key: cord-0762229-vblfew3o
authors: Cele, S.; Gazy, I.; Jackson, L.; Hwa, S.-H.; Tegally, H.; Lustig, G.; Giandhari, J.; Pillay, S.; Wilkinson, E.; Naidoo, Y.; Karim, F.; Ganga, Y.; Khan, K.; Balazs, A. B.; Gosnell, B. I.; Hanekom, W.; Moosa, M.-Y. S.; Lessells, R. J.; de Oliveira, T.; Sigal, A.
title: Escape of SARS-CoV-2 501Y.V2 variants from neutralization by convalescent plasma
date: 2021-01-26
journal: nan
DOI: 10.1101/2021.01.26.21250224
sha: 3bc987e82fba5a16cf8f347e73687000150b4b52
doc_id: 762229
cord_uid: vblfew3o

New SARS-CoV-2 variants with mutations in the spike glycoprotein have arisen independently at multiple locations and may have functional significance. The combination of mutations in the 501Y.V2 variant first detected in South Africa include the N501Y, K417N, and E484K mutations in the receptor binding domain (RBD) as well as mutations in the N-terminal domain (NTD). Here we address whether the 501Y.V2 variant could escape the neutralizing antibody response elicited by natural infection with earlier variants. We were the first to outgrow two variants of 501Y.V2 from South Africa, designated 501Y.V2.HV001dF and 501Y.V2.HV002. We examined the neutralizing effect of convalescent plasma collected from six adults hospitalized with COVID-19 using a microneutralization assay with live (authentic) virus. Whole genome sequencing of the infecting virus of the plasma donors confirmed the absence of the spike mutations which characterize 501Y.V2. We infected with 501Y.V2.HV001dF and 501Y.V2.HV002 and compared plasma neutralization to first wave virus which contained the D614G mutation but no RBD or NTD mutations. We observed that neutralization of the 501Y.V2 variants was strongly attenuated, with IC50 6 to 200-fold higher relative to first wave virus. The degree of attenuation varied between participants and included a knockout of neutralization activity. This observation indicates that 501Y.V2 may escape the neutralizing antibody response elicited by prior natural infection. It raises a concern of potential reduced protection against re-infection and by vaccines designed to target the spike protein of earlier SARS-CoV-2 variants.

3 . CC-BY-ND 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 January 26, 2021.

Figure 1: Study design and sequences of SARS-CoV-2 variants. (A) We obtained convalescent plasma and detected the matching infecting variant in the first SARS-CoV-2 infection wave in South Africa. A blood draw and nasopharyngeal/oropharyngeal was performed on study participants. First wave virus was outgrown from one of the participants and compared to two viruses outgrown from the second wave, which were 501Y.V2 variants. A focus forming microneutralization assay was used to quantify neutralization. (B) Phylogenetic tree and mutations of variant sequences. Variants which infected the study participants who were plasma donors only for this study are marked in blue. Sequences of variants which were outgrown are marked in yellow. Participant 039-13-0013 was both a plasma donor and the donor from whom the first wave virus was outgrown. Y-axis denotes time of sampling for viral sequencing. 

. CC-BY-ND 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 January 26, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 As we have entered the second year of the SARS-CoV-2 pandemic with high levels of transmission in 89 many parts of the world, variants with mutations at key residues in the spike glycoprotein have emerged.

Here we present clear evidence using authentic SARS-CoV-2 that the 501Y.V2 variant first detected 91 in South Africa is associated with reduced neutralization by plasma collected from patients infected 92 in the first wave with SARS-CoV-2 variants without the 501Y.V2 defining RBD and NTD mutations.

While our findings are based on plasma samples from six convalescent study participants, the relative 94 consistency of the effect argues that the potential to escape neutralizing antibodies elicited by prior 95 SARS-CoV-2 infection may be widespread.

The reduced neutralization is most likely related to the mutations in the spike RBD and NTD that 97 characterize the 501Y.V2 variant. While the E484K mutation has the clearest association with immune 98 escape, the other mutations in the RBD (K417N, N501Y) are also located within residues targeted 99 by some class 1 and class 2 NAbs [7] . Information about the significance of NTD mutations is also 100 emerging. NAbs targeting this site have been shown to be potent neutralizers of SARS-CoV-2 [5, 6].

The deletion at residues 242-244 is just outside an antigenic supersite loop (residues 245-264) and L18 102 also falls within the antigenic supersite. Furthermore, mutations at L18 and D80 have been selected 103 during passage with mAbs [5]. Our second variant contains the L18F mutation. This may be associated 104 with the trend to lower neutralization sensitivity relative to the first 501Y.V2 variant ( Figure S2 ). This 105 variant also has strikingly larger foci ( Figure 2A ).

The reasons for the rapid emergence and fixation of potential immune escape mutations in South In conclusion, we present data suggesting that the 501Y.V2 variant first detected in South Africa 131 is able to escape the neutralizing antibody response elicited by natural infection with earlier variants. 132 We expect data in the next weeks from phase 3 vaccine trials being conducted in South Africa. If the 133 variant does have an effect on vaccine efficacy, then there may be a signal in the data from these clinical 134 trials. 135 6 . CC-BY-ND 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 January 26, 2021. CC-BY-ND 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 January 26, 2021. ; https://doi.org/10.1101/2021.01.26.21250224 doi: medRxiv preprint HDM-Hgpm2, HDM-tat1b and pRC-CMV-Rev1b using TransIT-LT1 (Mirus) transfection reagent. Su-181 pernatant containing the lentivirus was harvested two days after infection, filtered through a 0.45µm 182 filter (Corning) and used to spinfect H1299-H2AZ at 1000 rcf for 2 hours at room temperature in the pres-183 ence of 5 µg/mL polybrene (Sigma-Aldrich). ACE-2 transduced H1299-H2AZ cells were then subcloned 184 at the single cell density in 96-well plates (Eppendorf) in conditioned media derived from confluent cells.

After 3 weeks, wells were trypsinized (Sigma-Aldrich) and plated in two replicate plates, where the first 186 plate was used to determine infectivity and the second was stock. The first plate was screened for the 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 January 26, 2021. ; https://doi.org/10.1101/2021.01.26.21250224 doi: medRxiv preprint a permiabilization buffer containing 0.1% saponin (Sigma-Aldrich), 0.1% BSA (Sigma-Aldrich), and 231 0.05% tween (Sigma-Aldrich) in PBS. Plates were incubated with primary antibody overnight at 4 • C, 232 then washed with wash buffer containing 0.05% tween in PBS. Secondary goat anti-rabbit horseradish 233 peroxidase (Abcam ab205718) was added at 1 µg/mL and incubated for 2 hours at room temperature 234 with shaking. The TrueBlue peroxidase substrate (SeraCare 5510-0030) was then added at 50µL per 235 well and incubated for 20 minutes at room temperature. Plates were then dried for 2 hours and imaged 236 using a Metamorph-controlled Nikon TiE motorized microscope with a 2x objective. Automated image 237 analysis was performed using a Matlab2019b (Mathworks) custom script, where focus detection was 238 automated and did not involve user curation. Image segmentation steps were stretching the image from 239 minimum to maximum intensity, local Laplacian filtering, image complementation, thresholding and 240 binarization. For the second 501Y.V2 variant, a dilation/erosion step was introduced to prevent the 241 large foci from fragmenting into smaller objects. Dilution Figure S 1: Fit of combined data for each plasma dilution to a normal distribution. The Matlab2019b function normplot was used to assess the fit of the data (blue crosses) to a normal distribution (solid red line).

Lack of pronounced curvature of the data in the range of the solid line indicates that a the data is a reasonably good fit to a normal distribution. see https://www.mathworks.com/help/stats/normplot.html for additional information.

. CC-BY-ND 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 January 26, 2021. . CC-BY-ND 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 January 26, 2021. ; https://doi.org/10.1101/2021.01.26.21250224 doi: medRxiv preprint . CC-BY-ND 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 January 26, 2021. ; https://doi.org/10.1101/2021.01.26.21250224 doi: medRxiv preprint 

Other amino acid substitutions N:L139F N:R203K N:G204R ORF14:G50N ORF1a:D1481N ORF1b:P314L E:P71L N:T205I ORF14:L52F ORF1a:T265I ORF1a:K1655N ORF1a:K3353R ORF1b:P314L ORF3a:Q57H ORF3a:S171L E:P71L N:T205I ORF14:L52F ORF1a:T265I ORF1a:K1655N ORF1a:K3353R ORF1b:P314L ORF3a:Q57H ORF3a:W131L ORF3a:S171L ORF7a:V93F N:L139F N:R203K N:G204R ORF14:G50N ORF1a:D1481N ORF1b:P314L E:L73P N:R203K N:G204R ORF14:G50N ORF1b:P314L ORF1b:T1522I E:L73P ORF1a:D3728N ORF1b:P314L N:T148A ORF10:A28V ORF1a:K2511R ORF1a:V3858I ORF1b:P314L ORF1a:F1178S ORF1b:P314L N:R203K N:G204R ORF14:G50N ORF1a:T1246I ORF1a:G3278S ORF1b:P314L Other deletions orf1ab

Lineage classification was performed by Pangolin software application version v2

Amino acid mutations are annotated based on mature protein region of coding sequence (CDS) of SARS-CoV-2 reference sequence NC_045512

Amino acid mutation nomenclature includes open reading frame, wild-type amino acid, ORF position and amino-acid mutation (e.g. S:D80A, Spike D to A substitution at position 80). del refers to deletion between stated positions. Amino acid mutations are annotated based on mature protein region of coding sequence (CDS) of SARS-CoV-2 reference sequence NC_045512.2. Substitutions and deletions in bold are those emerging during passage

C. C. Smith, M. D. Snape, R. Song, R. Tarrant, Y. Themistocleous, K. M. Thomas, T. L. Vil-