Single-dose SARS-CoV-2 vaccinations with either BNT162b2 or AZD1222 induce disparate Th1 responses and IgA production

Abstract

Background: While vaccination programs against the severe acute respiratory syndrome virus 2 (SARS-CoV-2) are globally ongoing, disparate strategies for the deployment of spike antigen show varying effectiveness.

Methods: In order to explore this phenomenon, we sought to compare the early immune responses against AZD1222 and BNT162b2. SARS-CoV-2 seronegative participants received a single dose of either vaccine and were analyzed for immune cell, effector T cell, and antibody dynamics.

Results: AZD1222 induced transient leukopenia and major changes among innate and adaptive subpopulations. Both vaccines induced spike protein-specific effector T cells which were dominated by type 1 helper T cell responses following AZD1222 vaccination. A significant reduction of anti-inflammatory T cells upon re-stimulation was also restricted to AZD1222 vaccinees. While IgM and IgG were the dominant isotypes elicited by AZD1222, BNT162b2 led to a significant production of IgG and IgA.

Conclusions: Our results suggest that the strategy for spike protein delivery impacts on how and to what extent immune priming against the main SARS-CoV-2 antigen proceeds.

Keywords: AZD1222; BNT162b2; COVID-19; Cytotoxic T cells; SARS-CoV-2; Type 1 helper T cells; Vaccination.

Fig. 1

Vaccination with AZD1222 induced a transient reduction of peripheral leukocytes and displacements of major immune cell populations. A Leukocyte counts in the peripheral blood after vaccination with AZD1222 (n = 13, left panel) and BNT162b2 (n = 15, right panel). p-values resulting from multiple group comparisons were 0.0005 (AZD1222/Kruskal-Wallis and Dunn’s multiple comparisons tests) and 0.5306 (BNT162b2/one-way ANOVA), respectively. **p < 0.01, ***p < 0.001. B UMAP of surface antigen expression and clustering of major immune cell populations for all time points after vaccination. C UMAPs on immune cell compositions for each time point

Fig. 2

AZD1222 vaccination induced the enrichment of pro-inflammatory monocytes. A Pseudocolor plots for the expression of CD14 and CD16 on monocytes are representative for the AZD1222 vaccination group. B Proportions of CD16⁺CD14⁺ pro-inflammatory monocytes after vaccination with AZD1222 (n = 18, left panel) or BNT162b2 (n = 18, right panel). All FACS analyses were on gated live monocytes. Asterisks indicate significant differences between time points. p-values resulting from one-way ANOVA and Tukey-Kramer multiple comparisons tests were < 0.0001 for AZD1222 and 0.0146 for BNT162b2 analyses. ***p < 0.001

Fig. 3

Vaccination with AZD1222 and BNT162b2 induced significant yet transient alterations among peripheral plasmablasts. AZD1222 also induced a significant increase in late memory B cells on days 13 and 20. A Pseudocolor plots for the expression of CD27 and CD38 on B cells are representative for the AZD1222 vaccination group. B Proportions of CD27⁺CD38^bright plasmablasts (top) and CD27⁺CD38^– late memory B cells (bottom) after vaccination with AZD1222 (n = 18, left) or BNT162b2 (n = 18, right). All FACS analyses were on CD19⁺CD45RA⁺ B cells. p-values resulting from one-way ANOVAs were < 0.0001 for both AZD1222 analyses. p-values resulting from Kruskal-Wallis and Dunn’s multiple comparisons tests were 0.0008 for the comparison of plasmablast and 0.6998 for the comparison of late memory B cells for the BNT162b2 analyses. *p < 0.05, **p < 0.01, ***p < 0.0001

Fig. 4

Spike-specific IFNγ-producing lymphocytes were significantly increased after vaccination with BNT162b2. PBMCs were isolated at the day of vaccination (day 0) and 20 days later. Cells were then stimulated with recombinant SARS-CoV-2 spike protein for 24 h. Subsequently, an ELISPOT assay for the detection of IFNγ-producing cells was performed. A Representative shots of spot-forming cells from individuals vaccinated with either AZD1222 (A9 and A12) or BNT162b2 (B4 and B20). PBMCs were stimulated with PMA as a positive control (bottom). B The counts of spot-forming cells from day 20 were normalized to counts from day 0. Data show the relative increase in IFNγ⁺ cells after vaccination (n = 16 for AZD1222 and n = 15 for BNT162b2). p-values result from one-sample t-tests assessing the differences of group means to the hypothetical value of 1

Fig. 5

Vaccination with single doses of AZD1222 or BNT162b2 induced the expansion of cytotoxic effector T cells. PBMCs were isolated on day 20 after vaccination and stimulated with (stim.) or without (unstim.) spike protein-encoding mRNA (BNT162b2) for 24 h and analyzed by flow cytometry for the expression of inducible activation markers and intracellularly trapped cytokines among CD8⁺ cells. Sample sizes were n = 9 for AZD1222 and n = 15 for BNT162b2, respectively. Paired samples are illustrated by color coding. In vitro re-stimulation significantly increased CD25⁺CD137⁺ (A) and FasL expressing cells in both vaccination groups (B). INFγ–producers were increased (C) and IL-2 and IL-10 co-expressing cells were significantly reduced among AZD1222 vaccinees only (D). *p < 0.05, **p < 0.01 resulting from paired t-tests. ^#p < 0.05, ^###p < 0.001 resulting from Wilcoxon signed rank tests for matched pairs

Fig. 6

AZD1222 vaccination supported the induction of type 1 helper T cells. PBMCs were isolated on day 20 after vaccination and stimulated with (stim.) or without (unstim.) spike protein encoding mRNA (BNT162b2) for 24 h and analyzed by flow cytometry for the expression of inducible activation markers and intracellularly trapped cytokines among CD4⁺ cells. Sample sizes were n = 9 for AZD1222 and n = 15 for BNT162b2, respectively. Paired samples are illustrated by color coding. In vitro re-stimulation significantly increased CD25⁺CD137⁺ cells in both vaccination groups (A) and TNFα-producing cells in the AZD1222 vaccinated group, only B. C Neither vaccination allowed for the re-stimulation of IL-4-producing cells. D IL-2 and IL-10 co-expressing cells were significantly reduced among AZD1222 vaccinees only. *p < 0.05, **p < 0.01, resulting from paired t-test. ^##p < 0.01, resulting from Wilcoxon signed rank test for matched pairs

Fig. 7

A single dose of BNT162b2 induced significantly higher amounts of spike protein binding IgG, IgA, and receptor binding domain neutralizing antibodies than AZD1222. A SARS-CoV-2 spike binding antibodies were detected by ELISA and absorbance readouts were normalized to calibrator values to obtain arbitrary units. Gray lines depict the positive response threshold within a range of 0.8 to 1.1 arbitrary units. Sample sizes were n = 18 for AZD1222 and n = 18 for BNT162b2, respectively. p-values resulting from one-way ANOVA and Tukey-Kramer multiple comparisons tests were < 0.0001 for all three isotype analyses. B Neutralizing antibodies were determined by the surrogate virus competitive ELISA. Absorbance readouts from BNT162b2 (n = 17) and AZD1222 (n = 17) were calibrated with WHO standards to obtain international units (IU)/mL. *p < 0.05 resulting from Mann-Whitney U test and ***p < 0.001 resulting from Dunn’s test