key: cord-0697128-86ubzx46
authors: Xiao, Chanchan; Mao, Lipeng; Wang, Zhigang; Zhu, Guodong; Gao, Lijuan; Su, Jun; Chen, Xiongfei; Yuan, Jun; Hu, Yutian; Luo, Oscar Junhong; Wang, Pengcheng; Chen, Guobing
title: SARS-CoV-2 variant B.1.1.7 caused HLA-A2+ CD8+ T cell epitope mutations for impaired cellular immune response
date: 2021-03-29
journal: bioRxiv
DOI: 10.1101/2021.03.28.437363
sha: bef5666b243d31e12550117abd8e54ecfbb07415
doc_id: 697128
cord_uid: 86ubzx46

The rapid spreading of the newly emerged SARS-CoV-2 variant, B.1.1.7, highlighted the requirements to better understand adaptive immune responses to this virus. Since CD8+ T cell responses play an important role in disease resolution and modulation in COVID-19 patients, it is essential to address whether these newly emerged mutations would result in altered immune responses. Here we evaluated the immune properties of the HLA-A2 restricted CD8+ T cell epitopes containing mutations from B.1.1.7, and furthermore performed a comprehensive analysis of the SARS-CoV-2 specific CD8+ T cell responses from COVID-19 convalescent patients recognizing the ancestral Wuhan strain compared to B.1.1.7. First, most of the predicted CD8+ T cell epitopes showed proper binding with HLA-A2, while epitopes from B.1.1.7 had lower binding capability than those from the ancestral strain. In addition, these peptides could effectively induced the activation and cytotoxicity of CD8+ T cells. Our results further showed that at least two site mutations in B.1.1.7 resulted in a decrease in CD8+ T cell activation and a possible immune evasion, namely A1708D mutation in ORF1ab1707-1716 and I2230T mutation in ORF1ab2230-2238. Our current analysis provides information that contributes to the understanding of SARS-CoV-2-specific CD8+ T cell responses elicited by infection of mutated strains. Graphical Abstract

The coronavirus disease 2019 pandemic has been sweeping the world. Its etiological agent, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) belongs to the Betacoronovirus genus of the Coronoviridae family, and this single-stranded positive-sensed RNA virus bears 11 protein-coding genes, including 4 for structural proteins: spike (S), envelope (E), membrane (M) and nucleocapsid (N), and 7 for nonstructural proteins: open reading frame (ORF) 1ab, ORF 3a, ORF 6, ORF 7a, ORF 7b, ORF 8 and ORF 10 (Wu et al., 2020) . It's believed that the viral clearance in SARS-CoV-2 infected individuals is mainly dependent on host immune system, especially adaptive immunity (Zhang et al., 2020) . Specific antibodies have been observed in virus infected individuals and convalescent COVID-19 patients, with S and N being the major viral proteins to elicit antibody production (Wheatley et al., 2021) . S protein bears the binding site to angiotensin converting enzyme 2 (ACE2) receptors on host cells and is crucial for viral infection.

So the neutralizing antibodies against S protein are believed to play an important role for the virus clearance (Bertoglio et al., 2021) . However, a couple of studies have shown that antibody titres decline fast in some convalescent patients, indicating a short duration of humoral immunity (Sette and Crotty, 2021) . On the other hand, current studies have demonstrated that specific T cell responses emerge in most of the COVID-19 patients during the early stage of the infection (Ferreras et al., 2021) . Although significant reduction in T cell counts was observed in severe patients, the revealed antigen specific T cell response indicated their important role in resolving SARS-CoV-2 infection (Grifoni et al., 2020; Le Bert et al., 2020; Weiskopf et al., 2020) . Furthermore, SARS-CoV-2 specific CD8 + T cells have been detected in convalescent COVID-19 patients (Braun et al., 2020; Grifoni et al., 2020; Sattler et al., 2020) and SARS-CoV-2 vaccinees (Jackson et al., 2020) . Recent studies have shown that specific CD8 + epitopes to SARS-CoV-2 are mainly located in ORF1ab, N protein, S protein, ORF 3, M protein and ORF 8(Ferretti et al., 2020; Grifoni et al., 2020) , and the identification of these epitopes will provide the basis for next-generation vaccine development and better understanding of CD8 + T cell immunity.

With the ongoing spreading of the virus all over the world, the genetic evolution in SARS-CoV-2 continues to provide the opportunities for the virus to obtain mutations which might contribute to the changes in viral transmissibility, infectivity, pathogenesis and even immune evasion (Neches et al., 2021; Rashid et al., 2021) . The D614G spike variant emerged in March 2020 was the earliest evidence of adaptive evolution of the virus in humans, which resulted in increased infectivity of the virus (Yurkovetskiy et al.) . Recently, a newer variant termed B.1.1.7 (also called VUI202012/01) was spreading rapidly in the United Kingdom (UK) and raised great concerns (Davies et al., 2021; Kirby, 2021) . This variant contains 17 non-synonymous mutations in ORF1ab, S protein, ORF8 and N proteins, some of which are of particular concerns, such as the D614G mutation and 8 additional mutations in S protein: ΔH69-V70, ΔY144, N501Y, A570D, P681H, T716I, S982A, and D1118H(Davies et al., 2021 . For example, N501Y is located in the receptor binding motif (RBM) and P681H is proximal to the furin cleavage site (Peacock et al., 2020; Starr et al., 2020) . ΔH69/ΔV70 deletion in S protein has evolved in other lineages of SARS-CoV-2, which enhances viral infectivity in vitro and is linked to immune escape in immunocompromised patients (Kemp et al., 2020a; Kemp et al., 2020b) .

There is strong evidence that variant B.1.1.7 is spreading substantially faster than preexisting SARS-CoV-2 variants (Davies et al., 2021; Kemp et al., 2020b; Volz et al., 2021) . The model analysis suggests that this difference could be explained by an overall higher infectiousness of variant B.1.1.7. However, it is not clear that this is due to the shorter generation time or immune escape (Davies et al., 2021) . Mutations in immune dominant epitopes might potentially alter their immunogenicity, and subsequently the immune responses of the host.

Our previous work has shown that the mutations in given CD8 + T cell epitopes resulted in antigen presentation deficiency and impaired antigen specific T cell function, indicating an immune evasion induced by viral evolution (Qiu et al., 2020) .

In this work, we predicted the potential CD8 + T cell epitopes within the areas where these mutations are located, and compared the immune properties of the ancestral and mutant peptides, including MHC I binding and activation of CD8 + T cells.

Furthermore, we detected the epitope specific CD8 + T cells in convalescent COVID-19 patients by using corresponding tetramers. The results showed that at least two site mutations in the variant B.1.1.7 resulted in a decrease in CD8 + T cell activation and a possible immune evasion, namely A1708D mutation in ORF1ab1707-1716 and I2230T mutation in ORF1ab2230-2238. Our current analysis provides useful information that helps for better understanding of the SARS-CoV-2-specific CD8 + T cell responses elicited by infection of mutated strains.

During late 2020, the World Health Organization (WHO) announced the emergence of a novel coronavirus variant B.1.1.7 in the UK. We immediately carried out HLA-A2-restricted T cell epitope screening and identification of all the possible peptides containing the mutations in B.1.1.7 by using the high-throughput screening platform and artificial antigen presentation system for epitopes ( Figure 1A Table 1 ). To validate these predicted epitopes, we first checked whether they could be presented by HLA-A2 on the antigen presenting cells (APC).

T2A2 is an APC with TAP deficiency and HLA-A2 expression on cell surface. The peptide-MHC complex would be more stabilized if the epitopes bind with HLA-A2

suitably. Compared with the negative control peptide GLQ (GLQRLGYVL) from Zika virus, most of the predicted SARS-CoV-2 epitopes showed reasonable HLA-A2 binding. However, the HLA-A2 binding of most of the epitopes from the variant B.1.1.7 was reduced ( Figure 1C -D). We further checked the direct binding of these epitopes to the purified HLA-A2 protein. According to the UV exchanged peptide-MHC assay, all the peptides, except for N S235F, exhibited strong binding to HLA-A2 ( Figure 1E ). The results indicated that majority of the predicted epitopes could form peptide-MHC complex (pMHC), and corresponding tetramers could be constructed next.

Activation and cytotoxicity of T cells stimulated with T cell epitopes containing mutations of B.1.1.7

To further analyze whether the epitope-bound T2A2 cells could activate T cells, we tested the expression level of T cell activation marker CD69 and the proportion of peptide-specific CD8 + T cells after stimulation with peptide-bound T2A2 cells. As shown in Figure 2A However, the mutant peptide group had a higher proportion of survival target cells compared to the ancestral group, suggesting a decreased cytotoxicity from mutant peptide specific CD8 + T cells ( Figure 2J -K). In addition, the proportion of CFSE-Annexin V + T2A2 was less for mutant peptide group than that for ancestral peptide group, indicating less T cell-mediated target cell apoptosis ( Figure 2L -M).

Finally, the CD8 + IFN- ( Figure 2N -O) and Granzyme B ( Figure 2P -Q) levels in mutant peptide group were significantly lower than those in ancestral peptide group.

All the above results suggested that, compared to the ancestral, T cell mediated immune responses induced by B.1.1.7 mutant peptides were impaired.

We recruited a cohort of 25 convalescent COVID-19 patients with 4 being identified as HLA-A2 positive. The demographic and clinical information of the patients with tetramer + cells were presented in Supplementary Table 2. We examined the ex vivo phenotypes of SARS-CoV-2 tetramer + CD8 + T cells in PBMC of the patients by assessing the expression levels of the chemokine receptor CCR7 and CD45RA. It's observed that the tetramers prepared with the above identified epitopes could recognize the specific memory T cells in convalescent patients ( Figure 3A -B).

(CCR7 -CD45RA -) phenotype, and roughly 20% of tetramer + cells were naїve T (CCR7 + CD45RA + ) cells ( Figure 3C -D). Meanwhile, in the same individual, the proportion of T cells recognized by the B.1.1.7 mutant epitope tetramers was lower than that by the ancesral epitope tetramers ( Figure 3B ). All above data indicated that these emerged mutations might have caused a deficiency in the antigen presentation of the dominant epitopes, which was required to rebuild a new CD8 + T cell immune response in COVID-19 patients.

To further explore the binding pattern between peptides and HLA-A2, molecular docking model was established with Galaxypepdock, and pair-wise comparison of pMHC structure was performed between ORF1ab1707-1716 and ORF1ab A1708D, ORF1ab2225-2234 and ORF1ab I2230T-1, ORF1ab2230-2238 and ORF1ab I2230T-2, respectively. It's observed that these mutations slightly decreased the interaction similarity of mutant peptides to HLA-A2 ( Figure 4A ). Interestingly, subtle structural alterations of peptides presented by HLA-A2 were observed before and after mutation.

Molecular docking comparison between ORF1ab 1707-1716 ( Figure 4B , blue) and

ORF1ab A1708D ( Figure 4B , red) showed A1708D mutation caused deflection of the benzene ring of the subsequent asparagine (N) ( Figure 4C , angle from 145.4 。 to 101.8 。 ). Modeling of ORF1ab2225-2234 ( Figure 4D , blue) and ORF1ab I2230T-1 ( Figure   4D , red) showed that the I2230T mutation herein might affect the later tryptophan (W), making the two benzene rings more convex ( Figure 4E , angle from 66.7 。 to 92.2 。 ).

Modeling of ORF1ab2230-2238 ( Figure 4F , blue) and ORF1ab I2230T-2 ( Figure 4F , red)

showed that tryptophan (W) was more convex, and its benzene ring tended to expand ( Figure 4G , angle from 0 to 66.6 。 ). These structural changes might provide the possible molecular basis for the altered antigen presentation and CD8 + T cell activation, while further protein crystallographic analysis is needed for confirmation.

The soaring rise of SARS-CoV-2 infection in the last months of 2020 has led to the evolution of several variants with related mutations or characteristics (Lauring and Hodcroft, 2021) . One such variant, designated B.1.1.7, was identified in the UK during late 2020 and continued to dominate the circulation in the region. Recent studies have reported longer persistence and higher viral loads in samples from B.1.1.7 infected individuals, indicating its association with the higher infectivity and transmissibility (Calistri et al., 2021; Parker et al., 2021) . It's also reported that B.1.1.7 might even lead to more severe illness (Challen et al., 2021) . Our study aimed to fill a key knowledge gap addressing the potential of SARS-CoV-2 variants to evade recognition by human immune responses. Based on the mutation sites in B.1.1.7, we performed computational prediction of HLA-A2-restricted CD8 + T cell epitopes, and obtained 19 potential epitopes for ancestral Wuhan strain and 20 for variant B.1.1.7, respectively. To validate the binding of these predicted epitopes, we then checked whether they could be presented on T2A2 cells, where the peptide-MHC complex 9 / 22 would be more stabilized if the epitopes bind with HLA-A2 suitably. Our results showed that most of the peptides had reasonable binding with HLA-A2, while the binding capability of most mutant peptides was lower than that of the ancestral.

Recently, Tarke et al. reported an identification of 523 CD8 + T cell epitopes associated with unique HLA restrictions (Tarke et al., 2021a) , and 508 (97.1%) of them were totally conserved within the B.1.1.7 mutant (Tarke et al., 2021b) , which might be a reason why they didn't see significant difference in the T cell reactivity to the ancestral and mutant peptides. By using computational prediction, they reported 73.3% of the mutations were not associated with decrease in binding capacity (Tarke et al., 2021b) . The difference of the results may be due to the different verification methods for the binding ability of SARS-CoV-2 T cell epitopes.

Up to date, SARS-CoV-2 mutations of most concern were existed in the viral (Collier et al., 2021; Muik et al., 2021) . Furthermore, slightly but significantly decreased sensitivity was observed to the sera from SARS-CoV-2 vaccinees and convalescent patients (Muik et al., 2021; Wang et al., 2021) .

SARS-CoV-2 vaccine clinical trial data demonstrated that specific CD8 + response was elicited as well as antibody production (Sahin et al., 2020) , and the rapid emergence of the protection at the time when antibodies were still low further supported the important role of cellular immunity (Polack et al., 2020) .

So far, the only report assessing the cellular immunity against B.1.1.7 is from Tarke et al, in which they evaluated the CD8 + T cell reactivity in convalescent patients by using proteome-wide overlapping peptide megapools, and reported similar responses between ancestral and B1.1.7 (Tarke et al., 2021b) . In our study, altered CD8 + T cell response was observed for particular CD8 + epitopes. Our results first did show that mixed epitope-loaded antigen presentation cells could activate T cells from healthy donors. Notably, the proportion of CD8 + T cells specific to certain mutant peptides was less than that to ancestral in the same host. In addition, the ancestral epitope specific CD8 + T cells could not be recognized by tetramers prepared with mutant epitopes, and vice versa. All these results showed that the T cell mediated immune responses induced by variant B.1.1.7 was decreased. Our previous work has also shown that the L>F mutations in spike protein epitope FVFLVLVPLV resulted in antigen presentation deficiency and reduced specific T cell function, indicating an immune evasion induced by viral evolution (Qiu et al., 2020) . The impaired immune responses were further confirmed with the epitope specific CD8 + T cells measurement 

The Institutional Review Board of the Affiliated Huaqiao Hospital of Jinan University approved this study. Unexposed donors were healthy individuals enrolled in Guangzhou Blood Center and confirmed with a negative report for SARS-CoV-2 RNA real-time reverse transcriptase polymerase chain reaction (RT-PCR) assay. These donors had no known history of any significant systemic diseases, including, but not limited to, hepatitis B or C, HIV, diabetes, kidney or liver diseases, malignant tumors, or autoimmune diseases. Convalescent donors included subjects who were hospitalized for COVID-19 or confirmed SARS-CoV-2 infection by RT-PCR assay (Supplementary Table 2 ). All subjects provided informed consent at the time of enrollment that their samples could be used for this study. Complete blood samples were collected in acid citrate dextrose tubes and stored at room temperature prior to peripheral blood mononuclear cells (PBMCs) isolation and plasma collection. PBMCs were isolated by density gradient centrifugation using lymphocyte separation medium (GE). After isolation, the cells were cryopreserved in fetal bovine serum (LONSERA ) with 10% dimethyl sulfoxide (DMSO) (Sigma-Aldrich) until use.

The spike (S), membrane (M), nucleocapsid (N) and ORF protein sequences of SARS-CoV-2 Wuhan-Hu-1 strain (NC_045512.2) were used for T cell epitope prediction with the "MHC I Binding"tool (http://tools.iedb.org/mhci). The prediction All predicted epitopes containing the same amino acid residue corresponding to the mutation from B.1.1.7 were compared. The peptide with the best prediction score was used as the candidate epitope for ancestral Wuhan strain. Meanwhile, peptides with identical amino acid sequences except for the mutated point were used as candidate epitopes for variant B.1.1.7.

The candidate peptides were synthesized in GenScript Biotechnology Co., Ltd (Nanjing, China) and resuspended in DMSO at a concentration of 10 mM, respectively. T2A2 cells were seeded into 96-well plates, and then incubated with peptides at a final concentration of 20 µM at 37°C for 4 hours. Set DMSO as blank control, Influenza A M1 peptide (GILGFVFTL) as positive control, and Zika virus peptide (GLQRLGYVL) as negative control. Cells were stained with PE anti-human HLA-A2 antibody (BioLegend) at 4°C in the dark for 30 min, and acquired in flow cytometer FACS Canto (BD).

10 mM peptide stock solution was diluted to 400 µM in PBS. 20 µL diluted peptide and 20µL 1µg/mL UV-sensitive peptide HLA-A2 monomer (BioLegend) were added into 96-well plates and mixed well by pipetting up and down. The plates were then exposed to UV light (365 nm) for 30 min on ice, and incubated for 30 min at 37°C in the dark. Finally, 40 µL of peptide-exchanged monomer was used for test. 

HLA-A2 expressing T2A2 cells were loaded with peptides for subsequent T cell activation. Briefly, T2A2 cells were treated with 20 µg/mL mitomycin C (Sinochem) for 30 min to stop cell proliferation, and loaded with given epitope peptides. 0.5×10 6 CD8 + T cells isolated from health donors were co-cultured with 0.5 × 10 6 peptide-loaded T2A2 cells stained with 5 µmol/L CFSE (TargetMol), and stimulated with 1 µg/mL anti-human CD28 antibodies (BioLegend) and 50 IU/mL IL-2 (SL PHARM, Recombinant Human Interleukin-2( 125 Ala) Injection). 50 IU/mL IL-2 and 20 µM mixed peptides were then supplemented every two days. The T cell activation marker CD69 (BioLegend), tetramer specific CD8 + T cells and apoptosis marker Annexin V-APC (BioLegend) on T2A2 cells were evaluated after 16 hours and 7 days, respectively. On day 7, cells were re-stimulated with peptides for 6 hours in the presence of Leuko Act Cktl with GolgiPlug (BD) plus 50 IU/mL IL-2, and the production of IFN-γ and Granzyme B was checked with PerCP anti-human IFN-γ (BioLegend) and FITC anti-human Granzyme B (BioLegend) staining.

To evaluate the binding pattern and affinity of peptides with HLA-A2, molecular docking simulation was carried out with Galaxypepdock. The available structure of HLA: 0201 (PDB ID: 3mrb) was downloaded from the RSCB PDB server (https://www.rcsb.org/) for modeling. Galaxypepdock is a template-based docking program for peptides and proteins, which can generate 10 models to evaluate the results of the docking (Hasup et al., 2015; Mani et al., 2020) . The top model with the highest interaction similarity score was selected and visualized by using Discovery Studio 4.5. PyMol 1.1 software was used to calculate the angle deflection of benzene ring in the polypeptide, and used the central atoms of three amino acids to calculate the angle.

The data were analyzed by one-way ANOVA and paired-samples t-tests for statistical significance by using Graphpad prism 8 and SPSS 22.0 software. P value less than 0.05 was considered to be statistically significant. assisted with data analysis. C.X., P.W. and G.C. wrote the manuscript.

The epitopes and tetramers from this study are the subjects of a patent application.

severe versus mild individuals. Signal transduction and targeted therapy 5, 156. Mitomycin pretreated T2A2 cells were loaded with mixed peptides from ancestral or mutant, and incubated with CD8 + T cell from health donors at 1:1 ratio, respectively.

Epitope specific CD8 + T cells were generated after 7 day stimulation. The p values were calculated by paired-samples T test. *p < 0.05, **p < 0.01, ***p < 0.001. 

HLA common and well-documented alleles in China

An mRNA vaccine against SARS-COV-2 -Preliminary report

Recurrent emergence and transmission of a SARS-CoV-2 Spike deletion ΔH69/V70

Neutralising antibodies drive Spike mediated SARS-CoV-2 evasion

New variant of SARS-CoV-2 in UK causes surge of COVID-19. The Lancet Respiratory Medicine

Genetic Variants of SARS-CoV-2-What Do They Mean?

SARS-CoV-2-specific T cell immunity in cases of COVID-19 and SARS, and uninfected controls

Molecular docking, validation, dynamics simulations, and pharmacokinetic prediction of natural compounds against the SARS-CoV-2 main-protease

Neutralization of SARS-CoV-2 lineage B.1.1.7 pseudovirus by BNT162b2 vaccine-elicited human sera

Atypical Divergence of SARS-CoV-2 Orf8 from Orf7a within the Coronavirus Lineage Suggests Potential Stealthy Viral Strategies in Immune Evasion

The furin cleavage site of SARS-CoV-2 spike protein is a key determinant for

Safety and Efficacy of the BNT162b2 mRNA Covid-19 Vaccine

CD8+ T Cell Epitope Variations Suggest a Potential Antigen Presentation Deficiency for Spike Protein of SARS-CoV-2

The ORF8 protein of SARS-CoV-2 induced endoplasmic reticulum stress and mediated immune evasion by

BNT162b2 induces SARS-CoV-2-neutralising antibodies and T cells in humans. medRxiv

-2 specific T-cell responses and correlations with COVID-19 patient predisposition

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

Deep mutational scanning of SARS-CoV-2 receptor binding domain reveals constraints on folding and ACE2 binding

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

Negligible impact of SARS-CoV-2 variants on CD4 + and CD8 + T cell reactivity in COVID-19 exposed donors and vaccinees

Transmission of SARS-CoV-2 Lineage B.1.1.7 in England: Insights from linking epidemiological and genetic data

Antibody Resistance of SARS-CoV-2 Variants B

Phenotype and kinetics of SARS-CoV-2-specific T cells in COVID-19 patients with acute respiratory distress syndrome

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

Genome Composition and Divergence of the Novel Coronavirus (2019-nCoV) Originating in China

Structural and Functional Analysis of the D614G SARS-CoV-2 Spike Protein Variant

SARS-CoV-2 RBD in vitro evolution follows contagious mutation spread, yet generates an able infection inhibitor

Adaptive immune responses to SARS-CoV-2 infection in