key: cord-0920657-57z0g4mq
authors: Yamasoba, Daichi; Kimura, Izumi; Nasser, Hesham; Morioka, Yuhei; Nao, Naganori; Ito, Jumpei; Uriu, Keiya; Tsuda, Masumi; Zahradnik, Jiri; Shirakawa, Kotaro; Suzuki, Rigel; Kishimoto, Mai; Kosugi, Yusuke; Kobiyama, Kouji; Hara, Teppei; Toyoda, Mako; Tanaka, Yuri L; Butlertanaka, Erika P; Shimizu, Ryo; Ito, Hayato; Wang, Lei; Oda, Yoshitaka; Orba, Yasuko; Sasaki, Michihito; Nagata, Kayoko; Yoshimatsu, Kumiko; Asakura, Hiroyuki; Nagashima, Mami; Sadamasu, Kenji; Yoshimura, Kazuhisa; Kuramochi, Jin; Seki, Motoaki; Fujiki, Ryoji; Kaneda, Atsushi; Shimada, Tadanaga; Nakada, Taka-aki; Sakao, Seiichiro; Suzuki, Takuji; Ueno, Takamasa; Takaori-Kondo, Akifumi; Ishii, Ken J; Schreiber, Gideon; Sawa, Hirofumi; Saito, Akatsuki; Irie, Takashi; Tanaka, Shinya; Matsuno, Keita; Fukuhara, Takasuke; Ikeda, Terumasa; Sato, Kei
title: Virological characteristics of SARS-CoV-2 BA.2 variant
date: 2022-02-15
journal: bioRxiv
DOI: 10.1101/2022.02.14.480335
sha: 020287fd586c0dc3c586b93fc0241286b077de69
doc_id: 920657
cord_uid: 57z0g4mq

Soon after the emergence and global spread of a new severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Omicron lineage, BA.1 (ref1, 2), another Omicron lineage, BA.2, has initiated outcompeting BA.1. Statistical analysis shows that the effective reproduction number of BA.2 is 1.4-fold higher than that of BA.1. Neutralisation experiments show that the vaccine-induced humoral immunity fails to function against BA.2 like BA.1, and notably, the antigenicity of BA.2 is different from BA.1. Cell culture experiments show that BA.2 is more replicative in human nasal epithelial cells and more fusogenic than BA.1. Furthermore, infection experiments using hamsters show that BA.2 is more pathogenic than BA.1. Our multiscale investigations suggest that the risk of BA.2 for global health is potentially higher than that of BA.1.

representing the epidemic dynamics of SARS-CoV-2 lineages. This hierarchical 138 model can estimate the global average of the relative effective reproduction 139 numbers of viral lineages (Fig. 1d) as well as those in each country (Extended 140 Data Fig. 2) . The effective reproduction number of BA.2 is 1.40-fold higher than 141 that of BA.1 on average in the world [95% confidence interval (CI), 1.29-1.52; 142 Fig. 1d] . Furthermore, the effective reproduction number of BA.2 was even 143

higher than that of BA.1.1, which spread more rapidly than BA.1 in several 144 countries such as the UK and USA ( Fig. 1d and Extended Data Fig. 2d) . These 145 results suggest that the BA.2 epidemic will more expand around the world, 146

raising the importance of elucidating virological features of BA.2 in depth. 147 148

Immune resistance of BA.2 149

Since the sequence of BA.2, particularly in S protein, is substantially different 150 from that of BA.1 ( Fig. 1b and Fig. 2a) , it is reasonable to assume that the 151 virological properties of BA.2, such as immune resistance and pathogenicity, are 152 different from those of BA.1. To reveal the virological features of BA.2, we set 153 out to perform neutralisation assay using pseudoviruses and the neutralising 154 antibodies elicited by vaccination. Consistent with recent studies 4-11,16-19 , BA.1 is 155 highly resistant to the antisera elicited by mRNA-1273 and ChAdOx1 vaccines 156 (Fig. 2b,c) . Similar to BA.1, BA.2 was also highly resistant to the 157 vaccine-induced antisera (Fig. 2b,c) . Also, BA.2 was almost completely 158 resistant to two therapeutic monoclonal antibodies, Casirivimab and Imdevimab, 159 and was 35-fold more resistant to another therapeutic antibody, Sotrovimab, 160 when compared to the ancestral D614G-bearing B.1.1 virus (Fig. 2d) . Moreover, 161 both BA.1 and BA.2 were highly resistant to the convalescent sera who had 162 infected with early pandemic virus (before May 2020; Fig. 2e ), Alpha (Extended 163 data Fig. 3a ) and Delta (Extended data Fig. 3b) . These data suggest that, 164 similar to BA.1, BA.2 is highly resistant to the antisera induced by vaccination 165 and infection with other SARS-CoV-2 variants as well as three antiviral 166 therapeutic antibodies. 167 We then tested the 17 sera infected with BA.1: 13 convalescents were 168 fully vaccinated (2 shots), 1 convalescent was 1-dose vaccinated, and 3 169 convalescents were not vaccinated. BA.1 convalescent sera exhibited the 170 strongest antiviral effect against BA.1 (Fig. 2f) . Although BA.2 was 1.4-fold more 171 resistant to the BA.1-infected sera than BA.1, there was no statistical difference 172 ( Fig. 2f; P=0 .091 by Wilcoxon signed-rank test). Importantly, the BA.1 173 convalescent sera with full vaccination exhibited significantly stronger antiviral 174 effects against all variants tested than unvaccinated or 1-dose vaccinated 175 convalescents (Extended Data Fig. 3c) . 176 To address the possibility that the BA.1-induced humoral immunity is 177 less effective against BA.2, we used the convalescent sera obtained from 178 infected hamsters at 16 days postinfection (d.p.i.). Similar to the results of 179 convalescent human sera ( Fig. 2e and Extended Data Fig. 2b) , both BA.1 and 180

BA.2 exhibited pronounced resistances against B.1.1-and Delta-infected 181 convalescent hamster sera ( Fig. 2g and Extended Data Fig. 3d) . Interestingly, 182

BA.2 was significantly (2.9-fold) more resistant to BA.1-infected convalescent 183 hamster sera than BA.1 (Fig. 2g) . To further verify the resistance of BA.2 184 against BA.1-induced immunity, mice were immunised with the cells expressing 185 the S proteins of ancestral B.1.1 and BA.1 and obtained murine antisera. Again, 186 the neutralisation assay using murine sera showed that BA.2 is more 187 significantly (6.4-fold) resistant to the BA.1 S-immunised sera than BA.1 (Fig.  188  2h) . These findings suggest that BA. 195 4) 21 . Although the growth of BA.1 and BA.2 was comparable in 196

VeroE6/TMPRSS2 cells, BA.2 was more replicative than BA.1 in Calu-3 cells 197 and primary human nasal epithelial cells (Fig. 3a) . Notably, the morphology of 198 infected cells was different; BA.2 formed significantly (1.52-fold) larger syncytia 199 than BA.1 ( Fig. 3b and Extended Data Fig. 5a) . Whereas the plaque size in 200

VeroE6/TMPRSS2 cells infected with BA.1 and BA.2 was significantly smaller 201 than those of cells infected with B.1.1, the plaques formed by BA.2 infection are 202

significantly (1.27-fold) larger than those by BA.1 infection ( Fig. 3c and  203 Extended Data Fig. 5b) . Moreover, the coculture of S-expressing cells with 204 HEK293-ACE2/TMPRSS2 cells showed that BA.2 S induces significantly 205

(2.9-fold) larger multinuclear syncytia formation when compared to BA.1 S 206 (Extended Data Fig. 5c ). These data suggest that BA.2 is more fusogenic than 207 BA.1. To further address this possibility, we analysed the fusogenicity of the S 208 proteins of BA.2 S by a cell-based fusion assay 12,22,23 . The expression level of 209 BA.2 S on the cell surface was significantly lower than that of BA.1 S (Extended 210 Data Fig. 6a) . Nevertheless, our fusion assay using VeroE6/TMPRSS2 cells and 211

Calu-3 cells showed that BA.2 S is significantly more fusogenic than BA.1 S (Fig.  212 3d). We then analysed the binding affinity of BA.2 S receptor binding domain 213

(RBD) to ACE2 by an yeast surface display assay 16 Fig. 6b ).

Because we have proposed that the SARS-CoV-2 S-mediated 218 fusogenicity is closely associated with the efficacy of S1/S2 cleavage 12,23 , we 219 hypothesized that BA.2 S is more efficiently cleaved than BA.1 S. However, an 220

western blotting analysis showed that BA.2 S is less cleaved than BA.1 S (Fig.  221 3e), suggesting that BA.2 S exhibits a higher fusogenicity independently of 222 S1/S2 cleavage. 223 We have recently revealed that BA.1 poorly utilizes TMPRSS2 for the 224 infection 11 . To analyse the TMPRSS2 usage by BA.2 S, we performed 225 cell-based fusion assay using 293-ACE2 cells with or without TMPRSS2 226 expression. We verified that 293-ACE2 cells do not express endogenous 227 TMPRSS2 on the cell surface (Extended Data Fig. 6c ). As shown in Fig. 3f enhanced pause (Penh) and the ratio of time to peak expiratory follow relative to 250 the total expiratory time (Rpef), and subcutaneous oxygen saturation (SpO 2 ), 251

whereas BA.1-infected hamsters exhibited no or weak disorders (Fig. 4a) . 252 Notably, all parameters routinely measured, including body weight, Penh, Rpef 253

and SpO 2 , of BA.2-infected hamsters were significantly different from uninfected 254 and BA.1-infected hamsters, and these values were comparable to those of 255 B.1.1-infected hamsters (Fig. 4a) . These data suggest that BA.2 is more 256 pathogenic than BA.1. 257

To analyse viral spread in the respiratory organs of infected hamsters, 258 viral RNA load and nucleocapside (N) expression were assessed by RT-qPCR 259 of viral RNA and immunohistochemistry (IHC), respectively. As shown in Fig. 4b , 260 viral RNA loads in the two lung regions, hilum and periphery, of BA.2-infected 261 hamsters were significantly higher than those of BA.1-infected hamsters. In the 262 lung periphery, the viral RNA load of BA.2 was significantly higher than that of 263 B.1.1, and the viral RNA load of BA.2 at 1 d.p.i. was 11-fold and 9.3-fold higher 264 than those of B.1.1 and BA.1 at the same timepoint, respectively (Fig. 4b) .

To 265 address the possibility that BA.2 more efficiently spreads than BA.1, we 266 investigated the positivity for N protein in the trachea and lung area close to the 267 hilum. At 1 d.p.i., N protein was detectable in the lower tracheal epithelium in all 268

infected hamsters, and particularly, was clearly detectable in the middle part of 269 trachea in BA. BA.2-infected hamsters was 5.4-fold lower than that of BA.1-infected hamsters 278 (Fig. 4d) . At 5 d.p.i., N protein was almost disappeared in BA.1-infected lungs, 279

whereas alveolar staining was still detectable in B.1.1-and BA.2-infected lungs 280 ( Fig. 4c and Extended Data Fig. 7c) . These data suggest that BA.2 is more 281 rapidly and efficiently spread in the lung tissues than BA.1.

Pathogenicity of BA.2 284

To investigate the pathogenicity of BA.2, the right lungs of infected hamsters 285

were collected at 1, 3, and 5 d.p.i. and used them for haematoxylin and eosin 286 (H&E) staining and histopathological analysis 12,23 . All histopathological 287 parameters including bronchitis/bronchiolitis, haemorrhage, alveolar damage, 288 and the levels of type II pneumocytes, of BA.2-infected hamsters were 289 significantly higher than those in BA.1 (Fig 4e and Extended Data Fig. 8a) . The 290 score indicating haemorrhage including congestive edema of BA.2 was 291 significantly higher than that of B.1.1 (Fig. 4e) . As shown in our previous 292 studies 12,23 , hyperplastic large type II pneumocytes suggesting the severity of 293 inflammation were observed in all infected hamsters at 5 d.p.i., and particularly, 294 the area of large type II pneumocytes in BA.2-infected hamsters was 295 significantly larger than those in B.1.1-and BA.1-infected hamsters (Fig. 4e) . 296 Total histology score of BA.2 was significantly higher than that of BA.1 (Fig. 4e) of concern, and this SARS-CoV-2 variant should be monitored in depth. 351

Ethics statement 354

All experiments with hamsters were performed in accordance with the Science 355

Council of Japan's Guidelines for the Proper Conduct of Animal Experiments. 356

The protocols were approved by the Institutional Animal Care and Use 357

Committee of National University Corporation Hokkaido University (approval ID:  358 20-0123 and 20-0060). All experiments with mice were also performed in 359 accordance with the Science Council of Japan's Guidelines for the Proper 360

Conduct of Animal Experiments. The protocols were approved by the 361

Institutional Animal Experiment Committee of The Institute of Medical Science, 362

The University of Tokyo (approval ID: PA21-39 Singapore, South Africa, Sweden, the UK, and the USA) ( Fig. 1 and Extended 482 Data Fig. 2) . Fig. 4) Complete removal of the CPER products (i.e., SARS-CoV-2-related DNA) from 681 the seed virus was verified by PCR. The working virus stock was prepared from 682 the seed virus as described below (see "SARS-CoV-2 preparation and titration" 683 section). The titre of the prepared working virus was measured as the 50% 695 tissue culture infectious dose (TCID 50 ). Briefly, one day before infection, 696

VeroE6/TMPRSS2 cells (10,000 cells) were seeded into a 96-well plate. Serially 697 diluted virus stocks were inoculated into the cells and incubated at 37°C for 4 d. 698

The cells were observed under microscopy to judge the CPE appearance. The 699

value of TCID 50 /ml was calculated with the Reed-Muench method 45 . 700

To verify the sequence of chimeric recombinant SARS-CoV-2, viral 701 RNA was extracted from the working viruses using a QIAamp viral RNA mini kit 702 (Qiagen, Cat# 52906) and viral genome sequence was analysed as described 703

above (see "Viral genome sequencing" section). In brief, the viral sequences of 704

GFP-encoding recombinant SARS-CoV-2 (strain WK-521; GISIAD ID: 705 EPI_ISL_408667) 21,29 that harbour the S genes of respective variants (B.1.1, 706

BA.1 or BA.2) were used for the reference. Information on the unexpected 707 mutations detected is summarized in Supplementary Table 6, and the raw data  708 are deposited in Gene Expression Omnibus (accession number: GSE196649). 709 710

One day before infection, Vero cells (10,000 cells), VeroE6/TMPRSS2 cells 712

(10,000 cells), Calu-3 cells (20,000 cells), HEK293-ACE2 cells (10,000 cells), 713

HEK293-ACE2/TMPRSS2 cells (10,000 cells), were seeded into a 96-well plate. 714 SARS-CoV-2 [100 TCID 50 for VeroE6/TMPRSS2 cells (Fig. 3a) , 1,000 TCID 50 715

for Vero cells (Fig. 3a) , HEK293-ACE2 cells (10,000 cells) (Fig. 3j) , and 716

HEK293-ACE2/TMPRSS2 cells (10,000 cells) (Fig. 3j) ; and 2,000 TCID 50 for 717

Calu-3 cells (Fig. 3a) ] was inoculated and incubated at 37°C for 1 h. The 718

infected cells were washed, and 180 µl of culture medium was added. The 719 culture supernatant (10 µl) was harvested at the indicated timepoints and used 720

for RT-qPCR to quantify the viral RNA copy number (see "RT-qPCR" section 721 below). 722

The infection experiment primary human nasal epithelial cells (Fig. 3a The viral RNA copy number was standardized with a SARS-CoV-2 direct 744 detection RT-qPCR kit (Takara, Cat# RC300A). Fluorescent signals were 745 acquired using QuantStudio 3 Real- Fluorescence microscopy ( Fig. 3b and Extended Data Fig. 5a ) was performed 752 as previously described 23 . Briefly, one day before infection, VeroE6/TMPRSS2 753 cells (10,000 cells) were seeded into 96-well, glass bottom, black plates and 754

infected with SARS-CoV-2 (100 TCID 50 ). At 24, 48, and 72 h.p.i., GFP 755 fluorescence was observed under an All-in-One Fluorescence Microscope 756 BZ-X800 (Keyence) in living cells, and a 13-square-millimeter area of each 757 sample was scanned. Images were reconstructed using an BZ-X800 analyzer 758 software (Keyence), and the area of the GFP-positive cells was measured using 759 this software. 760 761

Plaque assay 762

Plaque assay ( Fig. 3c and Extended Data Fig. 5b SARS-CoV-2 S-based fusion assay 795 SARS-CoV-2 S-based fusion assay ( Fig. 3d and 3f) Fig. 6a ) was measured using FACS Canto II (BD Biosciences) 823 and the data were analysed using FlowJo software v10.7.1 (BD Biosciences). 824

Gating strategy for flow cytometry is shown in Supplementary Fig. 1 Western blot (Fig. 3e) Animal experiments (Fig. 4) Lung function test (Fig. 4a) Fig. 7 ) was performed as previously described 12,23 using 939 an Autostainer Link 48 (Dako). The deparaffinized sections were exposed to 940

EnVision FLEX target retrieval solution high pH (Agilent, Cat# K8004) for 20 m at 941 97°C to activate, and mouse anti-SARS-CoV-2 N monoclonal antibody (R&D 942 systems, Clone 1035111, Cat# MAB10474-SP, 1:400) was used as a primary 943

antibody. The sections were sensitized using EnVision FLEX (Agilent) for 15 m  944  and  visualised  by  peroxidase-based  enzymatic  reaction  with  945 3,3'-diaminobenzidine tetrahydrochloride as substrate for 5 m. The N protein 946 positivity ( Fig. 4c and 4d) was evaluated by certificated pathologists as 947 previously described 12 . Images were incorporated as virtual slide by 948

NDRscan3.2 software (Hamamatsu Photonics). The N-protein positivity was 949 measured as the area using Fiji software v2.2.0 (ImageJ). 950 951 H&E staining 952 H&E staining (Extended Data Fig. 8 ) was performed as previously 953 described 12,23 . Briefly, excised animal tissues were fixed with 10% formalin 954 neutral buffer solution, and processed for paraffin embedding. The paraffin 955 blocks were sectioned with 3 µm-thickness and then mounted on silane-coated 956 glass slides (MAS-GP, Matsunami). H&E staining was performed according to a 957 standard protocol. 958 959

Histopathological scoring 960

Histopathological scoring (Fig. 4e and Extended Data Fig. 8a ) was performed 961 as previously described 12,23 . Pathological features including bronchitis or 962 bronchiolitis, haemorrhage with congestive edema, alveolar damage with 963 epithelial apoptosis and macrophage infiltration, hyperplasia of type II 964 pneumocytes, and the area of the hyperplasia of large type II pneumocytes were 965 evaluated by certified pathologists and the degree of these pathological findings 966

were arbitrarily scored using four-tiered system as 0 (negative), 1 (weak), 2 967 (moderate), and 3 (severe). The "large type II pneumocytes" are the hyperplasia 968 of type II pneumocytes exhibiting more than 10-μm-diameter nucleus. We 969 described "large type II pneumocytes" as one of the remarkable 970 histopathological features reacting SARS-CoV-2 infection in our previous 971 studies 12,23 . Total histology score is the sum of these five indices. 972

To measure the inflammation area in the infected lungs (Extended 973 Data Fig. 8b) , four hamsters infected with each virus were sacrificed at the 1, 3 974 and 5 d.p.i., and all four right lung lobes, including upper (anterior/cranial), 975 middle, lower (posterior/caudal), and accessory lobes, were sectioned along 976 with their bronchi. The tissue sections were stained by H&E, and the digital 977 microscopic images were incorporated into virtual slides using NDRscan3.2 978 software (Hamamatsu Photonics). The inflammatory area including type II 979 pneumocyte hyperplasia in the infected whole lungs was morphometrically 980 analysed using Fiji software v2.2.0 (ImageJ). 981 982

Statistics and reproducibility 983

Statistical significance was tested using a two-sided Student's t-test or a 984 two-sided Mann-Whitney U-test unless otherwise noted. The tests above were 985 performed using Prism 9 software v9.1.1 (GraphPad Software). 986

In the time-course experiments (Fig. 3a, 3d, 3f, 3h, 4a-4c, and 4e) , a 987 multiple regression analysis including experimental conditions (i.e., the types of 988 infected viruses) as explanatory variables and timepoints as qualitative control 989

variables was performed to evaluate the difference between experimental 990 conditions thorough all timepoints. P value was calculated by a two-sided Wald 991

test. Subsequently, familywise error rates (FWERs) were calculated by the Holm 992 method. These analyses were performed in v4.1.2 (https://www.r-project.org/). 993

In Extended Data Fig. 7 and 8 Data availability 1000

The raw data of virus sequences analysed in this study are deposited in Gene 1001

Expression Omnibus (accession number: GSE196649). All databases/datasets 1002 used in this study are available from GISAID database (https://www.gisaid.org) 1003

and Genbank database (https://www.ncbi.nlm.nih.gov/genbank/). The accession 1004

numbers of viral sequences used in this study are listed in Method section. 1005 1006

Code availability 1007

The computational code to estimate the relative effective reproduction number of 1008 each viral lineage ( Fig. 1) is available in the GitHub repository 1009

(https://github.com/TheSatoLab/Omicron_BA2/tree/main/lineage_growth_hierar 1010 chical_model). 1011 Asterisk denote nodes with

Classification of Omicron (B.1.1.529): SARS-CoV-2 variant of B.1.1.529 Omicron

SARS-CoV-2 Omicron virus causes attenuated 1054 disease in mice and hamsters

Receptor binding and complex structures of human ACE2 to 1057 spike RBD from omicron and delta SARS-CoV-2

SARS-CoV-2 Omicron-B.1.1.529 leads to 1060 widespread escape from neutralizing antibody responses

Broadly neutralizing antibodies overcome 1063 SARS-CoV-2 Omicron antigenic shift

Efficacy of Antibodies and Antiviral Drugs against 1066

Covid-19 Omicron Variant

An infectious SARS-CoV-2 B.1.1.529 Omicron 1069 virus escapes neutralization by therapeutic monoclonal antibodies

Rapid epidemic expansion of the SARS-CoV-2 Omicron 1072 variant in southern Africa

Establishment of a reverse genetics system for 1075 SARS-CoV-2 using circular polymerase extension reaction

SARS-CoV-2 spike L452R variant evades cellular 1078 immunity and increases infectivity

Enhanced fusogenicity and pathogenicity of SARS-CoV-2

Delta P681R mutation

The SARS-CoV-2 Lambda variant exhibits enhanced 1083 infectivity and immune resistance

Human serum from SARS-CoV-2 vaccinated and 1086 COVID-19 patients shows reduced binding to the RBD of SARS-CoV-2 1087 Omicron variant

SARS-CoV-2 Omicron RBD shows weaker binding affinity 1090 than the currently dominant Delta variant to human ACE2

SARS-CoV-2 B.1.617.2 Delta variant replication and 1093 immune evasion

fastp: an ultra-fast all-in-one 1096 FASTQ preprocessor

Enhanced isolation of SARS-CoV-2 by 1099 TMPRSS2-expressing cells

Fast and accurate short read alignment with 1102 Burrows-Wheeler transform

The Sequence Alignment/Map format and SAMtools

A program for annotating and predicting the effects of 1107 single nucleotide polymorphisms, SnpEff: SNPs in the genome of 1108 Drosophila melanogaster strain w1118

Minimap2: pairwise alignment for nucleotide sequences

trimAl: a tool 1113 for automated alignment trimming in large-scale phylogenetic analyses

RAxML version 8: a tool for phylogenetic analysis and 1116 post-analysis of large phylogenies

BEAST 2: a software platform for Bayesian 1119 evolutionary analysis

Dating of the human-ape splitting 1122 by a molecular clock of mitochondrial DNA

Genomic reconstruction of the SARS-CoV-2 1125 epidemic in England

Analysis of 2.1 million SARS-CoV-2 genomes 1128 identifies mutations associated with transmissibility

The Anticoagulant Nafamostat Potently Inhibits 1131 SARS-CoV-2 S Protein-Mediated Fusion in a Cell Fusion Assay System 1132 and Viral Infection In Vitro in a Cell-Type-Dependent Manner

SARS-CoV-2 D614G spike mutation increases entry 1135 efficiency with enhanced ACE2-binding affinity

The SARS-CoV-2 Delta variant is poised to acquire complete 1138 resistance to wild-type spike vaccines

Neutralization of the SARS-CoV-2 Mu Variant by 1141 Convalescent and Vaccine Serum

Super-rapid quantitation of the production of HIV-1 harboring a 1145 luminescent peptide tag

A Simple Method of Estimating Fifty Percent 1148 Endpoints

A Protein-Engineered, Enhanced Yeast Display 1150 Platform for Rapid Evolution of Challenging Targets

SARS-CoV-2 variant prediction and antiviral drug 1153 design are enabled by RBD in vitro evolution

Application of the restriction-free (RF) cloning for 1156 multicomponents assembly

AMED Research 1204 Program on HIV/AIDS (21fk0410034, to Akifumi Takaori-Kondo; 21fk0410033, 1205 to Akatsuki Saito; and 21fk0410039

AMED CRDF Global Grant 1206 (21jk0210039 to Akatsuki Saito)

AMED Japan Program for Infectious Diseases 1207 Research and Infrastructure (21wm0325009, to Akatsuki Saito; 21wm0125008, 1208 to Hirofumi Sawa and 21wm0225003

JST SICORP (e-ASIA) (JPMJSC20U1, to 1210 Kei Sato)

JST CREST 1211

Aid for 1213 Scientific Research B (21H02736

JSPS Fund for the Promotion of Joint 1215 International Research (Fostering Joint International Research) (18KK0447, to 1216 Kei Sato)

JSPS Core-to-Core Program (A. Advanced Research Networks) 1217 (JPJSCCA20190008

JSPS Leading Initiative for Excellent Young Researchers 1219 (LEADER) (to Terumasa Ikeda)

World-leading Innovative and Smart Education 1220 (WISE) Program 1801 from the Ministry of Education

The Tokyo Biochemical Research 1222 Foundation (to Kei Sato)

Mitsubishi Foundation (to Terumasa Ikeda)

Foundation of Advanced Medical Research (to Mako Toyoda and 1224

Osaka 1226 University (to Akatsuki Saito); an intramural grant from Kumamoto University 1227 COVID-19 Research Projects (AMABIE) (to Terumasa Ikeda); Intercontinental 1228 Research and Educational Platform Aiming for Eradication of HIV/AIDS (to 1229 Terumasa Ikeda); and Joint Usage/Research Center program of Institute for 1230 Frontier Life and Medical Sciences

Hiromi Mouri 9

Seiya 1237 Ozono

Japan 1246 1247 the panel indicate the number of GFP-positive cells counted. c, Plaque assay. 1289 Diameter of plaques (20 plaques per virus) are summarized. d,f, S-based fusion 1290 assay. The fusion activity (arbitrary units) is shown. e, Western blot. Left, 1291 representative blots of S-expressing cells. ACTB is an internal control. Right, the 1292 ratio of S2 to the full-length S plus S2 proteins. g, Fold increase of pseudovirus 1293 infectivity by TMPRSS2 expression. Assays were performed in quadruplicate 1294 (a,g,h), octuplicate (a, most left) or triplicate (d-f) and data are the average ± s.d. 1295 Each dot indicates the result from an individual plaque (c) and an individual 1296 replicate (e,g). In b and c, raw data are shown in Extended Data Fig. 5 and 6. 1297 Statistically significant differences between BA.2 and other variants through 1298 timepoints were determined by multiple regression (a

| Virological features of BA.2 in vivo

SpO 2 values routinely measured. Hamsters of the same age 1306 were intranasally inoculated with PBS (uninfected). b, Viral RNA load in the lung 1307 hilum (left) and periphery (right). c,d, Percentage of N-positive cells in the whole 1308 lobes of lung (c) and bronchiole in the frontal/upper lobe of lung (d) measured by 1309 IHC. e, Histopathological scoring of lung lesions. Representative pathological 1310 features are shown in our previous studies 12,23 . Data are the average (a, 6 1311 hamsters per each group; b-e, 4 hamsters per each group) ± s.e.m. In a-c,e, 1312 statistically significant differences between BA.2 and other variants or uninfected 1313 hamsters through timepoints were determined by multiple regression. The data 1314 at 0 d.p.i. was excluded from the analyses. FWERs calculated using the Holm 1315 method are indicated in the figures

All the BA.2 and BA.3 sequences and 200 1322 randomly sampled BA.1 (including 20 BA.1.1) sequences were used. The 1323 time-resolved trees were constructed by BEAST2. Regarding a node with ≥ 0.95 1324 posterior value (denoted by an asterisk), the 95% CI of the divergence time is 1325 shown. b, Estimated emergence dates of the Omicron lineages. The 95% CI 1326 (error bar) and posterior mean (dot) are shown. 1327 1328 Extended Data Fig. 2. Epidemic dynamics of SARS-CoV-2 lineages in 1329 countries with the BA.2 epidemic. 1330 a, Daily sequence frequency of each viral lineage in eleven countries where 1331 ≥ 100 BA.2 sequences have been reported by

Alpha, Delta, BA.1 and 1344 BA.2. Assay of each serum sample was performed in triplicate to determine 1345 NT50, and each dot represents each NT50 value. Geometric mean and 95% CI 1346 are shown. The number indicates the fold change of resistance versus each 1347 antigenic variant. In a and b, statistically significant differences between BA.1 1348 and BA.2 were determined by two-sided Wilcoxon signed-rank test. In c, 1349 statistically significant differences between fully-vaccinated (13 donors) and 1350 not-fully-vaccinated (4 donors) were determined by two-sided Mann-Whitney 1351 U-test. Information of vaccinated and convalescent donors are summarized in 1352

Additionally, the ORF7b was swapped with sfGFP gene. 1360 1361 Extended Data Fig. 5. Syncytia and plaque formations by BA.2. 1362 a, Fluorescence microscopy. GFP area of infected VeroE6/TMPRSS2 cells 1363 (m.o.i. 0.01) at 24, 48, and 72 h.p.i were measured. Higher-magnification views 1364 of the regions indicated by squares are shown at bottom. b, Plaque assay. c, 1365 Coculture of S-expressing cells with HEK293-ACE2/TMPRSS2 cells. Left, 1366 representative images of S-expressing cells cocultured with HEK293 cells (top) 1367 or HEK293-ACE2/TMPRSS2 cells (bottom)

Representative histograms stained with 1376 anti-S1/S2 polyclonal antibody (left) and the summarised data (right) are 1377 respectively shown. The number in the histogram indicates MFI. Grey 1378 histograms indicate isotype controls. b, Binding affinity of SARS-CoV-2 S RBD 1379 to ACE2 by yeast surface display. Left, The percentage of the binding of the 1380 SARS-CoV-2 S RBD expressed on yeast to soluble ACE2 (left) and the 1381 summarised data (right) are respectively shown. c, TMPRSS2 expression on the 1382 cell surface. Left, representative histograms stained with anti-TMPRSS2 1383 polyclonal antibody are shown. The number in the histogram indicates MFI. Grey 1384 histograms indicate isotype controls. Right, summarized data. Assays were 1385 performed in triplicate, and data are the average ± s.d. Each dot indicates the 1386 result from an individual replicate. A statistically significant difference between 1387 BA.1 an BA.2 was determined by two-sided unpaired

Each 1393 panel indicates the representative result from an individual infected hamster. c, 1394 Lung lobes of the hamsters infected with B.1.1, BA.1 or BA.2 (n = 4 for each 1395 virus) at 1, 3 and 5 d.p.i. were immunohistochemically stained with 1396 anti-SARS-CoV-2 N monoclonal antibody

Yeast surface display (Extended Data Fig. 6b ) was performed as previously 831 described as previously described 22, 24, 46 . Briefly, the peptidase domain of human 832 ACE2 (residues 18-740) was expressed in Expi293 cells and purified by a 5-ml 833HisTrap Fast Flow column (Cytiva, Cat# 17-5255-01) and Superdex 200 16/600 834 (Cytiva, Cat# 28-9893-35) using an ÄKTA pure chromatography system (Cytiva), 835 and the purified soluble ACE2 was labelled with CF640 (Biotium, Cat# 92108). 836Protein quality was verified using a Tycho NT.6 system (NanoTemper) and 837 ACE2 activity assay kit (SensoLyte, Cat# AS-72086