key: cord-0903805-4cwt0lme
authors: Riou, C.; Keeton, R.; Moyo-Gwete, T.; Hermanus, T.; Kgagudi, P.; Baguma, R.; Tegally, H.; Doolabh, D.; Iranzadeh, A.; Tyers, L.; Mutavhatsindi, H.; Tincho, M.; Benede, N.; Marais, G.; Chinhoyi, L.; Mennen, M.; Skelem, S.; Bruyn, E.; Stek, C.; SA-CIN,; de Oliveira, T.; Williamson, C.; Moore, P.; Wilkinson, R.; Ntusi, N.; Burgers, W.
title: Loss of recognition of SARS-CoV-2 B.1.351 variant spike epitopes but overall preservation of T cell immunity
date: 2021-06-06
journal: nan
DOI: 10.1101/2021.06.03.21258307
sha: e62ef469630b26666f8f31bcf42bbda0c850e173
doc_id: 903805
cord_uid: 4cwt0lme

SARS-CoV-2 variants have emerged that escape neutralization and potentially impact vaccine efficacy. T cell responses play a role in protection from reinfection and severe disease, but the potential for spike mutations to affect T cell immunity is poorly studied. We assessed both neutralizing antibody and T cell responses in 44 South African COVID-19 patients infected either with B.1.351, now dominant in South Africa, or infected prior to its emergence (first wave), to provide an overall measure of immune evasion. We show for the first time that robust spike-specific CD4 and CD8 T cell responses were detectable in B.1.351-infected patients, similar to first wave patients. Using peptides spanning only the B.1.351 mutated regions, we identified CD4 T cell responses targeting the wild type peptides in 12/22 (54.5%) first wave patients, all of whom failed to recognize corresponding B.1.351-mutated peptides (p=0.0005). However, responses to the mutated regions formed only a small proportion (15.7%) of the overall CD4 response, and few patients (3/44) mounted CD8 responses that targeted the mutated regions. First wave patients showed a 12.7 fold reduction in plasma neutralization of B.1.351. This study shows that despite loss of recognition of immunodominant CD4 epitope(s), overall CD4 and CD8 T cell responses to B.1.351 are preserved. These observations may explain why, despite substantial loss of neutralizing antibody activity against B.1.351, several vaccines have retained the ability to protect against severe COVID-19 disease.

High levels of ongoing SARS-CoV-2 transmission have led to the emergence of new viral variants, which now dominate the pandemic. Variants of concern have been characterized as having increased transmissibility, potentially greater pathogenicity, and the ability to evade host immunity (1) . Four such variants of concern now circulate widely, namely B.1.1.7 in the US and Europe, B.1.351 in southern Africa, P.1 in Brazil and South America, and B.1.617 in India (2) (3) (4) (5) (6) .

A primary concern is whether the immune response generated against ancestral SARS-CoV-2 strains, upon which all approved first generation vaccines are based, still confers protection against variants. The potential threat of reduced vaccine efficacy has prompted swift action from vaccine manufacturers, and an adapted mRNA vaccine based on B.1.351 has been developed and tested in clinical trials (7) .

SARS-CoV-2 variant B.1.351, which was first described in South Africa in October 2020 (5) , is now responsible for >95% of infections in the country, and has spread across much of southern Africa (6) . It is the most concerning of the variants, tending to demonstrate the greatest reduction in neutralization sensitivity to COVID-19 convalescent and vaccinee plasma (8) (9) (10) (11) (12) (13) , as well as reduced vaccine efficacy (14) (15) (16) . However, some vaccines have still demonstrated high efficacy against severe COVID-19 disease after B.1.351 infection (16, 17) , suggesting that T cell immunity plays an important role in immune protection, and may mitigate the effect of reduced neutralizing antibody activity.

To date, efforts to characterize immune evasion by SARS-CoV-2 variants have focused mainly on their ability to escape neutralization (8) (9) (10) (11) (12) (13) . There is limited data addressing whether SARS-CoV-2 variants can evade T cell immunity (18) (19) (20) (21) (22) (Figure S1 ). Although SARS-CoV-2 viral sequences were not available for patients recruited in June to August 2020 during the first wave, we assumed that all participants were infected with the ancestral virus, since B.1.351 was first detected in October 2020 in the Western Cape.

First, we compared the magnitude of CD4 and CD8 T cell responses directed at the spike protein of SARS-CoV-2 in first and second wave patients. Using flow cytometry, we measured the production of IFN-γ in response to a peptide pool covering the full ancestral spike protein ('Full spike') ( Figure 1C ). All participants tested exhibited a CD4 response, with a comparable frequency of spike specific-CD4 T cells in first and second wave patients (0.051% and 0.045%, respectively, Figure 1D ). As previously reported, the prevalence and magnitude of the SARS-CoV-2-specific CD8 T cell response was significantly lower compared to the CD4 response in the first wave (23), with 63.6% (14/22) of first wave patients and 81.8% (18/22) of second wave patients exhibiting a detectable spike-specific CD8 T cell response. The median frequency of spike-specific CD8 T cells was higher in second wave patients compared to first wave patients . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 6, 2021. ; https://doi.org/10.1101/2021.06.03.21258307 doi: medRxiv preprint (0.031 % and 0.007%, respectively, P = 0.037, Figure 1D ). This may be explained by the fact that patients from the second wave were sampled at a later time post-PCR positivity compared to the patients recruited during the first wave (median: 8 days vs. 4.5 days, respectively, P = 0.006, Figure 1A ). Overall, these data are in accordance with a recent report showing that T cell responses directed at the entire SARS-CoV-2 spike protein in convalescent COVID-19 donors were not substantially affected by mutations found in SARS-CoV-2 variants (22) . Table S1 ). In patients recruited during the first wave, IFN-γ CD4 T cell responses to the WT pool were detectable in 54.5% (12/22) patients (Figure 2A-B) . In those who mounted responses, the magnitude of the WT pool response was ~ 6.4-fold lower than full spike responses (median: 0.0075% vs 0.048%, respectively, P < 0.0001).

In the 12 participants responding to the WT pool, the overall median relative contribution of WT epitopes located at spike mutation sites to the total spike-specific CD4 T cell response was 15.7%, ranging from 5.7% to 24%. These results suggest that the majority of SARS-CoV-2 spike specific-CD4 T cell responses are directed against conserved epitopes between the ancestral and the B.1.351 lineage. When we tested the corresponding B.1.351 pool, all 12 of the first wave WT pool responders failed to cross-react with the mutated peptides from B.1.351 ( Figure 2B , left panel). These results show that B.1.351 spike mutated epitopes were no longer recognized by CD4 T cells targeting the WT epitopes, demonstrating that this loss of recognition by CD4 T cells is likely mediated by variant mutations. This is broadly consistent with recent data from mRNA vaccinees, where full spike pools containing B.1.351 mutated peptides detected T cell responses that were diminished by 30% compared to ancestral full spike, revealing that the mutated sequences mediate differential recognition but make up a minor contribution to the overall SARS-CoV-2 spike specific-T cell response (20) . is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 6, 2021. ; https://doi.org/10.1101/2021.06.03.21258307 doi: medRxiv preprint B.1.351 pool. These data suggest that mutations in B.1.351 spike epitopes abrogate epitope immunogenicity by altering their processing and/or presentation, consistent with the loss of recognition of B.1.351 mutated peptides by T cells in first wave patients. The few responders may represent individuals with uncommon HLA alleles able to present peptides within the mutated pool. Further analysis will be necessary to define the specific epitope(s) and presenting HLA alleles accounting for these CD4 responses.

In order to obtain an overall measure of immune escape in our participants, we measured their neutralizing antibody responses to the ancestral and B.1.351 spike protein, using a pseudotyped lentivirus neutralization assay (Figure 2C-D) . As we showed previously (11) in patients infected with the ancestral strains (first wave), a considerable loss of neutralization activity was observed against B.1.351 (median fold change: 12.7, IQR: 7.3-18.8). In contrast, patients infected with B.1.351 (second wave) retained a substantial capacity to neutralize the ancestral virus, as shown by a moderate reduction in neutralizing activity against the ancestral strain (median: 2.3, IQR: 1.3-3.9). Of note, in the six first wave patients where loss of crossneutralization was profound (titer <100), it is reassuring that the T cell response was relatively intact, thus providing some cross-protection. We found no association between the frequency of SARS-CoV-2 spike-specific CD4 T cell responses and neutralizing activity (data not shown), consistent with an earlier study (24) .

Finally, we also defined the recognition of WT and B.1.351 peptide pools by CD8 T cells in both patient groups (Figure 3) . Regardless of the infecting SARS-CoV-2 lineage, peptides covering the spike mutation sites were rarely recognized by CD8 T cells, with only 3/44 (6.8%) of patients exhibiting a CD8 response, one in the first wave cohort and two in the second wave cohort. Thus, in contrast to CD4 T cells, the regions in which B.1.351 mutations occur are not commonly targeted by CD8 T cells. Moreover, in the three patients with a CD8 response, the frequency of IFN-g producing CD8 T cells was comparable between the WT and the B.1.351 pool stimulation, indicating that mutations did not affect epitope recognition ( Figure 3B) . Overall, these data indicate that B.1.351 mutations do not affect CD8 T cell responses in these experiments.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 6, 2021. (25) . We also demonstrate for the first time that the recognition of epitopes by CD4 T cells targeting variable spike regions was affected by B.1.351 spike mutations in patients infected with ancestral lineages. However, it is reassuring that the loss of recognition of B.1.351 mutated spike epitopes has a minor impact on the overall CD4 Th1 cell response. Moreover, the CD8 T cell response to spike was unaffected.

In contrast to neutralizing antibody epitopes, T cell epitopes are located along the full length of the spike protein and T cell responses are broadly targeted in natural infection (22, (26) (27) (28) (29) . Thus, it is unsurprising that the B. Viral evasion of cytotoxic T lymphocyte or T helper recognition may result in delayed clearance of infected cells, or inadequate help provided to B cells, influencing the antibody response. Viral escape from specific SARS-CoV-2 CD8 epitopes has recently been described, in spike, nucleocapsid and ORF3a proteins (18, 19, 21) . Whilst CD8 T cells can exert selective pressure on viruses resulting in mutational escape, CD4 T helper responses act indirectly, and it is not clear how they could drive viral evolution. Mutations occurring in response to immune pressure from neutralizing antibodies, or associated with increased viral infectivity or greater protein stability (21) could coincide with CD4 epitopes, and thus represent 'collateral damage' for the CD4 response.

Further work in a greater number of participants may identify additional responders to epitopes in the mutated regions of spike, given polymorphism in HLA genes. It remains to be . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 6, 2021. ; https://doi.org/10.1101/2021.06.03.21258307 doi: medRxiv preprint determined which specific epitopes within spike variable regions are immunogenic. A number of shared mutations are found in several variants of concern or interest, having arisen through convergent evolution (31) . Our findings would have generalizability beyond the B.1.351 variant if loss of recognition occurred in mutated regions that occurred in multiple variants. Examining responses to B.1.351 in the context of full mutated spike (20, 22) would corroborate our findings regarding the degree to which the overall spike T cell response is affected by mutations.

In conclusion, these results advance our understanding of cross-reactive T cell immunity in the context of viral variability, and highlight the importance of monitoring both antibody and T cell responses to emerging SARS-CoV-2 variants. We demonstrate a limited effect of viral mutations on T cell immunity which may explain why, despite substantial loss of neutralizing antibody activity against B.1.351, some vaccines have retained the ability to protect against severe COVID-19 disease. Whilst second generation vaccines based on SARS-CoV-2 variants are desirable, they may not be needed to generate improved T cell responses, and therefore the rollout of present vaccines must continue apace.

Hospitalized patients with PCR-confirmed acute COVID-19 were enrolled at Groote Schuur Figure   S1 . T cell immune responses were assessed by stimulating PBMC with peptide pools spanning full-length spike or smaller pools covering the regions mutated in B.1.351, followed by intracellular cytokine staining and flow cytometry ( Figure S2) . The study was approved by the . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 6, 2021. ; https://doi.org/10.1101/2021.06.03.21258307 doi: medRxiv preprint University of Cape Town Human Research Ethics Committee (HREC: 207/2020 and R021/2020) and electronic or written informed consent was obtained from all participants.

Whole genome sequencing of SARS-CoV-2 was performed using nasopharyngeal swabs obtained from 19 of the hospitalized patients recruited during the second COVID-19 wave.

Sequencing was performed as previously published (25) . Briefly, cDNA was synthesized from RNA extracted from the nasopharyngeal swabs using the Superscript IV First Strand synthesis system (Life Technologies, Carlsbad, CA) and random hexamer primers. Whole genome amplification was then performed by multiplex PCR using the ARTIC V3 protocol The workflow performs alignment of genomes, phylogenetic tree inference, tree dating and ancestral state construction and annotation. The phylogenetic trees were visualized using ggplot and ggtree (34) .

To assess the overall response to the full length SARS-CoV-2 spike protein, we combined two commercially available peptide pools (PepTivator®, Miltenyi Biotech, Bergisch Gladbach, . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Table S1 , which also indicates where their recognition has been previously described (22, 27, 29) . Ancestral or B.1.351 pools (16 peptides) selectively spanning the mutated regions were created by pooling aliquots of these individual peptides at a final concentration of 160 µg/mL.

Blood was collected in heparin tubes and processed within 3 hours of collection. Peripheral Subsequently, 1x10 4 HEK293T cells engineered to over-express ACE-2, kindly provided by Dr Michael Farzan (Scripps Research Institute), were added and the incubated at 37°C, 5% CO 2 for 72 hours, upon which the luminescence of the luciferase gene was measured. CB6 and CA1monoclonal antibodies were used as controls.

Analyses were performed in Prism (v9; GraphPad Software Inc, San Diego, CA, USA). Nonparametric tests were used for all comparisons. The Mann-Whitney and Wilcoxon tests were used for unmatched and paired samples, respectively. P values less than 0.05 were considered to indicate statistical significance.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 

Authors declare that they have no competing interests.

All data are available in the main text or the supplementary materials.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 6, 2021. were tested for their neutralization cross-reactivity against the ancestral or B1.351 pseudoviruses. The threshold of detection for the neutralization assay was a 50% inhibitory dilution (ID 50 ) of 20. Gray dots indicate patients who displayed a detectable CD4 T cell response to the WT pool, selectively covering the variable regions of spike, and lost recognition to the B.1.351 pool. Neutralization data on the second wave cohort are from (25) . (D) Fold-change in neutralization titers is shown for the data in c. Bars represent medians. Statistical analyses were performed using the Wilcoxon test and the chi-squared test.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 6, 2021. ; https://doi.org/10.1101/2021.06.03.21258307 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 6, 2021. ; https://doi.org/10.1101/2021.06.03.21258307 doi: medRxiv preprint

The Variant Gambit: COVID's Next Move

European Centre for Disease Prevention and Control

Convergent evolution of SARS-CoV-2 spike mutations, L452R, E484Q and P681R, in the second wave of COVID-19 in Maharashtra

Preliminary Analysis of Safety and Immunogenicity of a SARS-CoV-2 Variant Vaccine

Escape of SARS-CoV-2 501Y.V2 from neutralization by convalescent plasma

Resistance of SARS-CoV-2 variants to neutralization by monoclonal and serum-derived polyclonal antibodies

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

SARS-CoV-2 501Y.V2 escapes neutralization by South African COVID-19 donor plasma

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

Evidence of escape of SARS-CoV-2 variant B.1.351 from natural and vaccine induced sera

Interim Results of a Phase 1-2a Trial of Ad26.COV2.S Covid-19 Vaccine

Effectiveness of the BNT162b2 Covid-19 Vaccine against the B.1.1.7 and B.1.351 Variants

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

The impact of viral mutations on recognition by SARS-CoV-2 specific T-cells

-19 C. & P. Team, M. V. Maus, SARS -CoV-2 T-cell immunity to variants of concern following vaccination

An emerging SARS-CoV-2 mutant evading cellular immunity and increasing viral infectivity

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

Targets of T Cell Responses to SARS-CoV-2 Coronavirus in Humans with COVID-19

Disease and Unexposed Individuals

Longitudinal analysis shows durable and broad immune memory after SARS-CoV-2 infection with persisting antibody responses and memory B and T cells

Cross-Reactive Neutralizing Antibody Responses Elicited by SARS-CoV-2 501Y.V2 (B.1.351)

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

Selective and crossreactive SARS-CoV-2 T cell epitopes in unexposed humans

Mapping the SARS-CoV-2 spike glycoprotein-derived peptidome presented by HLA class II on dendritic cells

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

The Control of the Specificity of CD4 T Cell Responses: Thresholds, Breakpoints, and Ceilings

The emergence and ongoing convergent evolution of the N501Y lineages coincides with a major global shift in the SARS-CoV-2 selective landscape

Genome Detective Coronavirus Typing Tool for rapid identification and characterization of novel coronavirus genomes

Nextstrain: real-time tracking of pathogen evolution

ggtree: an r package for visualization and annotation of phylogenetic trees with their covariates and other associated data

The authors thank the study participants and their families, and the clinical staff and personnel at Groote Schuur Hospital in Cape Town for their support and dedication. We thank the informal 501Y.V2 consortium of South African scientists, chaired by Drs Willem Hanekom and Tulio de Oliveira, for suggestions and discussion of data.