Limited induction of SARS-CoV-2-specific T cell responses in children with multisystem inflammatory syndrome compared with COVID-19

Abstract

Why multisystem inflammatory syndrome in children (MIS-C) develops after SARS-CoV-2 infection in a subset of children is unknown. We hypothesized that aberrant virus-specific T cell responses contribute to MIS-C pathogenesis. We quantified SARS-CoV-2-reactive T cells, serologic responses against major viral proteins, and cytokine responses from plasma and peripheral blood mononuclear cells in children with convalescent COVID-19, in children with acute MIS-C, and in healthy controls. Children with MIS-C had significantly lower virus-specific CD4+ and CD8+ T cell responses to major SARS-CoV-2 antigens compared with children convalescing from COVID-19. Furthermore, T cell responses in participants with MIS-C were similar to or lower than those in healthy controls. Serologic responses against spike receptor binding domain (RBD), full-length spike, and nucleocapsid were similar among convalescent COVID-19 and MIS-C, suggesting functional B cell responses. Cytokine profiling demonstrated predominant Th1 polarization of CD4+ T cells from children with convalescent COVID-19 and MIS-C, although cytokine production was reduced in MIS-C. Our findings support a role for constrained induction of anti-SARS-CoV-2-specific T cells in the pathogenesis of MIS-C.

Keywords: Adaptive immunity; COVID-19; Cellular immune response; Peptides.

Figure 1. SARS-CoV-2–specific T cell responses.

(A–C) SARS-CoV-2 CD4⁺ T cell data in MIS-C, convalescent COVID-19, and healthy children (HC) shown as AIM⁺ (OX40⁺41BB⁺) CD4⁺ T cells in response to CD4_S peptide MP (spike-containing) and CD4_R peptide MP (non-spike proteins) stimulation by (A) fold change (FC) over the DMSO condition, (B) frequency of CD4⁺OX40⁺41BB⁺ T cells after MP stimulation with DMSO subtraction, and (C) total (frequency of CD4_S plus CD4_R MP) CD4⁺ T cell responses with DMSO subtraction. (D–F) SARS-CoV-2 CD8⁺ T cell data in MIS-C, convalescent COVID-19, and HC shown as AIM⁺ (CD69⁺41BB⁺) CD8⁺ T cells in response to CD8-A peptide MP and CD8-B peptide MP stimulation by (D) FC over the DMSO condition, (E) frequency of CD8⁺CD69⁺41BB⁺ T cells after MP stimulation with DMSO subtraction, and (F) total (frequency of CD8-A plus CD8-B MP) CD8⁺ T cell responses with DMSO subtraction. DMSO, CD4_S, and CD8-A: MIS-C (n = 21), COVID-19 (n = 19), and HC (n = 20). CD4_R and CD8-B: MIS-C (n = 16), COVID-19 (n = 16), and HC (n = 20). Background-subtracted values ≤ 0 are represented as the lowest calculated AIM⁺ CD4⁺ or CD8⁺ T cell frequency in the data set. Geometric mean shown with statistical comparison by Kruskal-Wallis test. *P < 0.05, **P ≤ 0.01, ***P ≤ 0.001.

Figure 2. Serologic responses to SARS-CoV-2 antigens and correlations with T cell responses.

(A) Antibody titers to SARS-CoV-2 RBD, full-length spike, and nucleocapsid protein in MIS-C, convalescent COVID-19, and healthy children (HC). Geometric mean shown with statistical comparison by Kruskal-Wallis test. ****P < 0.0001. (B) Relationship between RBD IgG and spike IgG titers for MIS-C, convalescent COVID-19, and HC. (C, D, F, G, I, and J) Relationships between spike-specific CD4⁺ T cell responses (fold change, FC) and RBD IgG titers (C and F) and full-length spike IgG titers (D and G) in MIS-C (C and D), convalescent COVID-19 (F and G), and HC (I and J). (E, H, and K) Relationship between non-spike-specific CD4⁺ T cell responses (FC) and nucleocapsid IgG titers for MIS-C (E), convalescent COVID-19 (H), and HC (K). Statistical relationships in C–K were assessed with Spearman’s correlation.

Figure 3. SARS-CoV-2–specific CD4⁺ T cell cytokine profiling.

(A) Cytokine production for major T helper cytokines shown as fold change (FC) of cytokine concentration after CD4_ S MP (spike-containing) and CD4_R MP (non-spike) stimulation versus DMSO control measured on a subset of participants by MSD multiplex cytokine array. Individual results for MIS-C (n = 19), convalescent COVID-19 (n = 12), and HC (n = 20) are plotted with geometric mean ± SDs. Statistical comparison between groups was conducted by Kruskal-Wallis test. (B) Heatmap showing CD4⁺ T cell cytokine response by median FC in cytokine production for each peptide MP by clinical cohort. (C) Cytokine concentrations from DMSO control and pair-matched peptide stimulation (Top, spike-containing MP; bottom, non-spike MP) for MIS-C and convalescent COVID-19. Cytokines with significant increases after stimulation are shown. Statistical comparison using Wilcoxon matched-pairs signed rank test. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.