T-Cell Epitope Mapping of SARS-CoV-2 Reveals Coordinated IFN-γ Production and Clonal Expansion of T Cells Facilitates Recovery from COVID-19

Background: T-cell responses can be protective or detrimental during severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection; however, the underlying mechanism is poorly understood. Methods: In this study, we screened 144 15-mer peptides spanning the SARS-CoV-2 spike, nucleocapsid (NP), M, ORF8, ORF10, and ORF3a proteins and 39 reported SARS-CoV-1 peptides in peripheral blood mononuclear cells (PBMCs) from nine laboratory-confirmed coronavirus disease 2019 (COVID-19) patients (five moderate and four severe cases) and nine healthy donors (HDs) collected before the COVID-19 pandemic. T-cell responses were monitored by IFN-γ and IL-17A production using ELISA, and the positive samples were sequenced for the T cell receptor (TCR) β chain. The positive T-cell responses to individual SARS-CoV-2 peptides were validated by flow cytometry. Results: COVID-19 patients with moderate disease produced more IFN-γ than HDs and patients with severe disease (moderate vs. HDs, p < 0.0001; moderate vs. severe, p < 0.0001) but less IL-17A than those with severe disease (p < 0.0001). A positive correlation was observed between IFN-γ production and T-cell clonal expansion in patients with moderate COVID-19 (r = 0.3370, p = 0.0214) but not in those with severe COVID-19 (r = −0.1700, p = 0.2480). Using flow cytometry, we identified that a conserved peptide of the M protein (Peptide-120, P120) was a dominant epitope recognized by CD8+ T cells in patients with moderate disease. Conclusion: Coordinated IFN-γ production and clonal expansion of SARS-CoV-2-specific T cells are associated with disease resolution in COVID-19. Our findings contribute to a better understanding of T-cell-mediated immunity in COVID-19 and may inform future strategies for managing and preventing severe outcomes of SARS-CoV-2 infection.


Introduction
Coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is associated with significant morbidity and mortality worldwide.With the global spread of SARS-CoV-2, the emergence of variants of concern (VOCs) poses significant challenges to epidemic control and development of vaccines and therapeutics [1].For example, mutations in the spike protein in the Omicron strain, particularly within the Receptor Binding Domain (RBD), lead to a reduction in neutralizing antibody affinity by 36-40 times compared to the wild-type strain [2][3][4][5].Moreover, the levels of SARS-CoV-2-specific B cells and neutralizing antibodies declined significantly over time in COVID-19 convalescents and vaccine recipients [6][7][8].Thus, neutralizing antibodies offer limited protection against VOCs.
The antigen-specific T-cell response plays a critical role in clearing SARS-CoV-2 and controlling disease progression [9,10].Early studies have shown that robust T-cell responses are associated with milder disease courses in COVID-19.Individuals with mild or asymptomatic COVID-19 have demonstrated significant T-cell responses despite having low antibody titers [11].Moreover, the critical role of T-cell responses in controlling the virus has been shown in B-cell-depleted COVID-19 patients and SARS-CoV-2-infected non-human primate models [12][13][14].However, dysregulation of the T-cell response has also been observed in severe COVID-19 cases.Evidence of increased total activated CD8+ T cells has been observed in some individuals with severe COVID-19, possibly indicating bystander CD8+ T-cell activation [15].An imbalance between circulating regulatory and cytotoxic SARS-CoV-2-specific CD4+ T cells is associated with greater COVID-19 disease [16].Therefore, a comprehensive investigation of the T-cell immune response in individuals with natural SARS-CoV-2 infection is crucial for understanding viral control and COVID-19 disease progression.
Over several hundred T-cell epitopes of SARS-CoV-2 have been described, and SARS-CoV-2 infection induces vigorous CD4+ and CD8+ T-cell responses, which exhibit remarkable breadth in most subjects [17].Some mutated T-cell epitopes in SARS-CoV-2 VOCs were not recognized by CD8+ T cells in infected or vaccinated individuals, and some individuals even lost the CD8+ T-cell response to the Omicron strain [18,19].Tarke et al. showed that more than 80% of T-cell reactivity was preserved at the population level and that, in most cases, no decrease was apparent [20].These results indicate that the remarkable breadth of human T-cell responses makes it exceedingly difficult for variants to escape T-cell recognition at the population level [21].However, the phenotypic diversity of epitope-specific T cells and their association with disease outcomes are poorly understood.
Thus, to study the T-cell response to SARS-CoV-2 antigens in COVID-19 patients and its correlation with disease severity, we synthesized 144 SARS-CoV-2 peptides covering spike, NP, M, ORF8, and ORF10 proteins, as well as 39 reported SARS-CoV-1 peptides.Subsequently, we screened their reactivity with peripheral blood mononuclear cells (PBMCs) of COVID-19 patients with moderate or severe disease, as well as healthy donors (HDs), by measuring IFN-γ and IL-17A production.We also analyzed T-cell expansion through T-cell receptor (TCR) sequencing and flow cytometry.Our results will improve our understanding of T-cell responses to SARS-CoV-2 infection.

Study Subjects
Thirteen laboratory-confirmed COVID-19 patients were hospitalized at the Fifth Medical Center of the Chinese PLA General Hospital in Beijing between 23 January and 26 May 2020, and nine healthy donors (HDs) were enrolled before the COVID-19 pandemic.All the subjects had no vaccination and no previous infection history.These patients were classified into two clinical groups, namely moderate and severe groups, according to the World Health Organization guidelines.Patients in the moderate disease group exhibited obvious clinical symptoms and pneumonia and were admitted to general wards but did not require intensive care.Patients in the severe disease group required critical care and met at least one of these criteria: dyspnea and respiratory rate ≥ 30/min; blood oxygen saturation ≤ 93%; PaO 2 /FiO 2 ratio < 300 mmHg; lung infiltrates on CT scan >50% within 24-48 h; or who exhibited respiratory failure, septic shock, and/or multiple organ dysfunction/failure.Discharge criteria were consistent with the guideline, including the resolution of respiratory symptoms, substantial improvement in chest CT images, afebrile for more than 3 days, and two consecutive negative RT-PCR results from respiratory tract swab samples collected at least 24 h apart.

Sample Collection
Blood samples were collected from COVID-19 patients and HDs and centrifuged at 400× g for 5 min at room temperature.Plasma samples were stored at −80 • C until use.After plasma collection, PBMCs were isolated via Ficoll density gradient centrifugation and stored at −80 • C and in liquid nitrogen subsequently.

Epitopes Screening
One hundred and forty-four peptides spanning the spike (n = 83) and NP protein (n = 28) without overlap and the M (n = 7), ORF8 (n = 19), and ORF10 (n = 5) proteins with an 8-amino acid overlap were synthesized.Two ORF3a peptides were also synthesized.A total of 39 reported SARS-CoV-1 peptides were synthesized, including 17 spike protein peptides, 14 NP protein peptides, 7 M protein peptides, and 1 ORF3a protein peptide.The peptide sequence information is available in Table A1.All peptides were resuspended in DMSO and stored at −80 • C.
PBMCs were thawed and rested overnight in AIM medium (Gibco, New York, NY, USA) supplemented with 10% heat-inactivated human AB serum (Thermo Fisher Scientific, Waltham, MA, USA).PBMCs (1.5 × 10 5 ) were seeded in a 96-well U bottom plate in AIM medium supplemented with 100 U/mL penicillin (Gibco), 0.1 mg/mL streptomycin (Gibco), 1000 IU/mL IL-2 (PeproTech, Rocky Hill, NJ, USA), 5 ng/mL IL-15 (PeproTech), 0.5 ng/mL IL-21 (PeproTech), and 1 µg/mL of one peptide (Genescript, Nanjing, China) and cultured at 37 • C in 5% CO 2 .An equimolar amount of DMSO was used to stimulate the cells as a negative control, and 1 mg/mL phytohemagglutinin (PHA, Roche, Basel, Switzerland) was used as the positive control.Heat-inactivated human AB serum was added the following morning to a final concentration of 5%.Seven days after incubation, the supernatant was collected and stored at −80 • C for enzyme-linked immunosorbent assay (ELISA).IFN-γ and IL-17A in the supernatant were assessed via ELISA (DAKEWE, Shenzhen, China) following the manufacturer's instructions.The remaining cells were lysed using TRIzol reagent and stored at −80 • C for TCR sequencing and human leukocyte antigen (HLA) genotyping.

Next-Generation Sequencing (NGS) of the TCR β Gene
Total RNA was extracted from each sample and subjected to RT-PCR amplification of the TCRβ gene using a QIAGEN one-step RT-PCR kit (QIAGEN, Hilden, Germany), with Cβ anti-sense primers and 36 Vβ mix primers used for two rounds of PCR.The resulting size-corrected PCR products were purified using magnetic beads (BioMagBeads, Wuxi, China).A sequencing library was prepared and purified using a Thermo Fisher Ion Plus Fragment Library Kit (Thermo Fisher Scientific), Ion Xpress Barcode Adaptors 1-16 Kit (Thermo Fisher Scientific), and Agencourt AMPure XP (Beckman Coulter, Brea, CA, USA).NGS was performed on an Ion S5 system using an Ion 530 Chip kit (Thermo Fisher Scientific).

TCR β Gene Sequence Analyses
The quality of the sequences obtained was monitored by cross-analyzing for potential contamination.Sequences shorter than 150 bp were excluded from analysis.TCR Vβ, Dβ, and Jβ germline gene assignment was conducted using a locally operating IgBlast program at Chengdu ExAb Biotechnology, Ltd.Non-functional TCR β VDJ sequences were removed, and 30,000 randomly selected TCR genes were analyzed.

HLA Genotyping
Genomic DNA (200 ng) from each sample was sheared using a Biorupter (Diagenode, Liège, Belgium) to acquire 150~200 bp fragments.The ends of the DNA fragments were repaired, and an Illumina Adaptor was added (Fast Library Prep Kit; iGeneTech, Beijing, China).After constructing the sequencing library, the target region was captured using an AI-HLA-Cap Enrichment Kit (iGeneTech) and sequenced on an Illumina platform (Illumina, San Diego, CA, USA) with 150 base paired-end reads.The raw reads were filtered using FastQC to remove low-quality reads.Clean reads were then mapped to the reference sequences in the HLA dictionary and typed to generate HLA types for HLA-A, -B, -C, -DPB1, -DQB1, and -DRB1 using HLA-HD Software (Version 1.5.0,Tokyo, Japan).

Ethical Approval
This study was approved by the Ethics Committee of the Fifth Medical Center of the Chinese PLA General Hospital (S2020-044-02).Informed consent was obtained from all enrolled participants or their legal guardians.

Statistical Analysis
GraphPad Prism statistical software (version 8.0; GraphPad Software, USA) and SPSS version 25 (SPSS Inc., USA) were used for statistical analysis.Spearman's rank correlation was performed using the 'ggcor' package of the R software.Categorical variables were compared using the χ 2 test or Fisher's exact test and are reported as counts and percentages.Continuous variables were compared using the Mann-Whitney U test and Kruskal-Wallis H test and are presented as the median and interquartile range (IQR).Statistical significance was defined as a two-tailed p-value < 0.05.

T Cells of Patients with Moderate Disease Had a Robust Response towards SARS-CoV-2 Antigens
To reveal the spectrum of T-cell responses to SARS-CoV-2 in naturally infected patients, we synthesized 144 peptides spanning the spike, NP, M, ORF8, ORF10, and ORF3a proteins.As the SARS-CoV-2 genome is highly homogenous with that of the SARS-CoV-1 virus, 39 reported SARS-CoV-1 peptides were synthesized (Table A1).Nine laboratory-confirmed and hospitalized COVID-19 patients (four severe and five moderate cases) were enrolled for epitope screening.Blood samples from nine HDs collected before the SARS-CoV-2 pandemic were used as controls.Demographic information and HLA alleles are listed in Tables 1 and A2, and the timeline of the course of the disease is shown in Figure A1.PBMCs from HDs and COVID-19 patients were stimulated with each peptide and cultured for 7 days.IFN-γ in the supernatant was assessed using ELISA, and the experimental procedure is shown in Figure 1A.A positive response was defined as an IFN-γ concentration two-fold higher than that in the unstimulated sample.All samples that had a positive response were sequenced for the TCR β chain V-D-J gene usage and HLA allele.The IL-17A concentration in these positive samples was also analyzed using ELISA.The corresponding peptide responses were validated in short-term cells via flow cytometry.The IFN-γ secretion of PBMCs in patients with moderate COVID-19 was higher than that in patients with severe COVID-19 and HDs (moderate vs. HD, p < 0.0001; moderate vs. severe, p < 0.0001) (Figure 1B,C).Venn diagrams showed the number of responsive peptides detected in healthy donors (HDs), COVID-19 patients, as well as COVID-19 patients with moderate or severe disease.Of the 183 peptides, 92 exhibited reactivity in both HDs and COVID-19 patients, while 34 were recognized by HD and 23 were recognized by COVID-19 patients (Figure A2A).Notably, PBMCs from patients with moderate disease recognized more peptides than those from patients with severe disease (84 vs. 63) (Figure A2C).However, no obvious pattern of antigen specificity was observed between HDs and COVID-19 patients or between patients with moderate and severe disease (Figure A2B,D).

T-Cell Response of Patients with Severe Disease Produced Higher Levels of IL-17A
In the response-positive samples, the IFN-γ secreted by PBMCs of patients with moderate disease was higher compared to that of patients with severe disease (203.49pg/mL [102.45,286.08]vs. 50.22pg/mL [33.25, 47.38]; p < 0.0001) (Figure 2A).As IL-17A is reported to be involved in COVID-19 disease progression [22], we analyzed the levels of IL-17A in the supernatant of positive samples.Interestingly, PBMCs from patients with severe disease expressed higher levels of IL-17A than those with moderate disease (50.03 pg/mL [22.91, 79.75] vs. 21.00pg/mL [13.89, 29.93]; p = 0.0002) (Figure 2B).Then, we analyzed the bias of epitopes to induce IFN-γ or IL-17A production.The data in Figure 2C reveal a complex tendency in the secretion of IFN-γ or IL-17A induced by individual peptides, which is dependent on disease severity.IL-17A was more pronounced in patients with severe disease than in those with moderate disease.
To assess the antigen-specific T-cell expansion after antigen peptide stimulation, we  The IFN-γ secretion of PBMCs in patients with moderate COVID-19 was higher than that in patients with severe COVID-19 and HDs (moderate vs. HD, p < 0.0001; moderate vs. severe, p < 0.0001) (Figure 1B,C).Venn diagrams showed the number of responsive peptides detected in healthy donors (HDs), COVID-19 patients, as well as COVID-19 patients with moderate or severe disease.Of the 183 peptides, 92 exhibited reactivity in both HDs and COVID-19 patients, while 34 were recognized by HD and 23 were recognized by COVID-19 patients (Figure A2A).Notably, PBMCs from patients with moderate disease recognized more peptides than those from patients with severe disease (84 vs. 63) (Figure A2C).However, no obvious pattern of antigen specificity was observed between HDs and COVID-19 patients or between patients with moderate and severe disease (Figure A2B,D).

T-Cell Response of Patients with Severe Disease Produced Higher Levels of IL-17A
In the response-positive samples, the IFN-γ secreted by PBMCs of patients with moderate disease was higher compared to that of patients with severe disease (203.49pg/mL [102.45,286.08]vs. 50.22pg/mL [33.25, 47.38]; p < 0.0001) (Figure 2A).As IL-17A is reported to be involved in COVID-19 disease progression [22], we analyzed the levels of IL-17A in the supernatant of positive samples.Interestingly, PBMCs from patients with severe disease expressed higher levels of IL-17A than those with moderate disease (50.03 pg/mL [22.91, 79.75] vs. 21.00pg/mL [13.89, 29.93]; p = 0.0002) (Figure 2B).Then, we analyzed the bias of epitopes to induce IFN-γ or IL-17A production.The data in Figure 2C reveal a complex tendency in the secretion of IFN-γ or IL-17A induced by individual peptides, which is dependent on disease severity.IL-17A was more pronounced in patients with severe disease than in those with moderate disease.
To assess the antigen-specific T-cell expansion after antigen peptide stimulation, we sequenced the TCR β chain of the responsive sample and found that the proportion of the top immunodominant TCR β V-J gene pair (top TCR clone) was comparable between patients with moderate and severe disease (8.10% [4.97, 10.62] vs. 7.21% [5.62, 13.38], p = 0.47) (Figure 2D).The correlation between the proportion of top TCR clones and cytokine production was analyzed.The data in Figure 2E show a positive correlation between TCR expansion and IFN-γ production in patients with moderate disease (r = 0.38, p = 0.021) but not in those with severe disease (r = −0.17,p = 0.25) (Figure 2E, Left).No correlation between TCR expansion and IL-17A expression was observed in patients with moderate or severe disease (Figure 2E, Right).These data suggest that concerted IFN-γ production and TCR expansion in patients with moderate disease may favor disease resolution.
duction was analyzed.The data in Figure 2E show a positive correlation between TCR expansion and IFN-γ production in patients with moderate disease (r = 0.38, p = 0.021) but not in those with severe disease (r = −0.17,p = 0.25) (Figure 2E, Left).No correlation between TCR expansion and IL-17A expression was observed in patients with moderate or severe disease (Figure 2E, Right).These data suggest that concerted IFN-γ production and TCR expansion in patients with moderate disease may favor disease resolution.

COVID-19 Patients Broadly Recognized Conserved P120 Peptide
We then validated the T-cell response towards SARS-CoV-2 peptides in short-term T cells of COVID-19 patients using flow cytometry.Twenty broadly recognized peptides identified (Table A4) in Figure 1D were validated in short-term T-cell lines from six COVID-19 patients, including four with moderate and two with severe disease.The demographic information and HLA alleles are listed in Tables A2 and A3, and the timeline of the course of the disease is also shown in Figure A1.Compared to other peptides, the P120-peptide (P120) from the M protein (M148-162: HLRIAGHHLGRCDIK) elicited a broad and robust IFN-γ expression in CD8+ T cells in COVID-19 (Figure 3A).The P120 peptide was primarily recognized by CD8+ T cells of patients with moderate COVID-19 compared

COVID-19 Patients Broadly Recognized Conserved P120 Peptide
We then validated the T-cell response towards SARS-CoV-2 peptides in short-term T cells of COVID-19 patients using flow cytometry.Twenty broadly recognized peptides identified (Table A4) in Figure 1D were validated in short-term T-cell lines from six COVID-19 patients, including four with moderate and two with severe disease.The demographic information and HLA alleles are listed in Tables A2 and A3, and the timeline of the course of the disease is also shown in Figure A1.Compared to other peptides, the P120-peptide (P120) from the M protein (M 148-162 : HLRIAGHHLGRCDIK) elicited a broad and robust IFN-γ expression in CD8+ T cells in COVID-19 (Figure 3A).The P120 peptide was primarily recognized by CD8+ T cells of patients with moderate COVID-19 compared to those with severe COVID-19 (Figure 3B,C).It also induced a slight CD4+ T-cell response in patients with moderate COVID-19 (Figure 3D).Furthermore, the data in Figure 1B show that P120 induced higher IFN-γ secretion in the supernatant of PBMCs from patients with moderate disease.The median concentration of IFN-γ was 17.28 pg/mL in HDs, 64.78 pg/mL in patients with moderate disease, and 28.91 pg/mL in patients with severe disease.We used the Immune Epitope Database & Analysis Resource (IEDB) to predict potential epitopes within P120.Table A5 showed that P120 contains four epitopes (HLRIAGHHL, HLRIAGHHLGR, LRIAGHHLGR, and RIAGHHLGR) that could be presented by 15 A6).

Discussion
In this study, we evaluated the reactivity of 183 SARS-CoV-2 and SARS-CoV-1 derived peptides towards PBMCs from nine COVID-19 patients and nine HDs.We found that T cells in patients with moderate disease had strong IFN-γ production with SARS-CoV-2 peptides, while patients with severe disease had higher IL-17A production.Although TCR amplification following antigen restimulation in vitro did not differ by COVID-19 disease severity, a positive correlation between TCR expansion and IFN-γ secretion was observed in patients with moderate but not severe disease.We identified one peptide (P120, M 148-162 ) from the M protein broadly recognized by COVID-19 patients, especially those with moderate disease.The P120 peptide is conserved among SARS-CoV-2 variants, such as Delta and Omicron strains, and is presented by diverse HLA alleles.These results indicate that it is a promising target for promoting T-cell response and COVID-19 disease control.
Antigen-specific T cells are critical in the fight against viral infections.Associations between early T-cell responses and less severe COVID-19 outcomes have been observed [23,24].We found that PBMCs in patients with severe disease produced less IFN-γ but higher levels of IL-17A compared to those of patients with moderate disease, despite similar levels of T-cell clone expansion between these two groups.Furthermore, a positive correlation was observed between the abundance of the top TCR β chain and IFN-γ production in patients with moderate disease, suggesting that coordinated IFN-γ production and clonal expansion of SARS-CoV-2-specific T cells are associated with disease control in COVID-19 patients.Conversely, such correlation was absent in severe cases, indicating that Th17 may be involved in the expanded T cells' population, which may contribute to disease progression.COVID-19 patients with severe disease are characterized by inflammation and tissue damage in the respiratory tract, with monocyte and lymphocyte infiltration observed in the lungs [25][26][27][28].Th17 cells characterized by high IL-17A and GM-CSF expression are enriched and clonally expanded in the lungs of COVID-19 patients with severe disease, potentially contributing to lung damage through interacting with CD8+ T cells and macrophages [22].Additionally, IL-17A and Th17 cells are involved in the pathogenesis of MERS-CoV, SARS-CoV-1, and H1N1 infection [29][30][31].
Spike, NP, M, and NSP3 proteins are the major T-cell antigens of the SARS-CoV-2 virus, and differences exist in the phenotype and functional characteristics of T cells specific to different antigens [32].Spike-specific T cells tend to differentiate into T follicular helper (TFH) cells, suggesting a crucial role in generating potent antibody responses, while M protein-specific and NP-specific CD4+ T cells are skewed toward a Th1 or Th1/Th17 profile [11].Most studies have not distinguished the differences in T-cell responses induced by different SARS-CoV-2 antigens.We found that the IFN-γ and IL-17A production induced by the SARS-CoV-2 peptide was severity-dependent but not epitope-dependent.The bias of IL-17A production in patients with severe disease persisted even after several days of in vitro culture, and this phenomenon indicated that a stable program toward IL-17A production was established in patients with severe disease.This program may contribute to the elevated IL-17A in critical patients and be associated with fatal outcomes [22,33].Furthermore, IL-17A has been implicated in promoting alveolar epithelial cell apoptosis and exacerbating the progression of pulmonary fibrosis, disrupting normal alveolar architecture and impairing alveolar-capillary gas exchange.Consequently, these processes adversely affect the normal oxygenation process, contributing to the respiratory symptoms [34].
After careful screening and validation, we identified a peptide (P120, M 148-162 : HLRIA GHHLGRCDIK) from the M protein, which was broadly recognized by COVID-19 patients with moderate disease and may contribute to disease control in SARS-CoV-2 infection.Similar to a prior study, a prevalent HLA-A*24:02-restricted epitope (M 198-206 ) from the M protein was associated with convalescence in COVID-19 patients admitted to intensive care.While the reactivity of M 198-206 was observed in all COVID-19 patients, CD8+ T cells specific to the M 198-206 epitope exhibited an exhausted phenotype in severe cases compared to those in the moderate group [35].The dysfunction of T cells in SARS-CoV-2 patients with severe disease has also been reported in the HLA-B*07:02-restricted NP 105-113 epitope-responsive CD8+ T cells, which had a weaker response in patients who recovered from severe disease compared to those with mild disease [36].The mechanism underlying the diminished responses of antigen-specific T cells in COVID-19 patients with severe disease remains unclear.It is hypothesized that the dysregulation of Tumor Necrosis Factorα/Tumor Necrosis Factor Receptor 1 signaling, upregulation of inhibitory receptors, and prolonged antigen exposure may contribute to the dysfunction observed in antigen-specific T-cell responses in COVID-19 patients with severe disease [35,37,38].
The TCR β chain of a P120-responsive sample was sequenced, revealing that the predominant TCR V-J gene pair was TRBV18-TRBJ2-5, constituting 26% of the TCR β chains.TRBV18-TRBJ2-5 TCR was exclusively and prominently expanded in the P120-responsive sample compared with other positive samples in Pt02.We speculated that the HLRIAGHHL epitope in P120 might be presented by HLA-A*30:01 and drive the TRBV18-TRBJ2-5 TCR expansion in the in vitro culture assay.In a study on human immunodeficiency virus, researchers noted a significant TCR bias toward the HLA-B*0702-FPQGEAREL epitope, with TRBV18 and TRBJ2-5 being preferentially utilized.The TRBV18/CASSPRGREETQY/TRBJ2-5 clonotype retains functionality and persistence, even when antigen levels decay dramatically after antiretroviral therapy [48].Therefore, the T cell of TRBV18-TRBJ2-5 expanded by the P120 peptide may preferentially persist under conditions of limited antigenic stimulation and provide long-term protection.
Our study had several limitations.While our peptide synthesis strategy avoided the bioinformatics bias of the epitope prediction algorithm, we only screened peptides derived from SARS-CoV-2 spike, NP, M, ORF8, and ORF10 proteins, with no overlap in spike and NP protein peptides.In addition, the epitopes of the ORF1a and ORF1b antigens were not screened.We only analyzed the correlation between TCR expansion and cytokine production.A comprehensive and detailed analysis of the relationship among HLA alleles, individual peptide responses, and TCR β chain sequences is needed to elucidate the comprehensive immune response of SARS-CoV-2.Moreover, our epitope screening included only nine COVID-19 patients and nine healthy donors (HDs), with significant variation in sampling time among COVID-19 patients.Additionally, glucocorticoid usage in COVID-19 patients with severe disease may weaken our conclusions.Our findings need to be validated in a larger cohort of COVID-19 patients.
Funding: This research was funded by PLA General Hospital (22QNCZ011).

Institutional Review Board Statement:
The study was conducted in accordance with the Decla-ration of Helsinki and approved by the Ethics Committee of the Fifth Medical Center, Chinese PLA General Hospital (S2020-044-02).
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Viruses 2024 , 22 Figure 1 .
Figure 1.T cells of patients with moderate disease had a robust response towards SARS-CoV-2 antigens.(A) A schematic showing the overall study design.(B) IFN-γ in supernatant of cultured PBMCs from HD and COVID-19 patients with moderate or severe disease were assessed by ELISA, and antigen information was labeled along the X axis.(C) The mean value of IFN-γ concentration of cultured PBMCs from HD and COVID-19 patients with moderate or severe disease induced by each peptide assessed by ELISA.(D) Violin plot showing the response rate of each peptide in HD and COVID-19 patients with moderate or severe disease, two times higher IFN-γ concentration than the unstimulated sample was recognized as a positive response.S1: spike S1 domain; S2: spike S2 domain.Data are expressed as mean ± SD. and **** p < 0.0001 by two-tailed Mann-Whitney U-test.

Figure 1 .
Figure 1.T cells of patients with moderate disease had a robust response towards SARS-CoV-2 antigens.(A) A schematic showing the overall study design.(B) IFN-γ in supernatant of cultured PBMCs from HD and COVID-19 patients with moderate or severe disease were assessed by ELISA, and antigen information was labeled along the X axis.(C) The mean value of IFN-γ concentration of cultured PBMCs from HD and COVID-19 patients with moderate or severe disease induced by each peptide assessed by ELISA.(D) Violin plot showing the response rate of each peptide in HD and COVID-19 patients with moderate or severe disease, two times higher IFN-γ concentration than the unstimulated sample was recognized as a positive response.S1: spike S1 domain; S2: spike S2 domain.Data are expressed as mean ± SD. and **** p < 0.0001 by two-tailed Mann-Whitney U-test.ns, non-significant.

Figure 2 .
Figure 2. T-cell response of patients with severe disease produced higher levels of IL-17A.(A,B) IFN-γ (A) and IL-17A (B) of the positive-response sample was assessed by ELISA assay.(C) Spearman's rank correlation between IFN-γ and IL-17A concentration in the supernatant of 6 representative samples was shown in dot plot.The dot shapes indicated representative peptides, the red indicated patients with severe disease, and the blue indicated patients with moderate disease.(D) The top TCR clone (immunodominant TCR β chain V-J gene pair) was analyzed between patients with moderate and severe disease.(E) Spearman's rank correlation of the top TCR clone and IFN-γ or IL-17A was analyzed between patients with moderate and severe disease.Data are expressed as mean ± SD. **** p < 0.0001 by two-tailed Mann-Whitney U-test.

Figure 2 .
Figure 2. T-cell response of patients with severe disease produced higher levels of IL-17A.(A,B) IFN-γ (A) and IL-17A (B) of the positive-response sample was assessed by ELISA assay.(C) Spearman's rank correlation between IFN-γ and IL-17A concentration in the supernatant of 6 representative samples was shown in dot plot.The dot shapes indicated representative peptides, the red indicated patients with severe disease, and the blue indicated patients with moderate disease.(D) The top TCR clone (immunodominant TCR β chain V-J gene pair) was analyzed between patients with moderate and severe disease.(E) Spearman's rank correlation of the top TCR clone and IFN-γ or IL-17A was analyzed between patients with moderate and severe disease.Data are expressed as mean ± SD. **** p < 0.0001 by two-tailed Mann-Whitney U-test.

Figure 3 .Figure 3 .
Figure 3. COVID-19 patients broadly recognized the conserved P120 peptide.(A) The short-term cell line was incubated with 20 peptides (10 ug/mL) for 5 h, and IFN-γ expression was measured by Figure 3. COVID-19 patients broadly recognized the conserved P120 peptide.(A) The short-term cell line was incubated with 20 peptides (10 ug/mL) for 5 h, and IFN-γ expression was measured by flow cytometry.(B) Representative flow cytometry results of IFN-γ expression in CD4+ or CD8+ T cells induced by P120 peptide in patients with moderate and severe disease.(C,D) The P120-specific CD4+ or CD8+ T-cell level was analyzed by flow cytometry.The value of antigen-specific T cells was calculated by subtracting the negative control.Data are expressed as mean ± SD.

22 (
Figure4C), rendering it a promising candidate for T-cell activation to confer protection against prevalent SARS-CoV-2 variants.

Figure 4 .
Figure 4. P120 peptide was conserved among SARS-CoV-2 variants and induces strong TCR expansion.(A) Circos plot depicting each β VJ gene usage of P120 responsive sample.(B) The proportion of the top 4 CDR3 sequences of the short-term cell of the P120 peptide.(C) Alignment of the amino acid sequences of original SARS-CoV-2 P120 peptides with Delta, Omicron variant, SARS-CoV-1, and endemic human coronaviruses.

Figure 4 .
Figure 4. P120 peptide was conserved among SARS-CoV-2 variants and induces strong TCR expansion.(A) Circos plot depicting each β VJ gene usage of P120 responsive sample.(B) The proportion of the top 4 CDR3 sequences of the short-term cell of the P120 peptide.(C) Alignment of the amino acid sequences of original SARS-CoV-2 P120 peptides with Delta, Omicron variant, SARS-CoV-1, and endemic human coronaviruses.

Figure A3 .
Figure A3.Eight dominant TCRs among all positive samples of Pt02 are presented.Different colors of the dot represent different TCRs as indicated on the right side of the panel.
HLA alleles in bold and grey shadow represent predicted HLA alleles with P120 peptide binding activity.