key: cord-0786205-f970h5oo
authors: Tanaka, Shiho; Anders Olson, C.; Barnes, Christopher O.; Higashide, Wendy; Gonzalez, Marcos; Taft, Justin; Richardson, Ashley; Martin-Fernandez, Marta; Bogunovic, Dusan; Gnanapragasam, Priyanthi N.P.; Bjorkman, Pamela J.; Spilman, Patricia; Niazi, Kayvan; Rabizadeh, Shahrooz; Soon-Shiong, Patrick
title: Rapid Identification of Neutralizing Antibodies against SARS-CoV-2 Variants by mRNA Display
date: 2021-09-14
journal: bioRxiv
DOI: 10.1101/2021.09.14.460356
sha: 0ed4cbb1014acf873fade169ac009ee1017700f3
doc_id: 786205
cord_uid: f970h5oo

The increasing prevalence of SARS-CoV-2 variants with the ability to escape existing humoral protection conferred by previous infection and/or immunization necessitates the discovery of broadly-reactive neutralizing antibodies (nAbs). Utilizing mRNA display, we identified a set of antibodies against SARS-CoV-2 spike (S) proteins and characterized the structures of nAbs that recognized epitopes in the S1 subunit of the S glycoprotein. These structural studies revealed distinct binding modes for several antibodies, including targeting of rare cryptic epitopes in the receptor-binding domain (RBD) of S that interacts with angiotensin- converting enzyme 2 (ACE2) to initiate infection, as well as the S1 subdomain 1. A potent ACE2-blocking nAb was further engineered to sustain binding to S RBD with the E484K and L452R substitutions found in multiple SARS-CoV-2 variants. We demonstrate that mRNA display is a promising approach for the rapid identification of nAbs that can be used in combination to combat emerging SARS-CoV-2 variants.

The emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the 41 causative agent of the respiratory disease COVID-19, has resulted in a pandemic that brought the 42 world to a standstill (Zhou et al., 2020) . with IC50s between 0.076 -7.0 µg/mL. Structural analysis revealed a subset of these nAbs 80 recognize RBD and NTD epitopes, including a rare, cryptic, cross-reactive RBD epitope. 81

Moreover, we characterize a weakly neutralizing antibody that recognizes the S1 subdomain 1 82 (SD1), providing insight into a unique class of antibodies that are infrequently found among 83 convalescent individuals (Zost et al., 2020a (Zost et al., , 2020b ) that can be utilized in the fight against 84 COVID-19. Finally, we describe the utility of mRNA display for rapid identification of variant-85 resistant antibody clones. This powerful technique enabled the rapid selection of a discovered 86 SARS-CoV-2 nAb to extend its neutralizing capability to SARS-CoV-2 expressing the E484K 87 and L452R S RBD mutations found in multiple SARS-CoV-2 variants. 88

Identification of Anti-SARS-CoV-2 Spike Antibodies by mRNA Display 90

We utilized mRNA display to identify mAbs targeting the S protein of SARS-CoV-2 and 91 discovered 10 novel VH/VL sequences that bind various domains on S (Table S1 ). S comprises 92 an N-terminal fragment known as S1, which further divides into the NTD, ACE2 RBD, small C-93 terminal subdomains 1 and 2 (SD1 and SD2), and a C-terminal "S2" fragment (Figure 1A 1A-C). All 10 antibodies bind corresponding binding domains with low nM binding affinity (KD) 100 (Figure 1B and Table S2 ), but apparent affinities are far superior (KD < 3 pM) for S trimers 101 (Figure 1B and 1C, and Table S3 ). Among the 3 RBD binders, only N-612-017 showed 102 competition with ACE2 binding ( Figure 1D ). 103

To further map the binding regions of the 10 antibodies, we performed epitope binning 104 experiments using S1 and S2 fragments separately ( Figure 1E ). Two NTD binders blocked each 105 other but not RBD or SD1 binding antibodies. The 3 RBD binders competed with each other 106 (Figure 1E Epitope binning using the S2 domain revealed N-612-007 and N-612-044  110   are non-competing whereas N-612-086 competes with both N-612-007 and N-612-044,  111 suggesting they all bind distinct epitopes on S2 ( Figure 1E and 1F) . 112

In addition, multiple biophysical assays were carried out to determine the developability of 113 all 10 antibodies (Table S4 ) (Jain et al., 2017). All 10 mAbs displayed low polyreactivity scores 114 by meso scale diagnostic (MSD) analysis and low self-interaction scores by BLI-clone self-115 interaction (CSI) ( Table S4 ). Eight of the mAb candidates exhibited low hydrophobicity in the 116 hydrophobic interaction column (HIC) chromatography while higher hydrophobicity was 117 observed for N-612-041 and N-612-074 (Table S4) . N-612-041 also showed more rapid 118 aggregation in an accelerated stability assay system while the other 9 mAbs demonstrated long-119 term stability (Table S4) . Furthermore, all 10 mAbs exhibited desirable thermostability of Fab in 

Ten mAbs identified by mRNA display were assessed for neutralization activity against 142 authentic SARS-CoV-2 virus in a Vero E6 cell neutralization assay. The ACE2-blocking anti-143 RBD antibody N-612-017 demonstrated the highest neutralization activity and the non-ACE2-144 blocking RBD binder N-612-056 showed weaker but nearly complete neutralization of ~87%. 145 (Figure 2A To investigate the specificity of RBD-targeting for nAbs N-612-017 and N-612-056, we 171 determined a 3.2 Å single-particle cryo-electron microscopy (cryo-EM) structure of a complex 172 between SARS-CoV-2 S trimer and the N-612-017 Fab (Figure 3, Figure S2 and Table S5 ), and 173 a 2.9 Å X-ray crystal structure of a SARS-CoV-2 RBD -N-612-056 Fab complex (Figure 4 and 174 Table S6 ). The N-612-017 -S trimer complex structure revealed N-612-017 Fab binding to both 175 'up' and 'down' RBD conformations, and recognition of an epitope that partially overlapped 176 with the ACE2 receptor binding site ( Figure 3A and 3B), consistent with BLI competition data 177 ( Figure 1D ). N-612-017 uses five of its six complementarity-determining region (CDR) loops 178 and HC framework region 3 (FWR3) to interact with an epitope focused on RBD residues 179 adjacent to the ACE2 receptor binding ridge (Figure 3C Next, we analyzed the high-resolution X-ray crystal structure of the SARS-CoV-2 RBD -N-201 612-056 Fab complex ( Figure 4A ). This method was used rather than cryo-EM due to N-612-202 056's lack of binding to intact S trimers ( Figure 4A ; inset). Similar to the donor-derived antibody 203 COVOX-45 (Dejnirattisai et al., 2021), N-612-056 binds a rare cryptic epitope that is not readily 204 found in the repertoire of antibodies from convalescent donors ( Figure S4 ). Consistent with 205 observed binding to dissociated S1 protomers by single-particle cryo-EM (data not shown), the 206 N-612-056 cryptic epitope is inaccessible on an S trimer due to steric clashes with the 207 neighboring NTD, and does not overlap with the ACE2 binding site (Figure 4A and 4B) . N-612-208 056 HC and LC CDR loops participate equally to bury ~890 Å 2 of the RBD epitope surface area 209 that comprises residues 352-357 in the b1 strand, which is part of a structurally conserved 5-210 stranded RBD b-sheet, and residues 457-471 that comprise a disordered loop directly beneath the 211 ACE2 receptor binding ridge (Figure 4C and 4D) . In addition to N-612-014, we also identified antibody N-612-004, an S1-specific antibody 256 that was mapped to a domain outside of the NTD and RBD (Figure 1 ). Using single-particle 257 cryo-EM, we determined a 4.8Å structure of N-612-004 bound to a dissociated S1 protomer, 258 which revealed recognition of a SD1 epitope ( Figure 5F and Figure S2 ). Consistent with our 259 library design that varied CDR loops H2, H3 and L3, N-612-004 contacts were solely mediated 260 by these three regions, which led to recognition of loops 556-563 and 567-69 in the SD1 domain 261 ( Figure 5G ). The epitope for N-612-004 is not accessible on S trimers, which likely explains the 262 lack of N-612-004-like antibodies identified among a repertoire of antibodies found in 263 convalescent plasma ( Figure S4 variants with IC50 of 2-10 µg/mL as expected ( Figure 6C ). 296

The binding affinity of N-612-014 and N-612-004 against the recombinant S1 domain 297 containing B.1.1.7 mutations was tested and it was determined that 69-70del and Y144del on 298 NTD did not affect binding affinity of N-612-014 for S1 whereas these mutations moderately 299 lowered (by about 3-fold) the binding affinity of N-612-004 for S1 ( Figure S3B and C). type of discrepancy is rare, such inter-assay discrepancies have been described previously (Liu et  384 al., 2020). N-612-014 may require a longer incubation time to reach maximum neutralization 385 because this allows the opportunity for the S trimer to adopt a conformation that is more 386 susceptible to S1 shedding that is promoted by the antibody, thus destabilizing spike; this 387 hypothesis awaits experimental confirmation. 388

The SD1-targeting antibody N-612-004 displayed partial neutralization activity and was 389 only observed in complex with S1 domain dissociated from the spike trimer in cryo-EM. To our 390 knowledge, there have been no reports on SD1-targeting antibodies that display neutralization 391 activity. We also identified the S2-targeting antibody N-612-007 that displayed partial 392 neutralization activity in a live virus neutralization assay and while structural analysis was 393 attempted, we were unable to visualize/characterize an S trimer/N-612-007 complex. nAbs Corporation) for 15 sec and subsequently blocked with 5 µg/mL mAbs or BLI assay buffer for 5 812 min. BLI signal from ACE2 binding were measured by incubating RBD-coated/mAb blocked 813 biosensors in 25 nM ACE2-IgG1Fc for 3 min. 814

For epitope binning using S1 domain, biotinylated S1 binding mAbs at 25 µg/mL were first 816 loaded on High Precision Streptavidin SAX biosensors (Sartorius Corporation) for 10 sec. 3.75 817 µg/mL of recombinant SARS-CoV-2 S1 were used to bind mAb captured on biosensors for 3 818 min and subsequently 10 µg/mL S1 binding mAb were incubated with biosensors to observe 819 binding competition and signal was recorded for 3 min. For epitope binning using S2 domain, 820 recombinant SARS-CoV-2 S2-His-tag at 10 µg/mL was loaded on Anti-Penta-HIS (HIS1K) 821 biosensors (Sartorius Corporation) for 1 min. 10 µg/mL of S2 binding mAbs were sequentially 822 incubated with biosensors for 3-5 min to observe binding competition. 823

All aspects of the assay utilizing virus were performed in a BSL3 containment facility 825 according to the ISMMS Conventional Biocontainment Facility SOPs for SARS-CoV-2 cell 826 culture studies. Vero E6 cells were seeded into 96-well plates at 20,000 cells/well and cultured 827 overnight at 37°C. The next day, 3-fold serial dilutions of mAbs were prepared in DMEM 828 containing 2% FBS, 1% NEAAs, and 1% Pen-Strep (vDMEM). SARS-CoV-2 virus stock was 829 prepared in vDMEM at 10,000 TCID50/mL, mixed 1:1 (v:v) with the mAb dilutions, and 830 incubated for 30 min or 24 hr at 37°C. Media was removed from the Vero E6 cells, mAb-virus 831 complexes were added and incubated at 37°C for 48 hours before fixation with 4% PFA. movies by 2 (0.869 Å/pixel). The non-dose-weighted images were used to estimate CTF 866 parameters using Patch CTF in cryoSPARC, and micrographs with poor CTF fits, signs of 867 crystalline ice, and field of views that were majority carbon were discarded. Particles were 868 picked in a reference-free manner using Gaussian blob picker in cryoSPARC(Punjani et al., 869 2017) Initial particle stacks were extracted, binned x4 (3.48 Å/pixel), and subjected to ab initio 870 volume generation (4 classes) and subsequent heterogeneous refinement with all particles. The 871 3D classes that showed features for a Fab-S trimer complex or Fab-S1 protomer were 2D 872 classified to polish particle stacks. The resulting particle stacks were unbinned (0.869 Å/pixel) 873 and re-extracted using a 432 box size, and moved to Relion v3.1 (Zivanov et al., 2018) for 874 further 3D classification. Particles corresponding to distinct states were separately refined using 875 non-uniform 3D refinement imposing C1 symmetry in cryoSPARC and final resolutions were 876 estimated according to the gold-standard FSC (Bell et al., 2016) . 877 004 to determine whether each mAb was completely blocked, partially blocked, or non-blocked 968 by convalescent plasma from 4 different patients. 969 Supplementary Tables  970  971  Table S1 . Ten unique spike binding VH/VL sequences identified by mRNA display 972 Table S6 . X-ray crystallography data collection and refinement statistics (related to Figure  986  4) 

Immunodominant Sites on the SARS-CoV-2 Spike Receptor-Binding Domain by Structure-Guided 598

Cross-neutralization of SARS-CoV-2 by a human 601 monoclonal SARS-CoV antibody

Spike mutation D614G alters SARS-CoV-2 fitness. 604

cryoSPARC: algorithms for 606 rapid unsupervised cryo-EM structure determination

Adapt or perish: SARS-CoV-2 antibody escape 608 variants defined by deletions in the Spike N-terminal Domain

Convergent antibody responses to SARS-CoV-2 in 612 convalescent individuals

RNA-peptide fusions for the in vitro selection of 614 peptides and proteins

Deep Mutational Scanning of SARS-CoV

Receptor Binding Domain Reveals Constraints on Folding and ACE2 Binding

Prospective mapping of viral mutations that escape antibodies used to 621 treat COVID-19

Neutralizing and protective human 624 monoclonal antibodies recognizing the N-terminal domain of the SARS-CoV-2 spike protein

mRNA display: ligand discovery, 627 interaction analysis and beyond

Detection of a SARS-CoV-2 variant of concern in 630 South Africa

Cryo-electron microscopy structure of a coronavirus spike glycoprotein 633 trimer

Unexpected Receptor Functional Mimicry Elucidates 636 Activation of Coronavirus Fusion

Antigenicity of the SARS-CoV-2 Spike Glycoprotein

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

Broad neutralization of SARS-related viruses by 644 human monoclonal antibodies

REGN-COV2, a Neutralizing Antibody Cocktail, in Outpatients with 647 Covid-19

Escape from neutralizing 650 antibodies by SARS-CoV-2 spike protein variants

Bamlanivimab does not neutralize two SARS-CoV-2 653 variants carrying E484K in vitro

Overview of the CCP4 suite and current 656 developments

Cryo-EM structure of the 2019-nCoV spike in the prefusion conformation

SARS-CoV-2 and bat RaTG13 spike glycoprotein structures inform on virus 662 evolution and furin-cleavage effects

Structural basis of a shared antibody response to SARS-CoV-2

Structural and functional ramifications of antigenic drift in 668 recent SARS-CoV-2 variants

Cryo-EM structures of MERS-CoV and SARS-CoV spike glycoproteins reveal the dynamic 671 receptor binding domains

Structural and Functional Analysis of the 674 D614G SARS-CoV-2 Spike Protein Variant

A pneumonia outbreak associated with a new coronavirus of probable bat 677 origin

A protective broadly cross-reactive human antibody defines a conserved 680 site of vulnerability on beta-coronavirus spikes

New tools for automated high-resolution cryo-EM structure

ELife 7, e42166

Potently neutralizing and protective human antibodies 686 against SARS-CoV-2

Rapid isolation and profiling of a diverse panel of 689 human monoclonal antibodies targeting the SARS-CoV-2 spike protein

All expression plasmids generated in this study for CoV proteins, CoV pseudoviruses, human 696 Fabs and IgGs are available upon request through a MTA

2010) followed by sequence matching and repeated cycles of phenix.refine and 900 manual building in Coot (v0

Buried surface area estimates were made using PDBePISA with a 1.4Å probe

Potential hydrogen bonds were assigned using a distance of <3.6A ̊ and an A-D

H angle of >90˚, and the maximum distance allowed for a van der Waals

Structure figures were made using UCSF Chimera v1

Human IgG (Sigma) was 946 loaded to an AHQ biosensor (ForteBio) to ∼1 nm, followed by sensor blocking with human 947

The self-association was performed at 1 µM solution concentration of 948 antibodies for 300s on an Octet Red96e system (Sartorius Corporation). The binding response 949 from the association step was subtracted from that of a reference IgG

Accelerated Stability Assay: Antibody samples at 1 mg/mL were kept at 40°C for 30 days in 951 10 mM Hepes and 150 mM sodium chloride

SEC-300 size-exclusion column (Sepax) on HPLC at Day 0, 5, 20, and 30. A long-term stability 953 slope (% aggregation/day) was calculated from the percent aggregated measured on the SEC-954 HPLC at each time-point

Twenty 956 microliters of 1mg/mL antibody sample was mixed with 10 µL of 20x SYPRO Orange 957 (ThermoFisher) in a 96-well PCR plate (ThermoFisher). The plate was scanned from 40℃ to 958 95℃ at a rate of 0.5℃/ 2 min in a CFX96 Real-Time PCR system

Convalescent plasma blocking assay

Spike trimer with C-terminal biotin at 5 µg/mL was first loaded on High Precision 962

Sartorius Corporation) for 75 min. Spike coated biosensor was 963 subsequently blocked with 10-fold diluted SARS-CoV-2 convalescent plasma for 15 min. BLI 964 signal from 10 mAbs binding to available surface of spike timer were measure by incubating 965

10 µg/mL of mAbs for 3 min. BLI signal was 966 compared to self-blocking of

/Ms) koff (1/s) KD (nM) kon (1/Ms) koff (1/s) KD (nM) kon (1/Ms) koff (1/s) KD (nM)

Vero E6 live virus neutralization assay (A) Neutralization activity of N-612-014 in 1028 5 separate experiments. (B) Comparison of neutralization activity of N-612-014, convalescent 1029 (C+) serum, and control human IgG with 30-min vs 24-hr antibody-virus incubation

Change in neutralization potency by incubation time

Cryo-EM data processing and validation (related to Figures 3 and 5)

Representative micrograph, 2D class averages, data processing workflow, and Gold Standard 1038 FSC plots for the final reconstructions of (A) N-612-017 -S 6P

D-F) Local resolution estimates calculated in cryoSPARC v3

) BLI kinetic analysis of 1044 N-612-004 and N-612-014 against S1 domain from WT and B

CHO CD EfficientFeed™ A, 6.2% Glucose, 6.9% FunctionMax™ Titer Enhancer, 3.5% L-766 Glutamine) was added at 10% of the culture volume on Days 3 and Day 8. 767

FectoPRO® transfection cell culture medium was centrifuged and filtered through a 0.22 µm 769 filter to remove cells and debris, then loaded onto a HiTrap™ MabSelect SuRe™ column (GE 770 Healthcare Life Sciences) on the AKTA Pure system pre-equilibrated with 10 mM Na Phosphate 771 and 150 mM NaCl at pH 7.0. After loading, the column was washed with 10 column volumes of 772 the same buffer. The protein was eluted with 100 mM sodium acetate, pH 3.6, then immediately 773 neutralized using 2 M Tris pH 8.0. The elution fractions were pooled and dialyzed into 10 mM 774Hepes and 150 mM sodium chloride at pH 7.4. 775