key: cord-0877781-92rq5pvm
authors: Garcia-Valtanen, Pablo; Hope, Christopher M.; Masavuli, Makutiro G.; Yeow, Arthur Eng Lip; Balachandran, Harikrishnan; Mekonnen, Zelalem A.; Al-Delfi, Zahraa; Abayasingam, Arunasingam; Agapiou, David; Stella, Alberto Ospina; Aggarwal, Anupriya; Bouras, George; Gummow, Jason; Ferguson, Catherine; O’Connor, Stephanie; McCartney, Erin M.; Lynn, David J.; Maddern, Guy; Gowans, Eric J.; Reddi, Benjamin AJ.; Shaw, David; Kok-Lim, Chuan; Beard, Michael R.; Weiskopf, Daniela; Sette, Alessandro; Turville, Stuart G.; Bull, Rowena A.; Barry, Simon C.; Grubor-Bauk., Branka
title: SARS-CoV-2 Omicron variant escapes neutralising antibodies and T cell responses more efficiently than other variants in mild COVID-19 convalescents.
date: 2022-05-17
journal: Cell Rep Med
DOI: 10.1016/j.xcrm.2022.100651
sha: 91baa5fde5ed176d36c4ce22ee60f478dd534c80
doc_id: 877781
cord_uid: 92rq5pvm

COVID-19 convalescents living in regions with low vaccination rates rely on post-infection immunity for protection against re-infection with SARS-CoV-2. We evaluate humoral and T cell immunity against five Variants of Concern (VOCs) in mild-COVID-19 convalescents at 12 months after infection with ancestral virus. In this cohort, ancestral, receptor-binding domain (RBD)-specific antibody and circulating memory B cell levels are conserved in most individuals and yet, serum neutralisation against live B.1.1.529 (Omicron) is completely abrogated and significantly reduced for other VOCs. Likewise, ancestral SARS-CoV-2-specific memory T cell frequencies are maintained in >50% of convalescents, but the cytokine response in these cells to mutated Spike epitopes corresponding to B.1.1.529 and B.1.351 (Beta) VOC were impaired. These results indicate that increased antigen variability in VOCs impair humoral and Spike-specific T cell immunity post-infection, strongly suggesting that COVID-19 convalescents are vulnerable and at risk of re-infection with VOC. Thus stressing the importance of vaccination programes.

them the denomination variant of concern (VOC). 74 After primary infection and in parallel with the antibody response, symptomatic COVID-75 19 convalescents generate a robust CD4 + and CD8 + memory T cell response which targets a 76 wider range of antigens and epitopes than that covered by antibodies. 20-24 Importantly, the 77 breadth of SARS-CoV-2-specific T cell epitopes appears to be less sensitive to mutations 78 present in VOCs. 25, 26 It is unclear to what extent T cells can protect from re-infection and 79 progression to severe COVID-19. However, it is likely that T cell responses in convalescents, 80 which target most SARS-CoV-2 antigens, 20 could afford some of level of protection for many 81 J o u r n a l P r e -p r o o f months, even years. In fact, SARS-CoV-specific T cells can be detected in convalescents for 82 almost two decades. 27 83 While current vaccines are highly effective in preventing severe disease and death, and 84 booster vaccinations may temporarily circumvent dwindling efficacy over time, 28 next 85 generation vaccines, that can prevent virus transmission, are likely needed to end the 86 pandemic. 29, 30 Long-term studies of the evolution of immune correlates in convalescents, where the immune system has encountered an active live virus infection in 88 the presence of all its antigens, are necessary to elucidate the fine specificities and immune 89 functionality of antibody and T cell responses. In particular, the adaptability of pre-existing 90 immunity to mutated Spike antigens, present in VOC, is the key piece of information that is 91 still unanswered. 92 Compared to the majority of the world, South Australia is in an optimal position to 93 undertake studies on mid-to long-term immunity of COVID-19 due to 1) early and strict 94 border control measures with other countries, and other states within Australia, which were 95 enforced by health authorities in 2020-21, thus eliminating local transmission of the virus in 96 the community, and 2) South Australia has maintained a high testing rate with a total test 97 count of >2.2 M with only 899 positive cases of which only 9 were caused by unknown, locally-98 acquired contacts (accessed on 23/09/2021). 31 99 We present a COVID-19 immunity study at 12 months after PCR confirmed SARS-CoV-2 100 infection and in the complete absence of community transmission in a South Australian 101 cohort of 43 mild COVID-19 convalescents. An in-depth evaluation of multi-isotype antibody 102 responses, homologous pseudotyped virus, homologous and VOC live-virus serum 103 neutralisation activity, RBD-specific B cell populations and Spike and non-Spike SARS-CoV-2 104 J o u r n a l P r e -p r o o f specific CD8 + and CD4 + T cell immunity against ancestral and VOC antigenic epitopes, was 105 undertaken. Results were compared to age-and gender-matched COVID-19 naïve, healthy 106 individuals and to COVID-19 convalescent responses at 6 months after infection in the same 107 cohort.

Longitudinal SARS-CoV-2 humoral responses in mild-COVID-19 convalescents. 110 The receptor binding domain (RBD) of SARS-CoV-2 Spike protein is the main target of 111 neutralising antibodies (nAb), and nAb titers decline in the months after COVID-19 infection. term SARS-CoV-2-specific humoral immunity 12 months post-infection. Spike-specific IgG + B 123 cells were also elevated in COVID-19 convalescents, but healthy control background 124 frequencies were also higher ( Figure S2 ), likely due to cross-reactivity. RBD-and Spike-specific 125 non-IgG + B cell frequencies were present a low rates (figure S2). 126 J o u r n a l P r e -p r o o f Obtaining long-term serum neutralisation data from communities free of circulating CoV-2, such as in this study, is difficult in other cohorts in the context of this pandemic. In our 128 cohort, sera in 64% of convalescents at 12 months yielded ID50 neutralisation titers 129 significantly above healthy background levels against the pseudotyped virus bearing a 130 Wuhan-like Spike protein, which is the same as the prevalent virus present in the community 131 when study participants were infected ( Figure 1D ). 132 Since early 2020, when the study participants were recruited, five VOCs, namely, Alpha Figure 1E ). Values were similar to those against pseudotyped 137 virus particles bearing the same Spike protein sequence in Figure 1D Spike-and non-Spike-specific T cell responses. 32 These T cells can be detected and quantified 152 by means of activation-induced marker (AIM) assays using SARS-CoV-2 antigen peptide pools 153 and flow cytometric analysis ( Figure S3 ). 20, 22, 23 For accurate interpretation in SARS-CoV-2 AIM 154 assays, naïve healthy controls must be included in the analysis to establish assay baseline 155 levels that arise from previous immunity to unrelated antigens/pathogens, and particularly to 156 seasonal human coronaviruses. 33 In our cohort, the relative frequency of CD4 + and CD8 + T 157 cells did not differ between the two time points (6 and 12 months) or between COVID-19 158 convalescents and healthy controls ( Figure S4A ). 159 AIM assays revealed that the frequency of circulating Spike-specific CD4 + T cells did not 160 significantly decrease between 6 and 12 moths post-infection (95% CI, from 0.35-0.91 at 5-6 161 months to 0.29-0.68) (Figure 2A ). The same trend was observed for non-Spike antigen CD4 + T 162 cells (from 0.22-0.43 to 0.18-0.37) and for the combined CD4 + T cell response, Spike + non-163 Spike (from 0.60-1.33 to 0.49-1.04) ( Figure 2A ). 164 Comparatively, the reduction of Spike-specific CD8 + T cell frequency over time was more 165 pronounced and statistically significant (95% CI, from 0.28-1.19 to 0.11-0.35, p<0.01) ( Figure   166 3B). No statistically significant differences were observed for the frequencies of CD8 + T cell 167 reacting to whole SARS-CoV-2 proteome pools A and B ( Figure 2B ). Convalescent AIM results 168 were compared to naïve healthy controls in all instances, corroborating significantly lower T 169 cell frequencies (baseline levels), consistent with previous AIM studies using the same peptide subset within CD8 + CD69 + CD137 + cells at 6 months (>65%) and at 12 months (>60%), 204 irrespective of antigen specificity ( Figure 4B ). 205 The presence of significant frequencies of SARS-CoV-2 antigen-specific CD4 + TEM and CD8 + 206 TEMRA at 12 months, has important implications, particularly in the absence of virus re-207 exposure, given that, in principle, these cells have a limited life-span. Importantly, a significant 208 decrease in these populations from 6 to 12 months suggests that in the absence of 209 restimulation by either infection or vaccination, these populations will eventually not be 210 detectable in circulation, or plateau at low frequencies for extended periods of time. In future, 211 the renewal mechanisms of both of these cell populations is an important question that needs 212 to be addressed.

Functionality of Spike-specific T cells in 'high-responder' convalescents. For the context of this experiment and its results, it is important to note that our cohort was 257 infected in early 2020 in Australia, where SARS-CoV-2 with unmutated Spike amino acid 258 sequences was prevalent ( Figure 6A ). In order to clearly identify changes in functionality of T 259 cell immunity to Spike antigens present in VOCs, we calculated fold-change of cytokine Figure 6B ) and for IFNγ in CD8 + T cells compared to B.1.617.2 274 (p<0.001, Figure 6C ). Notably, GZMB and PRF1 expression in CD8 + T cells was completely 275 unaffected by changes in the Spike sequence ( Figure 6C ), suggesting that cellular cytotoxic 276 immune function may be most resistant to antigenic changes. 277 To further analyse our data, we used a cut-off fold-change value of -2, meaning a two-fold Figure S11C ). 290 Here, we report a loss of Spike-specific T cell functionality against VOCs, despite durable (12 291 months) maintainance in overall frequency, has been reported in COVID-19 convalescents. Circles represent AUC individual patient values (n=43 at 6 months, orange, and 12 months, yellow, 514 n=15 for healthy controls, blue), with mean value denoted by a horizontal black line. Seronegative 515 samples were assigned a value of 0.001 data visualisation purposes. 516 (C) SARS-CoV-2 RBD-specific (n=28) memory B cells (CD27 + ) were quantified 12 month post-infectoion 517 with corresponding specific tetramers and further characterised as IgG + . 518

Cell population-specific background ( Neutralisation activity was considered negative, value of zero, when neutralisation of initial serum 530 dilution was <50%. *, **, *** and **** denote P values < 0.05, 0.01, 0.001, 0.0001 respectively. ns = 531 not significant. The percentage of convalescents with neutralisation activity is indicated for each VOC 532

in the corresponding X-axis labels. 533 Spike-, non-Spike-and cumulative (Spike + non-Spike)-specific CD4 + T cells detected in Figure 2A . Cells 551 were classified as follows: naïve (TN, CCR7 + CD45RA + ), central memory (TCM, CCR7 + CD45RA -), effector 552 memory (TEM, CCR7 -CD45RA -) and terminally differentiated effector memory cells re-expressing 553 CD45RA (TEMRA, CCR7 -CD45RA + ). Frequencies are indicated as percentage of total SARS-CoV-2 antigen-554 specific CD4 + T cells within the total pool of immune cells with same phenotype in the patient's PBMCs. 555 (B) Doughnut charts indicating the proportion (%) of each immune phenotype in (A) of the total of 556 SARS-CoV-2-specific CD4 + T cells for each antigen. In (A) circles represent patient individual values. 557

Averages are denoted by a horizontal line and statistically significant differences between time points 558 indicated by asterisks. * and **denote P values < 0.05 and 0.01, respectively. ns = not significant. 559 Spike-, whole proteome pools A-and B-specific CD8+ T cells detected in Figure 2B . Immune 564 phenotypes were defined as in Figure 3 . Frequencies are indicated as percentage of total SARS-CoV-2 565

antigen-specific CD8+ T cells within the total pool of immune cells with same phenotype in the 566 patient's PBMCs. 

SARS-CoV-2 and CMV peptide megapools were kindly provided by Prof Alessandro Sette

CD8-specific 733 peptide pools, 628 peptides restricted to the 12 most common HLA-A and HLA-B alleles and 734 partially covering the sequences of nsp1, nsp2, PLpro, nsp4, nsp6, nsp7, nucleocapsid 735 phosphoprotein, 3CL, nsp8, nsp9, nsp10, nsp14, RdRpol, Hel, nsp15, nsp16, surface 736 glycoprotein, ORF3a, ORF10, ORF6, ORF7a, ORF8, envelope protein, and membrane 737 glycoprotein were predicted in silico as previously described. 21 Peptides were divided into two 738 separate megapools, CD8_A and CD8_B. In Spike peptide pool, 15-mer peptides overlapping 739 by 10 amino acids and covering the entire Spike protein sequence were used (total of 253 740 peptides). For the non-Spike SARS-CoV-2 CD4 megapool 221 15-mer restricted to seven 741 common HLA-DR (Class II HLA) alleles and covering the entire SARS-CoV-2 proteome

complete RPMI (cRPMI) medium 748 (40 U/mL penicillin, 40 ug/mL streptomycin, 2 mM L-Glutamine) with 5% (v/v) heat-749 inactivated human AB serum. Cells were then plated at 10 6 PBMC/well in u-bottom 96-well 750 plates and stimulated with 1 µg/mL of different SARS-CoV-2-megapools. Combined CD4 and 751 CD8 cytomegalovirus (CMV) megapool (1 µg/mL) and PHA 10µg/mL (Sigma Aldrich), were 752 included as positive controls. An equimolar amount of dimethyl sulfoxide (DMSO, vehicle) was 753 used as a negative control. PBMC were stimulated for 24 h at 37°C, 5% CO2

Cells were then washed and stained with (CD3 BUV737, CD4 BUV496, CD8

41-BB) BV421) for 20min, at room temperature in the dark. Fluorescence minus one (FMO) antigens: CD45RA, CCR7, CD134, CD69 and CD137 were added to PHA stimulated 759 cells. PBMC were washed and FACS Fix (0.4%PFA, 20g/L Glucose, Sodium Azide 0.02% in PBS) 760 was added for 20 min at room temperature, in the dark. Fixed cells were washed, 761 resuspended in FACS wash buffer and data was acquired on BD FACS Symphony. Data analysis 762 was performed using FCS Express TM (DeNovo Software

Variant of concern Spike-specific peptide pools

Gladbach, GER) were utilised to 767 test immune reactivity of COVID-19 convalescent T cells to mutated Spike epitopes present in 768 five VOCs. 26,75 All peptide pools consisted of 15-mer peptides with 11 aa overlap covering 769 Spike protein sequences affected by mutations in each VOC. Five VOC peptide pools 770 corresponding to mutated Spike sequences in SARS-CoV-2 variants B

34, 30, 41, 32 and 83 peptides, respectively) and five corresponding 772 control/reference pools with Wuhan aa sequences were compared in parallel along with 773 whole Spike peptide pool as positive control (pool described above). Mutations and deletions 774 represented in mutated pools are summarized in Table S1. For the assays, lyophilised peptides 775 were resuspended in sterile miliQ water as per the manufacturer's instructions at 30 nM (50 776 µg/mL

Spike-specific T follicular helper cell quantification and intracellular cytokine staining in 778 Spike high responder convalescents

Double responders (meeting criterion for both CD4 + and CD8 + T cells) and high CD4 + 782 with available PBMC samples were selected (n=15) for further analysis. Following methods 783 similar to other published prior to this study, 76 PBMCs were thawed and prepared for cell 784 culture as described for the AIM assay. Cells were pre-treated with 0.555 µg/mL of anti-CD40 785 blocking antibody (HB14, Miltenyi Biotec) for 15 min. Then, peptide pools were added to a 786 final concentration of 1 µg/mL (making anti-CD40 concentration 0.5µg/mL for the remainder 787 of the stimulation period)

BD, 555029) were added to the 789 cells and incubated for an additional 4 hrs. Cells were then co-incubated with Fixable Viability 790 stain 780 (BD) and Fc Block (BD) for 20 min, RT, in the dark, washed with FACS buffer solution 791 and stained with surface stain mix

CD14 APC-Cy7, CD20 APC-Cy7) for 20 min, RT, in the dark. Cells were washed with PBS and 793 subsequently fixed and permeabilized with Cytofix/Cytoperm TM (BD, 51-2090KZ) for 20 min

the dark. Cells were then washed with Perm/Wash TM (BD, 51-2091KZ) and stained with 795 ICS stain Mix (CD154 PE, IFNγ PE-Cy7, TNFα APC, PRF1 FITC, IL-2 BV711, GZMB BV421

RT, in the dark. Cells were then washed twice with Perm/Wash TM and once with PBS

A pneumonia outbreak associated with a new coronavirus of probable bat origin

The impact of lockdown 845 timing on COVID-19 transmission across US counties

COVID-19 dynamics 848 after a national immunization program in Israel

Impact of scaling up SARS-CoV-2 vaccination on COVID-19 hospitalizations in Spain

Impact of vaccination on new SARS-CoV-2 855 infections in the United Kingdom

Vaccination reduces need for emergency care in breakthrough COVID-19 infections: A 859 multicenter cohort study

The Combined Effect of Vaccination and Nonpharmaceutical Public 862 Health Interventions-Ending the COVID-19 Pandemic

Neutralizing antibody titres in SARS-CoV-2 866 infections

Twelve-month specific IgG response to SARS-CoV-2 receptor-binding domain among COVID-869 19 convalescent plasma donors in Wuhan

Dynamics of antibody response to BNT162b2 vaccine 873 after six months: a longitudinal prospective study. The Lancet regional health

Large-scale study of antibody titer decay following BNT162b2 mRNA 877 vaccine or SARS-CoV-2 infection. medRxiv

Neutralizing Activity of BNT162b2-Elicited Serum. The New 881 England journal of medicine

Neutralization of the 884 SARS-CoV-2 Delta variant after heterologous and homologous BNT162b2 or ChAdOx1 nCoV-885 19 vaccination

Reduced sensitivity of SARS-CoV-2 variant 888 Delta to antibody neutralization

SARS-CoV-2 variant of concern following vaccination. medRxiv

Neutralization of SARS-CoV-2 Variants B.1.429 and B.1.351. The New England journal 895 of medicine

Neutralization of Beta and Delta variant with sera 898 of COVID-19 recovered cases and vaccinees of inactivated COVID-19 vaccine BBV152/Covaxin

Neutralization of SARS-CoV-2 variants by 902 convalescent and BNT162b2 vaccinated serum

Efficacy of NVX-CoV2373 Covid-19 Vaccine 906 against the B.1.351 Variant

Comprehensive analysis of T cell immunodominance 910 and immunoprevalence of SARS-CoV-2 epitopes in COVID-19 cases

A 913 Sequence Homology and Bioinformatic Approach Can Predict Candidate Targets for Immune 914 Responses to SARS-CoV-2

Immunological memory to SARS-CoV-2 assessed for up to 8 months 918 after infection

Targets of T Cell Responses to 921 SARS-CoV-2 Coronavirus in Humans with COVID-19 Disease and Unexposed Individuals

CoV-2 Human T cell Epitopes: adaptive immune response against COVID-19. Cell host & 925 microbe

Impact of SARS-CoV-2 variants on the total CD4+ and 928 CD8+ T cell reactivity in infected or vaccinated individuals

SARS-CoV-2 variants of 932 concern partially escape humoral but not T-cell responses in COVID-19 convalescent donors 933 and vaccinees

Protection of BNT162b2 Vaccine 939 Booster against Covid-19 in Israel. The New England journal of medicine

Distinguishing features of current COVID-19 vaccines: 944 knowns and unknowns of antigen presentation and modes of action. npj Vaccines 6

Adaptive immunity to SARS-CoV-2 and COVID-19

SARS-CoV-2 genome-wide T cell epitope mapping reveals 951 immunodominance and substantial CD8(+) T cell activation in COVID-19 patients

Automatic Generation of Validated Specific Epitope Sets

SARS-CoV-2-specific T cell memory is sustained in 958 COVID-19 convalescent patients for 10 months with successful development of stem cell-like 959 memory T cells

Human T Cell Development, Localization, 961 and Function throughout Life

Humoral and circulating follicular helper T cell 964 responses in recovered patients with COVID-19

Mutations of SARS-CoV-2 variants of concern escaping Spike-968 specific T cells. bioRxiv

Assessment of SARS-CoV-2 Reinfection 1 Year After Primary Infection in a Population

Antibody Status and Incidence of SARS-CoV-974 2 Infection in Health Care Workers

SARS-CoV-2 infection rates of antibody-positive 978 compared with antibody-negative health-care workers in England: a large, multicentre, 979 prospective cohort study (SIREN)

Long-term persistence of RBD(+) 982 memory B cells encoding neutralizing antibodies in SARS-CoV-2 infection

Evolution of immune responses to SARS-CoV-2 in mild-986 moderate COVID-19

Neutralizing antibody levels are highly 989 predictive of immune protection from symptomatic SARS-CoV-2 infection

Primary exposure to SARS-CoV-2 protects against reinfection in rhesus macaques

SARS-CoV-2 neutralizing antibodies

Longevity, breadth, and evasion by emerging viral variants

The durability of immunity against reinfection by SARS-CoV-2: 1000 a comparative evolutionary study. The Lancet

Omicron Variant from Previous SARS-CoV-2 Infection

Major role of IgM in the neutralizing activity of 1007 convalescent plasma against SARS-CoV-2

SARS-CoV-2 escape <em>in 1011 vitro</em> from a highly neutralizing COVID-19 convalescent plasma. bioRxiv

Reduced neutralization of SARS-CoV-2 B.1.617 by 1015 vaccine and convalescent serum

Considerable escape of SARS-CoV

Omicron to antibody neutralization

Omicron extensively but incompletely escapes 1021

Pfizer BNT162b2 neutralization

Tissue reservoirs of antiviral T cell 1024 immunity in persistent human CMV infection

mRNA vaccines induce 1028 durable immune memory to SARS-CoV-2 and variants of concern

CD8+ T cells in HIV control, cure and 1031 prevention

CD4+ T-cell-epitope escape mutant virus selected in vivo

Viral escape by selection 1036 of cytotoxic T cell-resistant variants in influenza A virus pneumonia. The Journal of 1037 experimental medicine 191

Stable cytotoxic T cell escape mutation in hepatitis C 1040 virus is linked to maintenance of viral fitness

Laboratory-Confirmed COVID-19 Among Adults Hospitalized with 1043 COVID-19-Like Illness with Infection-Induced or mRNA Vaccine-Induced SARS

Immunity -Nine States

Association of Prior SARS-1048

Infection With Risk of Breakthrough Infection Following mRNA Vaccination in Qatar

Neutralization of the SARS-CoV-2 Mu Variant by Convalescent and Vaccine 1053 Serum

Emergence and 1056 spread of a SARS-CoV-2 lineage A variant (A.23.1) with altered spike protein in Uganda

Long-term perturbation of 1060 the peripheral immune system months after SARS-CoV-2 infection. medRxiv

Peptidase inhibitor 16 identifies a human 1064 regulatory T-cell subset with reduced FOXP3 expression over the first year of recent onset 1065 type 1 diabetes

Structure-based design of prefusion

CoV-2 spikes

A serological assay to 1071 detect SARS-CoV-2 seroconversion in humans

A Multibasic Cleavage Site in the 1074

Spike Protein of SARS-CoV-2 Is Essential for Infection of Human Lung Cells

Mutations in hepatitis C virus E2 located outside 1078 the CD81 binding sites lead to escape from broadly neutralizing antibodies but compromise 1079 virus infectivity

In vitro assay for neutralizing antibody to hepatitis C 1082 virus: evidence for broadly conserved neutralization epitopes

Optimised cell systems for the investigation of hepatitis C virus E1E2 1086 glycoproteins. bioRxiv, 2020

Protocol and Reagents for Pseudotyping 1089

Lentiviral Particles with SARS-CoV-2 Spike Protein for Neutralization Assays

Revising a display of multidimensional laboratory measurements to 1092 improve accuracy of perception

Comparative Anti-Inflammatory Effects 1095 of Salix Cortex Extracts and Acetylsalicylic Acid in SARS-CoV-2 Peptide and LPS

Low-dose mRNA-1273 COVID-19 vaccine generates 1100 durable memory enhanced by cross-reactive T cells

Long-term persistence of RBD-positive 1104 memory B cells encoding neutralising antibodies in SARS-CoV-2 infection

HIGHLIGHTS • Most mild COVID-19 convalescents maintain immunity at 12 months after disease onset. • B.1.1.529 escapes antibodies in convalescents infected with ancestral SARS-CoV-2

CoV-2 VoC can partially avoid recognition by antigen-specific T cells

Antigenic drift in SARS-CoV-2 VOCs significantly challenges convalescent immunity

et al investigate immunity in mild COVID-19 convalescents at 12 months after infection in the absence of re-exposure to the virus and vaccination. Both neutralising antibody and Spike-specific T cell responses were significantly affected by the Spike amino acid differences