key: cord-262145-i29e3fge
authors: Huang, Kuan-Ying A.; Tan, Tiong Kit; Chen, Ting-Hua; Huang, Chung-Guei; Harvey, Ruth; Hussain, Saira; Chen, Cheng-Pin; Harding, Adam; Gilbert-Jaramillo, Javier; Liu, Xu; Knight, Michael; Schimanski, Lisa; Shih, Shin-Ru; Lin, Yi-Chun; Cheng, Chien-Yu; Cheng, Shu-Hsing; Huang, Yhu-Chering; Lin, Tzou-Yien; Jan, Jia-Tsrong; Ma, Che; James, William; Daniels, Rodney S.; McCauley, John W.; Rijal, Pramila; Townsend, Alain R.
title: Breadth and function of antibody response to acute SARS-CoV-2 infection in humans
date: 2020-10-19
journal: bioRxiv
DOI: 10.1101/2020.08.28.267526
sha: 
doc_id: 262145
cord_uid: i29e3fge

Serological and plasmablast responses and plasmablast-derived IgG monoclonal antibodies (MAbs) have been analysed in three COVID-19 patients with different clinical severities. Potent humoral responses were detected within 3 weeks of onset of illness in all patients and the serological titre was elicited soon after or concomitantly with peripheral plasmablast response. An average of 13.7% and 13.0% of plasmablast-derived MAbs were reactive with virus spike glycoprotein or nucleocapsid, respectively. A subset of anti-spike (10 of 32) and over half of anti-nucleocapsid (19 of 35) antibodies cross-reacted with other betacoronaviruses tested and harboured extensive somatic mutations, indicative of an expansion of memory B cells upon SARS-CoV-2 infection. Fourteen of 32 anti-spike MAbs, including five anti-RBD, three anti-non-RBD S1 and six anti-S2, neutralised wild-type SARS-CoV-2 in independent assays. Anti-RBD MAbs were further grouped into four cross-inhibiting clusters, of which six antibodies from three separate clusters blocked the binding of RBD to ACE2 and five were neutralising. All ACE2-blocking anti-RBD antibodies were isolated from two patients with prolonged fever, which is compatible with substantial ACE2-blocking response in their sera. At last, the identification of non-competing pairs of neutralising antibodies would offer potential templates for the development of prophylactic and therapeutic agents against SARS-CoV-2.

In late 2019, a novel coronavirus emerged and was identified as the cause of a cluster 54 of respiratory infection cases in Wuhan, China. It spread quickly around the world. In 55

March of 2020 a pandemic was declared by the World Health Organization, the virus 56 was formally named as Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-57

CoV-2) and the resulting disease was named COVID-19. As of 1 October 2020, there (Table 1) , suggesting the presence of conserved 137 epitopes on the spike glycoproteins of betacoronaviruses. 138

Each of 32 anti-spike glycoprotein MAbs was encoded by a unique set of heavy chain 139 VDJ and light chain VJ rearrangements in the variable domain (Supplemental Table  140 2). OC43 virus and three of these also cross-reacted on MERS (Table 1 ). All five cross-146 reactive anti-S2 antibodies had high rates of somatic mutation (25±5), indicating a 147 memory phenotype, and three of the five were neutralising to a moderate level (half 148 maximal effective concentration, EC 50 , 36-133.33 nM, Table 1 ). 149

The CDR3 length varied among anti-spike glycoprotein antibodies (Supplemental 150 Table 2 ). No significant differences were found between anti-S2 and anti-S1 or anti-151 RBD subsets. Among anti-S2 MAbs, a significantly longer heavy chain CDR3 length 152 was found in the cross-reactive group compared to the specific group (Cross-reactive 153 20±2 versus Specific 12±4, p= 0.02, two-tailed Mann-Whitney test; Figure 2c) , 154

indicating that a long CDR3 may play a role in antigen binding, which is also found in 155 several broadly reactive human MAbs against human immunodeficiency virus and 156 influenza virus (9, 10).

The binding activities of 10 anti-RBD MAbs were further characterised in detail. 158

Using MDCK-SIAT1 cells transduced to express the RBD and flow cytometry, 159 binding activities of the anti-RBD MAbs were shown to vary with 50% binding 160 concentration from 0.10 to 1.83 µg/ml (Supplemental Figure 2 ). The MAbs with 161 strong anti-RBD binding have a relatively long heavy chain CDR3 length (50% 162 binding concentration <0.5 µg/ml versus >0.5 µg/ml, p=0.03, two-tailed Mann-163 Whitney test; Supplemental Figure 3 The 32 anti-spike glycoprotein MAbs were systematically examined by plaque 173 reduction neutralisation (PRNT) assay for neutralisation of wild type SARS-CoV-2 174 virus (see methods; summarised in Table 1 ). A total of 14 neutralising antibodies 175 distributed between different regions of the spike glycoprotein were identified: 5 of 10 176 to RBD, 3 of 13 to S1 (non-RBD), 6 of 9 to S2. The EC 50 concentrations, as a 177 measure of potency, ranged from 0.05 to ~133 nM (8 ng/ml -~20 µg/ml). 178 (see methods): inhibition of virus replication was measured by quantitative PCR in the 182 supernatant bathing the infected cells. This results corroborated that anti-RBD FD  183   11A, anti-RBD FI 3A, anti-RBD FD 5D, anti-RBD EY 6A and anti-S2 EW 9C, as  184 crude culture supernatants, reduced the virus signal from ~56-to ~10,085-fold 185 (Supplemental Figure 5) . 186

Potent neutralising antibodies to the RBD of SARS-CoV-2 spike glycoprotein were 188 identified and we thus analyse the blockade of the ACE2-RBD interaction by anti-189 RBD antibodies in two assays ( Figure 3 , Table 1 The structure of VHH72-Fc bound to RBD is known (17) and its footprint on the 198 RBD does not overlap that of ACE2, so inhibition is thought to occur by steric 199 hindrance. 200

In the second assay, we employed MDCK-SIAT1 cells overexpressing full-length 201 human ACE2 as a transmembrane protein. Unlabelled antibodies or ACE2-Fc were 202 mixed in excess with biotinylated RBD, and binding of RBD was detected with 203

Streptavidin-HRP in ELISA (Figure 3b ). The results of this assay mostly mirrored 204 those of the first assay and confirmed that in this orientation anti-RBD neutralising 205 antibodies FD 11A and FD 5D competed in excess with soluble RBD for binding to 206 ACE2 ( Figure 3b ). In addition, anti-RBD neutralising antibody EY 6A competed with 207 RBD for ACE2 binding. The binding pattern of EY 6A is analogous to a previously 208 described antibody CR3022 (Table 1 ) (11). These two antibodies are known to bind to 209 the same region of RBD away from the ACE2 binding site, but they influence the 210 binding kinetics of RBD to ACE2, presumably through steric effects (15). 211

The ten anti-RBD MAbs were then divided into cross-inhibiting groups as described 213

for human MAbs to Ebola (18) by assessing competition of unlabelled antibodies at 214 10-fold (or greater) excess over a biotin labelled target antibody by ELISA. Included 215 as controls were the VHH72-Fc (17) and H11-H4-Fc (14) The ten antibodies formed four cross-inhibiting clusters (Table 2) , represented by 222 antibodies EY 6A (cluster 1, which included CR3022), FI 3A (cluster 2, which 223 included H11-H4), FD 11A (cluster 3, which included S309) and FJ 10B (cluster 4). 224

The strongest inhibitors of ACE2-Fc binding were in clusters 2 and 3 (Tables 1 and 2) . 225

Neutralising antibodies were detected in clusters 1, 2 and 3, with the strongest 226 antibodies FI 3A and FD 11A being in clusters 2 and 3 (Tables 1 and 2) . Table 2 ) and did not cross-react strongly with other betacoronaviruses 259 (Table 1) . FD 10A exhibits the most potent neutralising activity in the PRNT assay 260 and also completely inhibits SARS-CoV-2-induced cytopathic effect (see methods) at 261 8.33 nM. 262

Thirteen MAbs were defined that bound the S1 region and three, close to germline in 264 sequence, were neutralising. FJ 1C showed strong neutralisation (EC 50 55.5 nM), 265 whilst FD 11E (EC 50 70 nM) and FD 1E (EC 50 110 nM) were moderately neutralising 266 (Table 1) The 35 MAbs were evolved from 33 clonal groups defined by their heavy chain VDJ 279 and light chain VJ rearrangements (Supplemental Table 3 The presence of pre-existing immune memory to betacoronavirus that cross-react with 322 SARS-CoV-2 is supported by the accumulation of somatic mutations in the genes 323 encoding cross-reactive antibodies isolated from COVID-19 patients (Figures 2c and  324 2d, Supplemental Tables 2 and 3 ). This situation is reminiscent of re-exposure to 325 immunogenic epitopes shared by closely related viruses leading to induction of 326 broadly cross-reactive antibodies in patients infected with influenza, dengue or Zika 327 viruses (29-31). 328

The 32 MAbs that bound to the spike glycoprotein were systematically tested for 329 neutralisation (summarised in Table 1 ). Results established that neutralising epitopes 330 were present on the RBD, S1-NTD, S1-non NTD/RBD, and S2 regions of the spike CD3 neg CD19 pos CD20 neg CD27 hi CD38 hi IgG pos plasmablasts were gated and isolated in 425 chamber as single cells as previously described (53) .

Sorted single cells were used to produce human IgG MAbs as previously described 428 (53) 

Confluent monolayers of Vero E6 cells in 96-well plates were incubated with ~14 501 plaque forming units (PFU) of SARS CoV-2 (hCoV-19/England/02/2020, 502 EPI_ISL_407073) and antibodies in a 2-fold dilution series (triplicates) for 3 hours at 503 room temperature. Inoculum was then removed, and cells were overlaid with plaque 504 assay overlay. Cells were incubated at 37°C, 5% CO 2 for 24 hours prior to fixation 505 with 4% paraformaldehyde at 4°C for 30 minutes. Fixed cells were then 506 permeabilised with 0.2% Triton-X-100 and stained with a horseradish peroxidase 507 conjugated-antibody against virus protein for 1 hour at room temperature. TMB 508 substrate was then added to visualise virus plaques as described previously for 509 influenza virus (54). Convalescent serum from COVID-19 patients was used as a 510 control. 511

In brief, this rapid, high-throughput assay determines the concentration of antibody 513 that produces a 50% reduction in infectious focus-forming units of authentic SARS- Eagle's Medium containing 2% FBS), two-fold serially diluted MAbs in VGM 541 starting at 100 µg/ml were added to each duplicated well. The plates were 542 immediately transferred to a BSL-3 laboratory and 100 TCID 50 SARS-CoV-2 (hCoV-543 19/Taiwan/4/2020, EPI_ISL_411927) in VGM was added. The plates were further 544 incubated at 37°C with 5% CO 2 for three days and the cytopathic morphology of the 545 cells was recorded using an ImageXpress Nano Automated Cellular Imaging System.

Competitive binding assays were performed as described previously (18) 

Two assays were used to determine the blocking of binding of ACE2 to RBD by 567

MAbs. RBD was anchored on the plate in the first assay whereas ACE2 was anchored 568 for the second assay. The second ACE2 blocking assay was performed as described previously (14, 15) . B non-NTD S1 pos 1.45 0.11 0.13 0.12 0.12 0.14 110.00 EW 8B B non-NTD S1 -ve 1.61 0.11 0.13 0.11 0.11 0.12 -ve FD 11D B NTD pos 1.42 0.18 0.22 0.42 0.25 0.29 -ve FD 11C B non-NTD S1 pos 1.20 0.14 0.20 0.11 0.11 0.12 -ve FD 7D

B non-NTD S1 -ve 1.44 0.12 0.14 0.11 0.11 0.12 -ve FD 8B

B non-NTD S1 -ve 1.10 0.14 0.12 0.08 0.10 0.09 -ve FD 7C B NTD pos 1.90 0.15 0.15 0.13 0.12 0.13 -ve FG 12C A non-NTD S1 pos 1.74 0.14 0.11 0.08 0.08 0.10 -ve FN 8C

C non-NTD S1 -ve 0.54 0.16 0.16 0.11 0.09 0.12 -ve FD 5E

B non-NTD S1 pos 0.34 0.16 0.16 0.13 0.11 0.11 -ve EW 9B B non-NTD S1 -ve 0.22 0.17 0.11 0.09 0.10 0.09 -ve 

Deployment of convalescent plasma for the prevention and 650 treatment of COVID-19

Effect of Convalescent Plasma Therapy on Time to Clinical 652 Improvement in Patients With Severe and Life-threatening

Use of convalescent plasma therapy in SARS patients in Hong Kong

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

Breadth of concomitant immune responses prior to patient 659 recovery: a case report of non-severe COVID-19

Neutralizing antibodies in patients with severe acute respiratory 661 syndrome-associated coronavirus infection

Antibody responses to SARS-CoV-2 in patients with COVID-19

Serology characteristics of SARS-CoV-2 infection since exposure and 665 post symptom onset

Cross-neutralization of influenza A viruses mediated by a single 667 antibody loop

Structural insights on the role of antibodies in HIV-1 vaccine and 669 therapy

Human monoclonal antibody combination against SARS 12. Huo, J. et al. Neutralization of SARS-CoV-2 by Destruction of the Prefusion Spike

A highly conserved cryptic epitope in the receptor binding domains of

Neutralizing nanobodies bind SARS-CoV-2 spike RBD and block 677 interaction with ACE2

Structural basis for the neutralization of SARS-CoV-2 by an antibody 679 from a convalescent patient

Rugged Nanoscaffold To Enhance Plug-and-Display Vaccination

Structural Basis for Potent Neutralization of Betacoronaviruses by

Single-Domain Camelid Antibodies

Therapeutic Monoclonal Antibodies for Ebola Virus Infection Derived 687 from Vaccinated Humans

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

CoV antibody

Protective humoral immunity in SARS-CoV-2 infected pediatric 691 patients

Antibody responses to SARS-CoV-2 in patients of novel coronavirus 693 disease 2019

Serologic cross-reactivity of SARS-CoV-2 with endemic and seasonal

An Outbreak of Human Coronavirus OC43 Infection and 697

Serological Cross-reactivity with SARS Coronavirus

Recovery in tracheal organ cultures of novel viruses from patients with respiratory 701 disease

Epidemiology of Seasonal Coronaviruses: Establishing the

Context for the Emergence of Coronavirus Disease

Human Coronavirus OC43 Associated with Fatal

Development of a nucleocapsid-based human coronavirus 708 immunoassay and estimates of individuals exposed to coronavirus in a U.S. 709 metropolitan population

The dominance of human coronavirus OC43 and NL63 infections 711 in infants

The human immune response to Dengue virus is dominated by 713 highly cross-reactive antibodies endowed with neutralizing and enhancing activity

Zika virus activates de novo and cross-reactive memory B cell 716 responses in dengue-experienced donors

Broadly cross-reactive antibodies dominate the human B cell 718 response against 2009 pandemic H1N1 influenza virus infection

Receptor-binding domain of severe

Convergent antibody responses to SARS-CoV-2 in 725 convalescent individuals

Potent Neutralizing Antibodies against SARS-CoV-2 Identified by

Single-Cell Sequencing of Convalescent Patients' B Cells

Potent neutralizing antibodies against multiple epitopes on SARS-CoV-730 2 spike

A noncompeting pair of human neutralizing antibodies block COVID-732 19 virus binding to its receptor ACE2

Potent neutralizing antibodies from COVID-19 patients 734 define multiple targets of vulnerability

Studies in humanized mice and convalescent humans yield a SARS

CoV-2 antibody cocktail

Isolation of potent SARS-CoV-2 neutralizing antibodies and 738 protection from disease in a small animal model

Human neutralizing antibodies elicited by SARS-CoV-2 infection

A human monoclonal antibody blocking SARS-CoV-2 infection

Human monoclonal antibodies block the binding of SARS-CoV-2

spike protein to angiotensin converting enzyme 2 receptor

Receptor-binding domain as a target for 747 developing SARS vaccines

The SARS-CoV-2 receptor-binding domain elicits a potent 749 neutralizing response without antibody-dependent enhancement

A vaccine targeting the RBD of the S protein of SARS-CoV-2 induces 752 protective immunity

Structural basis of receptor recognition by SARS-CoV-2

Structure of the SARS-CoV-2 spike receptor-binding domain bound to 756 the ACE2 receptor

Structural and Functional Basis of SARS-CoV-2 Entry by Using 758

SARS-CoV-2 neutralizing antibody structures inform therapeutic 760 strategies

Complete mapping of mutations to the SARS-CoV-2 spike 762 receptor-binding domain that escape antibody recognition

Antibody cocktail to SARS-CoV-2 spike protein prevents rapid 765 mutational escape seen with individual antibodies

A neutralizing human antibody binds to the N-terminal domain of the

Structure-function analysis of neutralizing antibodies to H7N9 769 influenza from naturally infected humans

Optimisation of a micro-neutralisation assay and its application in 771 antigenic characterisation of influenza viruses. Influenza Other Respir Viruses 9

Isolation and rapid sharing of the 2019 novel coronavirus

from the first patient diagnosed with COVID-19 in Australia

The data are presented 795 as specificity, number of antibodies, and the percentage of total antibodies isolated 796 from each patient. (b) The binding activity of anti-SARS-CoV-2 MAbs with spike 797 glycoprotein, RBD and the S2 subunit in ELISA. Anti-influenza H3 MAb BS-1A and 798 anti-SARS RBD CR3022 were included as controls. Each experiment was repeated 799 twice. The OD 450 values are presented as mean ± standard error of the mean. Panels 800 (c) and (d) show numbers of variable domain mutations in MAb genes and variation

Antibodies that 802 strongly cross-react with at least one betacoronavirus (SARS or MERS or OC43) 803 were defined as cross-reactive MAbs. CDR3 length and mutation numbers are 804 presented as mean ± standard error of the mean (anti-S2, specific

reactive, n=5; anti-N, specific, n=16 versus cross-reactive, n=19). The two-tailed

test was performed to compare the mutations between two groups

D, =Day ; ns, non-significant

hinge and Fc region of human IgG1 and ACE2-Fc were included as controls

The RBD was colored in 816 green. The epitopes recognized by EY 6A, CR3022 and VHH72 (cluster 1 MAb) (11, 817 15, 17) were colored in magenta. The epitopes recognized by ACE2 and H11-H4 818 (cluster 2 MAb) (14) were overlapping and colored in blue and light blue. The 819 epitopes recognized by S309

Convalescent sera were analysed in the ACE2-blocking (ACE2 anchored) assay

Anti-RBD antibody FD 11A and anti-822 influenza H3 antibody BS 1A were included as controls. Data are presented as mean ± 823 standard error of the mean

ACE2-blocking activity of anti-RBD antibody compared to ACE2-Fc (see methods): +, partial; ++

Abbreviations: IFA, immunofluorescence; RBD, receptor-binding domain; PRNT, plaque reduction 835 neutralisation assay