the second immunization, splenocytes were isolated and either unstimulated or stimulated with the ID93 protein in the presence of brefeldin A for 8 h at 37 °C. Cells were then stained for surface expression of CD4 and CD44, as well as intracellular expression of IFN-ƴ, TNF and/or CD154. c Splenocytes from immunized mice were stimulated with the ID93 protein and assessed for secretion of IFN-ƴ. d Serum was collected from immunized animals 3 weeks after the first immunization and assessed for ID93 binding antibody titers by ELISA for IgG1 and IgG2 subclasses. ***(P < 0.001), ****P < 0.0001 relative to the antigen alone group as determined by one-way ANOVA with Dunnett’s correction for multiple comparisons. Representative FACS gating is shown in Supplementary Fig. 3A. Data are representative of three experiments with similar results. a and d Bars are drawn to the means with whiskers to the standard deviations. c The line is drawn at the mean, the box at the interquartile range and the whiskers at the min and max

Fig. 5

PAA:Nanoalum creates an immunocompetent lymph node environment and activates the inflammasome. a The amount of IL-18, IL-12p70 and IFN-γ in the draining lymph node 6 h after immunization with saline, Alhydrogel or PAA:nanoalum. b At the same time, the numbers of neutrophils and subcapsular macrophages were enumerated and intracellular IL-1β and IFN-γ expression were determined by ICS. c Antigen-specific CD4 T cells were determined in wild-type (WT), ASC-/-, NLRP3-/-, or IL-18R-/- mice 1 week after the boosting immunization with either antigen alone or antigen adjuvanted with PAA:nanoalum using ICS. Representative FACS gatings are shown in Supplementary Fig. 3A, B. N = 5 mice per group. The data are representative of two experiments with similar results. Bars are drawn to the means with whiskers to the standard deviations

Fig. 6

Nanoalum significantly augments the immunogenicity and protective efficacy of a flu vaccine. C57Bl/6 mice were immunized with recombinant H1 antigen alone or adjuvanted as indicated. Three weeks after the immunization serum antibody titers a and HAI titers b. One week later the animals (N = 10/group) were intranasally challenged with 25LD50 of the PR8 strain of H1N1 influenza. Viral titers in the lungs were determined at day 4 after challenge (N = 5/group) c. The remaining 5/group were monitored for weight loss d and mortality e. a, b, and c *, **, ***, and **** indicate P < 0.05, 0.01, 0.001, and 0.0001 as determined by one-way ANOVA with Dunnett’s correction for multiple comparisons, compared to the rHA alone control, respectively, unless the comparison is otherwise indicated. e **indicates P < 0.01 as determined by Log-rank (Mantel-Cox) test compared to rHA alone. Data are representative of two individual experiments. The a, c, and d means and standard deviations or b geometric means are shown