SARS-CoV-2 variants of concern partially escape humoral but not T-cell responses in COVID-19 convalescent donors and vaccinees

Abstract

The emergence of SARS-CoV-2 variants harboring mutations in the spike (S) protein has raised concern about potential immune escape. Here, we studied humoral and cellular immune responses to wild type SARS-CoV-2 and the B.1.1.7 and B.1.351 variants of concern in a cohort of 121 BNT162b2 mRNA-vaccinated health care workers (HCW). Twenty-three HCW recovered from mild COVID-19 disease and exhibited a recall response with high levels of SARS-CoV-2-specific functional antibodies and virus-specific T cells after a single vaccination. Specific immune responses were also detected in seronegative HCW after one vaccination, but a second dose was required to reach high levels of functional antibodies and cellular immune responses in all individuals. Vaccination-induced antibodies cross-neutralized the variants B.1.1.7 and B.1.351, but the neutralizing capacity and Fc-mediated functionality against B.1.351 was consistently 2- to 4-fold lower than to the homologous virus. In addition, peripheral blood mononuclear cells were stimulated with peptide pools spanning the mutated S regions of B.1.1.7 and B.1.351 to detect cross-reactivity of SARS-CoV-2-specific T cells with variants. Importantly, we observed no differences in CD4⁺ T-cell activation in response to variant antigens, indicating that the B.1.1.7 and B.1.351 S proteins do not escape T-cell-mediated immunity elicited by the wild type S protein. In conclusion, this study shows that some variants can partially escape humoral immunity induced by SARS-CoV-2 infection or BNT162b2 vaccination, but S-specific CD4⁺ T-cell activation is not affected by the mutations in the B.1.1.7 and B.1.351 variants.

Fig. 1.. HCW study design.

N = 121 HCWs were enrolled in a prospective SARS-CoV-2 infection and vaccination study. Upon symptomatic presentation to occupational health services, a paired nasopharyngeal swab and EDTA blood sample was obtained (T0). A second EDTA blood sample was obtained 3 weeks after diagnostic RT-PCR (T3). On the basis of the diagnostic RT-PCR result at T0 and serology result at T3, 98 COVID-19–naive (yellow) and 23 COVID-19–recovered (blue) HCWs were enrolled in the vaccination study on average 50 days after inclusion. N = 13 COVID-19–recovered and N = 12 COVID-19–naive participants were randomly selected for in-depth analysis. Blood samples were collected after the first (Vx¹3) and second (Vx²3) vaccination, processed, and subsequently used for downstream serological and cellular assays.

Fig. 2.. Detection of SARS-CoV-2–specific humoral responses.

Total Ig levels were measured in COVID-19–naive (yellow) and –recovered (blue) donors at the acute, convalescent, post-vaccination 1, and post-vaccination 2 stage (T0, T3, Vx¹3, and Vx²3) by an (A) ELISA against nucleocapsid (N) and (B) RBD. (C) Quantitative IgG against S1 was measured by a Luminex bead assay. (D) Antibody binding to WT SARS-CoV-2 and VOC B.1.1.7 and B.1.351 was determined by end point titration in ELISA. Virus neutralization was measured by PRNT₅₀ against (E) WT SARS-CoV-2 (D614G) and (F) VOC. Analyses in (B) and (C) were performed on 121 participants, and in-depth analyses were performed in (A), (D), (E), and (F) on 25 participants. Time points in (A), (B), and (C) were compared by performing a nonparametric repeated measures Friedman test. End point titers between VOC in (D) were compared by RM one-way ANOVA or Friedman test. PRNT₅₀ titers in (D) and (E) were compared by RM one-way ANOVA. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Symbol shapes indicate individual donors and are consistent throughout the figures. Lines in (A) and (B) show the means; lines in (C), (D), (E), and (F) show geometric means. Dotted lines represent cutoff values for positivity [3× background OD₄₅₀ in (A), OD₄₅₀ ratio = 1 in (B), and 10.08 BAU/ml in (C)]. NT: not tested.

Fig. 3.. Detection of ADCC-mediating antibodies by measuring NK92.05 degranulation.

(A) Gating strategy for detection of degranulating NK cells: (i) NK92.05-CD16 cells are selected on the basis of size and granularity, (ii) exclusion of doublets, and (iii) selection of LIVE and CD56⁺ cells. Degranulation is measured as percentage CD107a⁺ cells within the NK fraction; PBS coating is included as background control. (B and C) ADCC-mediating antibodies were detected in COVID-19–naive (yellow) and –recovered (blue) donors at the acute, convalescent, post-vaccination 1, and post-vaccination 2 stage (T0, T3, Vx¹3, and Vx²3) against the WT N (B) and S (C) protein. (D) ADCC-mediating antibody reactivity with WT SARS-CoV-2 and VOC B.1.1.7 and B.1.351. These analyses were performed on 25 participants. Time points in (B) and (C) were compared by performing a nonparametric repeated measures Friedman test. Differences between variants were assessed by mixed-effect models. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. Symbol shapes indicate individual donors and are consistent throughout the figures. Lines indicate mean responses.

Fig. 4.. Detection of S-specific T cells by measuring up-regulation of AIM.

(A) Gating strategy for virus-specific T cells that up-regulate AIM: (i) Lymphocytes are selected on the basis of size and granularity, (ii) exclusion of doublets, (iii) selection of LIVE and CD3⁺ cells, and (iv) division into CD4⁺ and CD8⁺ T cells. Activation is measured as percentage CD69⁺/CD137⁺ double-positive cells within the CD4 or CD8 fraction; DMSO stimulation is included as background control. (B to D) Antigen-specific activation of CD4⁺ and CD8⁺ T cells in COVID-19–naive (yellow) and COVID-19–recovered (blue) donors at the acute, convalescent, post-vaccination 1, and post-vaccination 2 stage (T0, T3, Vx¹3, and Vx²3) by overlapping peptide pools covering the full WT S protein. Activation of SARS-CoV-2–specific CD4⁺ T cells is shown as percentage of AIM⁺ cells within the CD4⁺ subset after (B) subtraction of the DMSO background or (C) as an SI by dividing specific activation over background activation. (D) Activation of SARS-CoV-2–specific CD8⁺ T cells is shown as SI. An SI of 2 or higher is considered a positive T cell response. (E and F) Antigen-specific activation of CD4⁺ T cells by peptide pools exclusively covering mutational regions in VOC B.1.1.7 and B.1.351, compared against homologous WT peptide pools. Antigen-specific T cell responses are shown as SI. These analyses were performed in 20 participants. Time points in (B), (C), and (D) were compared by performing a Kruskall-Wallis test. Differences between variants were compared by performing Wilcoxon test. *P < 0.05 and **P < 0.01. Symbol shapes indicate individual donors and are consistent throughout the figures. Lines indicate mean (B) or geometric mean (C to F) responses. Low cell count samples (<10,000 or <5000 events within CD4⁺ or CD8⁺ gate, respectively) were excluded.