key: cord-0766077-t8ejlv94 authors: Nutalai, Rungtiwa; Zhou, Daming; Tuekprakhon, Aekkachai; Ginn, Helen M.; Supasa, Piyada; Liu, Chang; Huo, Jiandong; Mentzer, Alexander J.; Duyvesteyn, Helen M.E.; Dijokaite-Guraliuc, Aiste; Skelly, Donal; Ritter, Thomas G.; Amini, Ali; Bibi, Sagida; Adele, Sandra; Johnson, Sile Ann; Constantinides, Bede; Webster, Hermione; Temperton, Nigel; Klenerman, Paul; Barnes, Eleanor; Dunachie, Susanna J.; Crook, Derrick; Pollard, Andrew J.; Lambe, Teresa; Goulder, Philip; Paterson, Neil G.; Williams, Mark A.; Hall, David R.; Mongkolsapaya, Juthathip; Fry, Elizabeth E.; Dejnirattisai, Wanwisa; Ren, Jingshan; Stuart, David I.; Screaton, Gavin R.; Conlon, Christopher; Deeks, Alexandra; Frater, John; Frending, Lisa; Gardiner, Siobhan; Jämsén, Anni; Jeffery, Katie; Malone, Tom; Phillips, Eloise; Rothwell, Lucy; Stafford, Lizzie title: Potent cross-reactive antibodies following Omicron breakthrough in vaccinees date: 2022-05-20 journal: Cell DOI: 10.1016/j.cell.2022.05.014 sha: 452a1e37914e523bdd82202c440cff38713a931d doc_id: 766077 cord_uid: t8ejlv94 Highly transmissible Omicron variants of SARS-CoV-2 currently dominate globally. Here, we compare neutralization of Omicron BA.1, BA.1.1 and BA.2. BA.2 RBD has slightly higher ACE2 affinity than BA.1 and slightly reduced neutralization by vaccine serum, possibly associated with its increased transmissibility. Neutralization differences between sub-lineages for mAbs (including therapeutics) mostly arise from variation in residues bordering the ACE2 binding site, however, more distant mutations S371F (BA.2) and R346K (BA.1.1) markedly reduce neutralization by therapeutic antibody Vir-S309. In-depth structure-and-function analyses of 27 potent RBD-binding mAbs isolated from vaccinated volunteers following breakthrough Omicron-BA.1 infection reveals that they are focussed in two main clusters within the RBD, with potent right-shoulder antibodies showing increased prevalence. Selection and somatic maturation have optimized antibody potency in less-mutated epitopes and recovered potency in highly mutated epitopes. All 27 mAbs potently neutralize early pandemic strains and many show broad reactivity with variants of concern. There was a major reduction in neutralization titre for all Omicron shows that the bulky Phe protrudes outwards disturbing the glycan attached to residue 343 of 295 the RBD (Figure 4A ). This sugar is critical for S309 binding, explaining the 126-fold reduction 296 of neutralization titre to BA.2 compared to Victoria. Furthermore, neutralization of BA.1.1 is 297 4-fold worse than BA.1, due to the R346K mutation, since the shortened side chain cannot 298 interact as effectively with Asp 93 of the S309 heavy chain ( Figure 4B ). Neutralization of 299 BA.2 is approximately 20-fold worse than BA.1, consistent with SPR analysis which showed 300 that binding to BA.2 ~15-fold weaker than to BA.1 (Figure S2H,I) . Omi-9, which shows reduced neutralization of Beta and Gamma, positions 329 close to residue 484 which is mutated from Glu to Lys in Beta/Gamma and to Ala in Omicron. 330 The second, right shoulder, cluster was seen in the full set of early pandemic antibodies, above 331 the S309 site ( Figure 5A ). This region is occupied by 5 of the 6 IGHV1-69 mAbs, the other, give structural information on the binding of 11 of the 28 most potent antibodies, although for 341 several the resolution was limited, and for some a structurally characterised nanobody (Huo et 342 al., 2021) Omi-3 and -18 are representative of IGHV3-53 and IGHV3-66 antibodies that bind at the back 349 of the neck and account for 9/28 of the most potent antibodies. They show how these antibodies 350 can be adapted to broadly neutralize all major SARS-CoV-2 variants ( Figure 6B) Figure S3, Figure 6B) . 359 We have determined structures for five mAbs within the neck/left shoulder cluster, Omi-2, -9, 361 -12, -25 and -42. Some broadly neutralize all VoC while others are sensitive to the mutations 362 at residue 417 and 484 found in Beta and Gamma (explained for Omi-25 in Figure S4A ). In 363 terms of overall pose Omi-9 is an outlier, being perched upright on the RBD, whilst the others 364 approach from the back ( Figure 5E ). Omi-2 belongs to the IGHV1-69 gene family but has 365 features in common with Omi-12, the only member of the IGHV1-58 gene family found in the 366 set of 28 potent antibodies. In particular, Omi-2 and Omi-12 have a disulphide bond and Pro 367 and Phe residues at the same positions in the H3 loop which mediate interactions with F486 of 368 the RBD, these commonalities appear to drive Omi-2 to adopt almost exactly the same pose as 369 Omi-12, which differs from the other potent antibodies that bind in this region ( Figure 6A) . 370 Note that while Omi-12, like many other IGHV1-58 antibodies, is glycosylated in the H3 loop, 371 Omi-2 is non-glycosylated (Dejnirattisai et The emergence of the highly transmissible Omicron variant and its extremely rapid global 456 spread led to considerable concern, however early data from South Africa that Omicron led to 457 less severe disease has been borne out in waves of infection in other countries (Nealon and 2021b). We find that most of these bind in a similar way to the right shoulder, with several 511 affected by the R346K mutation on BA.1.1, presumably due to stabilizing contacts analogous 512 to that seen for S309 ( Figure 7F ). Interestingly the exception to this binding pattern is Omi-2 513 which binds in the other major cluster ( Figure 6A) . 514 515 Whilst the neutralization properties of most Omicron monoclonal antibodies isolated in this 516 study did not show differences against BA.1, BA.1.1 and BA.2, some clinical mAbs showed 517 differences, in particular REGN10987 regained some activity against BA.2 and AZD1061 518 regained most activity against BA.2 but lost activity against BA.1.1 compared to BA.1. Of 519 particular concern S309, the activity of which is already reduced 6-fold against BA.1 520 (Dejnirattisai et al., 2022) , was reduced a further 4-fold against BA.1.1 and a further 20-fold 521 against BA.2. Although in the short term, genotyping may allow more efficient targeting of 522 mAb therapy, there is a need to develop new Omicron specific antibodies to add to existing 523 SARS-CoV-2 monoclonal antibody cocktails, or to develop broadly cross-reactive antibodies, 524 to provide pre-exposure prophylaxis or post exposure treatment to the many 525 immunosuppressed patients unable to mount protective responses following vaccination. (A67V, Δ69-70, 919 T95I, G142D/Δ143-145, Δ211/L212I, ins214EPE, G339D, S371L, S373P, S375F, K417N, 920 N440K, G446S, S477N, T478K, E484A, Q493R, G496S, Q498R, N501Y, Y505H, T547K, 921 D614G, H655Y, N679K, P681H, N764K, D796Y, N856K, Q954H, N969K and L981F), 922 BA.1.1 (BA.1 as above plus R346K), BA.2 (T19I, LPPA24S, G142D, V213G, G339D, S371F, 923 S373P, S375F, T376A, D405N, R408S, K417N, N440K, S477N, T478K, E484A, Q493R, The gene for nanobody C1 (NbC1) and F2 (NbF2) and were codon-optimized using the IDT 1033 Codon Optimization Tool, synthesized as a ready-to-clone gene fragment (Integrated DNA 1034 Technologies), and cloned into the phagemid vector pADL-23c. The nanobodies were 1035 produced as previously described (Huo et al., 2021) . Briefly, the plasmid was transformed into 1036 the WK6 E. coli strain and protein expression induced by 1 mM IPTG grown overnight at 28 1037 °C. Periplasmic extract was prepared by osmotic shock, and the nanobody protein was purified 1038 with a 5 mL HisTrap nickel column (Cytiva), followed by size exclusion with a Hiload 16/60 1039 Superdex 75 column. 1040 1041 Surface Plasmon Resonance 1042 The surface plasmon resonance experiments were performed using a Biacore T200 (GE 1043 Healthcare). All assays were performed with a running buffer of HBS-EP (Cytiva) at 25 °C. 1044 To determine the binding kinetics between the SARS-CoV-2 RBDs and ACE2 / monoclonal 1045 antibody (mAb), a Protein A sensor chip (Cytiva) was used. ACE2-Fc or mAb was 1046 immobilized onto the sample flow cell of the sensor chip. The reference flow cell was left 1047 blank. RBD was injected over the two flow cells at a range of five concentrations prepared by 1048 serial twofold dilutions, at a flow rate of 30 μl min −1 using a single-cycle kinetics programme. 1049 Running buffer was also injected using the same programme for background subtraction. All 1050 data were fitted to a 1:1 binding model using Biacore Running buffer was also injected using the same programme for background subtraction. All 1057 data were fitted to a 1:1 binding model using Biacore T200 Evaluation Software 3.1. Οmi-41 08 1-18*04 4*02 3-9*01 11 K 4-1*01 2*02 () 5 Οmi-42 09 3-9*01 6*02 6-19*01 7 λ 2-8*01 2*01 , or 3*01 or 3*02 5 J o u r n a l P r e -p r o o f RBD proteins were deglycosylated with Endoglycosidase F1 before used for crystallization Initial screening of crystals was set up in Crystalquick 96-well X plates (Greiner Bio-One) with a Cartesian Robot using the nanoliter sitting-drop vapor-diffusion method, with 100 nL of 1076 protein plus 100 nL of reservoir in each drop, as previously described For crystallization, Omicron BA.1-RBD was mixed with Omi-25 Fab, and Omicron BA RBD was mixed with COVOX-150 and ACE2 separately, in a 1:1 molar ratio, with a final 1079 concentration of 13 mg ml -1 . Omicron BA.1-RBD was mixed with Omi-3 and EY6A Fabs Omi-6 and COVOX-150 Fabs, Omi-9 Fab and Nanobody F2 (NbF2), and Omi-12 and beta-54 Fabs separately, in a 1:1:1 molar ratio, with a final concentration of 7 mg ml -1 . Omicron BA.1-1082 RBD was mixed with Omi-32 Fab and NbC1 in a 1:1:1 molar ratio, with a final 1083 concentration of 11 mg/ml. Omi18 Fab :1:1 molar ratio, with a final 1085 concentration of 7 mg ml -1 .These complexes were separately incubated at room temperature 1086 for 30 min Crystals of BA.1-RBD/Omi-25 were obtained from Molecular Dimensions Proplex condition 1089 1-31, containing 3.0 M Sodium formate and 0.1 M Tris pH 7.5. BA crystals were obtained in 2 different space groups. Crystals of space group C2 were Hampton Research PEGRx condition 1-29, containing 0.1 M Sodium citrate tribasic dihydrate 1092 pH 5.5 and 18% (w/v) PEG 3350. Crystals of space group P21 SARS-CoV-2 Neutralizing Antibody LY-CoV555 in 1262 Outpatients with Covid-19 A neutralizing human antibody binds to the N-terminal domain of the 1266 Spike protein of SARS-CoV-2 Anti-SARS-CoV-2 receptor-binding 1270 domain antibody evolution after mRNA vaccination SARS-CoV-2 Omicron-B.1.1.529 1274 leads to widespread escape from neutralizing antibody responses The antigenic anatomy of SARS-CoV-1278 2 receptor binding domain Antibody evasion by the 1282 P.1 strain of SARS-CoV-2 Safety and 1289 immunogenicity of the ChAdOx1 nCoV-19 vaccine against SARS-CoV-2: a preliminary 1290 report of a phase 1/2, single-blind, randomised controlled trial Development of an inactivated vaccine candidate for SARS-CoV-2 Enzymatic assembly of overlapping DNA fragments Pre-clustering data sets using cluster4x improves the signal-to-noise ratio 1300 of high-throughput crystallography drug-screening analysis Receptor binding and complex structures of human ACE2 to spike RBD from 1305 omicron and delta SARS-CoV-2 A potent SARS-CoV-2 neutralising nanobody 1309 shows therapeutic efficacy in the Syrian golden hamster model of COVID-19 Neutralization of SARS-CoV-2 by 1314 Destruction of the Prefusion Spike Protein interfaces, surfaces and assemblies service 1317 PISA at European Bioinformatics Institute Structure of the SARS-CoV-2 spike receptor-binding domain bound to the ACE2 1321 receptor Macromolecular structure determination 1325 using X-rays, neutrons and electrons: recent developments in Phenix 1.617 by vaccine and convalescent serum The Beta mAb response underscores the 1334 antigenic distance to other SARS-CoV-2 1335 variants Covid-19: AstraZeneca says its antibody drug AZD7442 is effective for 1338 preventing and reducing severe illness Covid-19: Booster vaccine gives "significant increased protection" in 1341 over 50s Structural basis of SARS-CoV-2 1345 Omicron immune evasion and receptor engagement Phaser crystallographic software Increased Potency and Breadth of SARS-CoV-2 Neutralizing Antibodies After a Third mRNA Vaccine Dose Omicron severity: milder but not mild Application of the Restriction-Free (RF) cloning for 1359 multicomponents assembly UCSF ChimeraX: Structure visualization for researchers, 1363 educators, and developers Cross-neutralization of SARS-CoV-2 by a 1367 human monoclonal SARS-CoV antibody Safety and Efficacy of the 1371 BNT162b2 mRNA Covid-19 Vaccine cryoSPARC: algorithms 1374 for rapid unsupervised cryo-EM structure determination Modular basis for potent SARS-CoV-2 neutralization 1378 by a prevalent IGHV1-2-derived antibody class Immune imprinting, breadth of variant 1382 recognition, and germinal center response in human SARS-CoV-2 infection and vaccination Crystal structure of cat 1386 muscle pyruvate kinase at a resolution of 2.6 A Emerging antibody-based therapeutics against SARS-CoV-2 1389 during the global pandemic Structure-based development of three-and four-antibody cocktails 1393 against SARS-CoV-2 via multiple mechanisms Reduced neutralization of 1397 SARS-CoV-2 B.1.1.7 variant by convalescent and vaccine sera Safety and efficacy of the 1401 ChAdOx1 nCoV-19 vaccine (AZD1222) against SARS-CoV-2: an interim analysis of four 1402 randomised controlled trials in Brazil, South Africa, and the UK Tectonic conformational changes of a coronavirus spike glycoprotein promote 1406 membrane fusion A procedure for setting up high-throughput nanolitre crystallization experiments. I. 1410 Protocol design and validation 2021) mRNA vaccine-elicited 1414 antibodies to SARS-CoV-2 and circulating variants REGN-COV2, a Neutralizing Antibody Cocktail, in 1418 Outpatients with Covid-19 xia2: an expert system for macromolecular crystallography data reduction DIALS: 1425 implementation and evaluation of a new integration package Virological characteristics of SARS-CoV-2 BA.2 1430 variant bioXriv Structural basis of a shared antibody response to SARS-CoV-2 A Protein-Engineered, Enhanced Yeast Display Platform for Rapid 1438 Evolution of Challenging Targets Evidence of escape of 1442 SARS-CoV-2 variant B.1.351 from natural and vaccine-induced sera Structural basis for the neutralization 1447 of SARS-CoV-2 by an antibody from a convalescent patient A Novel Coronavirus from Patients with Pneumonia in China Rapid isolation and profiling of a diverse 1456 panel of human monoclonal antibodies targeting the SARS-CoV-2 spike protein Οmi-23 picked with the blob-picker module before spike-like particles from 2D classification of this 1178 initial set were used as a template for template-based picking. Maps and FSC curves for all 1179 analyses are shown in Figure S4I . For Omi-42 particles were sorted in two rounds of 2D 1180 classification followed by ab-initio reference classification into three classes, followed by a 1181 second classification into two classes. Particles from the best class, 106811 in total, were then 1182 further refined to 3.64 Å reported resolution (as determined within the cryoSPARC interface, 1183 AuFSC = 0.143). A second, somewhat lower resolution class, where RBDs were oriented 1184 slightly differently was also refined (see Figure S4I ). For Omi-2 182828 particles were 1185 derived from two rounds of classification, before further 3D classification and local 1186 refinement of the entire spike, but with the fulcrum focussed at the RBD/fab region to better 1187 resolve the interfaces of interest (various local refinements with masking and with/without 1188 subtracted densities failed to improve this region). For Omi-38, particles were sorted in two 1189 rounds of 2D classification before classification using three ab-initio models. The best class, 1190 with 201474 particles was then refined further, with global and local ctf refinement and no 1191 symmetry imposed, resulting in a final reported global reconstruction at AuFSC 0.143 of 2.90 1192Omi-38 with B.1.135 was performed also using cryoSPARC upon this particle set from 1194 which the areas outside of the area of interest (two upwards conformation RBDs in close 1195 proximity to each other and associated fabs) was subtracted. Areas were subtracted/refined 1196 using masks created in Chimera X (Pettersen et al., 2021) . Masks were created as follows, 1197within Chimera X, the area of interest was selected from the global spike map using the 1198 volume eraser tool, a gaussian filter was then applied, and the resulting volume imported into 1199 cryoSPARC with an additional dilation radius of 5 and soft padding width of 5 pixels. The Video S1 Antibody response correlation clustering. Related to Figure 4D . 1223 Video S2 Antigenic cartography three-dimensional analysis. Related to Figure 7E .