key: cord-0891422-72f64kpf
authors: Moriyama, Miyu; Lucas, Carolina; Monteiro, Valter Silva; Iwasaki, Akiko
title: SARS-CoV-2 variants do not evolve to promote further escape from MHC-I recognition
date: 2022-05-16
journal: bioRxiv
DOI: 10.1101/2022.05.04.490614
sha: 287427b6568ba7e92ea81ee0e96eb2295cf93fbf
doc_id: 891422
cord_uid: 72f64kpf

SARS-CoV-2 variants of concern (VOCs) possess mutations that confer resistance to neutralizing antibodies within the Spike protein and are associated with breakthrough infection and reinfection. By contrast, less is known about the escape from CD8+ T cell-mediated immunity by VOC. Here, we demonstrated that VOCs retain similar MHC-I downregulation capacity compared to the ancestral virus. However, VOCs exhibit a greater ability to suppress type I IFN than the ancestral virus. Although VOCs possess unique mutations within the ORF8 gene, which suppresses MHC-I expression, none of these mutations enhanced the ability of ORF8 to suppress MHC-I expression. Notably, MHC-I upregulation was strongly inhibited after the ancestral SARS-CoV-2 infection in vivo. Collectively, our data suggest that the ancestral SARS-CoV-2 already possesses an intrinsically potent MHC-I evasion capacity, and that further adaptation by the variants was not observed. Summary Moriyama et al. demonstrate that SARS-CoV-2 variants of concern retain similar MHC-I downregulation capacity compared to the ancestral virus. The results suggest that MHC-I evasion capacity is optimized in the ancestral virus and thus further adaptation was not observed.

Moriyama et al. demonstrate that SARS-CoV-2 variants of concern retain similar MHC-I downregulation capacity compared to the ancestral virus. The results suggest that MHC-I evasion capacity is optimized in the ancestral virus and thus further adaptation was not observed. Introduction 9 conserved among different lineages. To investigate the prevalence of mutations found in variants, we downloaded 3,059 SARS-CoV-2 genome sequence data from GISAID database (https://www.gisaid.org/). We found that the mutation in a particular amino acid is only exclusively seen in a single lineage ( Fig.2A) . ORF8 L84S mutation, which was detected within the first 2 months of pandemic (Ceraolo and Giorgi, 2020 ) and corresponding to clade S, was not observed in any of the variants. We also observed the mutations discovered by multiple sequence alignment are generally highly prevalent, and the proportions ranged from 12.5 to 100% of the lineage (Fig.2B) . These results indicate that the variantspecific mutations were acquired independently during SARS-CoV-2 evolution.

We next tested whether variant-specific mutations alter MHC-I downregulating capability of ORF8 protein. To this end, we generated expression plasmids encoding six ORF8 mutants from SARS-CoV-2 variants (Fig.2C) , and subsequently transfected HEK293T cells with these plasmids for the detection of its effect on the surface MHC-I expression levels. We included SARS-CoV ORF8a/b proteins as negative controls, as they have been shown not to affect MHC-I expression levels (Zhang et al., 2021) . Since ORF8 induces degradation of MHC-I via autophagy by interacting with MHC-I and localizing in LC3-positive puncta (Zhang et al., 2021) , ORF8 presumably acts on MHC-I downregulation in cell-intrinsic manner. Indeed, surface MHC-I levels of the cells expressing WT ORF8 protein were much lower than those of the cells without ORF8 expression ( Fig.2D) . Among six ORF8 mutants tested, five mutants including I121L, E92K, del119-120, V100L, and T11I downregulated surface MHC-I levels of the cells expressing those proteins to a similar extent to WT ORF8 protein did (Fig.2E ).

On the other hand, Q27Stop ORF8 mutant showed significantly attenuated MHC-I downregulation capability compared to the WT ORF8 protein. These results indicated that none of the variant-specific mutations enhanced the ability of ORF8 protein to downregulate MHC-I, and the ORF8 encoded by the B.1.1.7 lineage was attenuated in its ability to reduce surface MHC I expression.

Given that B.1.1.7 and P.1 variants were able to reduce MHC-I expression levels even though these lineages retain functionally defective ORF8 mutant or are less effective in reducing HLA-I mRNA levels, we examined whether SARS-CoV-2 variants suppress interferon-stimulated gene (ISG) induction more strongly than the ancestral strain. Interestingly, all variants tested significantly reduced upregulation of ISG expressions compared to the ancestral strain (Fig.3) .

These results indicated that SARS-CoV-2 variants have an increased capacity to suppress ISGs, which may include genes involved in MHC I processing and presentation pathways.

As another possible compensation mechanism, we investigated the possibility that SARS-CoV-2 encodes multiple viral genes that redundantly act on MHC-I downregulation. We generated expression plasmids encoding SARS-CoV-2 E, M, ORF7a, and ORF7b, and assessed the effect on the surface MHC-I and MHC-II expression levels of HEK293T cells following transfection with these plasmids. We also included Human Immunodeficiency Virus (HIV) Nef as a positive control for downregulating both MHC-I and MHC-II (Schwartz et al., 1996; Stumptner-Cuvelette et al., 2001) , and SARS-CoV ORF8a/b proteins as a negative control. As expected, HIV Nef protein downregulated both MHC-I and MHC-II levels, whereas SARS-CoV-2 ORF8 specifically targeted MHC-I ( Fig.4A-B ). We found that in addition to ORF8, SARS-CoV-2 E, M, and ORF7a substantially downregulated MHC-I within the cells expressing these viral proteins (Fig. 4C ). Significant reduction of surface MHC-II levels was also observed by expression of these viral proteins (Fig. 4C ), albeit to a lesser extent (~20%). These results suggested that SARS-CoV-2 encodes multiple viral genes that are redundantly downregulating MHC-I likely to ensure viral evasion from MHC-I-mediated CTL recognition.

In the experiments above, we have shown that SARS-CoV-2 encodes multiple viral proteins that are targeting MHC-I expression, which can synergistically strengthen the capability of the virus to avoid MHC-I presentation.

We also demonstrated that SARS-CoV-2 variants have an enhanced capacity to suppress ISG induction than the ancestral strain but not MHC-I, which raised the possibility that the MHC-I evasion strategy was already optimal in the ancestral strain. Moreover, we confirmed the previous finding that the MHC-I downregulation is a newly acquired function of SARS-CoV-2 ORF8 protein, which was not seen in SARS-CoV ORF8a/b proteins. Considering these results, we hypothesized that even the ancestral strain of SARS-CoV-2 possesses a superior MHC-I evasion strategy than other respiratory viruses. To assess this hypothesis, we infected C57BL/6J mice intranasally with a mouse-adapted strain of SARS- 

CD8 + T cell-mediated elimination of infected cells plays an important role in the antiviral adaptive immune response. Thus, many viruses have developed ways to avoid the efficient MHC-I mediated antigen presentation to CD8 + T cells.

In the current study, we uncovered the intrinsically potent ability of SARS-CoV-2 to shut down the host MHC-I system by using live, authentic SARS-CoV-2 variants as well as the functional analysis of variant-specific mutations in ORF8 gene, a key viral protein for both MHC-I evasion and adaptation to the host.

We demonstrated that most variants of concern/interest possess unique mutations within ORF8 gene. By functional analysis using mutant ORF8, we found that none of the mutations enhanced the ability of ORF8 to suppress MHC-I presentation. Rather, the mutation in B.1.1.7/Alpha lineage attenuated its function. These results raised the question if these mutations are beneficial for the virus. One possibility is that these mutations play roles in modifying ORF8 functions independent of MHC-I downregulation. Our data indeed demonstrated the enhanced ability of VOCs to suppress ISG expression. Multiple functions of SARS-CoV-2 ORF8 were reported so far, which include inhibition of type I IFN,

ISGs, or NF-kB signaling (Geng et al., 2021; Lei et al., 2020; Li et al., 2020) and modulation of cytokine expression from macrophages (Kriplani et al., 2021) .

Interestingly, several studies showed that SARS-CoV-2 ORF8 is actively secreted into the cell culture media in a signal peptide-dependent manner when it is overexpressed in vitro (Kriplani et al., 2021; Wang et al., 2020) . Furthermore, ORF8 peptides and anti-ORF8 antibodies can be detected abundantly in serum of patients, suggesting the relevance of the active secretion of ORF8 to actual infection in humans . Likewise, studies from early in the COVID-19 pandemic observed the variability and rapid evolution of SARS-CoV-2 ORF8 gene (Alkhansa et al., 2021; Velazquez-Salinas et al., 2020) . Notably, SARS-CoV-2 isolates with 382nt deletion spanning ORF7b-ORF8 gene region were observed in Singapore (Su et al., 2020) , which correlated with robust T cell response and mild clinical outcome (Fong et al., 2022; Young et al., 2020) . Mutations in ORF8 gene thus may play a key role in modulating viral pathogenesis and adaptation to the host by regulating MHC-I levels and ISGs.

The enhanced immune evasion by VOCs has been well documented for escape from neutralizing antibodies (Garcia-Beltran et al., 2021; Lucas et al., 2021; Planas et al., 2022; Wang et al., 2021) and from innate immune responses (Guo et al., 2021; Thorne et al., 2022) . Here we demonstrated that SARS-CoV-2 variants of concern have evolved to better limit the host type I IFN response. In lineage has been shown to express an increased subgenomic RNA and protein abundance of ORF6 (Thorne et al., 2022) , which suppresses MHC-I at the transcriptional level by interfering with STAT1-IRF1-NLRC5 axis (Yoo et al., 2021) .

The multi-tiered MHC-I evasion mechanisms thus work redundantly to ensure escape from CTL killing.

Second, MHC-I downregulation may not only impair CTL recognition of infected cells for killing but may also impair priming of CD8 T cells. Indeed, the frequency of circulating SARS-CoV-2 specific memory CD8 + T cells in SARS-CoV-2 infected individuals are ~10 fold lower than for influenza or Epstein-Barr virus-specific T cell populations (Habel et al., 2020) , which indicates the suboptimal induction of memory CD8 + T cells following SARS-CoV-2 infection in human.

Third, given that the variants of concern had not further evolved to downregulate MHC-I more strongly than the original strain, SARS-CoV-2 ancestral virus was already fully optimized for escape from CD8 + T cell-mediated immunity with respect to downregulation of MHC-I expression and is under no evolutionary pressure to further optimize the evasion strategy. However, mutations and evasion from particular HLA-restricted CTL epitopes have been observed in circulating SARS-CoV-2 and VOCs (Agerer et al., 2021; Motozono et al., 2021) . Genome-wide screening of epitopes suggested the CD8 + T and CD4 + T cell epitopes are broadly distributed throughout SARS-CoV-2 genome (Ferretti et al., 2020; Tarke et al., 2021a) , and the estimated numbers of epitopes per individual are at least 17 for CD8 + T and 19 for CD4 + T cells, respectively (Tarke et al., 2021a) , and thus functional T cell evasion by VOCs is very limited (Tarke et al., 2021b) . This in turn suggests that MHC-I downregulation may be a more efficient way for viruses to avoid CTL surveillance than introducing mutations in epitopes, a process that appears to be maximally present in the ancestral SARS-CoV-2 lineage. The importance of MHC-I evasion by SARS-CoV-2 is also highlighted by the fact that no genetic mutations or variations in the MHC-I pathway has thus far been identified as a risk factor for severe COVID (COVID-19 Host Genetics Initiative, 2021), unlike innate immune pathways involving TLRs and type I IFNs (Zhang et al., 2022) .

to induce antigen-specific CD8 + T cell responses (Grifoni et al., 2021; Joag et al., 2021) , and the early CTL response correlated with a milder disease outcome in human (Tan et al., 2021) . Adoptive transfer of serum or IgG from convalescent animals alone, however, is enough to reduce viral load in recipients after SARS-CoV-2 challenge in mice and non-human primates (Israelow et al., 2021; McMahan et al., 2021) and neutralizing antibody is shown to be a strong correlate of protection (Earle et al., 2021; Israelow et al., 2021; Khoury et al., 2021) . The protective roles of CD8 + T cell-mediated immunity appear to be more important in the absence of the optimal humoral responses/neutralizing antibody (Bange et al., 2021; Israelow et al., 2021) . Blood anti-ORF8 antibodies can be used as the highly sensitive clinical marker for SARS-CoV-2 infection early (~14 days) after symptom onset (Hachim et al., 2020; Wang et al., 2020) , which suggests the role of ORF8 in the very early stage of the disease. ORF8-mediated MHC-I downregulation can therefore precede antigen presentation and hinder priming of viral antigen-specific CD8 + T cell immune responses. Robust MHC-I shutdown by SARS-CoV-2 may explain in part the less effective protection by CD8 + T cells and the less impact of CD8 + T cell absence compared with humoral immunity (Israelow et al., 2021) .

Collectively, our data shed light on the intrinsically potent ability of SARS-CoV-2 to avoid the MHC-I mediated antigen presentation to CD8 + T cells.

Importantly, we observed a complete inhibition of MHC-I upregulation in lung epithelial cells infected with SARS-CoV-2 at the early stage of infection in a mouse model. Since the ability of ORF8 to downregulate MHC-I is a newly acquired feature in SARS-CoV-2 ORF8 and is absent in SARS-CoV ORF8 (Zhang et al., 2021) , it is possible that ORF8 played a role in the efficient replication and transmission of SARS-CoV-2 in human and contributed to its 20 pandemic potential. Our work provides insights into SARS-CoV-2 pathogenesis and evolution and predicts difficulty for CD8 T cell-based therapeutic approaches to COVID-19.

Six to ten-week-old male C57BL6 mice were purchased from the Jackson Laboratory. All animal experiments in this study complied with federal and institutional policies of the Yale Animal Care and Use Committee. 1.7, B.1.351a, P.1, B.1.617.2, B.1.427, B.1.429, and B.1.526 lineages were isolated and sequenced as part of the Yale Genomic Surveillance Initiative's weekly surveillance program in Connecticut, United States, as previously described (Mao et al., 2022) . Virus stocks were propagated and titered as previously described (Lucas et al., 2021; Perez-Then et al., 2022) .

Briefly, TMPRSS2-VeroE6 cells were infected at multiplicity of infection of 0.01 for 3 days and the cell-free supernatant was collected and used as working stocks.

All experiments using live SARS-CoV-2 were performed in a biosafety level 3 laboratory with approval from the Yale Environmental Health and Safety office. 

Lungs were harvested and processed as previously described (Israelow et al., 2021) . In Brief, lungs were minced with scissors and digested in RPMI1640 media containing 1mg/ml collagenase A, 30ug/ml DNase I at 37C for 45 min. Digested lungs were then filtered through 70um cell strainer and treated with ACK buffer for 2 min. After washing with PBS, cells were resuspended in PBS with 1% FBS. 

Statistical significance was tested using one-way analysis of variance (ANOVA) with Tukey's multiple comparison test. P-values of <0.05 were considered statistically significant.

We thank Melissa Linehan and Huiping Dong for technical and logistical assistance. We thank Ralph Baric for kindly providing SARS-CoV-2 MA10. We thank Ya-Chi Ho for kindly providing NL4-3-dE-EGFP. We thank Craig Wilen for his technical expertise. We thank Benjamin Israelow and Tianyang Mao for critical reading of the manuscript. We also give special recognition to the services of Ben Fontes and the Yale EH&S Department for their ongoing assistance in safely conducting biosafety level 3 research. This work was in part supported by the Fast Grant from Emergent Ventures at the Mercatus Center and 

SARS-CoV-2 mutations in MHC-I-restricted epitopes evade CD8(+) T cell responses

Mutational analysis of SARS-CoV-2 ORF8 during six months of COVID-19 pandemic

Robust Virus-Specific Adaptive Immunity in COVID-19 Patients with SARS-CoV-2 Delta382 Variant Infection

Multiple SARS-CoV-2 variants escape neutralization by vaccine-induced humoral immunity

SARS-CoV-2 ORF8 Forms Intracellular Aggregates and Inhibits IFNgamma-Induced

SARS-CoV-2 human T cell epitopes: Adaptive immune response against COVID-19

Isolation and characterization of viruses related to the SARS coronavirus from animals in southern China

Discovery of a rich gene pool of bat SARS-related coronaviruses provides new insights into the origin of SARS coronavirus

Mapping the human genetic architecture of COVID-19

Adaptive immune determinants of viral clearance and protection in mouse models of

Cutting Edge: Mouse SARS-CoV-2 Epitope Reveals Infection and Vaccine-Elicited CD8

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

2020. A Comprehensive, Flexible Collection of SARS-CoV-2 Coding Regions. G3 (Bethesda

A Mouse-Adapted SARS-CoV-2 Induces Acute Lung Injury and Mortality in Standard Laboratory Mice

2020. The ORF6, ORF8 and nucleocapsid proteins of SARS-CoV-2 inhibit type I interferon signaling pathway

Impact of circulating SARS-CoV-2 variants on mRNA vaccine-induced immunity

2022. A stem-loop RNA RIG-I agonist protects against acute and chronic SARS-CoV-2 infection in mice

Correlates of protection against SARS-CoV-2 in rhesus macaques

Considerable escape of SARS-CoV-2 Omicron to antibody neutralization

Resurgence of COVID-19 in Manaus, Brazil, despite high seroprevalence

Endocytosis of major histocompatibility complex class I molecules is induced by the HIV-1 Nef protein

HIV-1 Nef impairs MHC class II antigen presentation and surface expression

Discovery and Genomic Characterization of a 382-Nucleotide Deletion

CoV-2 epitopes in COVID-19 cases

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

Positive Selection of ORF1ab, ORF3a, and ORF8 Genes Drives the Early Evolutionary Trends of SARS-CoV-2 During the 2020 COVID-19 Pandemic

Antibody resistance of SARS-CoV-2 variants B.1.351 and B.1.1.7

Accurate Diagnosis of COVID-19 by a Novel Immunogenic Secreted SARS-CoV-2 orf8 Protein

SARS-CoV-2 inhibits induction of the MHC class I pathway by targeting the STAT1-IRF1-NLRC5 axis

Effects of a major deletion in the SARS-CoV-2 genome on the severity of infection and the inflammatory response: an observational cohort study

Human genetic and immunological determinants of critical COVID-19 pneumonia

The ORF8 protein of SARS-CoV-2 mediates immune evasion through down-regulating MHC-Iota