Immunodominance is a poor predictor of vaccine-induced T follicular helper cell quality

Abstract

Background: Rational engineering of vaccine immunogens to focus B cell responses on potently neutralising epitopes is a promising approach to improve the potency, breadth and durability of viral vaccines. Such strategies, however, can compromise vaccine immunogenicity through the unintended exclusion of CD4+ T cell epitopes, which are critical for the development of T follicular helper (TFH) cells and to support high affinity antibody production.

Methods: Using a prototypic influenza haemagglutinin (HA) stem immunogen lacking effective CD4+ T cell help in C57BL/6 mice, we interrogated the minimal requirements for T cell help needed to drive serological responses to vaccination.

Findings: We find that priming of naïve CD4+ T cells is markedly efficient, however the immunodominance of a given CD4+ T cell epitope is not predictive of the propensity to provide high quality help to antigen-specific B cells. In the context of soluble antigens, provision of a single MHC class II epitope is sufficient to drive robust germinal centre responses and serum IgG titres. However not all CD4+ epitopes provide equivalent levels of B cell help, despite priming comparable numbers of antigen-specific CD4+ T cells. Finally, we show multimerizing and arraying antigens on nanoparticle scaffolds unlocks highly subdominant, near-undetectable CD4+ T cell helper responses to support a T-dependent antibody response.

Interpretation: Our findings emphasise the importance of CD4+ T cell help for programming robust and durable humoural immunity, and provide crucial insights to guide the rational incorporation of favourable T cell epitopes into vaccines.

Funding: The study was funded by the NHMRC.

Keywords: CD4 T cell; Germinal centre; Immunodominance; T follicular helper cell; Vaccine.

Fig. 1

Nanoparticle display enhances stem immunogenicity in the absence of prominent CD4⁺ T cell help. Mice were vaccinated with 5 mg of full-length HA protein (HA-FL), HA-stem protein, or stem-ferritin nanoparticles (Stem-Fe) co-formulated 1:1 with Addavax adjuvant. Serum and draining LN were collected at day 14 post-vaccination. (A) Serum endpoint titres of stem-specific IgG (N = 10/group). (B) Frequency of GL7^hiCD38^lo GC B cells, (C) number of stem-specific GC B cells, or (D) frequency of CXCR5^hiPD-1^hi TFH in the draining LN (N = 10/group). (E) Representative staining and frequency of CD154⁺ stem or HA-specific memory CD4⁺ T cells following in vitro stimulation. Frequencies are background subtracted based on the DMSO control (N = 5/group). (F) CD4⁺ T cell proliferation following in vitro peptide stimulation with HA, stem or ferritin peptide pools. Splenocytes were harvested from vaccinated mice at day 14, or at day 79 after 3 vaccinations (N = 5/group). Lines indicate median and IQR. Statistics assessed by Mann–Whitney test comparing Stem and Stem-Fe groups. ∗p < 0.05, ∗∗∗∗p = 0.0001.

Fig. 2

Stem-Fe nanoparticle immunogenicity in SMARTA and BALB/c mice. Stem-specific (A) GC B cell frequencies, (B) GC B cell counts or (C) IgG titres at day 14 following Stem-Fe vaccination of SMARTA (N = 4), WT BL6 (N = 10) or BALB/c (N = 5) mice. (D) Stem- or ferritin-specific CD4⁺ T cell responses in the draining LN of Stem-Fe vaccinated BALB/c mice at day 14 (n = 4). (E) CD4⁺ T cell proliferation in BALB/c mice following in vitro peptide stimulation with DMSO, stem or ferritin peptide pools. Splenocytes were harvested from vaccinated mice at day 14 (N = 5/group). Lines indicate median and IQR.

Fig. 3

Rescue of stem immunogenicity by genetically fused CD4⁺ T cell epitopes. (A) Identification of immunogenic peptides following intranasal infection with PR8 virus. Mediastinal LN were collected on day 14 and restimulated in vitro with DMSO, ConA or indicated peptides (N = 4–5 mice per peptide). Immunogenic peptides are labelled with the median response and number of responding mice. (B) Design of trimeric stem antigens covalently linked to HA-derived peptides or prototypic OVA₃₂₃ and GP₆₁ peptides. (C) Stem-specific IgG endpoint titres at day 14 post-vaccination with 5 μg of HA-FL, stem, or stem-peptide antigens formulated with Addavax (N = 7–10 per group). Statistics assessed by Kruskal–Wallis test with Dunn's post-test compared to the stem control. (D) Number of stem-specific GC B cells (GL7⁺CD38^lo) at day 14 post-vaccination (N = 4–5 per group). (E) Longitudinal tracking of stem-specific GC B cells in the draining LN at days 4, 5, 6, 10 and 14 post-vaccination (N = 5 per group). (F) Durability of antigen-specific serum IgG following stem-GP₆₁ or HA-FL vaccination (N = 5 per group). Symbols or lines indicate median with IQR.

Fig. 4

Differential GC recruitment of HA₉₁ and GP₆₁-specific TFH populations. (A) Longitudinal tracking of total, (B) TFH (CXCR5^hiPD-1^hi) or (C) GC resident TFH (CD90^lo) antigen-specific CD4⁺ T cell numbers in the draining LN using IA^b tetramers. (D) Proportion of tetramer ⁺ TFH with CD90^lo phenotype at days 5, 6 and 10 post-vaccination. Plots show CD90 expression for the tetramer⁺ TFH gate. Symbols indicate median and IQR (N = 5 per group). Statistics assessed by Kruskal-Wallis test with Dunn's post-test. ∗p < 0.05, ∗∗p < 0.01.

Fig. 5

Modulation of stem-HA₉₁ immunogenicity through altered B cell antigen presentation. (A) Overview of immunisation timeline and groups. (B) Stem or (C) OVA IgG titres at day 14 post-vaccination, related to the groups shown in panel A. N = 3–4 per group from either one or two independent experiments. Statistics assessed by Mann–Whitney test. (D) Frequencies of total, TFH, or GC TFH HA₉₁Tet ⁺ T cells at day 6 post-vaccination with 5 μg stem-HA₉₁ + 5 μg OVA-HA₉₁ (light blue) or 5 μg stem-HA₉₁ + 5 μg OVA + 0.2 μg HA₉₁ (dark blue). N = 5 per group. Lines indicate median and IQR. ∗p < 0.05.

Fig. 6

Modulation of stem-GP₆₁immunogenicity via altered B cell antigen presentation. (A) Antigen doses for titration of GP₆₁ peptide availability. (B) Numbers of total and TFH GP₆₁ Tet ⁺ cells at day 6 post-vaccination (N = 5 per group). (C) Number of stem-specific GC B cells at day 6 post-vaccination (N = 5 per group). Symbols indicate median and IQR. (D) Correlation between number of Tet⁺ TFH or (E) total Tet⁺ cells with stem-specific GC B cells. Lymph nodes with no detectible stem-specific GC B cells were arbitrarily assigned a value of 0.5 (marked with a dashed line) for graphical display. Statistics assessed by Spearman correlation.