key: cord-0809627-jd9ako6c
authors: Kang, Sisi; Yang, Mei; He, Suhua; Wang, Yueming; Chen, Xiaoxue; Chen, Yao-Qing; Hong, Zhongsi; Liu, Jing; Jiang, Guanmin; Chen, Qiuyue; Zhou, Ziliang; Zhou, Zhechong; Huang, Zhaoxia; Huang, Xi; He, Huanhuan; Zheng, Weihong; Liao, Hua-Xin; Xiao, Fei; Shan, Hong; Chen, Shoudeng
title: A COVID-19 antibody curbs SARS-CoV-2 nucleocapsid protein-induced complement hyper-activation
date: 2020-09-11
journal: bioRxiv
DOI: 10.1101/2020.09.10.292318
sha: a844b369a861788ff713294f6e26f893ad9a8115
doc_id: 809627
cord_uid: jd9ako6c

Although human antibodies elicited by severe acute respiratory distress syndrome coronavirus-2 (SARS-CoV-2) nucleocapsid (N) protein are profoundly boosted upon infection, little is known about the function of N-directed antibodies. Herein, we isolated and profiled a panel of 32 N protein-specific monoclonal antibodies (mAb) from a quick recovery coronavirus disease-19 (COVID-19) convalescent, who had dominant antibody responses to SARS-CoV-2 N protein rather than to Spike protein. The complex structure of N protein RNA binding domain with the highest binding affinity mAb nCoV396 reveals the epitopes and antigen’s allosteric changes. Functionally, a virus-free complement hyper-activation analysis demonstrates that nCoV396 specifically compromises N protein-induced complement hyper-activation, a risk factor for morbidity and mortality in COVID-19, thus paving the way for functional anti-N mAbs identification. One Sentence Summary B cell profiling, structural determination, and protease activity assays identify a functional antibody to N protein.

mechanics despite the severity of hypoxemia (7). It is reported that complement-mediated 48 thrombotic microvascular injury in the lung may contribute to atypical ARDS features of 19, accompanied by extensive deposition of the alternative pathway (AP) and lectin pathway (LP) 50 complement components(8). Indeed, complement activation is found in multiple organs of severe 51 COVID-19 patients in several other studies(9, 10), as well as in patients with severe acute 52 respiratory distress syndrome (SARS) (11, 12) . A recent retrospective observational study of 53 11,116 patients revealed that complement disorder associated with morbidity and mortality of 54 . 55 Although systemic activation of complement plays a pivotal role in protective immunity against 56 pathogens, hyper-activation of complement may lead to collateral tissue injury. Severe acute 57 respiratory distress syndrome-associated coronavirus-2 (SARS-CoV-2) nucleocapsid (N) protein 58 is a highly immunopathogenic and multifunctional viral protein (14) (15) (16) (17) (18) (19) , which elicited high titers 59 of binding antibodies in humoral immune responses (20) (21) (22) . A recent preprint study found that 60 SARS-CoV-2 N protein bound to LP complement components MASP-2 (Mannan binding lectin-61 associated serine protease-2), and resulted in complement hyper-activation and aggravated 62 inflammatory lung injury (15) . Several studies have reported in isolations of human monoclonal 63 antibodies (mAbs) targeting SARS-CoV-2 Spike (S) protein, shedding the light of developing 64 4 therapeutic interventions of [23] [24] [25] [26] [27] . However, little is known about the potential 65 therapeutic applications of N protein-targeting mAbs in the convalescent B cell repertoire. Herein, 66 we report a human mAb derived from COVID-19 convalescent, with specific targeting to SARS-67 CoV-2 N protein and functionally compromising complement hyper-activation ex vivo.

Isolation of N protein-directed mAbs 69 To profile antibody response to SARS-CoV-2 N protein in early recovered patients, we collected 70 six convalescent blood samples at seven to 25 days after the onset of the disease symptoms. All 71 patients are recovered from COVID-19 during the outbreak in Zhuhai, Guangdong Province, 72 China, with age ranging from 23 to 66 years old ( Table S1 ). The SARS-CoV-2 nasal swabs reverse 73 transcription-polymerase chain reaction (RT-PCR) tests were confirmed being negative at the 74 points of blood collection for all of these six COVID-19 patients. Plasma samples and peripheral 75 blood mononuclear cells (PBMC) were isolated for serological analysis and antibody isolation. 76 Serum antibody titers to SARS-CoV-2 S and N proteins were measured by enzyme-linked 77 immunosorbent assays (ELISA) (Fig. 1A , B, Table S1 ). Serologic analysis demonstrated that 78 serum antibody titers to the N protein were substantially higher than to the S protein in most of the 79 patients. For example, ZD004 and ZD006 had only minimal levels of antibody response to the S 80 protein, while they had much higher antibody titers to the N protein. To be noted, the time from 81 the disease onset to complete recovery from clinical symptoms of COVID19 patient ZD006 was 82 only 9 days (Table S1) . 83 To take advantage of patient ZD006 that was still in the early recovery phase with high possibility 84 of high percentage of antigen-specific plasma cells, single plasma cells ( Fig. 1C) with phenotype 85 of CD3 -/CD14 -/CD16 -/CD235a -/CD19 + /CD20 low-neg /CD27 hi /CD38 hi , as well as antigen-specific 86 memory B cells with phenotype of CD19 + /CD27 + (Fig. 1D) were sorted from PBMC of patient 87 5 ZD006 by fluorescence activating cell sorter (FACS). To ensure an unbiased assessment, the 88 sorting of antigen-specific memory B cells was carried out with combined probes of both 89 fluorophore-labeled S and N recombinant proteins. Variable region of immunoglobulin (Ig) heavy-90 and light-chain gene segment (V H and V L ) pairs from the sorted single cells were amplified by RT-91 PCR, sequenced, annotated and expressed as recombinant mAbs using the methods as described 92 previously(28). Recombinant mAbs were screened against SARS-COV-2 S and N proteins. In 93 total, we identified 32 mAbs reacted with SARS-COV2 N protein including 20 mAbs from plasma 94 cells, and 12 mAbs from memory B cells (Table S2) . We found that IgG1 is the predominant 95 isotype at 46.9% followed by IgG3 (25.0%), IgA (18.8%), IgG2 (6.3%) and IgM (3.1) (Fig.1E ). 96 V H gene family usage in SARs-COV2 N protein-reactive antibodies was 18.8% V H1 , 62.5% V H3 , 97 9.4% V H4 , 6.2% V H5 and 3.1% V H7 , respectively (Fig. 1F) , which was similar to the distribution 98 of V H families collected in the NCBI database. Nine of 32 SARS-COV-2 N protein-reactive 99 antibodies had no mutation from their germline V H and V H gene segments (Fig. 1F, Table S2 ). 100 Average mutation frequency of the remaining mutated antibodies was 5.3 % (+/-3.6%) in V H and 101 3.5% (+/-2.7%) in V L . 102 In consistent with the lower serum antibody titers to SARS-COV-2 S protein, we identified only 103 eight SARS-COV-2 S protein-reactive mAbs including 5 antibodies from plasma cells and three 104 antibodies from memory B cells. V H gene segment of the S protein-reactive antibodies had either 105 no mutation (6/8) or minimal mutation (1/300) (Fig.1G ). There were no significant differences in 106 complementarity-determining region 3 (CDR3) length in amino acid residues between the N- Fig. 2A) . Among 32 mAbs binding to NFL; 13 antibodies bound to N-NTD; one 122 antibody bound to N-CTD (Fig. 2B) . Total of nine antibodies including one antibody (nCoV400) 123 recognizing N-CTD, seven mAbs binding N-NTD (nCoV396, nCoV416, nCoV424, nCoV425, 124 nCoV433, nCoV454, nCoV457) and one mAb (nCoV402) binding only to NFL but not to the other 125 variant N proteins were chosen as representatives for further study. Purified antibodies were 126 confirmed to bind the NFL protein by ELISA (Fig. 2C) . Affinity of these antibodies to the NFL 127 protein was measured by surface plasmon resonance (SPR) (Fig. 2D) . In an effort to further 128 characterize the function and structure relationship, three antibodies nCoV396, nCOV416 and 129 nCOV457 were selected for production of recombinant Fab antibodies based on their unique 130 characters. MAb nCoV396 has V H mutation frequency of 2.8%, but high binding affinity with KD 131 of 1.02 nM (Fig. 2D) to the N protein. MAbs nCOV416 and nCOV457 have high V H mutation at 132 7 11.1% and 8.7%, respectively, and have binding affinity to N protein with KD of 7.26 nM and 133 12.6 nM (Fig.2D, Table S3 ).

Complex structure of mAb with N-NTD 135 To investigate the molecular interaction mechanism of mAb nCoV396 with N protein, we next 136 solved the complex structure of SARS-CoV-2 N protein NTD (N-NTD) with nCoV396 Fab 137 fragments (nCoV396Fab) at 2.1 Å resolution by X-ray crystallography. The final structure is fitted 138 with visible electron density spanning residues 49-173 (SARS-CoV-2 N-NTD), 1-220 139 (nCoV396Fab, the heavy chain of Fab fragments), and 1-213 (nCoV396Fab, the light chain of Fab 140 fragments, except residues ranged 136-141), respectively. The complete statistics for data 141 collection, phasing, and refinement are presented in Table S4 . 142 With the help of the high-resolution structure, we were able to designate all complementarity 143 determining regions (CDRs) in the nCoV396Fab as L-CDR1 (light chain CDR1, residues 23-32), with unambiguous electron density map (Fig. 3A, Fig. S1A ). 149 The interacting CDRs pinch the C-terminal tail of SARS-CoV-2 N-NTD (residues range from 159 150 to 172), with extensive binding contacts of 1079 Å 2 burying surface area (Table S5) . Light chain 151 L-CDR1 and L-CDR3 of nCoV396Fab interact with residues ranging from 159-163 of N-NTD via 152 numerous hydrophilic and hydrophobic contacts (Fig. 3B, Fig. S1B ). Of note, SARS-CoV-2 N-153 NTD residue Q163 is recognized by L-CDR3 residue T95 via a hydrogen bond, simultaneously 154 stacking with L-CDR3 residue W96 and L-CDR1 residue Y31 (Fig. 3C) . Besides, a network of 155 8 interactions from heavy chain H-CDR2, H-CDR3 of nCoV396Fab to residues 165-172 of N-NTD 156 suggests that SARS-CoV-2 N-NTD conservative residue K169 has a critical role in nCoV396 157 antibody binding. The K169 is recognized via hydrogen bonds with residues E99 d-carboxyl group 158 and T100, D102, S105 main-chain carbonyl groups inside the H-CDR3 of nCoV396Fab (Fig. 3D) . 159 Besides, SARS-CoV-2 N-NTD L167 also interacts with I33, V50, N57, and A59 of H-CDR1 and 160 H-CDR2 of nCoV396Fab through hydrophobic interactions (Fig. 3E) . Interestingly, all three 161 residues (Q163, L167, and K169) of SARS-CoV-2 N-NTD are relatively conserved in the highly 162 pathogenic betacoronavirus N protein (Fig. S2B) , which implicated that the nCoV396 may cross-163 interact with SARS-CoV N protein or MERS-CoV N protein. Indeed, the binding affinities 164 measured by SPR analysis demonstrate that nCoV396 interacts to SARS-CoV N protein and 165 MERS-CoV N protein with KD of 7.4 nM (Fig. S2B, C) . 166 To discover the conformational changes between the SARS-CoV-2 N-NTD apo-state with the 167 antibody-bound state, we next superimposed the complex structure with the N-NTD structure 168 (PDB:6M3M)(17). The superimposition result suggests that the C-terminal tail of SARS-CoV-2 169 N-NTD unfold from the basic palm region upon the nCoV396Fab binding (Fig. 3F) , which likely 170 contributes to allosteric regulation of normal full-length N protein's function. Additionally, 171 nCoV396Fab binding results in a 7.4 Å movement of the b-finger region outward from the RNA 172 binding pocket, which may enlarge the RNA binding pocket of the N protein (Fig. 3F) . 173 To sum up, our crystal structural data demonstrated that the human mAb nCoV396 recognizes the and found to be linear (Fig. 4C) . Therefore, the additions of SARS-CoV-2 N protein do not change 203 the single substrate binding site characterization of the enzymatic reactions. To assess the 204 suppression ability of nCoV396 to the SARS-CoV-2 N protein-induced complement hyper-205 activation function, we next conducted the complement hyper-activation analysis in serial N 206 protein: nCoV396 ratios. As shown in Fig. 4D , the addition of N protein elevates V max value up to 207 40-folds (1:0 ratio), whereas the additions of antibody nCoV396 decline the V max in a dose-208 depended manner (Table S9) 

The epidemiological characteristics 271 of an outbreak of 2019 novel coronavirus diseases (COVID-19) in China

Treatment of Coronavirus Disease 2019 (COVID-19): A 275 Review

Abnormal coagulation parameters are associated 277 with poor prognosis in patients with novel coronavirus pneumonia

Hospitalized Patients With 2019 Novel Coronavirus-Infected Pneumonia in

A Novel Coronavirus from Patients with Pneumonia in China

Clinical course and risk 289 factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective 290 cohort study

Acute Respiratory Distress Syndrome

Complement associated microvascular injury and thrombosis in the 296 pathogenesis of severe COVID-19 infection: A report of five cases

Complement activation in patients with 300 COVID-19: A novel therapeutic target

The case of complement activation in COVID-19 302 multiorgan impact

Serum proteomic fingerprints of adult patients 305 with severe acute respiratory syndrome

Plasma proteome of severe acute respiratory syndrome 308 analyzed by two-dimensional gel electrophoresis and mass spectrometry

Immune complement 312 and coagulation dysfunction in adverse outcomes of SARS-CoV-2 infection

The coronavirus nucleocapsid is a multifunctional 315 protein

The origin, transmission and clinical therapies on coronavirus disease 2019 323 (COVID-19) outbreak -an update on the status

Crystal structure of SARS-CoV-2 nucleocapsid 326 protein RNA binding domain reveals potential unique drug targeting sites

The ORF6, ORF8 and 329 nucleocapsid proteins of SARS-CoV-2 inhibit type I interferon signaling pathway

Architecture and self-assembly of the 332 SARS-CoV-2 nucleocapsid protein

A neutralizing human antibody binds to the N-terminal 336 domain of the Spike protein of SARS-CoV-2

Antibodies to SARS-CoV-2 and their potential for therapeutic 338 passive immunization

343 Longitudinal Isolation of Potent Near-Germline SARS-CoV-2-Neutralizing Antibodies 344 from COVID-19 Patients

Potently neutralizing and protective human antibodies 352 against SARS-CoV-2

A noncompeting pair of human neutralizing antibodies block 356 COVID-19 virus binding to its receptor ACE2

A human monoclonal antibody 359 blocking SARS-CoV-2 infection

Human neutralizing antibodies elicited by SARS-CoV-2 363 infection

Potent Neutralizing Antibodies against SARS-CoV-2 Identified by High-368

Throughput Single-Cell Sequencing of Convalescent Patients' B Cells

High-throughput isolation of 373 immunoglobulin genes from single human B cells and expression as monoclonal antibodies

Multiple domains of MASP-2, an initiating complement protease, are required for 377 interaction with its substrate C4

Specificity, cross-reactivity, 382 and function of antibodies elicited by Zika virus infection

Delineating antibody recognition against 386 Zika virus during natural infection

Complement regulators and inhibitory proteins

COVID-19: Complement, Coagulation, and 390 Collateral Damage

Study on interaction between SARS-CoV N and MAP19. Xi 392 bao yu fen zi mian yi xue za zhi = Chinese journal of cellular and molecular immunology 393 25

for 398 technical assistants of mAbs isolation, production and characterization

Emerging Prevention Products

contributed to 406 protein purification and crystallization, in vitro protein-protein interaction analysis, and 407 complement activation analysis. Y. W. contributed to mAbs isolation, in vitro protein-protein 408 interaction analysis

We thank the staffs of the BL18U/19U/17U beamlines at SSRF for their help 396 with the X-ray diffraction data screening and collections. We thank Junlang Liang, Tong Liu, Nan 397

The authors declare no conflict of interest. are shown. P values: *P < 0.05; **P < 0.01; "-" means that the kinetics did not conform to 462 Michaelis-Menten kinetics. 463