We used peripheral blood mononuclear cells (PBMCs) and serological samples from 42 vaccinated convalescent COVID-19 individuals (further referred as C+V+), 8 vaccinated uninfected with SARS-CoV-2 adults (further referred as C-V+) and 7 unvaccinated convalescent (further referred as C+V-) individuals.

As we studied convalescent subjects, a direct molecular sequencing of the viral variant responsible for SARS-CoV-2 infection was not possible. Therefore, we determined the viral variants that the study participants were infected with, relying on temporal viral spreading defined by epidemiologcal trends in Germany. The SARS-CoV-2 diagnosis for the C+V+ and C+V- participants was set between June 2020 and May 2021. According to epidemiological data of the Robert Koch Institute alpha variant drove and overrepresented SARS-CoV-2 infections in Germany for the first half of 2021 (https://www.rki.de/DE/Content/InfAZ/N/Neuartiges_Coronavirus/Virologische_Basisdaten.html;jsessionid=93905B429A0980A796F15E1AA070EF2C.internet102?nn=13490888#doc14716546bodyText14).

The study was approved by the Ethics Committee of the University Hospital Essen (20-9753-BO). A part of the patient data was included in the research article "Increased migratory/activated CD8+ T cell and low avidity SARS-CoV-2 reactive cellular response in post-acute COVID-19 syndrome (PASC)" currently under consideration for publication. Written informed consent was obtained from all participants. Demographic and clinical characteristics are provided in Table 1 .

As previously described (20), peripheral blood was collected in S-Monovette K3 EDTA blood collection tubes (Sarstedt). Collected blood was prediluted in PBS/BSA (Gibco) at a 1:1 ratio and underlaid with 15 mL of Ficoll-Paque Plus (GE Healthcare). Tubes were centrifuged at 800g for 20 min at room temperature. Isolated PBMCs were washed twice with PBS/BSA and stored at -80°C until use. The cryopreserved PBMCs were thawed by incubating cryovials 2-3 minutes at 37°C in bead bath, washed twice in 37°C RPMI 1640 media (Life Technologies) supplemented with 1% penicillin-streptomycin-glutamine (Sigma-Aldrich), and 10% fetal calf serum (PAN-Biotech) medium, and incubated overnight at 37°C.

As previously described, PBMCs were plated in 96-U-Well plates in RPMI 1640 media (Life Technologies) (21, 22). Each well was stimulated with one of the following SARS-CoV-2 proteins: the pool of B1.617.2 (delta) S mutant peptides (Miltenyi Biotec), their corresponding position in wildtype (WT) pool of peptides (Dref) (Miltenyi Biotec), the pool of B.1.1529 (omicron) S mutant peptides (Miltenyi Biotec), their corresponding position in WT pool of peptides (Oref) (Miltenyi Biotec) and the complete sequence of B.1.1.7 D614G S mutant (JPT Peptide Technologies) or WT S-protein (Miltenyi Biotec) or left untreated as a control for 16 h. The proteins were dissolved per manufacturer’s directions. As a positive control, cells were stimulated with staphylococcal enterotoxin B (1 μg/mL, Sigma-Aldrich). After 2 h, brefeldin A (1 μg/mL, Sigma-Aldrich) was added. Detailed listing of the antibody panels for general phenotyping and T cell activation ex vivo is in Table S1 . The PBMCs stimulated overnight were stained with optimal concentrations of antibodies for 10 min at room temperature in the dark. Stained cells were washed twice with PBS/BSA before preparation for intracellular staining using the Intracellular Fixation & Permeabilization Buffer Set (Thermo Fisher Scientific) as per the manufacturer’s instructions. Fixed and permeabilized cells were stained for 30 min at room temperature in the dark with an optimal dilution of antibodies against the intracellular antigen. All samples were immediately acquired on a CytoFLEX flow cytometer (Beckman Coulter). Quality control was performed daily using the recommended CytoFLEX daily QC fluorospheres (Beckman Coulter). No modification to the compensation matrices was required throughout the study. Antigen-reactive responses were considered positive after the non-reactive background was subtracted, and more than 0.01% were detectable. Negative values were set to zero.

Human IgG autoantibodies against the SARS-CoV-2 nucleocapsid (N) protein were detected using commercial ELISA kits (EUROIMMUN) according to the manufacturer’s instructions.

As previously described (22), for the virus neutralization assay, sera were incubated for 30 min at 56°C in order to inactivate complement factors. Single cycle VSV∗ΔG(FLuc) pseudoviruses bearing the SARS-CoV-2 WT S (D614G) protein or SARS-CoV-2 alpha S protein, or SARS-CoV-2 delta or SARS-CoV-2 omicron S protein virus in the envelope were incubated with quadruplicates of two-fold serial dilutions of immune sera in 96-well plates prior to infection of Vero E6 cells (1x104 cells/well) in DMEM + 10% FBS (Life Technologies). At 18 hours post infection, firefly luciferase (FLuc) reporter activity was determined, based on CPE values, the 50% neutralization capacity (ND50 value) was calculated using non-linear regression.

Flow cytometry data were analyzed using FlowJo version 10.6.2 (BD Biosciences); gating strategies are presented in Figure S1 . For the analysis of anti-SARS-CoV-2 reactive T cells, a threshold of 0.01% was employed to define a detectable response. Single stains and fluorescence-minus-one controls were used for gating. Gates of each study participant were adjusted according to the negative control. CD4+ T cells expressing CD154 and CD137 and CD8+ T cells expressing CD137 were defined as reactive T cells. Statistical analysis was performed using GraphPad Prism v7. Categorical variables are summarized as numbers and frequencies; quantitative variables are reported as median and interquartile range. Normality tests were performed with the Shapiro-Wilk Test. All applied statistical tests are two-sided. Kruskal-Wallis Test and Mann-Whitney-Test were applied to perform comparisons. Spearman´s rank correlation coefficient was used to explore a possible correlational relationship between T cells and neutralizing antibodies. The age between the two cohorts was compared using unpaired two-tailed t-test, and gender was compared using two-tailed Fisher’s exact test. p values below 0.05 were considered significant; only significant p values are reported in the figures. p values were not corrected for multiple testing, as this study was of an exploratory nature.

Our study group comprised 42 vaccinated convalescent COVID-19 individuals (further referred as C+V+), 8 vaccinated individuals without previous COVID-19 (further referred as C-V+) and 7 unvaccinated COVID-19 convalescent (further referred as C+V-) individuals. All study participants had a negative SARS-CoV-2 nasal swab tested via RT-PCR on recruitment. The C-V+ study group was seronegative for N-SARS-CoV-2 IgG, excluding an asymptomatic SARS-CoV-2 infection in the past. Most individuals 93% (n=39) experienced moderate COVID-19 during the acute phase of COVID-19 disease without need for hospitalization, whereas only 7% (n=3) were severely or critically ill and hospitalized.

67% (n=28) of the C+V+ individuals received one anti-SARS-CoV-2 mRNA vaccination after the SARS-CoV-2 infection, 29% (n=12) 2 vaccinations and 4% (n=2) 3 vaccinations. All C-V+ individuals received 3 anti-SARS-CoV-2 mRNA vaccinations.

The median time since COVID-19 convalescence for the C+V+ study group was 11 months (range 7-16 months). The median time since the last vaccination was 4 and 5 months for the C+V+ and C-V+ study group, respectively (range 1-8 & 2-5 months respectively). 71% (n=30) of the C+V+ study group was infected with the WT whereas 29% (n=12) with the alpha variant.

The median age of the C+V+ study group was 48.5 years (range 19-68 years), whereas the control cohort was significantly younger, with a median age of 30 years (range 22-46 years, p=0.003 two tailed unpaired t test). The C+V+ and C-V+ cohorts comprised of 69% (n=29) and 88% (n=7) female participants, respectively and showed no significant gender difference (Fisher´s exact test, p<0.05). The demographic and clinical characteristics of the study cohorts are presented in Table 1 .

Up to date, humoral immunity and more concrete high titers of neutralizing antibodies are considered the pylons of protection against SARS-CoV-2 infection (1). Therefore, we performed a pseudovirus neutralization assay against WT, alpha, delta and omicron VOC. First, we compared the neutralization capacity of WT, alpha, delta and omicron VOC in C+V+ cohort. WT NAb titers were significantly higher compared to the alpha, delta and omicron VOC ( Figure S2A ) (Kruskal-Wallis test, p<0.0001). Furthermore, we observed a steady decline of NAb titers within the known SARS-CoV-2 variant evolution from WT to the most current VOC omicron. To quantitatively assess the loss of neutralizing capacity following the viral evolution, we normalized the median NAb of VOCs to WT: alpha/WT 72% (range 0-352%); delta/WT 54% (range 0-237); omicron/WT 31% (range 0-100%). Especially for the most current VOC omicron, this calculation demonstrates an approximately 70% loss of neutralizing potential among the C+V+ cohort ( Figure S2A ).

A similar pattern of humoral immunity was found for the C-V+ cohort( Figure S2B ). A comparable loss of neutralizing capacity was also demonstrated for the C-V+ group: alpha/WT 84% (range 29-148%), delta/WT 43% (range 28-43), omicron/WT 21% (range 0-89%). All in all, we observed a steady decline of NAb titers across viral evolution with the lowest neutralization capacity against omicron VOC.

Next, we compared the neutralizing capacity between the C-V+ und and C+V+ study groups. The C+V+ cohort possessed significantly higher titers of neutralizing antibodies compared to the C-V+ cohort against all VOC (Mann Whitney test, WT NAb C+V+ vs C-V+ p=0.0049) (Mann Whitney test, alpha NAb C+V+ vs C-V+ p=0.0068) (Mann Whitney test, delta NAb C+V+ vs C-V+ p=0.0065) (Mann Whitney test, omicron NAb C+V+ vs C-V+ p=0.035) ( Figure 1 ). However, the absolute loss of neutralizing capacity among the C+V+ and C-V+ cohorts did not show any significant differences (Fisher´s exact test, C+V+ vs C-V+ omicron/WT ratio p=0.1464) (Fisher´s exact test, C+V+ vs C-V+ delta/WT ratio p=0.1569) (Fisher´s exact test, C+V+ vs C-V+ alpha/WT ratio p=0.0597) among the two cohorts ( Figure 1 ). These results indicate that the immunocompetent vaccinated convalescent adults generate higher NAb titers compared to immunocompetent uninfected vaccinated adults, however the loss of neutralization potential is proportionally equal.

At last, we compared the S-reactive T cell responses among the C+V+ and C-V+ study groups for all analyzed VOCs. We stimulated overnight PBMCs with the pool of delta S mutant peptides, their corresponding position in WT pool of peptides (Dref), the pool of omicron S mutant peptides, their corresponding position in WT pool of peptides (Oref), the complete sequence of alpha S mutant or WT S-protein. First, we analyzed the frequencies of SARS-CoV-2 reactive CD4+ T cells. We found similar frequencies of WT, delta and omicron reactive CD4+ T cells( Figures 2A–C ) with comparable avidity ( Figures 2D–F ), however the alpha reactive CD4 T cells showed higher frequencies among the C+V+ group ( Figure 2A ) (Mann-Whitney and Fisher´s exact test). The frequencies of SARS-CoV-2 reactive CD8+ T cells were similar against all tested peptides ( Figures 3A–C ), as well as their avidity( Figures 3D–F ).

To be able to integrate our findings into the current immune dynamics and to avoid one sided interpretation of our results, we additionally analyzed the humoral and cellular SARS-CoV-2 response of 7 unvaccinated convalescent COVID-19 adults with a median convalescent time of 9 months (range 5-12 months), further referred as C+V-. We found significantly higher titers of NAb against all studied VOC among the C+V+ individuals compared to the C+V- (Mann Whitney test, WT NAb C+V+ vs C+V- p<0.0001) (Mann Whitney test, alpha NAb C+V+ vs C+V- p<0.0001) (Mann Whitney test, delta NAb C+V+ vs C+V- p<0.0001) (Mann Whitney test, omicron NAb C+V+ vs C+V- p=0.0028) ( Figure 1 ). Similar data were observed comparing C+V- to C-V+ group ( Figure 1 ).

Next, we analyzed the SARS-CoV-2 reactive cellular response among the C+V+ and C+V-. We detected WT, alpha, delta and omicron reactive CD4+ and CD8+ T cells with similar frequencies between the two cohorts( Figures 4A, B ). The CD3low analysis showed similar functional avidity among the cohorts ( Figures 4C, D ). These results demonstrate a superior humoral response among the C+V+ adults compared to C+V-, however similar cellular response with comparable cross-recognition potential against the studied VOC. Comparing C+V- group to C-V+ demonstrated similar results ( Figure 4 ).

To address the interesting question of VOC cross-recognition inside each study group separately we performed a reanalysis of our data. We demonstrate similar frequencies of WT, alpha, delta and omicron CD4+ & CD8+ reactive T cells in C+V+ individuals (Mann Whitney test, p<0.05) ( Figures S3A, C ). Furthermore, we performed an analysis of the CD3low subset among reactive CD4+ and CD8+ T cells (gating strategy, Figure S1 ) to address their avidity against the different variants (22, 23). We detected similar frequencies of reactive CD4+ and CD8+CD3low T cells against all studied VOC, except for the alpha variant where alpha CD4+CD3low reactive T cells showed significantly higher frequencies compared to WT CD4+CD3low reactive T cells (Mann Whitney test, p=0.022) ( Figures S3B, D ). Our results demonstrate that SARS-CoV-2 reactive CD4+ and CD8+ T cells recognize alpha, delta and omicron VOC 11 months after infection in vaccinated adults with equal avidity.

As next we assessed the reactive T cell response among the C-V+ study group. In the same line with the C+V+ group, we detected similar frequencies of WT, alpha, delta and omicron reactive CD4+ and CD8+ T cells ( Figures S3E, G ). The frequencies of CD3low reactive T cells showed no statistical difference ( Figures S3F, H ).

In addition to the quantification, avidity and the phenotypic differentiation analysis, we aimed for the analysis of the functional activity as defined by cytokine production. We found similar frequencies of monofunctional IL2, IFNg, TNFa or GrB producing WT, alpha, delta and omicron reactive CD4+ and CD8+ T cells ( Figures S4, S5 ). Our results suggest that SARS-CoV-2 specific T cells preserve their functionality against the studied VOC.

To evaluate the SARS-CoV-2 T cell memory in details, we analyzed differentiation phenotype using the markers CCR7 and CD45RA (T_CM=CD45RA-CCR7+, T_NAIVE=CD45RA+CCR7+, T_EM=CD45RA-CCR7- T_EMRA=CD45RA+CCR7-). Overall, the analysis of the C+V+ and C-V+ cohort independent from each other, revealed cross-reactive T_CM, T_NAIVE, T_EM, T_EMRA CD4+ and CD8+ memory subsets, recognizing equally all tested peptides ( Figures S6, S7 ). Among the C+V+ cohort we documented a clear predominance of the T_CM CD4+ cells against all tested VOC ( Figure S8 ) and significantly higher frequencies of A T_EM CD8+ among the rest CD8+ memory subsets ( Figure S9 ) (Kruskal-Wallis test and Mann-Whitney Test). The C-V+ memory CD4+ and CD8+ T cell subsets were equally represented ( Figure S8 ), except for the O T_CM CD4+ cells which showed significantly higher frequencies compared to T_EM CD4+ cells ( Figure S8D ) (Mann-Whitney Test).

The direct comparison of the memory subsets of C+V+ and C-V+ individuals revealed similar memory T cell frequencies ( Figure S8 and S9 ), except for significantly higher WT T_EM CD4+ frequencies among the C-V+ cohort compared to the C+V+ ( Figure S8A ).