Longitudinal isolation of potent near-germline SARS-CoV-2 ... · 108 points spanning 8 to 69 days post diagnosis (Figure 2A, Table S1). Across the 109 different individuals, EC 50

Longitudinal isolation of potent near-germline SARS-CoV-2-neutralizing 1

antibodies from COVID-19 patients 2

Christoph Kreer1,*, Matthias Zehner1,*, Timm Weber1, Cornelius Rohde2,3, Sandro 3

Halwe2,3, Meryem S. Ercanoglu1, Lutz Gieselmann1, Michael Korenkov1, Henning 4

Gruell1,4, Philipp Schommers1,4,5, Kanika Vanshylla1, Veronica Di Cristanziano6, 5

Hanna Janicki1, Reinhild Brinker7,8, Artem Ashurov1, Verena Krähling2,3, Alexandra 6

Kupke2,3, Hadas Cohen-Dvashi9, Manuel Koch10,11, Simone Lederer12, Nico 7

Pfeifer13,14,15, Timo Wolf16, Maria J.G.T. Vehreschild16, Clemens Wendtner17, Ron 8

Diskin9, Stephan Becker2,3, and Florian Klein1,4,11 9

1 Laboratory of Experimental Immunology, Institute of Virology, Faculty of Medicine and University Hospital Cologne, University 10 of Cologne, 50931 Cologne, Germany 11 2 Institute of Virology, Faculty of Medicine, Philipps University Marburg, 35043 Marburg, Germany 12 3 German Center for Infection Research, Partner Site Gießen-Marburg-Langen, 35043 Marburg, Germany 13 4 German Center for Infection Research, Partner Site Bonn-Cologne, 50931 Cologne, Germany 14 5 Department I of Internal Medicine, Faculty of Medicine and University Hospital Cologne, University of Cologne, 50937 15 Cologne, Germany 16 6 Institute of Virology, Faculty of Medicine and University Hospital Cologne, University of Cologne, 50931 Cologne, Germany 17 7 Department I of Internal Medicine, Center for Integrated Oncology (CIO) Aachen Bonn Cologne Düsseldorf, University Hospital 18 of Cologne, University of Cologne, 50937 Cologne, Germany 19 8 Cologne Excellence Cluster for Cellular Stress Responses in Ageing-Associated Diseases (CECAD), University of Cologne, 20 50931 Cologne, Germany 21 9 Department of Structural Biology, Weizmann Institute of Science, 76100 Rehovot, Israel 22 10 Institute for Dental Research and Oral Musculoskeletal Biology and Center for Biochemistry, University of Cologne, 50931 23 Cologne, Germany 24 11 Center for Molecular Medicine Cologne (CMMC), University of Cologne, 50931 Cologne, Germany 25 12 Institute for Translational Bioinformatics, University Hospital Tübingen and University of Tübingen, 72076 Tübingen, 26 Germany. 27 13 Faculty of Medicine, University of Tübingen, 72076, Germany 28 14 Methods in Medical Informatics, Department of Computer Science, University of Tübingen, 72076 Tübingen, Germany 29 15 German Center for Infection Research, Partner Site Tübingen, 72076 Tübingen, Germany 30 16 Department of Internal Medicine, Infectious Diseases, University Hospital Frankfurt, Goethe University Frankfurt, 60590 31 Frankfurt/Main, Germany 32 17 Department of Infectious Diseases and Tropical Medicine, Munich Clinic Schwabing, Academic Teaching Hospital, Ludwig-33 Maximilians-University, 80804, Munich, Germany. 34 * These authors contributed equally 35 Correspondence: [email protected] 36

37

(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted June 12, 2020. . https://doi.org/10.1101/2020.06.12.146290doi: bioRxiv preprint

https://doi.org/10.1101/2020.06.12.146290

2

SUMMARY 38

The SARS-CoV-2 pandemic has unprecedented implications for public health, social 39

life, and world economy. Since approved drugs and vaccines are not available, new 40

options for COVID-19 treatment and prevention are highly demanded. To identify 41

SARS-CoV-2 neutralizing antibodies, we analysed the antibody response of 12 42

COVID-19 patients from 8 to 69 days post diagnosis. By screening 4,313 SARS-43

CoV-2-reactive B cells, we isolated 255 antibodies from different time points as early 44

as 8 days post diagnosis. Among these, 28 potently neutralized authentic SARS-45

CoV-2 (IC100 as low as 0.04 µg/ml), showing a broad spectrum of V genes and low 46

levels of somatic mutations. Interestingly, potential precursors were identified in 47

naïve B cell repertoires from 48 healthy individuals that were sampled before the 48

COVID-19 pandemic. Our results demonstrate that SARS-CoV-2 neutralizing 49

antibodies are readily generated from a diverse pool of precursors, fostering the hope 50

of rapid induction of a protective immune response upon vaccination. 51

52

KEYWORDS 53

SARS-CoV-2; 2019-nCoV; COVID-19; neutralizing antibody; monoclonal antibody; 54

single B cell analysis 55

56


https://doi.org/10.1101/2020.06.12.146290

3

INTRODUCTION 57

By mid-May 2020 over 4.5 million severe acute respiratory syndrome coronavirus 2 58

(SARS-CoV-2) infections and over 300,000 casualties of the associated coronavirus 59

disease 2019 (COVID-19) were reported (Dong et al., 2020; Huang et al., 2020; Zhou 60

et al., 2020; Zhu et al., 2020). The exponential spread of the virus has caused 61

countries to shut down public life with unprecedented social and economic 62

consequences. Therefore, decoding SARS-CoV-2 immunity to promote the 63

development of vaccines as well as potent antiviral drugs is an urgent health need 64

(Sanders et al., 2020). 65

Monoclonal antibodies (mAbs) have been demonstrated to effectively target 66

and neutralize viruses such as Ebola virus (EBOV; Ehrhardt et al., 2019; Flyak et al., 67

2016; Saphire et al., 2018), respiratory syncytial virus (RSV; Kwakkenbos et al., 68

2010), influenza virus (Corti et al., 2011; Joyce et al., 2016; Kallewaard et al., 2016), 69

or human immunodeficiency virus 1 (HIV-1; Burton et al., 2009; Huang et al., 2016a, 70

2016b; Scheid et al., 2011; Schommers et al., 2020; Wu et al., 2010). The most 71

prominent target for an antibody-mediated response on the surface of SARS-CoV-2 72

virions is the homotrimeric spike (S) protein. The S protein promotes cell entry 73

through the interaction of a receptor-binding domain (RBD) with angiotensin-74

converting enzyme 2 (ACE2; Hoffmann et al., 2020; Walls et al., 2020). Antibodies 75

that target the S protein are therefore of high value to prevent and treat COVID-19 76

(Burton and Walker, 2020). 77

78


https://doi.org/10.1101/2020.06.12.146290

4

RESULTS 79

SARS-CoV-2-infected individuals develop a polyclonal memory B cell response 80

against the S protein 81

To investigate the antibody response against SARS-CoV-2, we collected blood 82

samples from seven COVID-19 patients (aged 38 to 59 years) between 8 and 36 83

days post diagnosis (Figure 1A, Table S1). Five patients presented with mild 84

symptoms including dry cough, fever, and dyspnoea, while two patients were 85

asymptomatic (Table S1). Purified plasma immunoglobulin G (IgG) of all seven 86

individuals showed binding to the full trimeric S-ectodomain (Wrapp et al., 2020), with 87

half maximal effective concentrations (EC50) ranging from 3.1 to 96.1 µg/ml (Figure 88

1B, Table S2). Moreover, neutralizing IgG activity was determined against authentic 89

SARS-CoV-2, showing 100% inhibitory concentrations (IC100) between 78.8 and 90

1,500 µg/ml in five out of seven patients (Figure 1B, Table S2). In order to decipher 91

the SARS-CoV-2 B cell and antibody response on a molecular level, we performed 92

single B cell sorting and sequence analysis of all individuals. Using flow cytometry, 93

we detected between 0.04% (± 0.06) and 1.02% (± 0.11) IgG+ B cells that reacted 94

with the S-ectodomain (Figure 1C, Figure S1). From these we isolated a total of 95

1,751 single B cells and amplified IgG heavy and light chains using optimized PCR 96

protocols (Figure 1C, Table S3; Kreer et al., 2020a; Schommers et al., 2020). 97

Sequence analysis revealed a polyclonal antibody response with 22% to 45% 98

clonally related sequences per individual and 2 to 29 members per identified B cell 99

clone (Figure 1D, Table S3). We conclude that a polyclonal B cell response against 100

the SARS-CoV-2 S protein was initiated in all studied COVID-19 patients. 101

102

103

104


https://doi.org/10.1101/2020.06.12.146290

5

Longitudinal analysis of the SARS-CoV-2 antibody response 105

To delineate the dynamics of the SARS-CoV-2 antibody response, we obtained 106

longitudinal blood samples from an additional five infected individuals at three time 107

points spanning 8 to 69 days post diagnosis (Figure 2A, Table S1). Across the 108

different individuals, EC50 (S-ectodomain binding) and IC100 (SARS-CoV-2 109

neutralization) values of plasma IgG ranged from 1.54 to 129 µg/ml and 78.8 to 1,500 110

µg/ml, respectively (Figure 2B, Table S2). For each individual, however, this 111

response remained almost unchanged over the studied period (Figure 2A, B). 112

To investigate B cell clonality and antibody characteristics on a single cell level, 113

we proceeded to sort S-ectodomain-reactive IgG+ B cells from all five subjects at the 114

different time points (t1, t2, t3). We found up to 0.65% SARS-CoV-2-reactive B cells, 115

with a tendency towards higher frequencies at later time points (Figure 2C). From a 116

total of 2,562 B cells, we detected 254 B cell clones (Table S3). 51% of these clones 117

(129) were recurrently detected, suggesting the persistence of SARS-CoV-2 reactive 118

B cells over the investigated period of 2.5 months. When separated by individual time 119

points, the fraction of clonally related sequences ranged from 18% to 67% across 120

patients and remained constant or showed only moderate decreases over time 121

(Figure 2D). 122

Next, we analysed the single cell Ig sequences (6,587 productive heavy and 123

light chains) from all 12 patients (Figure 2E-F, Figure S2). Here, clonally related and 124

non-clonal sequences similarly presented a broad spectrum of VH gene segments, 125

normally distributed heavy chain complementarity-determining region 3 (CDRH3) 126

lengths, symmetrical CDRH3 hydrophobicity distributions, and a predominance of the 127

IgG1 isotype (Figure 2E). However, in comparison to repertoire data from healthy 128

individuals, IgVH 3-30 was overrepresented and clonal sequences more often 129

facilitated κ over λ light chains (3/4 in clonal versus 2/3 in non-clonal, p = 0.0027; 130


https://doi.org/10.1101/2020.06.12.146290

6

Figure 2F, Figure S2). Finally, VH genes of S-reactive B cells were on average less 131

mutated than VH genes from healthy IgG+ repertoires (median identity of 98.3 vs. 132

94.3, p < 0.0001; Figure 2E, Figure S2). We concluded that a SARS-CoV-2-reactive 133

IgG+ B cell response readily develops after infection with the same B cell clones 134

detectable over time and a preference for facilitating the VH gene segment 3-30. 135

136

Isolation of highly potent near-germline SARS-CoV-2-neutralizing antibodies 137

from COVID-19 patients 138

To determine antibody characteristics and to isolate potent neutralizing antibodies, 139

we cloned a total of 312 matched heavy and light chain pairs (70% clonal, 30% non-140

clonal) from all 12 patients. From 255 successfully produced IgG1 antibodies, 79 141

(31%) bound to the full trimeric S-ectodomain (Wrapp et al., 2020) with EC50 values 142

ranging between 0.02 µg/ml and 5.20 µg/ml (Figure 3A). Of these, 30 antibodies 143

showed SARS-CoV-2 reactivity by a commercial diagnostic system (Euroimmun IgG 144

detection kit; Figure 3A and B, Table S4). Surface plasmon resonance (SPR) 145

analyses using the RBD as analyte for 13 SARS-CoV-2 interacting antibodies gave 146

dissociation constant (KD) values as low as 0.02 nM (Table S4). By determining the 147

neutralization activity against authentic SARS-CoV-2, we found 28 neutralizing 148

antibodies in 9 out of 12 patients with IC100 values ranging between 100 µg/ml (assay 149

limit) and 0.04 µg/ml (Figure 3C and D). Of note, neutralizing activity was mainly 150

detected among high affinity antibodies (Figure 3B, Table S4) and a positive 151

correlation between neutralization and binding could be detected (rs = 0.429, p = 152

0.023; Figure 3E). 153

To better characterize the interaction between SARS-CoV-2 S protein and 154

reactive antibodies, we determined binding to a truncated N-terminal S1 subunit 155

(including the RBD), the isolated RBD, and a monomeric S ectodomain. We found 27 156


https://doi.org/10.1101/2020.06.12.146290

7

out of 28 neutralizing antibodies binding to the RBD, but only 31% of the non-157

neutralizing antibodies, suggesting that the RBD is a major site of vulnerability on the 158

S protein. Epitopes for non-neutralizing antibodies included the N-terminal S1 domain 159

and conformational epitopes (Figure 3F, Table S4). Notably, both neutralizing and 160

non-neutralizing antibodies were characterized by a broad distribution of VH as well 161

as VL gene segments and a preference for κ light chains (Figure 3G, Figure S4). 162

Moreover, 32 of 79 binding and 11 of 28 neutralizing antibodies demonstrated 163

germline identities of 99% to 100 % and no correlation was detected between 164

neutralizing activity and the level of somatic mutation (Figure 3G, Supplementary 165

Table. 4, Figure S3). 166

Finally, we performed a HEp-2 cell autoreactivity assay. 4 out of 28 neutralizing 167

antibodies showed low to moderate signs of autoreactivity (Figure S5, Table S4) and 168

2 of them also reacted with other proteins (i.e. Ebola glycoprotein, HIV-1 gp140; 169

Table S4). In summary, these data show that SARS-CoV-2 neutralizing antibodies 170

develop from a broad set of different V genes and are characterized by a low degree 171

of somatic mutations. Moreover, we were able to isolate highly potent neutralizing 172

antibodies that present promising candidates for antibody mediated prevention and 173

therapy of SARS-CoV-2 infection. 174

175

Investigating ongoing somatic hypermutation in SARS-CoV-2 binding and 176

neutralizing antibodies 177

To investigate the development of somatic mutations over time, we longitudinally 178

analysed 129 recurring B cell clones that comprised 17 binding and six neutralizing 179

antibodies. To this end, we phylogenetically matched all members of a B cell clone at 180

a given time point with the most closely related member at the consecutive time point 181

(331 pairings in total). Mean mutation frequencies in either direction (i.e., towards 182


https://doi.org/10.1101/2020.06.12.146290

8

higher or lower V gene germline identities) were 0.51±0.61%, 0.08±0.51%, and 183

0.01±0.19% per week for all, binding, and neutralizing clonal members, respectively 184

(Figure 4A upper panels). When averaging VH gene germline identity of concurrent 185

clonal members, we found a moderate increase in somatic mutations over time 186

(Figure 4A lower panels). Changes were similar for binding and neutralizing 187

subsets with one exception among the neutralizing antibodies that accumulated 188

about 5% nucleotide mutations over the investigated period (Figure 4A lower 189

panel). In line with this finding, neutralizing antibodies isolated at days 8 to 17 and 190

days 34 to 42 post diagnosis showed VH gene germline identities of 97.5% and 191

97.0%, respectively (Figure 4B). We concluded that SARS-CoV-2 neutralizing 192

antibodies carry similar levels of somatic hypermutation independently of the time of 193

isolation. 194

195

Potential precursor sequences of SARS-CoV-2-neutralizing antibodies can be 196

identified among healthy individuals 197

The low rate of somatic mutations in the majority of binding and neutralizing 198

antibodies emphasizes the requirement for the presence of distinct germline 199

recombinations in the naïve human B cell repertoire. To estimate the frequency of 200

potential precursor B cells, we performed unbiased heavy and light chain next 201

generation sequencing (NGS) of the naïve B cell receptor repertoires from 48 healthy 202

donors (Table S5). All samples were collected before the SARS-CoV-2 outbreak and 203

comprised a total of 1,7 million collapsed reads with 455,423 unique heavy, 170,781 204

κ, and 91,505 λ chain clonotypes (defined as identical V/J pairing and the same 205

CDR3 amino acid sequence). Within this data set we searched for heavy and light 206

chains that resemble the 79 SARS-CoV-2 binding antibodies (Figure 5A). For 14 out 207

of 79 tested antibodies, we found 61 heavy chain clonotypes with identical V/J pairs 208


https://doi.org/10.1101/2020.06.12.146290

9

and similar (± 1 aa in length and up to 3 aa differences) CDRH3s in 28 healthy 209

individuals (Figure 5B and 4C), including one exact CDRH3 match 210

(MnC2t1p1_C12). For light chains, we identified 1,357 κ chain precursors with exact 211

CDR3 matches that cover 41 of 62 antibodies and 109 λ chain precursors that 212

represent 7 of 17 antibodies (Figure 5B and 4C). All 48 naive repertoires included at 213

least one κ and one λ chain precursor. When combining heavy and light chain data, 214

we found both precursor sequences of 9 antibodies in 14 healthy individuals (Figure 215

5C). Importantly, among these potential precursor pairs, we found three potent 216

neutralizing antibodies (CnC2t1p1_B4, HbnC3t1p1_G4, and HbnC3t1p2_B10). While 217

the NGS repertoire data did not include pairing information of heavy and light chain 218

combinations, we found matched heavy and light chain sequences despite small 219

sample sizes of on average 9,500 heavy and 2,000 to 3,500 light chain clonotypes 220

per individual. We thus conclude that potential SARS-CoV-2 binding and neutralizing 221

antibody precursors are likely to be abundant in naïve B cell repertoires. 222

223


https://doi.org/10.1101/2020.06.12.146290

10

DISCUSSION 224

Neutralizing antibodies can effectively target pathogens and their induction is a key 225

objective of vaccination strategies (Fauci and Marston, 2015; Mascola and 226

Montefiori, 2010; Walker and Burton, 2018; Zolla-Pazner et al., 2019). A detailed 227

understanding of the human antibody response to SARS-CoV-2 is therefore critical 228

for the development of effective immune mediated approaches against the continuing 229

pandemic (Burton and Walker, 2020; Koff et al., 2013; Kreer et al., 2020b). Through 230

the single cell analysis of >4,000 SARS-CoV-2-reactive B cells from 12 infected 231

individuals, we identified highly potent human monoclonal SARS-CoV-2-neutralizing 232

antibodies. These antibodies block authentic viral infection at concentrations as low 233

as 0.04 µg/ml and provide a novel option for prevention and treatment of SARS-CoV-234

2 infection. For many viral pathogens, the development of antibody potency is 235

dependent on prolonged affinity maturation (Abela et al., 2019; Andrews et al., 2019; 236

Davis et al., 2019; Wec et al., 2020). In contrast, high SARS-CoV-2-neutralizing 237

activity can be observed for antibodies that show little if any deviation from their 238

germline precursors outside of their CDR3s. Our longitudinal analysis of the B cell 239

response spanning a period of more than 2.5 months after SARS-CoV-2 240

transmission reveals that the development of a neutralizing antibody response is 241

followed by limited additional somatic mutation. Thus, vaccine efficacy may be more 242

dependent on the engagement of naïve B cells rather than an extended presence of 243

antigen to enable the accumulation of multiple antibody mutations. Importantly, we 244

observed potential heavy and light chain precursors of potent SARS-CoV-2-245

neutralizing antibodies among the naïve B cell repertoires of healthy individuals that 246

were sampled before the pandemic. Given the broad gene distribution among SARS-247

CoV-2-neutralizing antibodies, our findings therefore indicate the potential for a 248

broadly active SARS-CoV-2 vaccine. 249


https://doi.org/10.1101/2020.06.12.146290

11

ACKNOWLEDGEMENTS 250

We thank all study participants who devoted time to our research; Jan Mathis Eckert, 251

Ralf Ortmanns, and Heidrun Schößler of the health department of Heinsberg 252

for patient enrollment; all members of the Klein and Becker Laboratories for helpful 253

discussion and support; Jason McLellan, Nianshuang Wang, and Daniel Wrapp for 254

sharing the SARS-CoV-2 S-ectodomain plasmid; Florian Krammer for sharing the 255

RBD plasmid; Simon Pöpsel and Robert Hänsel-Hertsch for helpful discussion and 256

technical support; as well as Daniela Weiland and Nadine Henn for lab management 257

and assistance. This work was funded by grants from the German Center for 258

Infection Research (DZIF to F.K. and S.B.), the German Research Foundation (DFG; 259

CRC 1279, F.K.; CRC 1310, F.K.; FOR2722, M.K.), the European Research Council 260

(ERC-StG639961, F.K.), the German Federal Ministry of Education and Research 261

(BMBF) within the 'Medical Informatics Initiative' (DIFUTURE, reference number 262

01ZZ1804D, S.L., N.P.), the Ben B. and Joyce E. Eisenberg Foundation (R.D.), and 263

the Ernst I Ascher Foundation and from Natan Sharansky (R.D.). 264

265

AUTHOR CONTRIBUTIONS 266

Conceptualization, F.K; Methodology, F.K., S.B., C.K., M.Z., M.S.E., L.G., C.R., S.H., 267

S.L., N.P.; Investigation, C.K., M.Z., T.W., L.G., M.S.E., C.R., S.H., M.Kor., H.G., 268

P.S., K.V., V.D.C., H.J., R.B., A.A., V.K., A.K., H.C.D., M.Ko., T.Wo., M.J.G.T.V., 269

C.W.; Software, C.K., S.L., N.P.; Formal Analysis, C.K., M.Z., S.L., N.P., and F.K.; 270

Resources, F.K., S.B., R.D.; Writing - original Draft, F.K., C.K., M.Z., T.W., H.G.; 271

Writing - review and editing, all authors; Supervision, F.K., S.B., R.D. 272

273

DECLARATION OF INTERESTS 274

Reported antibodies are in the process of being patented. 275


https://doi.org/10.1101/2020.06.12.146290

12

FIGURE LEGENDS 276

Figure 1 SARS-CoV-2 infection induces a polyclonal B cell and antibody response. 277

(A) Scheme of cross-sectional sample collection. (B) Binding to the trimeric SARS-278

CoV-2 S-ectodomain (ELISA, EC50) and authentic SARS-CoV-2 neutralization activity 279

(complete inhibition of VeroE6 cell infection, IC100) of cross-sectional poly-IgG 280

samples. Bar plots show arithmetic or geometric means ± SD of duplicates or 281

quadruplicates for EC50 and IC100, respectively. n.n., not neutralizing. (C) Dot plots of 282

IgG+ B cell analysis. Depicted numbers (%) indicate average frequencies of S-283

reactive B cells across several experiments (see also Table S2 and Figure S1). (D) 284

Clonal relationship of S-ectodomain-reactive B cells. Individual clones are coloured in 285

shades of blue and green. Numbers of productive heavy chain sequences are given. 286

Clone sizes are proportional to the total number of productive heavy chains per 287

clone. 288

289

Figure 2 SARS-CoV-2-specific IgG+ B cells readily develop after infection with 290

recurring B cell clones and a preference for the VH gene segment 3-30. 291

(A) Scheme of longitudinal sample collection. Viral RNA load from nasopharyngeal 292

swabs is indicated in red (cp/ml, right Y axis). *Viral load for IDFnC1 is given as 293

positive/negative result. (B) Binding to trimeric SARS-CoV-2 S-ectodomain (ELISA, 294

EC50) and authentic SARS-CoV-2 neutralization activity (complete inhibition of 295

VeroE6 cell infection, IC100) of longitudinal poly-IgG samples. Bar plots show 296

arithmetic or geometric means ± SD of duplicates or quadruplicates for EC50 and 297

IC100, respectively. n.n., not neutralizing. (C) Percentage of SARS-CoV-2 S-298

ectodomain-reactive IgG+ B cells over time (mean ± SD). (D) Clonal relationship over 299

time. Individual clones are coloured in shades of blue and green. Numbers of 300

productive heavy chain sequences per time point are given. (E) Frequencies of VH 301


https://doi.org/10.1101/2020.06.12.146290

13

gene segments (top), CDRH3 length and CDRH3 hydrophobicity (lower left), as well 302

as VH gene germline identity and IgG isotype of clonal and non-clonal sequences 303

(lower right) from all 12 subjects and time points. NGS reference data from 48 304

healthy individuals (collected before the outbreak of SARS-CoV-2) are depicted in 305

red. (F) Ratio of κ and λ light chains in non-clonal (top, grey) and in clonal (bottom, 306

blue) sequences. 307

308

Figure 3 Infected individuals develop potent near-germline SARS-CoV-2-neutralizing 309

antibodies that preferentially bind to the S-protein receptor binding domain 310

(A) Interaction of isolated antibodies with SARS-CoV-2 S-ectodomain by ELISA. 311

Binding antibodies (blue) were defined by an EC50 < 30µg/ml and an OD415-695 > 0.25 312

(not shown). (B) EC50 values (mean of duplicates) of SARS-CoV-2 S-ectodomain 313

interacting antibodies per individual. Neutralizing antibodies are labelled in shades of 314

red. (C) Authentic SARS-CoV-2 neutralization activity (complete inhibition of VeroE6 315

cell infection, IC100, in quadruplicates) of S-ectodomain-specific antibodies (red). (D) 316

Geometric mean potencies (IC100) of all neutralizing antibodies. (E) Correlation 317

between S-ectodomain binding (EC50) and neutralization potency (IC100). Correlation 318

coefficient rS and approximate p-value were calculated by Spearman's rank-319

order correlation. (F) Epitope mapping of SARS-CoV-2 S-ectodomain-specific 320

antibodies against the RBD, truncated N-terminal S1 subunit (aa 14-529), and a 321

monomeric S ectodomain construct by ELISA. S2 binding was defined by interaction 322

with monomeric S but not RBD or S1. Antibodies interacting with none of the 323

subdomains were specified as conformational epitopes or not defined. (G) Top: 324

Frequencies of VH gene segments for non-neutralizing and neutralizing antibodies. 325

Clonal sequence groups were collapsed and treated as one sample for calculation of 326

the frequencies. Bottom: CDRH3 length (left) and VH gene germline identity (right) of 327


https://doi.org/10.1101/2020.06.12.146290

14

non-neutralizing and neutralizing antibodies. 328

329

Figure 4 Dynamics of somatic mutations for SARS-CoV-2-specific antibodies 330

(A) Distribution of mutation rates per week for clonal members and median change in 331

VH germline identity normalized by the first measurement for each longitudinal clone. 332

(B) VH gene germline identity of neutralizing antibodies from different time points. 333

Upper panel shows mean ± SD for groups of antibodies from early or late time points 334

(two-tailed unpaired t-test). Lower panel shows VH germline identities of all isolated 335

neutralizing antibodies depending on the time between diagnosis and blood sample 336

collection. 337

338

Figure 5 Precursor frequencies of SARS-CoV-2-specific antibodies in naïve 339

repertoires of healthy individuals 340

(A) Strategy for precursor identification from healthy naïve B cell receptor (BCR) 341

repertoires. HC, heavy chain; KC, κ chain; LC, λ chain; VH/VL, heavy and light chain 342

V gene; CDRH3/CDRL3, heavy and light chain CDR3. (B) Number of clonotypes in 343

healthy naïve B cell repertoires (n=48) with matched V/J genes from SARS-CoV-2 344

binding antibodies (n=79), plotted against the CDR3 difference. Bars of included 345

potential precursors are highlighted in shades of blue. For heavy chains, CDR3s 346

were allowed to differ one amino acid in length and contain up to 3 amino acid 347

mutations. For light chains, only identical CDR3s were counted. (C) Number of 348

different antibody heavy and light chains for which precursors have been identified 349

and number of different individuals from which precursor sequences have been 350

isolated. Numbers in overlapping circles represent matched heavy and light chain 351

combinations. 352

353


https://doi.org/10.1101/2020.06.12.146290

15

SUPPLEMENTARY FIGURE LEGENDS 354

Figure S1 Gating strategy for single cell sort. 355

CD19+ B cells isolated by MACS were used and cell aggregates were excluded by 356

FSC. Living CD20+ IgG+ cells were gated and cells with a positive SARS-CoV-2 S 357

ectodomain staining were selected for single cell sort. 358

359

Figure S2 Light chain characteristics of sorted single cells. 360

Left and middle panel: Frequencies of VL gene segments of clonal and non-clonal 361

sequences are shown (κ left, λ middle). Right panel: Ratios of κ and λ within the 362

single sample sets in clonal and non-clonal sequences. A two-tailed ratio paired t-test 363

was performed on κ / λ ratios to test for significance. 364

365

Figure S3 Correlation of binding and neutralization with VH gene characteristics 366

Correlation plots of EC50 values of binding or neutralizing antibodies or IC100 values 367

of neutralizing antibodies with CDRH3 lengths or VH germline identities. Spearman 368

correlation coefficient rS and approximate p values are given. 369

370

Figure S4 VL gene distribution in non-neutralizing and neutralizing antibodies 371

(A) Frequencies of VL gene segments for non-neutralizing (left, grey) and neutralizing 372

antibodies (right, red). Clonal sequence groups were collapsed and treated as one 373

sample for calculation of the frequencies. (B) Ratio of λ and κ light chains for 374

neutralizing (left) and non-neutralizing S-ectodomain-specific antibodies (bottom, 375

blue). 376

Figure S5 Autoreactivity of selected SARS-CoV-2 binding and neutralizing 377

antibodies. 378


https://doi.org/10.1101/2020.06.12.146290

16

HEp-2 cells were incubated with SARS-CoV-2 S-ectodomain antibodies at 379

concentrations of 100 µg/ml and analysed by indirect immunofluorescence. 380

Representative pictures of the scoring system are shown. 381

382


https://doi.org/10.1101/2020.06.12.146290

17

METHODS 383

CONTACT FOR REAGENT AND RESOURCE SHARING 384

Further information and requests for resources and reagents should be directed to 385

and will be fulfilled by the Lead Contact, Florian Klein ([email protected]). 386

387

EXPERIMENTAL MODELS AND SUBJECT DETAILS 388

SARS-CoV-2 infected individuals and sample collection 389

Samples were obtained under a study protocol approved by the Institutional Review 390

Board of the University of Cologne and respective local IRBs (study protocol 16-054). 391

All participants provided written informed consent and were recruited at hospitals or 392

as outpatients. Sites of recruitment were Munich Clinic Schwabing for IDMnC1,2,4 393

and 5, the University Hospital of Frankfurt for patients IDFnC1, 2, and University 394

Hospital Cologne for patient IDCnC2. Patients IDHbnC1-5 were recruited as 395

outpatients in the county Heinsberg. 396

397

METHOD DETAILS 398

Isolation of peripheral blood mononuclear cells (PBMCs), plasma and total IgG 399

from whole blood 400

Blood draw collection was performed using EDTA tubes and/or syringes pre-filled 401

with heparin. PBMC isolation was performed using Leucosep centrifuge tubes 402

(Greiner Bio-one) prefilled with density gradient separation medium (Histopaque; 403

Sigma-Aldrich) according to the manufacturer’s instructions. Plasma was collected 404

and stored separately. For IgG isolation, 1 ml of the collected plasma was heat-405

inactivated (56°C for 40 min) and incubated with Protein G Sepharose (GE Life 406

Sciences) overnight at 4°C. The suspension was transferred to chromatography 407

columns and washed with PBS. IgGs were eluted from Protein G using 0.1 M glycine 408


https://doi.org/10.1101/2020.06.12.146290

18

(pH=3.0) and buffered in 0.1 M Tris (pH=8.0). For buffer exchange to PBS, 30 kDa 409

Amicon spin membranes (Millipore) were used. Purified IgG concentration was 410

measured using a Nanodrop (A280) and samples were stored at 4°C. 411

412

SARS-CoV-2 S protein expression and purification 413

The construct encoding the prefusion stabilized SARS-CoV-2 S ectodomain (amino 414

acids 1−1208 of SARS-CoV-2 S; GenBank: MN908947) was kindly provided by 415

Jason McLellan (Texas, USA) and described previously (Wrapp et al., 2020). In 416

detail, two proline substitutions at residues 986 and 987 were introduced for 417

prefusion state stabilization, a “GSAS” substitution at residues 682–685 to eliminate 418

the furin cleavage site, and a C-terminal T4 fibritin trimerization motif. For purification, 419

the protein is C-terminally fused to a TwinStrepTag and 8XHisTag. Protein 420

production was done in HEK293-6E cells by transient transfection with 421

polyethylenimine (PEI, Sigma-Aldrich) and 1 µg DNA per 1 mL cell culture medium at 422

a cell density of 0.8 106 cells/mL in FreeStyle 293 medium (Thermo Fisher Scientific). 423

After 7 days of culture at 37°C and 5% CO2, culture supernatant was harvested and 424

filtered using a 0.45 µm polyethersulfone (PES) filter (Thermo Fisher Scientific). 425

Recombinant protein was purified by Strep-Tactin affinity chromatography (IBA 426

lifescience, Göttingen Germany) according to the Strep-Tactin XT manual. Briefly, 427

filtered medium was adjusted to pH 8 by adding 100 mL 10x Buffer W (1 M Tris/HCl, 428

pH 8.0, 1.5 M NaCl, 10 mM EDTA, IBA lifescience) and loaded with a low pressure 429

pump at 1 mL/min on 5 mL bedvolume Strep-Tactin resin. The column was washed 430

with 15 column volumes (CV) 1x Buffer W (IBA lifescience) and eluted with 6 x 2.5 431

mL 1x Buffer BXT (IBA lifescience). Elution fractions were pooled and buffer was 432

exchanged to PBS pH 7.4 (Thermo Fisher Scientific) by filtrating four times over 100 433

kDa cut-off cellulose centrifugal filter (Merck). 434


https://doi.org/10.1101/2020.06.12.146290

19

Cloning and expression of different SARS-CoV-2 S protein subunits and Ebola 435

surface glycoprotein 436

The RBD of the SARS-CoV-2 spike protein (MN908947; aa:319-541) was expressed 437

in 293T cells from a plasmid kindly provided by Florian Krammer and purified using 438

Ni-NTA Agarose (Macherey-Nagel), as previously published (Stadlbauer et al., 439

2020). SARS-CoV-2 S ectodomain “monomer” without trimerization domain 440

(MN908947; aa:1-1207) and S1 subunit (MN908947; aa:14-529) regions of the spike 441

DNA were amplified from a synthetic gene plasmid (furin site mutated; Wrapp et al., 442

2020) by PCR. PCR products were cloned into a modified sleeping beauty 443

transposon expression vector containing a C-terminal thrombin cleavage and a 444

double Strep II purification tag. For the S1 subunit, the tag was added at the 5’ end 445

and a BM40 signal peptide was included. For recombinant protein production, stable 446

HEK293 EBNA cell lines were generated employing the sleeping beauty transposon 447

system (Kowarz et al., 2015). Briefly, expression constructs were transfected into the 448

HEK293 EBNA cells using FuGENE HD transfection reagent (Promega). After 449

selection with puromycin, cells were induced with doxycycline. Supernatants were 450

filtered and the recombinant proteins purified via Strep-Tactin®XT (IBA Lifescience) 451

resin. Proteins were then eluted by biotin-containing TBS-buffer (IBA Lifescience), 452

and dialyzed against TBS-buffer. Ebola surface glycoprotein (EBOV Makona, 453

GenBank KJ660347) and HIV-gp140 (strain YU2), both lacking the transmembrane 454

domain and containing a GCN4 trimerization domain, were produced and purified as 455

previously described (Ehrhardt et al., 2019). 456

457

Isolation of SARS-CoV S ectodomain-specific IgG+ B cells 458

B cells were isolated from PBMCs using CD19-microbeads (Miltenyi Biotec) 459

according to the manufacturer’s instruction. Isolated B cells were stained for 20 460


https://doi.org/10.1101/2020.06.12.146290

20

minutes on ice with a fluorescence staining-mix containing 4’,6-Diamidin-2-461

phenylindol (DAPI; Thermo Fisher Scientific), anti-human CD20-Alexa Fluor 700 462

(BD), anti-human IgG-APC (BD), anti-human CD27-PE (BD) and DyLight488-labeled 463

SARS-CoV-2 spike protein (10µg/mL). Dapi-, CD20+, IgG+, SARC-CoV-2 spike 464

protein positive cells were sorted using a FACSAria Fusion (Becton Dickinson) in a 465

single cell manner into 96-well plates. All wells contained 4 µl buffer, consisting of 466

0.5x PBS, 0.5 U/µl RNAsin (Promega), 0.5 U/µl RNaseOUT (Thermo Fisher 467

Scientific), and 10 mM DTT (Thermo Fisher Scientific). After sorting, plates were 468

immediately stored at -80°C until further processing. 469

470

Antibody heavy/light chain amplification and sequence analysis 471

Single cell amplification of antibody heavy and light chains was mainly performed as 472

previously described (Kreer et al., 2020a; Schommers et al., 2020). Briefly, reverse 473

transcription was performed with Random Hexamers (Invitrogen), and Superscript IV 474

(Thermo Fisher Scientific) in the presence of RNaseOUT (Thermo Fisher Sicentific) 475

and RNasin (Promega). cDNA was used to amplify heavy and light chains using 476

PlatinumTaq HotStart polymerase (Thermo Fisher Scientific) with 6% KB extender 477

and optimized V gene-specific primer mixes (Kreer et al., 2020a) in a sequential 478

semi-nested approach with minor modifications to increase throughput (Manuscript in 479

preparation). PCR products were analyzed by gel electrophoresis for correct sizes 480

and subjected to Sanger sequencing. For sequence analysis, chromatograms were 481

filtered for a mean Phred score of 28 and a minimal length of 240 nucleotides (nt). 482

Sequences were annotated with IgBLAST (Ye et al., 2013) and trimmed to extract 483

only the variable region from FWR1 to the end of the J gene. Base calls within the 484

variable region with a Phred score below 16 were masked and sequences with more 485

than 15 masked nucleotides, stop codons, or frameshifts were excluded from further 486


https://doi.org/10.1101/2020.06.12.146290

21

analyses. Clonal analysis was performed separately for each patient. All productive 487

heavy chain sequences were grouped by identical VH/JH gene pairs and the pairwise 488

Levenshtein distance for their CDRH3s was determined. Starting from a random 489

sequence, clone groups were assigned four sequences with a minimal CDRH3 490

amino acid identity of at least 75% (with respect to the shortest CDRH3). 100 rounds 491

of input sequence randomization and clonal assignment were performed and the 492

result with the lowest number of remaining unassigned (non-clonal) sequences was 493

selected for downstream analyses. All clones were cross-validated by the 494

investigators taking shared mutations into account. V gene usage, CDRH3 length 495

and V gene germline identity distributions for all clonal sequences (Figure 2) were 496

determined for all input sequences without further collapsing. CDRH3 hydrophobicity 497

was calculated based on the Eisenberg-scale (Eisenberg et al., 1984). V gene 498

statistics for neutralizer and non-neutralizer (Figure 3) were calculated from collapsed 499

clonal sequences. 500

For longitudinal analyses on mutation frequencies of recurring clones, a multiple 501

sequence alignment for the B cell sequences was calculated with Clustal Omega 502

(version 1.2.3; Sievers et al., 2011) using standard parameters. From this, a 503

phylogenetic tree of the sequences was estimated with RAxML through the raxmlGUI 504

(version 2.0.0-beta.11; Edler et al., 2019) using the GTRGAMMA substitution model 505

(RAxML version 8.2.12; Stamatakis, 2014). Based on the phylogenetic tree 506

distances, all variants of a clone at a given time point were matched to variants at the 507

consecutive time point and the slope between the pairs was computed. Hamming 508

distances between the pairs were determined and normalized for sequence length 509

and time difference to calculate the mean mutation frequency per day. Given the 510

median slope per clone a one-sided Wilcoxon Signed Rank Test was applied to test 511

whether the slopes are equal to zero, with the alternative hypothesis that the slopes 512


https://doi.org/10.1101/2020.06.12.146290

22

are smaller than zero. For visualizing the change of VH gene germline identity over 513

time, the germline identity for each clone was normalized by its median value at the 514

first-time measurement and the median slope was plotted. 515

516

Next generation sequencing and evaluation of healthy control IgG+ and naïve B 517

cell repertoires 518

B cell receptor repertoire sequence data was generated by an unbiased template-519

switch-based approach as previously described (Ehrhardt et al., 2019; Schommers et 520

al., 2020). In brief, PBMCs from 48 healthy individuals (samples taken before the 521

SARS-CoV-2 outbreak) were enriched for CD19+ cells with CD19-microbeads 522

(Miltenyi Biotec). For each individual, 100,000 CD20+IgG+ and 100,000 523

CD20+IgD+IgM+CD27-IgG- B cells were sorted into FBS (Sigma-Aldrich) using a BD 524

FACSAria Fusion. RNA was isolated with the RNeasy Micro Kit (Qiagen) on a 525

QiaCube (Qiagen) instrument. cDNA was generated by template-switch reverse 526

transcription according to the SMARTer RACE 5’/3’ manual using the SMARTScribe 527

Reverse Transcriptase (Takara) with a template-switch oligo including an 18-528

nucleotide unique molecular identifier (UMI). Heavy and light chain variable regions 529

were amplified in a constant region-specific nested PCR and amplicons were used 530

for library preparation and Illumina MiSeq 2 x 300 bp sequencing. Raw NGS reads 531

were pre-processed and assembled to final sequences as previously described 532

(Ehrhardt et al., 2019). To minimize the influence of sequencing and PCR errors, 533

NGS-derived sequences were only evaluated when UMIs were found in at least three 534

reads. For the identification of overlapping clonotypes in healthy individuals a 535

maximum of one amino acid length difference and three or less differences in 536

absolute amino acid composition of CDR3s were considered as similar. 537

538


https://doi.org/10.1101/2020.06.12.146290

23

Cloning and production of monoclonal antibodies 539

Antibody cloning from 1st PCR products was performed as previously described 540

(Schommers et al., 2020) by sequence and ligation-independent cloning (SLIC; Von 541

Boehmer et al., 2016) with a minor modification. In contrast to the published protocol, 542

PCR amplification for SLIC assembly was performed with extended primers based on 543

2nd PCR primers (Kreer et al., 2020a) covering the complete endogenous leader 544

sequence of all heavy and light chain V genes (Manuscript in preparation). Variable 545

regions with endogenous leader sequences were assembled into mammalian 546

expression vectors for IgH, IgK, or IgL and transfected into HEK293-6E cells for 547

expression, followed by Protein G-based purification of monoclonal antibodies from 548

culture supernatants as previously described (Schommers et al., 2020). 549

550

ELISA analysis to determine antibody binding activity to SARS-CoV-2 S and 551

subunit binding 552

ELISA plates (Corning 3369) were coated with 2 µg/ml of protein in PBS (SARS-553

CoV-2 spike ectodomain, RBD, or n-terminal truncated S1) or in 2 M Urea (SARS-554

CoV-2 spike ectodomain “monomer” lacking the trimerization domain) at 4°C 555

overnight. For SARS-CoV-2 spike ectodomain ELISA, plates were blocked with 5% 556

BSA in PBS for 60 min at RT, incubated with primary antibody in 1% BSA in PBS for 557

90 min, followed by anti-human IgG-HRP (Southern Biotech 2040-05) diluted 1:2500 558

in 1% BSA in PBS for 60 min at RT. SARS-CoV-2 spike subunit ELISAs were done 559

following a published protocol (Stadlbauer et al., 2020). ELISAs were developed with 560

ABTS solution (Thermo Fisher 002024) and absorbance was measured at 415 nm 561

and 695 nm. Positive binding was defined by an OD>0.25 and an EC50<30 µg/ml. 562

The commercial anti-SARS-CoV-2 ELISA kit for immunoglobulin class G was 563

provided by Euroimmun (Euroimmun Diagnostik, Lübeck, Germany). Antibody 564


https://doi.org/10.1101/2020.06.12.146290

24

detection was done according to manufacturer’s instructions and a concentration of 565

50 µg/ml of antibodies and 2 mg/ml of plasma IgG was used. The samples were 566

tested using the automated platform Euroimmun Analyzer 1. 567

568

Virus neutralization test 569

SARS-CoV-2 neutralizing activity of poly-IgG samples or human monoclonal 570

antibodies was investigated based on a previously published protocol for MERS-571

CoV39. Briefly, samples were serially diluted in 96-well plates starting from a 572

concentration of 1,500 µg/ml for poly-IgG and 100 µg/ml for monoclonal antibodies. 573

Samples were incubated for 1 h at 37°C together with 100 50% tissue culture 574

infectious doses (TCID50) SARS-CoV-2 (BavPat1/2020 isolate, European Virus 575

Archive Global # 026V-03883). Cytopathic effect (CPE) on VeroE6 cells (ATCC CRL-576

1586) was analysed 4 days after infection. Neutralization was defined as absence of 577

CPE compared to virus controls. For each test, a positive control (neutralizing 578

COVID-19 patient plasma) was used in duplicates as an inter-assay neutralization 579

standard. 580

581

Surface Plasmon Resonance (SPR) measurements 582

For SPR measurement, the RBD was additionally purified by size exclusion 583

chromatography (SEC) purification with a Superdex200 10/300 column (GE 584

Healthcare). Binding of the RBD to the various mAbs was measured using single-585

cycle kinetics experiments with a Biacore T200 instrument (GE Healthcare). Purified 586

mAbs were first immobilized at coupling densities of 800-1200 response units (RU) 587

on a series S sensor chip protein A (GE Healthcare) in PBS and 0.02% sodium azide 588

buffer. One of the four flow cells on the sensor chip was empty to serve as a blank. 589

Soluble RBD was then injected at a series of concentrations (i.e. 0.8, 4, 20, 100, and 590


https://doi.org/10.1101/2020.06.12.146290

25

500 nM) in PBS at a flow rate of 60 µL/min. The sensor chip was regenerated using 591

10 mM Glycine-HCl pH 1.5 buffer. A 1:1 binding model was used to describe the 592

experimental data and to derive kinetic parameters. For some mAbs, a 1:1 binding 593

model did not provide an adequate description for binding. In these cases, we fitted a 594

two-state binding model that assumes two binding constants due to conformational 595

change. In these cases, we report the first binding constants (KD1). 596

597

HEp-2 Cell Assay 598

Monoclonal antibodies were tested at a concentration of 100 µg/ml in PBS using the 599

NOVA Lite HEp-2 ANA Kit (Inova Diagnostics) according to the manufacturer’s 600

instructions, including positive and negative kit controls on each substrate slide. HIV-601

1-reactive antibodies with known reactivity profiles were included as additional 602

controls. Images were acquired using a DMI3000 B microscope (Leica) and an 603

exposure time of 3.5 s, intensity of 100%, and a gain of 10. 604

605

QUANTIFICATION AND STATISTICAL ANALYSIS 606

Flow cytometry analysis and quantifications were done by FlowJo10. Statistical 607

analyses were performed using GraphPad Prism (v7), Microsoft Excel for Mac 608

(v14.7.3), Python (v3.6.8), and R (v4.0.0). 609

610

DATA AND SOFTWARE AVAILABILITY 611

All data supporting the findings of this study are available within the paper and its 612

supplementary information files. 613

614


https://doi.org/10.1101/2020.06.12.146290

26

REFERENCES 615

Abela, I.A., Kadelka, C., and Trkola, A. (2019). Correlates of broadly neutralizing antibody 616 development. Curr. Opin. HIV AIDS. 617

Andrews, S.F., Chambers, M.J., Schramm, C.A., Plyler, J., Raab, J.E., Kanekiyo, M., 618 Gillespie, R.A., Ransier, A., Darko, S., Hu, J., et al. (2019). Activation Dynamics and 619 Immunoglobulin Evolution of Pre-existing and Newly Generated Human Memory B cell 620 Responses to Influenza Hemagglutinin. Immunity 51, 398-410.e5. 621

Von Boehmer, L., Liu, C., Ackerman, S., Gitlin, A.D., Wang, Q., Gazumyan, A., and 622 Nussenzweig, M.C. (2016). Sequencing and cloning of antigen-specific antibodies from 623 mouse memory B cells. Nat. Protoc. 11, 1908–1923. 624

Burton, D.R., and Walker, L.M. (2020). Rational Vaccine Design in the Time of COVID-19. 625 Cell Host Microbe 27, 695–698. 626

Burton, D.R., Walker, L.M., Phogat, S.K., Chan-Hui, P.Y., Wagner, D., Phung, P., Goss, J.L., 627 Wrin, T., Simek, M.D., Fling, S., et al. (2009). Broad and potent neutralizing antibodies from 628 an african donor reveal a new HIV-1 vaccine target. Science (80-. ). 629

Corti, D., Voss, J., Gamblin, S.J., Codoni, G., Macagno, A., Jarrossay, D., Vachieri, S.G., 630 Pinna, D., Minola, A., Vanzetta, F., et al. (2011). A neutralizing antibody selected from 631 plasma cells that binds to group 1 and group 2 influenza A hemagglutinins. Science (80-. ). 632

Davis, C.W., Jackson, K.J.L., McElroy, A.K., Halfmann, P., Huang, J., Chennareddy, C., 633 Piper, A.E., Leung, Y., Albariño, C.G., Crozier, I., et al. (2019). Longitudinal Analysis of the 634 Human B Cell Response to Ebola Virus Infection. Cell 177, 1566-1582.e17. 635

Dong, E., Du, H., and Gardner, L. (2020). An interactive web-based dashboard to track 636 COVID-19 in real time. Lancet. Infect. Dis. 3099, 19–20. 637

Edler, D., Klein, J., Antonelli, A., and Silvestro, D. (2019). raxmlGUI 2.0 beta: a graphical 638 interface and toolkit for phylogenetic analyses using RAxML. BioRxiv 800912. 639

Ehrhardt, S.A., Zehner, M., Krähling, V., Cohen-Dvashi, H., Kreer, C., Elad, N., Gruell, H., 640 Ercanoglu, M.S., Schommers, P., Gieselmann, L., et al. (2019). Polyclonal and convergent 641 antibody response to Ebola virus vaccine rVSV-ZEBOV. Nat. Med. 25, 1589–1600. 642

Eisenberg, D., Schwarz, E., Komaromy, M., and Wall, R. (1984). Analysis of membrane and 643 surface protein sequences with the hydrophobic moment plot. J. Mol. Biol. 179, 125–142. 644

Fauci, A.S., and Marston, H.D. (2015). Toward an HIV vaccine: A scientific journey. Science 645 (80-. ). 646

Flyak, A.I., Shen, X., Murin, C.D., Turner, H.L., David, J.A., Fusco, M.L., Lampley, R., Kose, 647 N., Ilinykh, P.A., Kuzmina, N., et al. (2016). Cross-Reactive and Potent Neutralizing Antibody 648 Responses in Human Survivors of Natural Ebolavirus Infection. Cell. 649

Hoffmann, M., Kleine-Weber, H., Schroeder, S., Krüger, N., Herrler, T., Erichsen, S., 650 Schiergens, T.S., Herrler, G., Wu, N.H., Nitsche, A., et al. (2020). SARS-CoV-2 Cell Entry 651 Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor. 652 Cell. 653

Huang, C., Wang, Y., Li, X., Ren, L., Zhao, J., Hu, Y., Zhang, L., Fan, G., Xu, J., Gu, X., et al. 654 (2020). Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. 655 Lancet 395, 497–506. 656


https://doi.org/10.1101/2020.06.12.146290

27

Huang, J., Kang, B.H., Ishida, E., Zhou, T., Griesman, T., Sheng, Z., Wu, F., Doria-Rose, 657 N.A., Zhang, B., McKee, K., et al. (2016a). Identification of a CD4-Binding-Site Antibody to 658 HIV that Evolved Near-Pan Neutralization Breadth. Immunity. 659

Huang, Y., Yu, J., Lanzi, A., Yao, X., Andrews, C.D., Tsai, L., Gajjar, M.R., Sun, M., Seaman, 660 M.S., Padte, N.N., et al. (2016b). Engineered Bispecific Antibodies with Exquisite HIV-1-661 Neutralizing Activity. Cell. 662

Joyce, M.G., Wheatley, A.K., Thomas, P. V., Chuang, G.Y., Soto, C., Bailer, R.T., Druz, A., 663 Georgiev, I.S., Gillespie, R.A., Kanekiyo, M., et al. (2016). Vaccine-Induced Antibodies that 664 Neutralize Group 1 and Group 2 Influenza A Viruses. Cell. 665

Kallewaard, N.L., Corti, D., Collins, P.J., Neu, U., McAuliffe, J.M., Benjamin, E., Wachter-666 Rosati, L., Palmer-Hill, F.J., Yuan, A.Q., Walker, P.A., et al. (2016). Structure and Function 667 Analysis of an Antibody Recognizing All Influenza A Subtypes. Cell. 668

Koff, W.C., Burton, D.R., Johnson, P.R., Walker, B.D., King, C.R., Nabel, G.J., Ahmed, R., 669 Bhan, M.K., and Plotkin, S.A. (2013). Accelerating next-generation vaccine development for 670 global disease prevention. Science (80-. ). 340, 1232910–1232910. 671

Kowarz, E., Löscher, D., and Marschalek, R. (2015). Optimized Sleeping Beauty transposons 672 rapidly generate stable transgenic cell lines. Biotechnol. J. 10, 647–653. 673

Kreer, C., Döring, M., Lehnen, N., Ercanoglu, M.S., Gieselmann, L., Luca, D., Jain, K., 674 Schommers, P., Pfeifer, N., and Klein, F. (2020a). openPrimeR for multiplex amplification of 675 highly diverse templates. J. Immunol. Methods 480. 676

Kreer, C., Gruell, H., Mora, T., Walczak, A.M., and Klein, F. (2020b). Exploiting B cell 677 receptor analyses to inform on HIV-1 vaccination strategies. Vaccines 8. 678

Kwakkenbos, M.J., Diehl, S.A., Yasuda, E., Bakker, A.Q., Van Geelen, C.M.M., Lukens, M. 679 V., Van Bleek, G.M., Widjojoatmodjo, M.N., Bogers, W.M.J.M., Mei, H., et al. (2010). 680 Generation of stable monoclonal antibody-producing B cell receptor-positive human memory 681 B cells by genetic programming. Nat. Med. 682

Mascola, J.R., and Montefiori, D.C. (2010). The Role of Antibodies in HIV Vaccines. Annu. 683 Rev. Immunol. 28, 413–444. 684

Sanders, J.M., Monogue, M.L., Jodlowski, T.Z., and Cutrell, J.B. (2020). Pharmacologic 685 Treatments for Coronavirus Disease 2019 (COVID-19): A Review. JAMA. 686

Saphire, E.O., Schendel, S.L., Fusco, M.L., Gangavarapu, K., Gunn, B.M., Wec, A.Z., 687 Halfmann, P.J., Brannan, J.M., Herbert, A.S., Qiu, X., et al. (2018). Systematic Analysis of 688 Monoclonal Antibodies against Ebola Virus GP Defines Features that Contribute to 689 Protection. Cell. 690

Scheid, J.F., Mouquet, H., Ueberheide, B., Diskin, R., Klein, F., Oliveira, T.Y.K., Pietzsch, J., 691 Fenyo, D., Abadir, A., Velinzon, K., et al. (2011). Sequence and Structural Convergence of 692 Broad and Potent HIV Antibodies That Mimic CD4 Binding. Science (80-. ). 693

Schommers, P., Gruell, H., Abernathy, M.E., Tran, M.K., Dingens, A.S., Gristick, H.B., 694 Barnes, C.O., Schoofs, T., Schlotz, M., Vanshylla, K., et al. (2020). Restriction of HIV-1 695 Escape by a Highly Broad and Potent Neutralizing Antibody. Cell 180, 471-489.e22. 696

Sievers, F., Wilm, A., Dineen, D., Gibson, T.J., Karplus, K., Li, W., Lopez, R., McWilliam, H., 697 Remmert, M., Söding, J., et al. (2011). Fast, scalable generation of high-quality protein 698 multiple sequence alignments using Clustal Omega. Mol. Syst. Biol. 7. 699


https://doi.org/10.1101/2020.06.12.146290

28

Stadlbauer, D., Amanat, F., Chromikova, V., Jiang, K., Strohmeier, S., Arunkumar, G.A., Tan, 700 J., Bhavsar, D., Capuano, C., Kirkpatrick, E., et al. (2020). SARS‐CoV‐2 Seroconversion in 701 Humans: A Detailed Protocol for a Serological Assay, Antigen Production, and Test Setup. 702 Curr. Protoc. Microbiol. 57. 703

Stamatakis, A. (2014). RAxML version 8: A tool for phylogenetic analysis and post-analysis 704 of large phylogenies. Bioinformatics 30, 1312–1313. 705

Walker, L.M., and Burton, D.R. (2018). Passive immunotherapy of viral infections: “super-706 antibodies” enter the fray. Nat. Rev. Immunol. 18, 297–308. 707

Walls, A.C., Park, Y.J., Tortorici, M.A., Wall, A., McGuire, A.T., and Veesler, D. (2020). 708 Structure, Function, and Antigenicity of the SARS-CoV-2 Spike Glycoprotein. Cell 181, 281-709 292.e6. 710

Wec, A.Z., Haslwanter, D., Abdiche, Y.N., Shehata, L., Pedreño-Lopez, N., Moyer, C.L., 711 Bornholdt, Z.A., Lilov, A., Nett, J.H., Jangra, R.K., et al. (2020). Longitudinal dynamics of the 712 human B cell response to the yellow fever 17D vaccine. Proc. Natl. Acad. Sci. U. S. A. 117, 713 6675–6685. 714

Wrapp, D., Wang, N., Corbett, K.S., Goldsmith, J.A., Hsieh, C.L., Abiona, O., Graham, B.S., 715 and McLellan, J.S. (2020). Cryo-EM structure of the 2019-nCoV spike in the prefusion 716 conformation. Science (80-. ). 717

Wu, X., Yang, Z.Y., Li, Y., Hogerkorp, C.M., Schief, W.R., Seaman, M.S., Zhou, T., Schmidt, 718 S.D., Wu, L., Xu, L., et al. (2010). Rational design of envelope identifies broadly neutralizing 719 human monoclonal antibodies to HIV-1. Science (80-. ). 720

Ye, J., Ma, N., Madden, T.L., and Ostell, J.M. (2013). IgBLAST: an immunoglobulin variable 721 domain sequence analysis tool. Nucleic Acids Res. 722

Zhou, P., Yang, X.-L., Wang, X.-G., Hu, B., Zhang, L., Zhang, W., Si, H.-R., Zhu, Y., Li, B., 723 Huang, C.-L., et al. (2020). A pneumonia outbreak associated with a new coronavirus of 724 probable bat origin. Nature 579, 270–273. 725

Zhu, N., Zhang, D., Wang, W., Li, X., Yang, B., Song, J., Zhao, X., Huang, B., Shi, W., Lu, 726 R., et al. (2020). A novel coronavirus from patients with pneumonia in China, 2019. N. Engl. 727 J. Med. 382, 727–733. 728

Zolla-Pazner, S., Alvarez, R., Kong, X.P., and Weiss, S. (2019). Vaccine-induced V1V2-729 specific antibodies control and or protect against infection with HIV, SIV and SHIV. Curr. 730 Opin. HIV AIDS. 731

732


https://doi.org/10.1101/2020.06.12.146290

ASARS-CoV-2+

20 400Days

Days from diagnosisto blood draw

n=7

Time of diagnosis

Period of sample collection

Figure 1

IDCnC

2

IDFnC

2

IDHbn

C1

IDHbn

C2

IDHbn

C3

IDHbn

C4

IDHbn

C5

µg/m

lIC100 (µg/ml)EC50 (µg/ml)

102

101

100

B103

IDCnC2 IDFnC2 IDHbnC1 IDHbnC2

IDHbnC3 IDHbnC4 IDHbnC5 Control0.01 %

0.67 ± 0.16 % 0.04 ± 0.06 % 0.23 ± 0.06 % 0.21 ± 0.07 %

1.02 ± 0.11 % 0.25 ± 0.03% 0.22 ± 0.02 %

IgG

SA

RS

-CoV

-2 S

ect

odom

ain

100

102

104

100

102

104

100 102 104

C

100 102 104 100 102 104 100 102 104

D IDCnC2

IDHbnC3

181

324

IDFnC2

IDHbnC4

119

177

IDHbnC1

IDHbnC5

280

208

IDHbnC2

Non-clonal

178

Clonal

n.n.

n.n.


https://doi.org/10.1101/2020.06.12.146290

ASARS-CoV-2+

IDMnC1

IDFnC1

IDMnC2

IDMnC4

IDMnC5

Time ofdiagnosis

Time ofblood draw

(days)Viral RNAload (cp/ml)

Blood draws &viral load (cp/ml)

1010

105

100

0 20 40 60 80

1010

105

100

1010

105

100

1010

105

100

1010

105

100

*

Figure 2

B

102

101

102

101

102

101

102

101

102

101

103

103

103

103

103

IC100 (IgG µg/ml)EC50 (IgG µg/ml)

EC50 & IC100(µg/ml)

t1 t2 t3

n.n.

n.n.

n.n.

0.0

0.5

1.0

0.0

0.5

1.0

% reactive B cells

0.0

0.5

1.0

0.0

0.5

1.0

0.0

0.5

1.0

t1 t2 t3

C

t1 t2 t3

ClonalityD

125 191 246

26 1730

131 229 210

12 134 151

68 132 313

Non-clonalClonal

Non-clonalClonal

Freq

uenc

y (%

)

VH gene germline identity (%)CDRH3 length (aa)

Freq

uenc

y (%

)

Healthy reference

IGHV

CDRH3 Hydrophobicity

%

E

1-2

1-3

1-8

1-18

1-45

1-46

1-58

1-69

1-69

-2 2-5

2-26

2-70 3-

73-

93-

113-

133-

153-

203-

213-

233-

303-

30-3

3-33

3-43

3-43

D3-

483-

493-

533-

643-

64D

3-66

3-72

3-73

3-74 4-

44-

30-2

4-30

-44-

314-

344-

38-2

4-39

4-59

4-61

5-10

-15-

51 6-1

7-4-

1

05

1015203060

5 10 15 20 25 300

1020304050

80 85 90 95 1000

1020304050

-10-8 -6 -4 -2 0 2 4 6 8 100

2040

1 2 3 40

4080

IgG Isotype

%

Non-clonal

Clonal

F

VκVλ

VκVλ

900

1931


https://doi.org/10.1101/2020.06.12.146290

A

<0.1 µg/ml>0.1 µg/ml>1.0 µg/mlNot binding

SARS-CoV-2 S-ectodomain binding

EC50

255

36

24

19

Figure 3

B

CnC2

FnC1

HbnC1

HbnC2

HbnC3

HbnC4

HbnC5

MnC1MnC

2MnC

4MnC

510-2

10-1

100

101

EC50

(µg/

ml)

FnC2

IC100<1.0 µg/ml IC1001.0 - 20 µg/mlIC100>20 µg/ml

IC10

0 <100 µg/ml

Not analyzedNot neutralizing

SARS-CoV-2 neutralization

79

C

D

10-1

100

101

102

MnC

4t1p

1_A

11M

nC4t

2p1_

F5M

nC4t

2p1_

E6

MnC

4t1p

1_A

10M

nC4t

2p2_

A4

MnC

4t2p

1_D

10Fn

C1t

1p2_

A5

CnC

2t1p

1_G

6C

nC2t

1p1_

B10

Hbn

C4t

1p1_

D5

MnC

1t3p

1_G

9C

nC2t

1p1_

E8

MnC

2t1p

1_C

5C

nC2t

1p1_

D6

CnC

2t1p

1_E

12M

nC5t

2p1_

G1

Hbn

C2t

1p2_

D9

CnC

2t1p

1_B

4H

bnC

3t1p

1_F4

MnC

2t1p

1_A

3M

nC4t

2p1_

B3

Hbn

C3t

1p2_

C6

FnC

1t2p

1_G

5Fn

C1t

2p1_

D4

MnC

2t2p

1_C

11H

bnC

3t1p

2_B

10H

bnC

3t1p

1_G

4H

bnC

3t1p

1_C

6

IC10

0 (µg

/ml)

EC50 (µg/ml)

IC10

0 (µg

/ml)

E103

rS= 0.429p = 0.023

102

101

100

10-1

10-2

10010-1 10110-2

non-neutralizing

neutralizingRBD

S2 (+C-terminal S1 aa 530-1207)

conformation/not defined

S1 (N-terminal;aa14-529)

28

F

41

IGHV

Freq

uenc

y (%

)

1-2

1-8

1-18

1-46

1-58

1-69 3-7

3-9

3-15

3-21

3-23

3-30

3-30

-33-

333-

483-

493-

533-

663-

73 4-4

4-31

4-34

4-39

4-61

7-4-

1

0

10

20G

Non-neutralizingNeutralizing

VH germline identity (%)

8 10 12 14 16 18 20 22 24 26 32

0

10

20

CDRH3 length (aa)

Freq

uenc

y (%

)

86 88 90 92 94 96 98 100


https://doi.org/10.1101/2020.06.12.146290

n=6 clones

● ●●

●

● ●●

●

●●

● ●

● ●

● ●●

●

● ●●

●

● ●●

●

●●

● ●

● ●

● ●●

●

n=17 clones

● ●● ●●

●

● ●●

●

●

●

●

●

● ●●

●

●

●

● ●●●

●

●

●●

● ●

● ●

● ●●●

●

●

●

●●

●

●

●●

●

● ●● ●●

●

● ●●

●

●

●

●

●

● ●●

●

●

●

● ●●●

●

●

●●

● ●

● ●

● ●●●

●

●

●

●●

●

●

●●

●

● ●●

●

● ●●

●

●●

● ●

● ●

● ●●

●

● ●●

●

● ●●

●

●●

● ●

● ●

● ●●

●

n=129 clones

Nor

mal

ized

pair

coun

t (%

)

Mutation rate per week (%)

n=331 pairs n=111 pairs n=59 pairsA

Days post diagnosis

80

85

90

95

100

105

0 20 40 60 80 20 40 60 80 20 40 60 800 0

NeutralizerAll BinderN

orm

aliz

ed m

edia

n V H

gene

ger

mlin

e Id

entit

y (%

) B

FnC1

MnC4MnC2

CnC2

HbnC3HbnC2

HbnC4

MnC1

MnC5

Early: day 8-17 Late: day 34-42

V H g

ene

germ

line

Iden

tity

(%)

0 10 20 30 40 5080

85

90

95

100

105

90

95

100

105

Days post diagnosis

n.s.

Early Late0.0 1.0 2.0 3.00.1

1

10

100

0.0 1.0 2.0 3.0 0.0 1.0 2.0 3.0

Figure 4


https://doi.org/10.1101/2020.06.12.146290

Repertoires (clonotypes)

455,423

91,505

170,781

ANaive B cells

HC

KC

BCR sequencingHealthy

n=48

n=12

LC

n=79

S-reactive antibodiesSARS-CoV-2+ Germline precursorVH

VL

JH

JL

CDRH3

CDRL3

Precursorstatistics

B Heavy chains Kappa chains Lambda chains

100101102103104105

CDR3 difference

Uni

que

clon

otyp

es

with

mat

ched

V/J

gen

e

0 5 10 15 20 25 0 5 10 15 20 25 0 5 10 15 20 2548

Heavychains

Lightchains

C

14

9 14

28

48

Figure 5


https://doi.org/10.1101/2020.06.12.146290

Documents

Longitudinal isolation of potent near-germline SARS-CoV-2 ... · 108 points spanning 8 to 69 days post diagnosis (Figure 2A, Table S1). Across the 109 different individuals, EC 50