SARS-CoV-2-specific CD4+ T cells are associated with long-term persistence of neutralizing antibodies

Abstract

Understanding the decay and maintenance of long-term SARS-CoV-2 neutralizing antibodies in infected or vaccinated people and how vaccines protect against other SARS-CoV-2 variants is critical for assessing public vaccination plans. Here, we measured different plasm antibody levels 2 and 12 months after disease onset, including anti-RBD, anti-N, total neutralizing antibodies, and two neutralizing-antibody clusters. We found that total neutralizing antibodies declined more slowly than total anti-RBD and anti-N IgG, and the two neutralizing-antibody clusters decayed even more slowly than total neutralizing antibodies. Interestingly, the level of neutralizing antibodies at 12 months after disease onset was significantly lower than that at 2 months but more broadly neutralized SARS-CoV-2 variants, including Alpha (B.1.1.7), Beta (B.1.351), Gamma (P.1), Delta (B.1.617.2), and Lambda (C.37). Significant immune escape by the Omicron variant (B.1.1.529) was also observed 2 months post-recovery. Furthermore, we revealed that a high percentage of virus-specific CD4⁺ T cells and cTfh1 were associated with a slower decline in humoral immunity, accompanied by higher levels of CXCR3 ligands such as CXCL9 and CXCL10, higher frequency of cTfh1, and lower levels of cTfh2 and cTfh17. Our data highlight the importance of coordinating T-cell and humoral immunity to achieve long-term protective immunity.

Fig. 1

Neutralizing antibodies show a slower decay rate than binding antibodies. Two blood samples were collected from convalescent individuals with COVID-19 at approximately 2 (M2) and 12 months (M12) after disease onset (n = 46). The antibody levels at M2 and M12 were measured (n = 46). Total IgG to RBD and N in the plasma samples were examined using ELISA, and nAb levels were examined using the focus reduction neutralization test (FRNT₅₀). a Pair comparisons of IgG to RBD, N, and nAbs between M2 and M12. b Pair comparisons of antibody levels to 13G2 and 08B3 clusters. Each dot represents an individual participant. ***p < 0.001 and ****p < 0.0001. Wilcoxon rank-sum test was used to compare paired continuous variables that were not normally distributed. The percentage above the dot plots indicates the detection rate of each group.

Fig. 2

Broadly neutralizing antibodies against different SARS-CoV-2 strains 1 year after infection. Blood samples were collected from convalescent patients with COVID-19 at 2 (M2) and 12 months (M12) after disease onset (M2, n = 23; M12, n = 45). a, b Each dot represents an individual participant. Violin plot analysis comparing the distribution of nAb titers of M2 (a) and M12 (b) recovered plasma against WH-1 and VOCs, including B.1.1.7, B.1.351, P.1, B.1.617.2, and C.37, determined by using a pseudovirus-based neutralization assay. c The EC₅₀ of neutralizing antibodies against WH-1 and B.1.1.529 (Omicron) was determined in the plasma of convalescence at M2 (n = 32). The fold change was expressed as the ratio of WH-1 to B.1.1.529. Each dot represents an individual participant. ***p < 0.001 and ****p < 0.0001. Wilcoxon rank-sum test was used to compare paired continuous variables that were not normally distributed

Fig. 3

High levels of virus-specific CD4⁺ T cell at early convalescent phase correlate with long-term maintenance neutralizing antibodies. Antibodies present in plasma of the high- (n = 23) and low-CD4⁺ groups (n = 23) against RBD and N, and nAbs in the plasma at M2 and M12, were examined using ELISA and FRNT₅₀ assays. a, b Pair comparisons of IgG levels against RBD and N and nAbs between M2 and M12 for the high-and low-CD4 + groups; b Pair comparisons of IgG levels against the 13G2 and 08B3 clusters. Percentages on top of dot plots represent frequencies of samples with antibody titers above the cutoff. Each dot represents an individual participant. **p < 0.01 and ****p < 0.0001. Student’s t-test was used to analyze differences in the mean values between groups. The Mann–Whitney test was used to compare the central tendencies of the two groups (mean or median). Wilcoxon rank-sum test was used to compare paired continuous variables that were not normally distributed. The percentage above the dot plots indicates the detection rate of each group.

Fig. 4

Peripheral chemokines related to CD4⁺ T cells in convalescent patients with COVID-19. The levels of CXCL9, CXCL10, and CXCL11 were determined for high- (n = 18) and low-CD4⁺ (n = 22) convalescent individuals with COVID-19 and close contacts (n = 4) at M2 using cytometric Beads array. a–c Dot plot analysis of (a) CXCL9, (b) CXCL10, and (c) CXCL11 levels among high-convalescent and low-CD4⁺convalescent patients with COVID-19, and close contacts, respectively. Each dot represents an individual participant. Bars represent mean values. *p < 0.05 and **p < 0.01. Student’s t-test was used to analyze differences in the mean values between groups. The Mann–Whitney test was used to compare the central tendencies of the two groups (mean or median). Data are presented as mean ± SEM.

Fig. 5

Peripheral CD4⁺ T-cell subsets in convalescent patients with COVID-19. T-cell subsets were analyzed by fluorescence-activated cell sorting (FACS) in peripheral blood mononuclear cell (PBMCs) from high- (n = 10) and low-CD4⁺ (n = 6) convalescent individuals with COVID-19 and close contacts (n = 9). a Gating strategies for Treg cells (CD25⁺CD45RA⁻CD127⁻) and cTfr cells (CD25⁺CD45RA^–CD127^–CXCR5^hiPD-1^hi). b Dot plot analysis of the frequencies of Treg and cTfr cells in high- and low-CD4⁺ convalescent individuals with COVID-19 and close contacts. c Gating strategies for different peripheral circulating Tfh cell subsets, including CXCR3⁺CCR6⁻cTfh (cTfh1) cells, CXCR3^–CCR6^– cTfh (cTfh2) cells, and CXCR3^–CCR6⁺ cTfh (cTfh17) cells. d Frequencies of cTfh1, cTfh2, and cTfh17 cells in high-and low-CD4⁺ COVID-19 convalescent individuals with COVID-19 and close contacts. e–g Correlation analysis of (e) cTfh2, (f) cTfh17, and (g) cTfh1 cells with SARS-CoV-2-specific neutralizing-antibody titers in high-and-low-CD4⁺ convalescent individuals with COVID-19 (n = 16). Each dot represents an individual participant. Bars represent mean values. *p < 0.05 and **p < 0.01. Student’s t-test was used to analyze differences in the mean values between groups. The Mann–Whitney test was used to compare the central tendencies of the two groups (mean or median). Data are presented as mean ± SEM.