key: cord-0909617-89b20ylg
authors: Li, Qianqian; Nie, Jianhui; Wu, Jiajing; Zhang, Li; Ding, Ruxia; Wang, Haixin; Zhang, Yue; Li, Tao; Liu, Shuo; Zhang, Mengyi; Zhao, Chenyan; Liu, Huan; Nie, Lingling; Qin, Haiyang; Wang, Meng; Lu, Qiong; Li, Xiaoyu; Liu, Junkai; Liang, Haoyu; Shi, Yi; Shen, Yuelei; Xie, Liangzhi; Zhang, Linqi; Qu, Xiaowang; Xu, Wenbo; Huang, Weijin; Wang, Youchun
title: No higher infectivity but immune escape of SARS-CoV-2 501Y.V2 variants
date: 2021-02-23
journal: Cell
DOI: 10.1016/j.cell.2021.02.042
sha: 09f2189621bfa0b0daa7f735a51289c773f3bc50
doc_id: 909617
cord_uid: 89b20ylg

The 501Y.V2 variants of SARS-CoV-2 containing multiple mutations in Spike are now dominant in South Africa and are rapidly spreading to other countries. Here, experiments with 18 pseudotyped viruses showed that the 501Y.V2 variants do not confer increased infectivity in multiple cell types except for murine ACE2-overexpressing cells, where a substantial increase in infectivity was observed. Notably, the susceptibility of the 501Y.V2 variants to 12 of 17 neutralizing monoclonal antibodies was substantially diminished, and the neutralization ability of the sera from convalescent patients and immunized mice was also reduced for these variants. The neutralization resistance was mainly caused by E484K and N501Y mutations in the receptor-binding domain of Spike. The enhanced infectivity in murine ACE2-overexpressing cells suggests the possibility of spillover of the 501Y.V2 variants to mice. Moreover, the neutralization resistance we detected for the 501Y.V2 variants suggests the potential for compromised efficacy of monoclonal antibodies and vaccines.

worldwide and killed more than 2 million people (https://covid19.who.int). SARS-CoV-2 48 is a member of the coronavirus family, which carries the largest genome among 49 single-stranded RNA viruses. Although the replication-dependent RNA polymerase in 50 most RNA viruses has no proofreading activity, the coronavirus genome encodes a 3'-5' 51 exonuclease (ExoN, nsp14) with proofreading activity that can partially correct mutations 52 introduced during virus replication (Smith and Denison, 2013 To study the effects of 501Y.V2 related mutations we generated a total of 18 118 pseudotyped viruses. The 501Y.V2 variants, derived from B.1 (Tegally et al., 2020a) , 119 have the D614G S protein mutation. Note that all of the pseudotyped viruses in this study 120 were generated in the 614G background using site-directed mutagenesis, and 614G was 121 used as the reference pseudotyped virus for our experimental infectivity assays with 122 diverse host cells and antigenicity assays with various antibodies and sera. We first 123 constructed a set of 10 pseudotyped viruses carrying the single-site mutations in 501Y.V2 124 variants in a 614G background ( Figure 1A ). Then we generated the three main variants, 125 501Y.V2-1, 501Y.V2-2, and 501Y.V2-3 ( Figure 1B (Figure 3 ). No alteration of neutralization sensitivity was observed for 5 of the 17 176 monoclonal antibodies (2F7, P2C-1F11, H014, 4E5, and 7B8). 177

We found that an increasing number of mutation sites in the RBD was correlated 178 with immune escape from a steadily increasing number of monoclonal antibodies ( Figure  179 3), clearly suggesting a superposition effect. Conversely, monoclonal antibodies that do 180 not neutralize any of the three RBD site mutations were also ineffective in neutralizing 181 variant in the second wave of this epidemic (Tegally et al., 2020a) ; it carries the E484K 183 and N501Y mutations but not the K417N mutation. We found that the antibody escape 184 spectrum of our pseudotyped virus 501Y.V2-1 was essentially the same as for the 185 614G+E484K+N501Y triple RBD mutation variants. However, and recalling that the 186 501Y.V2-2 pseudotyped virus carries two additional mutations (L18F and K417N), it is 187 consistent that 501Y.V2-2's escape spectrum is wider than 501Y.V2-1's spectrum for 188 this panel of neutralizing antibodies ( Figure 3 ). Finally, our finding that 501Y.V2-3's 189 escape spectrum for this RBD--targeting antibody panel is identical to 501Y.V2-2's 190 spectrum fit with our expectations, because these two pseudotyped variants contain the 191 same mutations in their RBDs ( Figure S2 ). 192

We also obtained convalescent sera from 15 SARS-CoV-2-infected patients with 194 high neutralizing antibody titers and obtained three pooled sera samples from a total of 195 nine mice immunized with the RBD to further investigate how these mutations affect 196 antigenicity. ( Figure 4A ). Neutralization assays with the pseudotyped viruses showed that 197 mutations at a single site did not lead to significant alteration of the neutralization activity 198 of polyclonal antibodies; only the simultaneous presence of the E484K and N501Y 199 mutations resulted in a significant decrease in neutralization (p<0.05) ( Figure 4B ). 200

Among the 501Y.V2 pseudotyped viruses, 501Y.V2-1 showed the greatest decrease in 201 neutralization by polyclonal antibodies, displaying a 3.9-fold reduction compared to the 202 reference 614G variant ( Figure 4B ). Note that 501Y.V2-1 lacks the K417N mutation, so 203 increases susceptibility to neutralization by polyclonal antibodies. 205

To determine how the mutations in the 501Y.V2 variants may affect neutralization 206 activity in the sera with differing levels of neutralizing antibodies, we obtained 207 longitudinal sera from ten SARS-CoV-2-infected patients at 2, 5, and 8 months after 208 onset ( Figure 5A ). The pseudotyped viruses with 501Y.V2 related RBD mutations and 209 the 614G control virus then were used in assays with these 30 longitudinal sera samples. 210

We found that the E484K and N501Y mutations led to a decrease in neutralization and 211 that the combination of these two mutations resulted in an apparently superimposed 212 resistance to neutralization ( Figure 5B ). Further, it was again conspicuous that the K417N 213 mutation increased viral susceptibility to neutralization. 214

Taking the reference 614G pseudotyped virus as an example: compared to assays for 215 the sera collected at 2 months, neutralization titers for sera collected at 5-and 8-months 216 post-onset decreased by 2.2-and 2.5-fold, respectively ( Figure 5C ). We noted that the 217 trends for detected decreases varied consistently within sera of differing antibody titers: 218 the higher the antibody titer, the greater the reduction in the neutralizing activity ( Figure  219 5C). The most pronounced differences from the reference 614G pseudotyped virus were 220 detected for 501Y.V2-3, which exhibited reduced neutralization at antibody titers >1000, 221 500-1000, and <500 by an average of 5.3-, 3.1-, and 1.8-fold, respectively. Some samples 222 with low antibody titers (median effect dose, ED 50 for 614G<100) were not able to 223 neutralize 501Y.V2-3 (ED 50 <30). 224 225 DISCUSSION 226 SARS-CoV-2 is no exception. The emergence of a variety of SARS-CoV-2 mutants has 228 become a major concern during the ongoing pandemic. Mutants may be more 229 transmittable, or may be able to evade neutralizing monoclonal antibodies, or even the S protein's conformation is also more likely to expose epitopes to neutralizing 286 antibodies, which would increase the likelihood of virus neutralization by sera containing 287 polyclonal antibodies. Since both the E484 and N501 residues are fully exposed, it is 288 reasonable to speculate that mutations to these sites may weaken antibody binding 289 ( Figure 5D and XhoI sites to get ACE2 expression plasmids from the different species. 479

Based on pcDNA3.1.S2, 18 mutant plasmids were constructed. The point mutation 481 method was conducted as described in our previous study (Nie et al., 2020a, b 

The pseudotyped viruses of the SARS-CoV-2 variants and the point mutation 501 pseudotyped viruses were constructed using the methods reported in our previous study 502 (Nie et al., 2020a, b) . Briefly, one day prior to transfection for virus production, 503 HEK293T cells were digested and adjusted to a concentration of 5-7×10 5 cells/mL in a 504 15ml culture medium and incubated overnight in an incubator at 37°C with 5% CO 2 . 505

When cells reached 70%-90% confluence, the culture medium was discarded and 15 mL 506 G*∆G-VSV virus (VSV G pseudotyped virus, Kerafast) with a concentration of 7.0×10 4 507 TCID 50 /mL was used for infection. At the same time, 30 µg of the S protein expression 508 plasmid was transfected according to the instructions of Lipofectamine 3000 (Invitrogen), 509 and then the cells were cultured in an incubator at 37°C and 5% CO 2 . After 4-6 hours, the 510 cell medium was discarded, and the cells were gently washed two times with PBS+1% 511 was placed in an incubator at 37°C with 5% CO 2 for 24 h. After that, the SARS-CoV-2 513 pseudotyped virus containing the culture supernatant was harvested, filtered, aliquoted, 514 and frozen at -70°C for further use. System for RT-PCR kit reagent (Invitrogen). RT-PCR was performed using TB Green 520

Premix Ex Taq II (Takara). The plasmid containing the P protein gene of the VSV virus 521 was used as the standard to calculate the viral copy number. See the primers in the Key 522

Resources Table. 523

After quantification by RT-PCR, the pseudotyped virus was diluted to the same particle 525 number, and 100 µl aliquots were added into 96-well cell culture plates. Cells of the 526 assayed cell lines were then digested with trypsin and added into each well at 2×10 4 /100 527 µl. Chemiluminescence monitoring was carried out after a 24 h incubation with 5% CO 2 528 at 37°C. The supernatant was adjusted to 100 µl for each sample. Luciferase substrate 529 was mixed with cell lysis buffer (Perkinelmer, Fremont, CA) and was added to the plate 530 (100 µl/well). Two minutes later, 150 µl of lysate was transferred to opaque 96-well 531 plates. The luminescence signal was detected using a PerkinElmer Ensight luminometer, 532 with data collected in terms of relative luminescence unit (RLU) values. Each group 533 contained two replicates, and these experiments were repeated three times. were then digested and added to each well (2×10 4 /100 µl). After incubation with 5% CO 2 543 at 37°C for 24 hours, luminescence was measured as described above. The sample EC 50 544 (median effect concentration) was calculated using the Reed-Muench method ( 

Structures of Human Antibodies Bound

Structural basis for 642 neutralization of SARS-CoV-2 and SARS-CoV by a potent therapeutic 643 antibody

South Africa responds to new SARS-CoV-2 645 variant

Establishment and validation of a 648 pseudovirus neutralization assay for SARS-CoV-2

Quantification of SARS-CoV-2 652 neutralizing antibody by a pseudotyped virus-based assay

Spike mutation D614G alters SARS-CoV-2 fitness

Measuring SARS-CoV-2 neutralizing 661 antibody activity using pseudotyped and chimeric viruses

Structural basis of receptor 665 recognition by SARS-CoV-2

A human neutralizing antibody targets 668 the receptor-binding site of SARS-CoV-2

Coronaviruses as DNA 670 wannabes: a new model for the regulation of RNA virus replication 671 fidelity

Characterization and structural 674 basis of a lethal mouse-adapted SARS-CoV-2. BioRxiv

Introduction of the South African SARS-CoV-2 variant 501Y.V2 into 682 the UK

Emergence and rapid spread of a new severe acute 687 respiratory syndrome-related coronavirus 2 (SARS-CoV-2) lineage 688 with multiple spike mutations in South Africa. medRxiv

Major new lineages of SARS-CoV-2 emerge and spread in South 694

Africa during lockdown. medRxiv

Circulating SARS-CoV-2 spike N439K variants 699 maintain fitness while evading antibody-mediated immunity

Structure, Function, and Antigenicity of the 704 SARS-CoV-2 Spike Glycoprotein

A noncompeting pair of human neutralizing 719 antibodies block COVID-19 virus binding to its receptor ACE2

SARS-CoV-2 spike-protein D614G mutation increases virion 724 spike density and infectivity

Neutralization Assay with SARS-CoV-1 and SARS-CoV-2 Spike 728

Pseudotyped Murine Leukemia Virions. bioRxiv