Skip to main content

Identification of SARS-CoV-2-specific T cell and its receptor


The T-cell receptor (TCR) repertoires exhibits distinct signatures associated with COVID-19 severity. However, the precise identification of vaccine-induced SARS-CoV-2-specific TCRs and T-cell immunity mechanisms are unknown. We developed a machine-learning model that can differentiate COVID-19 patients from healthy individuals based on TCR sequence features with an accuracy of 95.7%. Additionally, we identified SARS-CoV-2-specific T cells and TCR in HLA-A*02 vaccinated individuals by peptide stimulation. The SARS-CoV-2-specific T cells exhibited higher cytotoxicity and prolonged survival when targeting spike-pulsed cells in vitro or in vivo. The top-performing TCR was further tested for its affinity and cytotoxic effect against SARS-CoV-2-associated epitopes. Furthermore, single-cell RNA sequencing (scRNA-seq), immune repertoire sequencing (IR-seq) and flow cytometry were used to access vaccine-induced cellular immunity, which demonstrated that robust T cell responses (T cell activation, tissue-resident memory T cell (Trm) generation, and TCR clonal expansion) could be induced by intranasal vaccination. In summary, we identified the SARS-CoV-2-associated TCR repertoires profile, specific TCRs and T cell responses. This study provides a theoretical basis for developing effective immunization strategies.

To the editor

Coronavirus Disease 2019 (COVID-19) is a global public health concern caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Despite significant emphasis on vaccine inoculation globally, vaccine-induced neutralizing antibody immunity alone has proven insufficient to prevent SARS-CoV-2 infection [1, 2]. Accumulating evidences demonstrate the critical role of coronavirus-specific T lymphocytes for recovery and long-term protection [3]. SARS-CoV-2 vaccination has triggered a robust and enduring T cell response that can effectively recognize variants from Alpha to Omicron [4]. Recent study indicates a disease severity-dependent TCR clonal expansion pattern in COVID-19 patients, demonstrating that the disease-specific TCRs is required for symptomatic relief [5]. However, the landscape of T-cell receptor (TCR) repertoires in COVID-19 and the TCRs responsible for recognizing SARS-CoV-2 remain uncertain [6, 7].

This study, as illustrated in Additional information, Fig. S1, was designed to address these uncertainties. Initially, we conducted a comprehensive analysis of peripheral blood TCR repertoire in various groups, including healthy controls and individuals at different stages of SARS-CoV-2 infection (comprising 54 healthy, 103 acute, 90 transition, and 108 convalescent patients), utilizing data from the ImmuneACCESS and ImmuneCODE databases (Additional information, Table S1). Comparison of TCR repertoire differences, the overlap ratio and complementary-determining region 3 (CDR3) amino acid usages between acute and transition groups (0.110) was more similar than others (Additional information, Fig. S2A, B). Besides, infected patients revealed significant differences in TCR repertoire distribution compared with healthy controls (Fig. 1A, B). Notably, TCR patterns in patients indicated a predilection for high-frequency clusters, while controls exhibited different TCR usage profiles characterized by a predilection for low-frequency clusters, attributed to increased TCRs diversity following SARS-CoV-2 infection (Fig. 1C; Additional information, Fig. S2C-E). Moreover, we developed a machine-learning model that could accurately differentiate COVID-19 patients from healthy individuals based on TCR sequence features, achieving an impressive area under the receiver operating characteristics (ROC) curve value of 95.7% (Fig. 1D, E). Intriguingly, we observed similarities in TCR repertoires when comparing TCR sequences after SARS-CoV-2 infection and vaccination, suggesting the potential for specific T cell and TCR identification post-SARS-CoV-2 vaccination (Additional information, Fig. S2F).

Fig. 1
figure 1

The characteristics of SARS-CoV-2-specific T cell and TCR repertoire. A Principal Component Analysis (PCA) visualization of TCR sequences obtained from the ImmuneACCESS and ImmuneCODE databases, comparing healthy donors (n = 54) with patients at different stages of infection (acute n = 103, transition n = 90, and convalescent n = 108). B Levenshtein distances depict TCR clone similarities between acute and transition groups and diversity in TCR clones between acute and healthy groups. C Rank-abundance curve illustrating TCR diversity. D Machine learning framework for analyzing TCR sequences from COVID-19 patients and healthy individuals. E Performance evaluation of machine learning models for predicting SARS-CoV-2 infection. F Bar graphs representing the proportion of peptide-specific CD8+ T cells in HLA-A*02+ healthy donors before and after immunization. G Representative flow plot showing the percentage of specific T cells (CD8+XG2+) after peptide stimulation. H Cytotoxic activities of XG2+ T cells against BEAS-2B-spike cells assessed at different effector/target (E/T) ratios. I Quantification of lysis rates when co-cultured with BEAS-2B-spike cells (n = 4). J Top 5 CDR3 amino acid sequences predicted by IR-seq and DLpTCR. K Cytotoxic activities of TCR-T cells against BEAS-2B-spike cells assessed by flow cytometry. L Quantification of TCR-T cell lysis rates when co-cultured with BEAS-2B-spike cells (n = 8). M Comparison of phospho-ZAP70, ZAP70, phospho-AKT, AKT, and β-Tubulin expression in TCR-T cells by immunoblot analysis (n = 3). Data are representative of at least three independent experiments

To identify SARS-CoV-2-specific T cells and TCRs, we employed a multiplexed peptide-MHC tetramer staining approach to screen 8 spike or nucleocapsid protein (XG1-XG8) for recognition by T cell responses with HLA allele HLA-A*02, the most common HLA class I allele in China [8]. Booster vaccinations notably enhanced T cell activation (Fig. 1F), with the SLSSTASAL peptide (XG2 peptide, one of peptide from spike protein) demonstrating the most robust expansion of CD8+XG2+ T cells and heightened cytokine expression (IL2, GZMB, GZMK, IFNG and TNF) (Fig. 1G; Additional information, Fig. S3A). To assess the cytotoxicity of XG2+ T cells, we co-cultured them with epithelial cells (BEAS-2B or SV-HUC-1) expressing spike protein by lentivirus (pCDH-EF1a-spike-GFP) infection. (Additional information, Fig. S3B). Compared to XG2 T cells, XG2+ T cells exhibited higher cytotoxicity and prolonged survival when targeting spike-pulsed cells (Fig. 1H, I; Additional information, Fig. S3C). Immune repertoire sequencing (IR-seq) and a deep learning framework for predicting immunogenic peptide recognized by TCR (DLpTCR) approaches were used to determine specific TCR clonotype from XG2+ T cells (Additional information, Table S2). Compared with XG2 T cells, XG2+ T cells showed the significant decrease in VJ and CDR3 amino acid usage after vaccination (Additional information, Figs. S3D-G). We identified the top 5 high-probability CDR3 amino acid sequences binding to the SLSSTASAL peptide (Fig. 1J). Subsequently, one of these high-probability TCRs (TRA CDR3, CILNNNNDMRF; TRB CDR3, CASSEFSGRMNTEAFF) was overexpressed in CD8+ T cells (Additional information, Fig. S3H, I), leading to enhanced cytolytic activity against target cells (Fig. 1K, L) with the elevated phospho-ZAP70 (Tyr319) and phospho-AKT (Ser473) (downstream of TCR signaling) (Fig. 1M).

To evaluate the T cell responses in the lower respiratory tract elicited by specific peptides, we immunized mice intranasally with the SLSSTASAL peptide (Additional information, Fig. S4A). Lung mononuclear cells were collected at 1, 7 and 30 days post-immunization for scRNA-seq and IR-seq (Additional information, Table S3, S4). Compared to non-immunized individuals, peptide-stimulated pulmonary tissues displayed increased fractions of total, central memory (Tcm), effector memory (Tem), and tissue-resident memory T cells (Trm) in the early days (1 and 7 days) (Fig. 2A, B; Additional information, Fig. S4B) without inducing tissue injury or inflammatory responses (Additional information, Fig. S4C-E). These T cells also exhibited high activation genes and various cytokine genes expressions (Ccl5, Cxcl10, Cxcl16, Gzmb, Gzmk, Ifng, and Nkg7) after 7 days post-immunization, similar to XG2+ T cells from humans (Fig. 2C; Additional information, Fig. S4F, G). Flow cytometry further confirmed a significant increase in the percentage of memory T cells and T cell activation (Fig. 2D, E; Additional information, Fig. S4H). Although the effect of T cell activation diminished after 30 days post-immunization, Trm cells were still detectable (Fig. 2D, E; Additional information, Fig. S4I-K). Additionally, we evaluated the pulmonary TCR repertoire on 0 day, 7 days, 30 days after intranasal immunization. Vaccination enhanced TRBV12-1 usage and reduced TRBV1 usage (Fig. 2F). Similar with TCRs expansion in COVID-19 patients, antigenic stimulation significantly augmented TCRs diversity on 7 day post-immunization (Fig. 2G, H), leading to similar CDR3 amino acids usage (including SHDR%TE, SD%RNTE, SDH%NTE, and S%HRNTE) (Fig. 2I-L). Taken together, antigen exposure induced significant expansion of TCR clonotypes in local pulmonary tissues, suggesting that epitope-specific Trm responses could provide long-term protection against SARS-CoV-2 infection.

Fig. 2
figure 2

Intranasal immunization enhances T cell response. A UMAP plot of scRNA-seq displaying pulmonary T cell subsets at 0, 1, 7, and 30 days post intranasal immunization in mice (n = 3). B Proportion of T cell subsets in pulmonary tissue at indicated time-point. C Bubble plot showing the expression of T cell activation genes at 0 and 7 days post-immunization. D Representative flow plots depicting CD8+ memory T cells and Trm in pulmonary tissues. E The percentage of various T cell subsets in pulmonary tissue after vaccination (n = 8). F Analysis of TCR repertoires in pulmonary tissues by IR-seq (n = 5). Pie graph displaying the top10 V and J gene usages after 0 (left), 7 (middle), and 30 days (right) immunization. G TCR diversity based on CDR3 amino acid clonotypes (left), Chao1 (middle), d50 (right). H Rank abundance analysis of TCR clonotypes. I-K Bias analysis of CDR3 amino acid motif. L Heatmap hierarchical clustering of CDR3 amino acid sequences. Data are representative of three independent experiments. Data are representative of at least three independent experiments

In summary, our study introduces a machine-learning approach capable of accurately predicting COVID-19 infection severity based on TCR sequence features. We successfully identified SARS-CoV-2-specific T cells and their CDR3 sequences from human peripheral blood and observed a robust memory T cell response in local pulmonary tissues. Furthermore, we cloned specific TCR sequences in CD8+ T cells and established highly efficient TCR-T cells. Our research introduces an autonomous TCR screening platform capable of identifying precise TCR sequences that bind to specific HLA-peptide complexes. Leveraging this platform, we can similarly pinpoint neoantigen-associated TCRs in various diseases, including cancer, infections, and autoimmune conditions.

Data availability

No datasets were generated or analysed during the current study.



Coronavirus Disease 2019


Severe acute respiratory syndrome coronavirus 2


T cell receptor


Complementary-determining region 3


Receiver operating characteristics


Principal component analyses


Immune repertoire sequencing


Deep learning framework for predicting immunogenic peptide recognized by TCR


Single-cell RNA sequencing


Central memory T cell


Effector memory T cell


Tissue-resident memory T cells


  1. Garcia-Beltran WF, Lam EC, St Denis K, Nitido AD, Garcia ZH, Hauser BM, et al. Multiple SARS-CoV-2 variants escape neutralization by vaccine-induced humoral immunity. Cell. 2021;184:2372–e23839.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  2. Rooney A, Bivona C, Liu B, Streeter D, Gong H, Khan Q. Risk of SARS-CoV-2 breakthrough infection in Vaccinated Cancer patients: a retrospective cohort study. J Hematol Oncol. 2022;15:67.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  3. Heitmann JS, Bilich T, Tandler C, Nelde A, Maringer Y, Marconato M, et al. A COVID-19 peptide vaccine for the induction of SARS-CoV-2 T cell immunity. Nature. 2022;601:617–22.

    Article  CAS  PubMed  Google Scholar 

  4. Tarke A, Coelho CH, Zhang Z, Dan JM, Yu ED, Methot N, et al. SARS-CoV-2 vaccination induces immunological T cell memory able to cross-recognize variants from alpha to Omicron. Cell. 2022;185:847–e85911.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Zhang J-Y, Wang X-M, Xing X, Xu Z, Zhang C, Song J-W, et al. Single-cell landscape of immunological responses in patients with COVID-19. Nat Immunol. 2020;21:1107–18.

    Article  CAS  PubMed  Google Scholar 

  6. Schultheiß C, Paschold L, Simnica D, Mohme M, Willscher E, von Wenserski L, et al. Next-generation sequencing of T and B cell receptor repertoires from COVID-19 patients showed signatures Associated with Severity of Disease. Immunity. 2020;53:442–e4554.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Zhang T, Tian W, Wei S, Lu X, An J, He S, et al. Multidisciplinary recommendations for the management of CAR-T recipients in the post-COVID-19 pandemic era. Exp Hematol Oncol. 2023;12:66.

    Article  PubMed  PubMed Central  Google Scholar 

  8. Wada D, Nakamori Y, Maruyama S, Shimazu H, Saito F, Yoshiya K, et al. Novel treatment combining antiviral and neutralizing antibody-based therapies with monitoring of spike-specific antibody and viral load for immunocompromised patients with persistent COVID-19 infection. Exp Hematol Oncol. 2022;11:53.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

Download references


We are grateful to Professor Guo Fu for providing SARS-CoV-2 associated tetramers. The author would like to thank MUSI biotech Co., Ltd (Shanghai, China) and PUCG biotech Co., Ltd (Shanghai, China) for technical support. We thank Home for Researchers editorial team ( for language editing service.


This work was supported by the National Natural Science Foundation of China (NO. 92374105, NO. 32322029 and NO. 82372785), the Natural Science Foundation of Fujian Province (NO. 2021J01017 and NO. 2022J01011), the Fundamental Research Funds for the Central Universities (NO. 20720230069 and NO. 20720220003), and Funds from Xiang An Biomedicine Laboratory (NO. 2023XAKJ0101034).

Author information

Authors and Affiliations



K. W. conceived and designed the study; Q. Z., Q. L., R. Z. performed the experiments; N. W. and J. S. analyzed the bulk IR-seq data; X. X. analyzed the scRNA-seq data. Q. Z. and K. W. wrote the manuscript.

Corresponding author

Correspondence to Kejia Wang.

Ethics declarations

Ethics approval and consent to participate

Informed consent was obtained from all individuals, and the procedures were approved by the Ethics Committee of Xiamen University in China.

Consent for publication

Written informed consent was obtained from all authors for publication of this study.

Competing interests

All authors declare that they have no conflicts of interest.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhang, Q., Liang, Q., Zhang, R. et al. Identification of SARS-CoV-2-specific T cell and its receptor. J Hematol Oncol 17, 15 (2024).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: