key: cord-0295196-wxor6kg3
authors: Lu, Zhongyan; Laing, Eric D.; Pena-Damata, Jarina; Pohida, Katherine; Tso, Marana S.; Samuels, Emily C.; Epsi, Nusrat J.; Dorjbal, Batsukh; Lake, Camille; Richard, Stephanie A.; Maves, Ryan C.; Lindholm, David A.; Rozman, Julia; English, Caroline; Huprikar, Nikhil; Mende, Katrin; Colombo, Rhonda E.; Colombo, Christopher J.; Broder, Christopher C.; Ganesan, Anuradha; Lanteri, Charlotte A.; Agan, Brian K.; Tribble, David; Simons, Mark P.; Dalgard, Clifton L.; Blair, Paul W.; Chenoweth, Josh; Pollett, Simon D.; Snow, Andrew L.; Burgess, Timothy H.; Malloy, Allison M.W.
title: Durability of SARS-CoV-2-specific T cell responses at 12-months post-infection
date: 2021-08-11
journal: bioRxiv
DOI: 10.1101/2021.08.11.455984
sha: 493b831a695d102cb0f0984be8aa08cbfd66ec26
doc_id: 295196
cord_uid: wxor6kg3

Background Characterizing the longevity and quality of cellular immune responses to SARS-CoV-2 is critical to understanding immunologic approaches to protection against COVID-19. Prior studies suggest SARS-CoV-2-specific T cells are present in peripheral blood 10 months after infection. Further analysis of the function, durability, and diversity of the cellular response long after natural infection, over a wider range of ages and disease phenotypes, is needed to further identify preventative and therapeutic interventions. Methods We identified participants in our multi-site longitudinal, prospective cohort study 12-months post SARS-CoV-2 infection representing a range of disease severity. We investigated the function, phenotypes, and frequency of T cells specific for SARS-CoV-2 using intracellular cytokine staining and spectral flow cytometry. In parallel, the magnitude of SARS-CoV-2-specific antibodies was compared. Results SARS-CoV-2-specific antibodies and T cells were detected at 12-months post-infection. Severity of acute illness was associated with higher frequencies of SARS-CoV-2-specific CD4 T cells and antibodies at 12-months. In contrast, polyfunctional and cytotoxic T cells responsive to SARS-CoV-2 were identified in participants over a wide spectrum of disease severity. Conclusions Our data show that SARS-CoV-2 infection induces polyfunctional memory T cells detectable at 12-months post-infection, with higher frequency noted in those who originally experienced severe disease.

Understanding the development and durability of protective immune responses against SARS-68

CoV-2 remains critical as we seek global reduction in disease burden. Antibody responses induced 69 during a primary SARS-CoV-2 infection have been shown to wane, but may be present in the 70 circulation up to twelve months post symptoms onset (PSO) [1, 2] . Thus far, SARS-CoV-2-71 specific T cells have been detected up to 10 months PSO [3, 4] . Intriguingly, SARS-CoV-1-72 specific T cells have been identified 17 years post-infection, suggesting the potential for very long-73 lived T cell memory [5] . Antibodies have been effective in reducing the disease burden of SARS-74

In order to utilize the specificity and potent viral clearance of T cells, further knowledge is required 76 regarding viral antigen specificity, memory differentiation and longevity [6] . 77

antigen-specific T cells are associated with mild disease, whereas a lack of these antiviral cells or 79 a delay in development is associated with severe disease [7, 8] . Peng et al. [9] showed that 80 individuals who had mild disease had a higher ratio of polyfunctional CD8 T cells compared to 81 CD4 T cells around 42 days PSO, suggesting that potent SARS-CoV-2-specific CD8 T cells may 82 be protective. Analysis of functional T cell responses at memory time points is needed to provide 83 insight into their cytolytic potential and role in protection upon re-infection that prior studies 84 relying on activation-induced markers (AIM) [6] could not. 85

The T cell response to SARS-CoV-2 has been shown to recognize epitopes across multiple 86 study. Receipt of a SARS-CoV-2 vaccine was identified by report and confirmed by medical data 113 repository review. Study participants who received a SARS-CoV-2 vaccine prior to the 12-month 114 peripheral blood collection were excluded from S-specific T cell and antibody analysis. Study 115 participants were evaluated for re-infection by documented PCR positive for SARS-CoV-2 or a 116 significant rise in both S and N-specific antibody responses. Participants who were re-infected 117 were excluded for comparative T cell and antibody analysis in the current study. The protocol was 118 approved by the Uniformed Services University Institutional Review Board (IDCRP-085), and all 119 subjects or their legally authorized representative provided informed consent to participate. 120

PBMCs were isolated from peripheral whole blood collected in acid citrate dextrose (ACD) tubes 123 at 12-months PSO. PBMCs were purified using a Ficoll-Histopaque (Fisher Scientific, NH) 124 gradient. Cells were preserved in 90% fetal bovine serum (Sigma-Aldrich, MO) and 10% dimethyl 125 sulfoxide (DMSO, Sigma-Aldrich, MO) and stored in liquid nitrogen. Serum was isolated from 126 blood collection in EDTA tubes. 127 128

One million thawed PBMCs were stained in PBS at 4°C for 20 minutes for antigen presenting cells 130 (APCs) using the antibodies listed in Supplemental Table 1 and were acquired on a CYTEK Aurora 131 5-laser spectral flow cytometer (CYTEK Biosciences, CA). 132

For identification of antigen-specific T cells, isolated PBMCs were cultured in 96-well plates and 135 stimulated overnight with peptide pools derived from selected viral proteins. Peptide pools were 136 comprised of 15-mer peptides overlapping by 11 amino acid residues covering the S, M, N, and E 137 proteins of SARS-CoV-2 (JPT, Germany), the S protein of endemic human coronavirus (hCoV) 138 strains: HKU1, 229E, NL63 and OC43; and a CMV peptide pool including pp50, pp65, IE1, IE2 139 and envelope glycoprotein B (Mabtech, OH) at a final concentration of 1 mg/ml for each individual 140 peptide (Supplemental Table 2 ). Monensin (BD Biosciences, CA; BioLegend, CA) was added to 141 the wells 1 hour after peptide addition to prevent cytokine secretion as recommended by the 142 manufacturer. Ten minutes after addition of the peptide pools, CD107a antibody (BioLegend, CA) 143 was added. Cell surface markers were identified with the antibodies listed in Supplemental Table  144 3. Surface staining was followed by cell permeabilization using FoxP3/transcription factor staining 145 buffer set (Thermo Fisher Scientific, MA) at 4°C for at least 30 minutes. Antibodies used for 146 intracellular cytokine staining (ICS) (Supplemental Table 3 ) were then added. PBMCs cultured in 147 medium without peptide stimulation served as the negative control, or with CytoStim (Miltenyi 148 Biotec, CA) as the positive control for the assay. Samples were acquired on a CYTEK Aurora 149 (CYTEK Biosciences, CA). All cytometric data were analyzed using FlowJo software (BD 150

Biosciences, CA). Samples were considered positive if the frequency of IFNg+ T cells was 2-fold 151 higher than the medium control and greater than 0.01% of CD4 or CD8 T cells after subtracting 152 the medium control value [14] . 153 154 samples were added to 96-well microtiter plates containing antigen-coupled microspheres and 158 tested in technical duplicates. After 45 minutes of agitation, wells were washed, and biotin-159 conjugated goat anti-human IgG (Thermo Fisher Scientific, Waltham, MA) diluted in PBS + 160 0.05% Tween20 (PBST) was added to each well. Wells were subsequently washed again, then 

Data were analyzed using GraphPad Prism 9 (GraphPad Software Inc, CA). Two-group test 171 significance levels were calculated using Mann-Whitney analysis. Correlation coefficients and 172 significance levels were calculated using Spearman rank correlation. A p-value < 0.05 was 173 considered statistically significant. 174 175

From our cohort, we identified 29 patients who had COVID-19 approximately 1 year prior, and 9

The age range of the study participants was 20-72 years of age and the overall distribution of age 181 did not vary significantly (p = 0.1) between inpatients (median age 50.6 with a range of 21.4 to 182 72.4) and outpatients (median age 44.6 with a range of 20.1 to 60.1), nor did sex ( and CD8 T cell responses in 73.9% of study participants. Overall, SARS-CoV-2-specific CD4 T 201 cells were more frequently identified in the peripheral blood compared to SARS-CoV-2-specific 202 CD8 T cells. In addition, T cell responses to endemic hCoV strains; HKU1, 229E, OC43, and NL63 were sporadic, with no differential distribution or magnitude difference between inpatients 204 and outpatients identified (Table 2) . 205

The SARS-CoV-2-specific antibody response against S and N proteins were also present 206 in both unvaccinated inpatient and outpatient groups at 12-months PSO ( Figure 1B , Supplemental 207 Figure 1B ). We found that the antibody response against N in inpatients was higher than in 208 outpatients, in whom it waned to levels below the threshold of endemic hCoV N responses (P = 209 0.01). These findings demonstrate that the humoral and cellular responses to SARS-CoV-2 exhibit 210 durability at 12-months PSO, but vary by severity of disease and antigen specificity. exhibited T cell specificity for either N or M at 12-months were also more likely (P = 0.0007) to 218 simultaneously recognize other SARS-CoV-2 epitopes in N or M (Supplemental Figure 1C ). 219

Compared with the CD4 T cell response, the CD8 T cell response, as measured by IFNg 220 production, was lower in magnitude and did not correlate with severity of disease ( Figure 2D ). 221

Quantification of the APCs and frequency of total T cells at 12-months PSO were measured 222 by flow cytometry (Supplemental Figure 1D ) to determine whether differences exhibited between 223 study groups were influenced by APC or T cell availability. Our data show that the ratio of total 224 CD4 and CD8 T cells (Supplemental Figure 1E The frequency of SARS-CoV-2-specific CD4, but not CD8, T cells was higher at 12-months PSO 281 in individuals who experienced severe disease compared to those who had mild disease at the acute 282 phase. Importantly, memory SARS-CoV-2-specific CD4 T and CD8 T cells exhibited 283 polyfunctionality and cytotoxicity, respectively, suggesting strong recall responses. Our findings 284

and that of others also show that antibodies to SARS-CoV-2 persist 12-months PSO, but wane 285 rapidly [1, 2, 18] and in the case of N-specific antibodies may be undetectable. 286

Overarchingly, we identified SARS-CoV-2-specific T cell responses in 75.9% of study 287 participants at 12-months PSO using peptide pools derived from SARS-CoV-2 structural proteins 288 N, M, E, and S. Other studies have shown that epitopes in SARS-CoV-2 N, M, and S, together 289 with nsp3, 4, ORF3a are recognized by CD4 and CD8 T cells representing the majority of T cell 290 responses at convalescence, while SARS-CoV-2 E was less frequently recognized [11, 19, 20] . 291

Although we did not map T cell epitopes, we overall observed the same trend of T cell responses 292 to corresponding peptide pools used in this study, but at a much later time point. As the magnitude 293

CoV-2-specific CD8 T cells in the peripheral blood are of low frequency at one year PSO. 296 Importantly, T cell recognition of multiple epitopes was common within study participants, 297

suggesting broad epitope recognition. 298

Study participants who were hospitalized during acute infection demonstrated the highest 299 frequency of SARS-CoV-2-specific CD4 T cell responses and antibodies to SARS-CoV-2 N. Why 300 severe COVID-19 is associated with a higher memory response for CD4 T cells remains unclear. 301

Since severe disease is associated with higher viral burden [24] [25] [26] , one hypothesis is that higher 302

more durable responses. However, due to differences in timing of nasal swab acquisition and the 304 size of this subcohort, we were unable to correlate T cell longevity and viral load at acute infection. 305

Published data have yet to show that initial SARS-CoV-2 viral load correlates to the longevity of 306 Limitations to our study include the challenge of obtaining biological samples at 12-months 338 PSO prior to vaccination in our MHS cohort, which had timely and reliable access to vaccines. 339

Fifty percent of individuals in our cohort were vaccinated prior to a 12-month draw (Table 1) . For these individuals we did not include their S-specific T cell or antibody responses, but did include 341 their N, M and E humoral or cellular responses since these are not components of the mRNA-342 based vaccines (the vaccines primarily available to our study population). A strength of our cohort 343 is that study participants are followed within the MHS and their epidemiologic, clinical and 344 COVID-19 testing records, and vaccination status are maintained primarily from electronic health 345 record. In addition, sera were analyzed longitudinally to avoid re-infected participants for 12-346 months T cell analysis. 347

In summary, our data show that both SARS-CoV-2 humoral and cellular responses are 348 

Naturally acquired SARS-CoV-2 immunity persists 465 for up to 11 months following infection

Antigen-based multiplex strategies to discriminate

SARS-CoV-2 natural and vaccine induced immunity from seasonal human coronavirus humoral 468 responses

Longitudinal Analysis of Human Memory T-Cell Response 470 according to the Severity of Illness up to 8 Months after SARS-CoV-2 Infection

SARS-CoV-2-specific T cell memory is sustained in COVID-19 472 convalescent patients for 10 months with successful development of stem cell-like memory T cells

SARS-CoV-2-specific T cell immunity in cases of

Immunological memory to SARS-CoV-2 assessed for up to 8 477 months after infection

Antigen-Specific Adaptive Immunity to

SARS-CoV-2 in Acute COVID-19 and Associations with Age and Disease Severity

Acute SARS-CoV-2 Infection Impairs Dendritic Cell and T

Broad and strong memory CD4(+) and CD8(+) T cells induced 484 by SARS-CoV-2 in UK convalescent individuals following COVID-19

SARS-CoV-2 genome-wide T cell epitope mapping 487 reveals immunodominance and substantial CD8(+) T cell activation in COVID-19 patients

Comprehensive analysis of T cell immunodominance and 490 immunoprevalence of SARS-CoV-2 epitopes in COVID-19 cases

Selective and cross-reactive SARS-CoV-2 T cell epitopes 492 in unexposed humans

SARS-CoV-2 variants of concern partially escape 494 humoral but not T-cell responses in COVID-19 convalescent donors and vaccinees

Long-lived poxvirus immunity, robust CD4 497 help, and better persistence of CD4 than CD8 T cells

SARS-CoV-2 Infections and Serologic Responses Among

Military Personnel Deployed on the USNS COMFORT to New York City During the COVID-19

Two subsets of memory T 502 lymphocytes with distinct homing potentials and effector functions

Differentiation of human CD8(+) T cells from a 504 memory to memory/effector phenotype

CoV-2 in Convalescent COVID-19 Patients One Year Post Symptom Onset

SARS-CoV-2-derived peptides define heterologous and 509 COVID-19-induced T cell recognition

Targets of T Cell Responses to SARS-CoV-2

Coronavirus in Humans with COVID-19 Disease and Unexposed Individuals

Initial viral load determines the magnitude of the 514 human CD8 T cell response to yellow fever vaccination

Human effector and memory CD8+ T cell 517 responses to smallpox and yellow fever vaccines

Characterization of pre-existing and induced SARS

Viral load dynamics and disease severity in patients infected with

SARS-CoV-2 in Zhejiang province, China

Temporal profiles of viral load in posterior oropharyngeal 524 saliva samples and serum antibody responses during infection by SARS-CoV-2: an observational 525 cohort study

SARS-CoV-2 viral load predicts COVID-19 527 mortality

Homeostasis of naive and memory T cells

How Does B Cell Antigen Presentation Affect Memory

Influenza-specific lung-resident memory T cells are 532 proliferative and polyfunctional and maintain diverse TCR profiles

Robust T Cell Immunity in Convalescent 535 Individuals with Asymptomatic or Mild COVID-19

High levels of SARS-CoV-2-specific T cells with 537 restricted functionality in severe courses of COVID-19

SARS-CoV-2-reactive T cells in healthy donors and 539 patients with COVID-19

Sprouty-2 regulates HIV-specific T cell polyfunctionality

In vitro cytokine induction by TLR-activating 546 vaccine adjuvants in human blood varies by age and adjuvant

Expanding roles for CD4(+) T cells in immunity to 548 viruses

Healthy 550 donor T cell responses to common cold coronaviruses and SARS-CoV-2

V367F mutation in SARS-CoV-2 spike RBD emerging during the 553 early transmission phase enhances viral infectivity through increased human ACE2 receptor 554 binding affinity

SARS-CoV-2 spike E484K mutation reduces antibody 556 neutralisation

Effect of natural mutations of SARS-CoV-2 on 558 spike structure, conformation, and antigenicity