Clonal analysis of immunodominance and cross-reactivity of the CD4 T cell response to SARS-CoV-2

Abstract

The identification of CD4⁺ T cell epitopes is instrumental for the design of subunit vaccines for broad protection against coronaviruses. Here, we demonstrate in COVID-19-recovered individuals a robust CD4⁺ T cell response to naturally processed severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike (S) protein and nucleoprotein (N), including effector, helper, and memory T cells. By characterizing 2943 S-reactive T cell clones from 34 individuals, we found that the receptor-binding domain (RBD) is highly immunogenic and that 33% of RBD-reactive clones and 94% of individuals recognized a conserved immunodominant S346-S365 region comprising nested human leukocyte antigen DR (HLA-DR)- and HLA-DP-restricted epitopes. Using pre- and post-COVID-19 samples and S proteins from endemic coronaviruses, we identified cross-reactive T cells targeting multiple S protein sites. The immunodominant and cross-reactive epitopes identified can inform vaccination strategies to counteract emerging SARS-CoV-2 variants.

Fig. 1. Robust T cell response to SARS-CoV-2 S and N in CD4⁺ memory T cell subsets.

Total CD4⁺ memory T cells from seven COVID-19 recovered patients and six unexposed (pre-COVID-19) healthy donors (HD) or CD4⁺ Tcm, Tem and cTfh cells from seven COVID-19 recovered patients were labeled with CFSE and cultured with autologous monocytes in the presence or absence of recombinant SARS-CoV-2 S or N. (A) CFSE profiles on day 7 and percentage of CFSE^lo proliferating Tcm, Tem, and cTfh cells in a representative recovered patient. Negative controls of T cells cultured with monocytes in the absence of antigen are shown as red lines. (B) Individual values, median and quartile values of percentage of CFSE^loCD25⁺ICOS⁺ cells in total CD4⁺ memory and CD4⁺ Tcm, Tem and cTfh subsets in recovered patients and healthy donors. Shown are also IFN-γ concentrations in the day 7 culture supernatants of SARS-CoV-2 S- or N-stimulated memory T cell subsets. IFN-γ concentrations were below detection limit in HD and in negative control cultures. ****, P-value < 0.0001 ***, P-value < 0.001 **, P-value < 0.01 as determined by two-tailed unpaired t test (total CD4⁺ memory and IFN-γ) or by two-tailed paired t test (CD4⁺ Tcm, Tem, and cTfh). (C) Pairwise comparison of TCR Vβ clonotype frequency distribution in samples of T cells isolated from S-stimulated Tcm, Tem, or cTfh subsets (initial input 5×10⁵ cells per subset) from P33. Frequencies are shown as percentage of productive templates. The total number of clonotypes is indicated in the x- and y-axes. Values in the upper right corner represent the number of clonotypes shared between two samples. The Venn diagrams show the number of unique and shared clonotypes between Tcm, Tem and cTfh subsets. (D) The bar histograms show the Chao–Jaccard similarity index between pairs of TCR Vβ repertoires in three donors.

Fig. 2. CD4⁺ T cell clones target multiple sites on the S protein.

(A and B) CD4⁺ T cell clones (n=2943) were isolated from S-reactive cultures of 34 COVID-19 individuals and their specificity was mapped by stimulation with autologous B cells and three pools of 15-mer peptides overlapping of 10 spanning the S1-325 and S536-685 sequences (S1_ΔRBD, 91 peptides), the S316-545 sequence (RBD, 44 peptides), and the S676-1273 sequence (S2, 118 peptides), using as readout ³H-thymidine incorporation. (A) Characterization of representative T cell clones (n = 72) from P20. Proliferation was assessed on day 3 after a 16-hours pulse with ³H-thymidine and expressed as counts per min after subtraction of the unstimulated control value (Δcpm). (B) Percentage of T cell clones specific for S1_ΔRBD (white), RBD (black), and S2 (gray) in the 34 individuals tested. The number of clones tested is indicated on the right. The distribution of all S-reactive T cell clones isolated from all 34 individuals (ALL, n=2943) is also indicated. (C) RBD-specific T cell clones (n=1254) isolated from 32 individuals were further characterized for their epitope specificity using 15-mer peptides overlapping by 10 spanning the S316-545 RBD sequence. The 20-mer specificity of each clone is represented by a horizontal line and the total number of clones mapped for each individual is indicated on the right. (D) Percentage of clones specific and percentage of individuals carrying T cells specific for different 20-mer segments of the RBD. Data for the immunodominant region S346-365 is shown in black. (E) Sequence alignments of the SARS-CoV-2 immunodominant region S346-365 with homologous sequences in different Sarbecoviruses, human and animal SARS-related coronaviruses, and alpha and beta coronaviruses. Dots indicate identity to SARS-CoV-2 reference strain; dashes indicate deletions.

Fig. 3. The immunodominant S346-365 RBD region contains nested epitopes targeted by a diverse repertoire of T cells restricted by HLA-DR and HLA-DP.

(A and B) Rearranged TCR Vβ sequences of S346-365-reactive CD4⁺ T cell clones (n=329) isolated from 25 COVID-19-recovered individuals as determined by RT-PCR and Sanger sequencing. (A) TCR Vβ gene usage of the 206 unique clonotypes. Slices in the chart represent different Vβ genes and their size is proportional to the number of clonotypes using that particular gene. Color-coded legend is reported for the top 18 Vβ genes (used by at least five different TCR Vβ clonotypes). (B) Number of S346-365-reactive T cell clones and clonotypes identified in the 25 individuals. Slices in the charts represent different TCR Vβ clonotypes and their size is proportional to the number of sister clones bearing the same sequence. The number of clones is reported on top and the number of clonotypes at the center of the pie chart. (C) Frequency distribution of TCR Vβ clonotypes from CD4⁺ Tcm Tem and cTfh subsets sequenced directly after ex vivo isolation from P31 and P33. Colored circles mark the TCR Vβ clonotypes found among the S346-365-specific T cell clones isolated from the same individual. Dotted lines in the graphs indicate the frequency threshold of the top 5% expanded clonotypes. (D) HLA class II isotype restriction of S346-365-specific T cell clones (n=10) isolated from P33 as determined by stimulation with peptide-pulsed autologous APCs in the absence (control) or in the presence of blocking antibodies to HLA-DR, -DP, -DQ or pan HLA class II. Proliferation was assessed on day 3 after a 16-hour pulse with ³H-thymidine. Data are expressed as percentage of control counts per minute (cpm). (E) HLA class II isotype usage by S346-365-reactive CD4⁺ T cell clones (n=298) from 24 individuals, as determined by >80% inhibition of proliferation. (F) Identification of the minimal peptide recognized by HLA-DR or HLA-DP-restricted S346-365-reactive CD4⁺ T cell clones (n=23) isolated from seven individuals, as determined by stimulation with autologous APCs pulsed with a panel of truncated peptides. Proliferation was assessed on day 3 and expressed as counts per minute (cpm). Bars indicate mean + SD; circles indicate individual clones. The minimal amino acid sequences recognized by T cell clones are highlighted with colored shading.

Fig. 4. Identification of coronavirus S protein cross-reactive T cell clones.

(A and B) CFSE-labeled CD4⁺ memory T cells from P28 and P33 were stimulated with recombinant SARS-CoV-2 S protein in the presence of autologous monocytes. CFSE^lo cells were expanded with IL-2 for 10 days and relabeled and restimulated with S protein from human beta (SARS-CoV, HKU1, and OC43) or alpha (NL63 and 229E) coronaviruses. T cell clones from proliferating cultures were isolated and tested for cross-reactivity against different S proteins. (A) CFSE profiles from primary and secondary stimulation in the absence or in the presence of the indicated antigens. (B) Proliferative response (day 3 cpm) of representative T cell clones isolated from secondary cultures to autologous APCs pulsed with titrated doses of recombinant S proteins from SARS-CoV-2, SARS-CoV, or HKU1. (C to G) Multiple blood samples were obtained from donor P34 several years before and 1.5 months after COVID-19 and characterized as far as memory T cells and serum antibody levels. (C) T cell proliferation (CFSE dilution) in pre-COVID-19 (2018) and post-COVID-19 samples in response to autologous monocytes pulsed with different S proteins. (D) Time course of serum IgG antibodies against different coronavirus S proteins as determined by ELISA (EC50 values). These data demonstrate that together with a strong induction of serum antibodies to SARS-CoV-2, antibody titers against HKU1 and OC43 also increased in the post-COVID-19 sample. (E) Proliferative response to a pool of SARS-CoV-2 peptides of T cell clones obtained from post-COVID-19 CFSE^lo cultures stimulated by SARS-CoV-2, SARS-CoV, OC43 or NL63. Pie charts show the total number of clones tested and the fraction of responsive clones. (F) Reactivity of T cell clones isolated from each culture (in E) was further mapped by stimulation with pools of peptides spanning the S1_ΔRBD, RBD, and S2 regions of SARS-CoV-2 S protein. The histograms show the percentage of clones specific for each region. Total number of clones tested is indicated on top. (G) Characterization of cross-reactive T cell clones specific for S proteins isolated from P34 post-COVID-19 sample. The peptides recognized are indicated on the right panels. Shown are sequence alignment of the recognized SARS-CoV-2 epitopes (S816-830 and S981-1000) with homologous sequences of endemic alpha and beta coronaviruses. Dots indicate identity to SARS-CoV-2 reference strain.