Vaccine Adjuvants Differentially Affect Kinetics of Antibody and Germinal Center Responses

Abstract

Aluminum salts and squalene based oil-in-water emulsions (SE) are widely used adjuvants in licensed vaccines, yet their mechanisms are not fully known. Here we report that induction of antibody responses displays different kinetics dependent on the adjuvant used. SE facilitated a rapid antibody response in contrast to aluminum hydroxide (AH) and the depot-forming cationic liposome-based adjuvant (CAF01). Antigen given with the SE adjuvant rapidly reached follicular B cells in the draining lymph node, whereas antigen formulated in AH or CAF01 remained at the site of injection as a depot. Removal of the injection site early after immunization abrogated antibody responses only when antigen was given in the depot-forming adjuvants. Despite initial delays in B cell activation and germinal center (GC) formation when antigen was given in depot-forming adjuvants, the antibody levels reached higher magnitudes than when the antigen was formulated in SE. This study demonstrates that the kinetic aspect of antibody responses is critical in adjuvant benchmarking and suggests that the optimal vaccination regime depends on the adjuvant used.

Keywords: CAF01; adjuvant; alum; antibody; germinal center (GC); kinetics; squalene emulsion; vaccine.

Figure 1

Distinct kinetics of antibody responses dependent on the adjuvant. Mice were vaccinated subcutaneously with 5 μg of the recombinant protein antigen (A,B) CTH522 or (C) H56 either alone or in the presence of CAF01, SE (squalene emulsion), or AH (aluminum hydroxide). Antigen-specific IgG antibody responses were measured at the indicated time points. (C) Antibody response kinetics to immunization with NP-OVA. Mice were immunized s.c. with NP-OVA (5 μg) alone or in the presence of the indicated adjuvant and total and high affinity NP-specific IgG antibodies were measured by coating with NP₁₃-BSA or NP₂-BSA, respectively. Data show titers expressed as geometric mean+95% CI or ratios of NP₂/NP₁₃ titers+SEM of four mice in the non-adjuvanted groups or eight (A–C) to 10 (D) mice in the adjuvanted groups. Statistically significant differences between the SE and AH groups are indicated by *, **, and ***, whilst significant differences between the SE and CAF01 groups are indicated by ^# and ^### (Kruskal Wallis test, using the SE group as reference and significance levels of p < 0.05 and p < 0.001, respectively). Data are representative of one (C) or two (A,B,D) experiments.

Figure 2

Persistence of antigen at the site of injection is influenced by adjuvant system. Mice were injected into the quadriceps muscle with ovalbumin (OVA) coupled to AF647 either alone or in the presence of the indicated adjuvant. (A) Total cell number (left panel) and the numbers of the indicated cell subsets (right panels) in the injected muscles at various time points after injection. (B) OVA-positive cells after injection. (C) Quantity of the indicated cytokines at various time points after injection. Mouse groups consisted of 12 mice per group with 3 mice per group sacrificed at each time point. Each point represents mean+ SEM. Statistically significant differences between the SE and AH groups are indicated by *, **, and ***, whilst significant differences between the SE and CAF01 groups are indicated by ^{#, ##}, and ^### (One-way ANOVA with Tukey's correction for multiple group comparison, using the SE group as reference and significance levels of p < 0.05, p < 0.01, and p < 0.001, respectively). (D) Mice were vaccinated with CTH522 in the presence of the indicated adjuvants. The mice were scored by HE histology and immunofluorescent staining for mononuclear and inflammatory (CD64⁺) cells infiltrating the muscle at various time points after immunization. Representative plots display the 7 day time point. Three (antigen alone) or 8 mice (adjuvanted groups) were sacrificed at each time point.

Figure 3

Reduced antigen drainage to proximal lymph nodes in the presence of depot-inducing adjuvants. Mice were injected subcutaneously at the base of tail with ovalbumin (OVA) coupled to AF647 either alone or in the presence of the indicated adjuvant and the draining inguinal lymph nodes were collected at various time points thereafter. (A) Cells binding to OVA in the draining lymph nodes. Representative plots display the 6-h-time-point (upper panels). Lower panels display the numbers of cells within the indicated subsets binding to OVA. (B) Fraction of cells within the indicated subsets binding to OVA at 6 and 24 h post administration. The total numbers of OVA⁺ cells (surface-adsorbed or internalized) are displayed below the pies. (C) Representative plots of B cells binding to OVA at 24 h post injection (left panel) and summarized for the different time points (right panel). (D) The total cell number (left panel) and the percentage and numbers of B220⁺ B cells (right panels) at the indicated time points after injection. Each group consisted of 3 (naïve and antigen alone) or 4 (for each adjuvant) mice evaluated at each time point. Each point represents mean+ SEM. Statistically significant differences between the SE and AH groups are indicated by *, **, and *** whilst significant differences between the SE and CAF01 groups are indicated by ^{#, ##}, and ^### (One-way ANOVA with Tukey's correction for multiple group comparison, using the SE group as reference and significance levels of p < 0.05, p < 0.01, and p < 0.001, respectively).

Figure 4

Adjuvants differentially influence germinal center kinetics. Mice were immunized subcutaneously with CTH522 protein antigen either alone or in the presence of the indicated adjuvant. (A) Representative plots of antigen-specific germinal center (GC) B cells (B220⁺CD38-GL7⁺ cells binding CTH522 coupled to AF488) in the draining inguinal lymph nodes at days 7 and 14 post immunization. Lower panels: percentages (of B220+) and numbers of antigen-specific germinal center B cells. Each group consisted of 2 (antigen alone) or 4 (antigen +adjuvants) mice. Data are representative of two independent experiments. (B) Draining lymph node GC areas (upper panel) and number of GCs as indicated by clusters of Ki67⁺ cells. Groups consisted of 3 (antigen alone) or 8 (antigen +adjuvants) mice. The experiment was performed once. Each point represents mean+ SEM. Statistically significant differences between the SE and AH groups are indicated by * and *** whilst significant differences between the SE and CAF01 groups are indicated by ^# and ^### (One-way ANOVA with Tukey's correction for multiple group comparison, using the SE group as reference and significance levels of p < 0.05 and p < 0.001, respectively).

Figure 5

Removal of the injection site impairs antibody responses when antigen is formulated in depot-inducing adjuvants. Mice were immunized intradermally (i.d.) with CTH522 protein antigen either alone or in the presence of the indicated adjuvant. At various time points after injection, mice were anesthetized and the injection site either surgically removed or left intact. (A,B) Antigen-specific IgG antibody responses were measured at the indicated time points. Each vaccine group consisted of three to four mice and data are displayed as geometric mean+95% CI. Data in which the injection site was left intact (left panels) were compiled from three experiments. Statistically significant differences between the SE and AH groups are indicated by * and ** whilst significant differences between the SE and CAF01 groups are indicated by ^#, ^##, and ^### (One-way ANOVA with Tukey's correction for multiple group comparison, using the SE group as reference and significance levels of p < 0.05, p < 0.01, and p < 0.001 respectively).

Figure 6

Schematic of depot formation affecting induction and persistence of germinal centers. Depot-forming adjuvants sequester antigen at the site of injection. Early after immunization this leads to reduced B cell activation and GC formation in the draining LNs, compared to when antigen is given in an adjuvant which does not form an antigen depot. The slow release of antigen from the depot may help sustain germinal centers to promote a higher magnitude long-term antibody response as illustrated by data adapted from Figure 1 (right panel).