The development of humoral and cellular immunity against SARS-CoV-2, the causative agent of new coronavirus disease (COVID-19) (1), has been the subject of numerous studies, given its importance in the pathogenesis of the disease and its usefulness from a diagnostic and epidemiological perspective (2, 3). SARS-CoV-2-reactive T lymphocytes predominantly recognize peptides belonging to spike (S) protein, followed by membrane protein (VME1) and nucleoprotein (NCAP), although reactivity against other viral proteins has been described. Phenotypic characterization of T lymphocytes indicates they are mostly CD4+ Th1 (3, 4), with characteristics of effector (T_EM) or central (T_CM) memory cells (5, 6). IgG against S protein was found in 90% of patients 6-8 months after infection (7, 8). Similarly, specific CD4+ and CD8+ cells were found respectively in 89% and 50-74% of patients 6-9 months after infection (9).

Numerous studies show the presence of T lymphocytes reactive against SARS-CoV-2 epitopes in a variable percentage of healthy individuals (3, 4, 10–13). In most cases, these epitopes have a high degree of homology with sequences present in common cold coronaviruses. These lymphocytes could be responsible for heterologous immunity, conferring resistance to infection by SARS-CoV-2 or leading to milder COVID-19 symptoms, but no studies confirm this hypothesis.

Officially approved COVID-19 mRNA vaccines include BNT162b2 and mRNA-1273, which require two doses administered 3-4 weeks apart. Recent reports describe kinetics of humoral and cellular responses after vaccine administration (14–17); however, there are no detailed analysis of immune profile changes in patients before and after vaccine administration.

In the present study, we conducted a comprehensive characterization of CD4+T and B cell response in healthy donors (HD) and convalescents (CD) pre-and post-vaccination. We provide evidence for specific changes in T cell reactivity profile, which include increased S1 and mainly S2 specific CD4+ lymphocytes, and a loss of reactivity against NCAP in vaccinated subjects. Likewise, we detected an increment in anti-S and anti-RBD antibodies that peaked after first dose administration in CD. Interestingly, we did not detect MPro-elicited CD4+ activation despite the overt humoral response against this protein in CD. Conversely, we found a strong significant correlation between S1-induced CD4+ response and anti-S IgA in CD, which was lost after vaccination.

In the present study, we sought to investigate changes in immune specificity profile after vaccination against SARS-CoV-2. We first analyzed pre-vaccine specific immune adaptive responses, and identified reactive T lymphocytes in both HD and CD. We selected NCAP and S proteins because they contain peptides that elicit CD4+ T cell response in a great proportion of HD and CD (4, 11). In fact, several HLA-DRB1-restricted immunodominant peptides have been described for both proteins (4). Conversely, VME1 protein has been reported to contain HLA-DRB1-restricted immunodominant peptides and drive CD4+ activation only in CD (4). However, other authors detected overt VME1-induced CD4+ reactivity also in HD (3, 19). We also included peptides from Mpro, a non-structural protein contained in Orf1, because of its high homology with the main protease of SARS-CoV and the strong humoral response evoked by this protein in CD (20). The heterogeneous specificity observed in CD4+ responses in both HD and CD ( Figures 1 , 2 ) could be due to differences in class II HLA profile (4). However, high-resolution analysis of HLA-DRB1 did not show a clear relationship between donor genotype and CD4+ activation profile ( Supplementary Table 2 ).

The detection of SARS-CoV-2-specific T cells in HD has been attributed to cross-reaction with peptides from common cold coronaviruses (3, 5, 12, 21), and cross-reactive CD4+ T cell receptors for S protein peptides have been recently characterized (22). HD recruited for our study were either healthcare workers or related contacts, so exposure to SARS-CoV-2 cannot be completely ruled out. In fact, we detected T cell response against NCAP in a percentage of HD ( Figures 1A, C ), which has been connected by several authors with exposure to SARS-CoV-2. In this regard, T cell activation induced by NCAP was detected both in exposed individuals and in HD recruited in 2020, but not in HD recruited in 2019 (10). Likewise, NCAP-specific T cells were identified only in exposed subjects and CD, but not in prepandemic or unexposed individuals (23). In accordance with our data, a CD4+ specificity profile that includes T cell reactivity against S, VME1 and NCAP peptides has been reported in CD (3, 19).

A recent paper shows that the percentage of anti-S specific CD4+ peaks 7 days after BNT162b2 first-dose administration in CD, decreasing 7 days later and not recovering with a second dose. Conversely, HD showed a significant increase in S-specific CD4+ cells after completing the vaccination schedule, reaching the same level of response as CD (14). However, this study did not analyze the CD4+ response specifically against S1 and S2. Interestingly, we have found a strong correlation between S1- and S2-specific CD4+ lymphocytes in vaccinated CD, which show a relatively higher response for S2 in both HD and CD ( Figure 3A ). Thus, a qualitative change in T cell response against S protein takes place in vaccinated donors; nevertheless, further studies are required to establish the actual significance of these findings in terms of protection against SARS-COV-2 infection. On the other hand, NCAP-induced T cell activation observed in some patients is no longer detectable after vaccination ( Figures 1A, 2A–C and Supplementary Figure 3 ). It is conceivable that vaccine-induced expansion of S-specific CD4+ clones dilutes the relative frequency of NCAP-activated T lymphocytes, hampering their detection in our short-term in vitro activation assay. In addition, T cells specific for NCAP epitope Nuc322-331 in CD undergo a gradual decrease in their activation state, together with an increment in lymph node homing and proliferation potential, which could influence their identification in short-term activation-based assays (24).

In addition to mRNA vaccines, SARS-CoV-2 vaccines based on adenoviral vectors (ChAdOx1 nCoV-19, Ad26.COV2.S, Ad5-nCoV, etc), synthetic S protein (NVX-CoV2373), and whole-cell inactivated vectors (BBV152) were found to elicit a robust CD4+ T cell response after first and/or second dose, mainly composed of IFNγ− and TNFα-producing TH₁ cells (25). However, there is no detailed characterization of T cell activation profile in these patients.

Our data on humoral response in CD, are also in line with previous reports showing constant levels of anti-S IgG up to 6 months after infection, together with progressive decrease of anti-RBD and anti-NCAP IgG (7). Despite the strong antibody response against Mpro in CD, we did not detect Mpro-induced T cell activation in these subjects ( Figure 1B ). A possible explanation is that Mpro protein processing does not yield peptides immunogenic enough to elicit a T cell response detectable in our assay. It is also possible that peptide pool composition does not adequately reflect actual in vivo Mpro protein processing.

Following our results, some authors have recently reported a maximum increase of anti-S and anti-RBD IgG with a single dose of BNT162b2 vaccine in CD (14–17). Nevertheless, in contrast to our data, an increase of anti-S IgM and IgA levels upon first dose administration was described in HD, which was further enhanced 1 week after the second dose (14). However, we determined antibody titers 5-7 weeks after complete vaccination, thus we cannot rule out that IgM and IgA levels had already decreased at this time point. As expected, anti-Mpro antibodies did not increase after vaccination, whereas at the end of the study, IgG, and to a lesser extent IgA, were readily detectable with significantly higher levels than in HD ( Figure 4A and Supplementary Figure 4B ). The presence of detectable titers of anti-Mpro IgG in CD but not in HD may provide a useful tool to reliably discriminate between vaccinated healthy population and COVID-19 patients.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

This study was approved by the local Research Ethics Committee (CEIm Hospital Universitario de La Princesa; register number 4070), and it was carried out following the ethical principles established in the Declaration of Helsinki. The patients or their representatives provided their written informed consent to participate in this study.

AA, FS-M, CM-C, and EM-G contributed to research design and data analysis, and either wrote or revised the manuscript. LE-P and PM-F contributed to data acquisition and analysis. NR and PV-T contributed to acquisition of the data. CV performed HLA genotyping and revised the manuscript. HR and JC generated recombinant proteins and revised the manuscript. MV-G and JR produced ELISA plates and revised the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by grants to AA: FIS PI19/01491 (Fondo de Investigación Sanitaria del Instituto de Salud Carlos III with co-funding from the Fondo Europeo de Desarrollo Regional FEDER) and Sociedad Cooperativa de Viviendas Buen Suceso, S.Coop.Mad. To AA and FS-M: CIBER Cardiovascular from the Instituto de Salud Carlos III (Fondo de Investigación Sanitaria del Instituto de Salud Carlos III with co-funding from the Fondo Europeo de Desarrollo Regional; FEDER). To FS-M: SAF2017-82886-R (Spanish Ministry of Economy and Competitiveness MINECO)), HR17-00016 (“La Caixa” Banking Foundation), “Fondos Supera COVID19” (Banco de Santander and CRUE), “Ayuda Covid 2019” and “Inmunovacter” REACT-UE (Comunidad de Madrid). To MV: Spanish National Research Council (CSIC, project number 202020E079 and CSIC-COVID19-028).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.