key: cord-0960830-b3ktsj47
authors: Shiakolas, Andrea R.; Kramer, Kevin J.; Wrapp, Daniel; Richardson, Simone I.; Schäfer, Alexandra; Wall, Steven; Wang, Nianshuang; Janowska, Katarzyna; Pilewski, Kelsey A.; Venkat, Rohit; Parks, Rob; Manamela, Nelia P.; Raju, Nagarajan; Fechter, Emilee Friedman; Holt, Clinton M.; Suryadevara, Naveenchandra; Chen, Rita E.; Martinez, David R.; Nargi, Rachel S.; Sutton, Rachel E.; Ledgerwood, Julie E.; Graham, Barney S.; Diamond, Michael S.; Haynes, Barton F.; Acharya, Priyamvada; Carnahan, Robert H.; Crowe, James E.; Baric, Ralph S.; Morris, Lynn; McLellan, Jason S.; Georgiev, Ivelin S.
title: Cross-reactive coronavirus antibodies with diverse epitope specificities and extra-neutralization functions
date: 2020-12-20
journal: bioRxiv
DOI: 10.1101/2020.12.20.414748
sha: 5c1fa758f72e2472931e2115da89640e505715e3
doc_id: 960830
cord_uid: b3ktsj47

The continual emergence of novel coronavirus (CoV) strains, like SARS-CoV-2, highlights the critical need for broadly reactive therapeutics and vaccines against this family of viruses. Coronavirus spike (S) proteins share common structural motifs that could be vulnerable to cross-reactive antibody responses. To study this phenomenon in human coronavirus infection, we applied a high-throughput sequencing method called LIBRA-seq (Linking B cell receptor to antigen specificity through sequencing) to a SARS-CoV-1 convalescent donor sample. We identified and characterized a panel of six monoclonal antibodies that cross-reacted with S proteins from the highly pathogenic SARS-CoV-1 and SARS-CoV-2 and demonstrated a spectrum of reactivity against other coronaviruses. Epitope mapping revealed that these antibodies recognized multiple epitopes on SARS-CoV-2 S, including the receptor binding domain (RBD), N-terminal domain (NTD), and S2 subunit. Functional characterization demonstrated that the antibodies mediated a variety of Fc effector functions in vitro and mitigated pathological burden in vivo. The identification of cross-reactive epitopes recognized by functional antibodies expands the repertoire of targets for pan-coronavirus vaccine design strategies that may be useful for preventing potential future coronavirus outbreaks.

). After bioinformatic processing, we recovered 2625 cells with paired heavy/light 120 chain sequences and antigen reactivity information (Supplemental Figure 1B) . Overall, LIBRA-121 seq enabled rapid screening of PBMCs from a patient sample, with recovery of paired 122 heavy/light chain sequences and antigen reactivity for thousands of single B cells. 123 124

With a goal of identifying antibodies that were cross-reactive to multiple coronavirus S proteins, 126

we prioritized lead candidates based on their sequence features and LIBRA-seq scores. We 127 selected 15 antibody candidates that exhibited diverse sequence features and utilized a number 

To elucidate the epitopes targeted by the cross-reactive antibodies, we performed binding 144 assays to various structural domains of S as well as binding-competition experiments. First, we 7 assessed antibody binding to the S1 and S2 subdomains of SARS-CoV-2. Antibodies 46472-1, 146 46472-2, 46472-3, and 46472-4 bound to the S2 domain, whereas 46472-6 and 46472-12 147 recognized the S1 domain but targeted different epitopes, the NTD and RBD, respectively 148 (Figure 2A-C, Supplemental Figure 2A-B) . Although 46472-12 bound to the RBD, it did not 149 compete with ACE2 for binding to SARS-CoV-2 S (Supplemental Figure 2C) . To determine 150 whether the antibodies targeted overlapping or distinct epitopes, we performed competition 151 ELISA experiments and found that the S2-directed antibodies 46472-1, 46472-2, and 46472-4 152 competed for binding to S ( Figure 2D ). This pattern was observed for both SARS-CoV-2 and 153 SARS-CoV-1 S. Of note, this competition group did not include S2-directed antibody 46472-3, 154 revealing the identification of multiple cross-reactive epitope targets on S2 ( Figure 2D ). Further, 155 binding assays with glycan knockout mutants and mannose competition experiments revealed 156 no notable glycan dependence for antibody reactivity to S (Supplemental Figure 2D-E) . Lastly, 157

we measured antibody autoreactivity, and found that with the exception of 46472-6 binding to 158 Jo-1, none of the antibodies showed autoreactivity against the tested antigens ( Figure 2E) . 159

Together, these data suggest that these cross-reactive antibodies are coronavirus-specific and 160 target multiple, diverse epitopes on the S protein ( Figure 2F) . Here, we described a set of cross-reactive Betacoronavirus antibodies isolated from a 208 convalescent SARS-CoV-1 donor. The antibodies targeted diverse epitopes on S, including the 209 S2 subdomain as well as the RBD and NTD on S1, and were shown to be functional in vitro. previously mentioned. S proteins were purified using StrepTrap HP columns and RBD 337 constructs were purified over protein A resin, respectively. Each resulting protein was further 338 purified to homogeneity by size-exclusion chromatography on a Superose 6 10/300 GL column. 339 340 SARS-CoV-2 S1, SARS-CoV-2 S1 D614G, SARS-CoV-2 S2, and SARS-CoV-2 NTD truncated 341 proteins were purchased from the commercial vendor, Sino Biological. 342

We used oligos that possess 15 bp antigen barcode, a sequence capable of annealing to the 345 template switch oligo that is part of the 10X bead-delivered oligos, and contain truncated We utilized and modified our previously described pipeline to use paired-end FASTQ files of 392 oligo libraries as input, process and annotate reads for cell barcode, UMI, and antigen barcode, 393 and generate a cell barcode -antigen barcode UMI count matrix 24 . BCR contigs were processed 394 using Cell Ranger (10X Genomics) using GRCh38 as reference. Antigen barcode libraries were 395 also processed using Cell Ranger (10X Genomics). The overlapping cell barcodes between the 396 two libraries were used as the basis of the subsequent analysis. We removed cell barcodes that 397 had only non-functional heavy chain sequences as well as cells with multiple functional heavy 398 chain sequences and/or multiple functional light chain sequences, reasoning that these may be 399 multiplets. Additionally, we aligned the BCR contigs (filtered_contigs.fasta file output by Cell Quest was parsed using ChangeO 39 and merged with an antigen barcode UMI count matrix. 402

Finally, we determined the LIBRA-seq score for each antigen in the library for every cell as 403 previously described 24 To assess antibody binding, soluble protein was plated at 2 μg/ml overnight at 4°C. The next 447 day, plates were washed three times with PBS supplemented with 0.05% Tween-20 (PBS-T) 448 and coated with 5% milk powder in PBS-T. Plates were incubated for one hour at room 449 temperature and then washed three times with PBS-T. Primary antibodies were diluted in 1% 450 milk in PBS-T, starting at 10 μg/ml with a serial 1:5 dilution and then added to the plate. The 451

plates were incubated at room temperature for one hour and then washed three times in PBS-T. dilution in 1% milk in PBS-T to the plates, which were incubated for one hour at room 454 temperature. Goat anti-mouse secondary was used for SARS-CoV-1 specific control antibody 455

adding TMB substrate to each well. The plates were incubated at room temperature for ten 457 minutes, and then 1N sulfuric acid was added to stop the reaction. Plates were read at 450 nm. Antibody-virus complexes were incubated at 37C with 5% CO2 for 1 hour. Following incubation, 523 growth media was removed and virus-antibody dilution complexes were added to the cells in 524 duplicate. Virus-only controls and cell-only controls were included in each neutralization assay 525 plate. Following infection, plates were incubated at 37C with 5% CO2 for 48 hours. After the 48 526 hour incubation, cells were lysed and luciferase activity was measured via Nano-Glo Luciferase 527

Assay System (Promega) according to the manufacturer specifications. SARS-CoV and SARS-528

CoV-2 neutralization titers were defined as the sample dilution at which a 50% reduction in RLU 529 was observed relative to the average of the virus control wells. 530

His-tagged SARS-CoV-2 RBD-SD1 was immobilized to a NiNTA sensorchip to a level of ~150 533 RUs using a Biacore X100. Serial dilutions of purified Fab 46472-12 were evaluated for binding, 534 ranging in concentration from 1 to 0.25 μM. The resulting data were fit to a 1:1 binding model 535 using Biacore Evaluation Software. 536

Antibody-dependent cellular phagocytosis (ADCP) was performed using biotinylated SARS-539

CoV-2 or SARS-CoV-1 S coated fluorescent neutravidin beads as previously described 43 . 540

Briefly, beads were incubated for two hours with antibodies at a starting concentration of 541 50μg/ml and titrated five fold. CR3022 was used as a positive control while Palivizumab was 542 used as a negative control. Antibodies and beads were incubated with THP-1 cells overnight, 543 fixed and interrogated on the FACSAria II. Phagocytosis score was calculated as the 544 percentage of THP-1 cells that engulfed fluorescent beads multiplied by the geometric mean 545 fluorescence intensity of the population in the FITC channel less the no antibody control. 546 547 Antibody-dependent Cellular Trogocytosis (ADCT) 548 ADCT was performed as described in and modified from a previously described study 26 . 549

HEK293T-ACE2 expressing cells were pulsed with SARS-CoV-2 S protein (10μg/ml) for 75 550 minutes or HEK293T cells transfected with a SARS-CoV-2 spike pcDNA vector were surface 551 biotinylated with EZ-Link Sulfo-NHS-LC-Biotin as recommended by the manufacturer. Fifty-552 thousand cells per well were incubated with antibody for 30 minutes starting at 25μg/ml and 553 titrated 5 fold. CR3022 was used as a positive control with Palivizumab as a negative. Following 554 a RPMI media wash, these were then incubated with carboxyfluorescein succinimidyl ester 555 EDTA/PBS followed by PBS. Cells were then stained for biotin using Streptavidin-PE and read 557 on a FACSAria II. Trogocytosis score was determined as the proportion of CFSE positive THP-1 558 cells also positive for streptavidin-PE less the no antibody control. 559 560

Antibody-dependent complement deposition was performed as previously described 44 Mice were monitored daily for weight loss, morbidity, and mortality, and after four days one lung 575 lobe was taken for pathological analysis and the other lobe was processed for qPCR and viral 576 load determination as previously described 28 displayed as a heatmap of the AUC analysis calculated from the data in Figure S2A and (B) for 638 SARS-CoV-2 S1 reactive antibodies, ELISA binding data against the RBD and NTD are 639 displayed as a heatmap of the AUC analysis calculated from the data in Figure S2B . Area under the curve of the phagocytosis score is shown, calculated from data in Figure S3D . 

Genomic characterisation and epidemiology of 2019 novel coronavirus: 712 implications for virus origins and receptor binding

Recombination, Reservoirs, and the Modular

Mechanisms of Coronavirus Cross-Species Transmission

Coronavirus Spike Protein Is a Class I Virus Fusion Protein: Structural and Functional 717 Characterization of the Fusion Core Complex. JVI

Structural insights into coronavirus entry

Cryo-EM structure of the 2019-nCoV spike in the 722 prefusion conformation

SARS-CoV-2 vaccines in development

Structure, Function, and Evolution of Coronavirus Spike Proteins

Potent 730 neutralizing antibodies from COVID-19 patients define multiple targets of vulnerability

Rapid isolation and profiling of 737 a diverse panel of human monoclonal antibodies targeting the SARS-CoV-2 spike 738 protein

Cross-neutralization of SARS-CoV-2 by 741 a human monoclonal SARS-CoV antibody

Isolation of potent SARS-CoV-2 neutralizing 744 antibodies and protection from disease in a small animal model

Cross-Neutralization of a SARS

CoV-2 Antibody to a Functionally Conserved Site Is Mediated by Avidity

Broad neutralization of

SARS-related viruses by human monoclonal antibodies

Cross-reactive Antibody Response 754 between SARS-CoV-2 and SARS-CoV Infections

Dissecting antibody-mediated protection against SARS

Phagocytic cells contribute to the 762 antibody-mediated elimination of pulmonary-infected SARS coronavirus

Antibody 766 potency, effector function, and combinations in protection and therapy for SARS-CoV-2 767 infection in vivo

Distinct Early Serological Signatures Track 770 with SARS-CoV-2 Survival

Evolution of Early SARS-CoV-2 and 773

Characterization of spike glycoprotein of SARS-CoV-2 on virus entry and its 776 immune cross-reactivity with SARS-CoV

Mapping of B Cell Receptor Sequences to Antigen Specificity

A highly conserved cryptic epitope in the receptor binding domains of SARS

Measuring the ability 784 of HIV-specific antibodies to mediate trogocytosis

A mouse-adapted model of

SARS-CoV-2 to test COVID-19 countermeasures

A Mouse-Adapted SARS-CoV

Induces Acute Lung Injury and Mortality in Standard Laboratory Mice

Broadly Neutralizing Anti-HIV-1 Antibodies Require Fc Effector Functions for

Broadly neutralizing 800 anti-influenza antibodies require Fc receptor engagement for in vivo protection

Beyond binding: antibody 803 effector functions in infectious diseases

Swine acute diarrhea 806 syndrome coronavirus replication in primary human cells reveals potential susceptibility 807 to infection

A SARS-like cluster 813 of circulating bat coronaviruses shows potential for human emergence

From SARS to MERS, Thrusting Coronaviruses into the Spotlight

Immunogenicity and 820 structures of a rationally designed prefusion MERS-CoV spike antigen

IMGT/V-QUEST and IMGT/HighV-825 QUEST for NGS

Change-O: a toolkit for analyzing large-scale B cell 828 immunoglobulin repertoire sequencing data

Enhancing dengue virus maturation using a stable furin over-831 expressing cell line

Reverse genetics with a full-length 838 infectious cDNA of severe acute respiratory syndrome coronavirus

A robust, high-throughput 842 assay to determine the phagocytic activity of clinical antibody samples

A high-throughput, bead-based, antigen-specific assay to assess the 846 ability of antibodies to induce complement activation

The EMBL-EBI search and 850 sequence analysis tools APIs in 2019

Amino acid substitution matrices from protein 852 blocks

n=4 or 5