The dichotomous and incomplete adaptive immunity in COVID-19 patients with different disease severity

Abstract

The adaptive immunity that protects patients from coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is not well characterized. In particular, the asymptomatic patients have been found to induce weak and transient SARS-CoV-2 antibody responses, but the underlying mechanisms remain unknown; meanwhile, the protective immunity that guide the recovery of these asymptomatic patients is elusive. Here, we characterized SARS-CoV-2-specific B-cell and T-cell responses in 10 asymptomatic patients and 64 patients with other disease severity (mild, n = 10, moderate, n = 32, severe, n = 12) and found that asymptomatic or mild symptomatic patients failed to mount virus-specific germinal center (GC) B cell responses that result in robust and prolonged humoral immunity, assessed by GC response indicators including follicular helper T (T_FH) cell and memory B cell responses as well as serum CXCL13 levels. Alternatively, these patients mounted potent virus-specific T_H1 and CD8⁺ T cell responses. In sharp contrast, patients of moderate or severe disease induced vigorous virus-specific GC B cell responses and associated T_FH responses; however, the virus-specific T_H1 and CD8⁺ T cells were minimally induced in these patients. These results, therefore, uncovered the protective immunity in asymptomatic patients and also revealed the strikingly dichotomous and incomplete humoral and cellular immune responses in COVID-19 patients with different disease severity, providing important insights into rational design of effective COVID-19 vaccines.

Fig. 1

Virus-specific B cell responses to SARS-CoV2 in COVID-19 patients with different severity. Samples of a, b and g were from Healthy (#5), Asymptomatic (#7), Mild (#3), Moderate (#22), Severe (#4). a FACS plot examples of SARS-CoV2 S1- or S2- specific B cells (S1⁺/S2⁺CD19⁺) of live lymphocytes in PBMCs. b FACS plot examples of SARS-CoV2 S1- or S2- specific memory B cells (CD27^hiIgD^low), gated on S1 or S2 specific B cells (S1⁺/S2⁺CD19⁺) shown in (a). c, d Percentages of S1 or S2 specific B cells (c), summarized from (a); and S1 or S2 specific memory B cells (d), summarized from (b), in COVID-19 recovered patients with different disease severity. e, f ELISA analysis of S1- and S2- specific IgG levels in convalescent-phase COVID-19 patients’ sera with various groups. Dilution of 1:100 was used for serum samples. g FACS plot examples of cT_FH (PD1^hiCXCR5^hi), gated on live CD4⁺ T cells. h The summarization of percentages of cT_FH from (g). i CXCL13 protein level in convalescent-phase COVID-19 patients’ serum with various groups, measured by ELISA. Bars represent the mean ± SEM. P values were calculated based on Bonferroni of one-way ANOVA analysis. ***p < 0.0001, **p < 0.001, and *p < 0.05

Fig. 2

Correlations between B cells frequency, serum CXCL13 concentration, cTFH cells frequency and S1- or S2- specific IgG levels. a, b Correlations between B cells frequency and S1- (a) or S2- (b) specific IgG levels. c, d Correlations between serum CXCL13 concentration and S1- (c) or S2- (d) specific IgG levels. e, f Correlations between cT_FH cells frequency and S1- (e) or S2- (f) specific IgG levels. g Correlations between cT_FH cells frequency and serum CXCL13 concentration. Statistical comparisons were performed using Spearman correlation analysis

Fig. 3

Kinetics of virus-specific B cell responses to SARS-CoV2 in COVID-19 Asymptomatic patients. Sample of a, b and f was from Asymptomatic (#6). a, b FACS plot examples of SARS-CoV2 S1- or S2- specific B cells and plasmablasts (CD19⁺CD20^-CD27^hiCD38^hi) cells’ percentage in early phase (0–3 days post hospitalization), acute phase (5–10 days post hospitalization), convalescent-phase, (more than 10 days after COVID-19 nucleic acid test shown negative) in the asymptomatic cohort. c, d Percentages of S1- or S2- specific B cells (c), summarized from (a), and plasmablasts cells (d), summarized from (b), in the asymptomatic cohort. e Kinetics of S1- and S2- specific IgG levels in asymptomatic patients’ serum, measured by ELISA. f FACS plot examples of cT_FH cells’ percentage of asymptomatic patients. g Percentages of cT_FH, summarized from (f). h Kinetics of CXCL13 level in asymptomatic patients’ serum, measured by ELISA. Bars represent the mean ± SEM. P values were calculated based on Bonferroni of one-way ANOVA analysis. ***p < 0.0001, **p < 0.001, and *p < 0.05

Fig. 4

SARS-CoV2 specific CD4⁺ and CD8⁺ T cell responses in COVID-19 convalescent patients. PBMCs of healthy control (n=8) and recovered COVID-19 patients (n=64) were stimulated with SARS-CoV-2 dominant antigen (S1, S2 and nucleoprotein, N) cocktails for 44h, Golgi-Plug containing Golgi-stop and DNase were added into cell culture for another 4h. Samples of a, c, e, and g were from Healthy (#3), Asymptomatic (#4), Mild (#8), Moderate (#13), severe (#6). a FACS plot examples of IFNγ⁺CD4⁺ T cells in total live CD4⁺ T cells, gated on total live CD4⁺ T cells. b Bar graph shows the frequency of IFNγ⁺CD4⁺ T cells in total CD4⁺ T cells after stimulation, summarized from (a). c FACS plot examples of GZMB⁺Perforin⁺CD4⁺ T cells in total IFNγ⁺CD4⁺ T cells, gated on total live IFNγ⁺CD4⁺ T cells. d Frequency of GZMB⁺Perforin⁺CD4⁺ T cells in total IFNγ⁺CD4⁺ T cells, summarized from (c). e FACS plot examples of IFNγ⁺CD8⁺ T cells in total live CD8⁺ T cells, gated on total live CD8⁺ T cells. f Bar graph shows the frequency of IFNγ⁺CD8⁺ T cells in total CD8⁺ T cells after stimulation, summarized from (e). g FACS plot examples of GZMB⁺Perforin⁺CD8⁺ T cells in total IFNγ⁺CD8⁺ T cells, gated on total live IFNγ⁺CD8⁺ T cells. h Frequency of GZMB⁺Perforin⁺CD8⁺ T cells in total IFNγ⁺CD8⁺ T cells, summarized from (g). Bars represent the mean±SEM. P values were calculated based on Bonferroni of one-way ANOVA analysis. ***p<0.0001, **p<0.001, and *p<0.05

Fig. 5

Kinetics of SARS-CoV2 specific CD4⁺ and CD8⁺ T cell responses in COVID-19 asymptomatic patients. PBMCs of healthy control (n = 8) and asymptomatic COVID-19 patients (n = 10) were stimulated with SARS-CoV-2 dominant antigen (S1, S2, and nucleoprotein, N) cocktails for 44 h, Golgi-Plug containing Golgi-stop and DNAase were added into cell culture for another 4 h. Representative samples were from Healthy (#3), Asymptomatic (#3). a FACS plot examples of IFNγ⁺CD4⁺ T cells in total live CD4⁺ T cells, gated on total live CD4⁺ T cells. b Bar graph shows the frequency of IFNγ⁺CD4⁺ T cells in total CD4⁺ T cells after stimulation, summarized from (a). c FACS plot examples of GZMB⁺Perforin⁺CD4⁺ T cells in total IFNγ⁺CD4⁺ T cells, gated on total live IFNγ⁺CD4⁺ T cells. d Frequency of GZMB⁺Perforin⁺CD4⁺ T cells in total IFNγ⁺CD4⁺ T cells, summarized from (c). e FACS plot examples of IFNγ⁺CD8⁺ T cells in total live CD8⁺ T cells, gated on total live CD8⁺ T cells. f Bar graph shows the frequency of IFNγ⁺CD8⁺ T cells in total CD8⁺ T cells after stimulation, summarized from (e). g FACS plot examples of GZMB⁺Perforin⁺CD8⁺ T cells in total IFNγ⁺CD8⁺ T cells, gated on total live IFNγ⁺CD8⁺ T cells. h Frequency of GZMB⁺Perforin⁺CD8⁺ T cells in total IFNγ⁺CD8⁺ T cells, summarized from (g). Bars represent the mean ± SEM. P values were calculated based on Bonferroni of one-way ANOVA analysis. There was no statistically significant difference among different stage of asymptomatic patients

Fig. 6

Schematic summary of adaptive immunity of COVID-19 patients with different disease severity. (Up panel) Patients of asymptomatic or mild disease mount a predominant cellular immunity but low to no humoral immunity. (Lower panel) Patients of moderate or severe disease induce a potent humoral immunity, while fail to elicit an effective cellular immunity