key: cord-300707-k9uk14b3
authors: Bouwman, Kim M.; Tomris, Ilhan; Turner, Hannah L.; van der Woude, Roosmarijn; Bosman, Gerlof P.; Rockx, Barry; Herfst, Sander; Haagmans, Bart L.; Ward, Andrew B.; Boons, Geert-Jan; de Vries, Robert P.
title: Multimerization- and glycosylation-dependent receptor binding of SARS-CoV-2 spike proteins
date: 2020-09-04
journal: bioRxiv
DOI: 10.1101/2020.09.04.282558
sha: 
doc_id: 300707
cord_uid: k9uk14b3

Receptor binding studies using recombinant SARS-CoV proteins have been hampered due to challenges in approaches creating spike protein or domains thereof, that recapitulate receptor binding properties of native viruses. We hypothesized that trimeric RBD proteins would be suitable candidates to study receptor binding properties of SARS-CoV-1 and -2. Here we created monomeric and trimeric fluorescent RBD proteins, derived from adherent HEK293T, as well as in GnTI mutant cells, to analyze the effect of complex vs high mannose glycosylation on receptor binding. The results demonstrate that trimeric fully glycosylated proteins are superior in receptor binding compared to monomeric and immaturely glycosylated variants. Although differences in binding to commonly used cell lines were minimal between the different RBD preparations, substantial differences were observed when respiratory tissues of experimental animals were stained. The RBD trimers demonstrated distinct ACE2 expression profiles in bronchiolar ducts and confirmed the higher binding affinity of SARS-CoV-2 over SARS-CoV-1. Our results show that fully glycosylated trimeric RBD proteins are attractive to analyze receptor binding and explore ACE2 expression profiles in tissues.

afford additional means for fluorescent-based experiments [13] , and thus are 101 attractive to be fused to RBD proteins. The resulting proteins were analyzed for 102 binding to cell culture cells and paraffin-embedded tissues of various hosts 103 including susceptible and non-susceptible animals. The results demonstrate 104 that fully glycosylated trimeric SARS-CoV-2 RBD proteins reveal the 105 differences in ACE2 expression between cell cultures and tissue sections. 106

These trimeric RBD proteins bind ACE2 efficiently in a species-dependent 107 manner and can be used to profile ACE2 tissue expression. Finally, we 108 without the GCN4 trimerization domain, fused to either sfGFP or mOrange (Fig  116   1A ). The monomeric and trimeric RBDs were efficiently expressed in both HEK 117 293T as well as GnTI cells (data not shown), with an increased yield up to 2-to 118 5-fold when fused to a C-terminal sfGFP ( Fig 1B) . Expression yields of the 119 mOrange2 fusions were comparable to that of sfGFP fusions (data not shown). 120

To illustrate the expression yields of SARS spike proteins or domains thereof 121 we measured the fluorescence in the cell culture supernatant (Fig 1C) . The 122 wild-type full-length ectodomains were difficult to express even with the addition 123 of sfGFP or mOrange2 fusion (Fig 1B) . To increase yields for the full-length 124 ectodomain we introduced the 2P and additional hexapro mutations [14] , and 125 analyzed the fluorescence in cell culture supernatants five days post-126 transfection after incubation at 33 or 37°C. Although we did not observe a large 127 increase in yields, we were able to purify sufficient protein to compare full-length 128 ectodomain trimers vs monomeric and trimeric RBD and NTD proteins. 129 130 Spike RBD domains in frame with a C-terminal GCN4 and fluorescent reporter 131 protein display multimeric features on gel and maintain antigenicity 132

After purification, all RBD proteins were analyzed on gel under non-and 133 reducing conditions (Fig 2A) . Without reducing agent, monomeric RBD proteins 134 revealed dimeric fractions which could be reduced to a single monomeric form. 135

The NTD trimers were reduced under non-reducing conditions, thus solely by 136 SDS. The trimeric RBD variants, on the other hand, revealed dimers and trimers 137 that could be reduced. Besides, the NTD of prototypical γ-coronavirus IBV-M41 138 and influenza A virus HA PR8 as control proteins were included. Finally, we 139 determined the extent of N-glycosylation maturation on purified proteins 140 expressed in either GnTI or 293T by subjecting the monomeric and trimeric 141 proteins to PNGaseF and EndoH treatment ( Fig S1) Next, we examined the antigenicity of the SARS-CoV-1 and -2 proteins using 149 serum collected from macaques 21 days post-infection with SARS-CoV-2 [15] . 150

Both SARS-CoV-2 RBD monomers and trimers derived from 293T cells were 151 efficiently recognized, indicating proper folding ( Fig 2B) . As expected, SARS-152

CoV-1 RBD proteins were poorly recognized, and the negative controls M41 153 NTD and PR8 HA displayed baseline binding identical to pre-infection serum. 154

The NTD trimers were likewise not recognized by the serum (not shown), 155

indicating that the majority of antibodies in naïve animals after infection are 156 directed against the SARS-CoV RBD [16] . Similar results were obtained using 157

GnTI-derived proteins, with the RBD trimer being less efficiently recognized by 158 the macaque serum than its monomeric counterpart. This is in line with recent 159 observations that insect cell-derived proteins are less well bound by serum To determine whether the fluorescent RBD trimers are indeed structured in a 165 trimeric manner we subjected these proteins to negative stain single-particle 166 EM. The EM data revealed that the RBD proteins form stable trimers that 167 resemble known spike structures (Fig. 2C) . Initially, 58,018 individual particles 168 were picked, placed into a stack, and submitted to reference-free two-169 dimensional (2C) classification. From the initial 2D classes, particles that did 170 not resemble RBD were removed, resulting are final particle stacks of 32,152 171 particles, which were then subject to Relion 2D classification. All resultant 172 classes demonstrated evident and distinct trimeric RBD, GCN4, and three 173 sfGFP protein structures that could be identified in the EM images. From the 174 EM images, we generated a model in which we took the crystal structures of 175 sfGFP, the GCN4 trimerization domain (PDB:2O7H), and the SARS-CoV-2 176 RBD (PDB: 6XM4) to demonstrate the likely structure of our RBD trimer (Fig  177   2D) . To determine the biological activity of our RBD proteins we stained A549 and 182 VERO cells that are reported to support SARS-CoV replication, with the latter 183 being more susceptible [17] . However, A549 cells were bound by all our RBD 184 proteins with a slight increase in intensity from monomeric GnTI derived RBD 185 proteins to trimeric 293T derived RBDs (Fig 3) . Trimeric 293T RBD binding was 186 efficiently blocked using 4μM recombinant ACE2 whereas 400nM ACE2 pre-187 incubation was not sufficient to prevent binding of fully glycosylated trimeric 188 RBD proteins to cells completely. SARS-CoV-2 RBD proteins bound slightly 189 more intensely to A549 cells compared to the same SARS-CoV-1 RBD 190 proteins. A similar pattern was observed for VERO-E6 cells, however, the fully 191 glycosylated SARS-CoV-2 RBD trimer bound markedly stronger compared to 192 the other RBD preparations ( Fig S2A) . Importantly, the full-length ectodomain 193 also bound efficiently to A549 cells ( Fig S2B) . We did not observe any binding 194 of the trimeric NTD domains to A549 cells ( Fig S2B) . MDCK cells, derived from 195 canine kidney, served as negative controls, to which we indeed did not observe 196 any binding with any of the indicated proteins ( Fig S2C) . applied. In all cases, SARS-CoV-2 displayed a higher avidity compared to 213 SARS-CoV-1. A similar trend of binding intensities was observed for 214 monomeric, trimeric, and different N-glycosylated SARS-CoV-RBD proteins 215 fused to mOrange2 ( Fig S3A) . Again specific binding was seen to the epithelium 216 of terminal bronchioles and, to a much lower extent, to alveoli and endothelium. 217

The results were confirmed using horseradish peroxidase readout with a 218 hematoxylin counterstain (Fig S3B) , which output is enzyme driven and purely 219 qualitative, however, we did observe similar differences in staining intensities. 220

Here, very minimal staining using the SARS-CoV NTD domains was observed 221 (Fig S3B) , which we did not detect using a fluorescent readout (Fig S3C) . To 222 determine if the binding was ACE2 dependent we pre-incubated trimeric RBD 223 proteins with recombinant ACE2. While 4 μM was sufficient to block binding to 224 cell culture cells (Fig 4C) , 8μM was needed to prevent all detectable binding to 225 ferret lung tissue. 226 227 To confirm our observations of different binding on tissues, we quantified the 228 intensities of the ACE2 antibody and SARS-CoV-1 and -2 RBD proteins, except 229 for the monomeric GnTI derived proteins as these were almost at the 230 background ( Fig 4D) . As expected a noteworthy trend was observed of 231 increasing binding strength from SARS-CoV1 GnTI derived monomers to 232 SARS-CoV-2 fully glycosylated RBD trimers. Interestingly multimerization 233 appears to be more important for strong ACE2 interaction to tissue compared 234 to the glycosylation status. Viroscience, Erasmus University, The Netherlands, respectively. Tissue 374 sections were rehydrated in a series of alcohol from 100%, 96% to 70%, and 375 lastly in distilled water. Tissues slides were boiled in citrate buffer pH 6.0 for 10 376 min at 900 kW in a microwave for antigen retrieval and washed in PBS-T three 377 times. Endogenous peroxidase activity was blocked with 1% hydrogen peroxide 378 for 30 min. Tissues were subsequently incubated with 3% BSA in PBS-T 379 overnight at 4 °C. The next day, the purified viral spike proteins (50μg/ml) were 

Human monoclonal 416 antibodies block the binding of SARS-CoV-2 spike protein to angiotensin 417 converting enzyme 2 receptor

Structural and 420 Functional Basis of SARS-CoV-2 Entry by Using Human ACE2

Functional assessment of cell entry and 424 receptor usage for SARS-CoV-2 and other lineage B betacoronaviruses

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

Epub 2020/03/11

A serological assay to detect SARS-CoV-2 seroconversion 433 in humans

A highly 435 conserved cryptic epitope in the receptor-binding domains of SARS-CoV-2 436 and SARS-CoV

Potent neutralizing antibodies from COVID-19 patients 441 define multiple targets of vulnerability

Conformational 445 dynamics of SARS-CoV-2 trimeric spike glycoprotein in complex with receptor 446 ACE2 revealed by cryo-EM

Closing coronavirus spike 449 glycoproteins by structure-guided design

Structures, 452 conformations and distributions of SARS-CoV-2 spike protein trimers on intact 453 virions

Stabilizing the Closed SARS-CoV-2 Spike Trimer

Structure-based Design of Prefusion-stabilized SARS-459

Fluorescent Trimeric Hemagglutinins Reveal Multivalent 463 Receptor Binding Properties

Structure-based design of prefusion

CoV-2 spikes. Science. 2020. Epub 2020/07/25

Comparative pathogenesis of COVID-19, MERS, and 472 SARS in a nonhuman primate model

Structural 476 basis of a shared antibody response to SARS-CoV-2

Severe Acute Respiratory Syndrome Coronavirus 2 from Patient with

Infection 483 and Rapid Transmission of SARS-CoV-2 in Ferrets

Okba 486 NMA, et al. SARS-CoV-2 is transmitted via contact and via the air between 487 ferrets

SARS-CoV-2 infection in farmed minks, the 492 Netherlands

Syrian hamsters as a small animal model for SARS-CoV-2 497 infection and countermeasure development

Severe Acute Respiratory Syndrome Coronavirus Infection of Golden Syrian 502

Pathogenesis and transmission of SARS-CoV-2 in golden hamsters

Deducing the N-and 508

O-glycosylation profile of the spike protein of novel coronavirus SARS-CoV-2

Site-specific 512 glycan analysis of the SARS-CoV-2 spike

Epub 2020/05/06

Site-516 specific N-glycosylation Characterization of Recombinant SARS-CoV-2 Spike 517

Glycans on the SARS-CoV-2 Spike Control the Receptor Binding 520 Domain Conformation

SARS-CoV-2 receptor ACE2 and TMPRSS2 are primarily expressed in 525 bronchial transient secretory cells

The protein expression profile 528 of ACE2 in human tissues

The SARS-CoV-2 Spike protein has a broad tropism for mammalian ACE2 532 proteins

Cell 534 entry of SARS-CoV-2 conferred by angiotensin-converting enzyme 2 (ACE2) 535 of different species

The 538 protein expression profile of ACE2 in human tissues

Tissue distribution of ACE2 protein, the functional receptor for SARS 542 coronavirus. A first step in understanding SARS pathogenesis

de Haan 546 CA. The influenza A virus hemagglutinin glycosylation state affects receptor-547 binding specificity

Improving the photostability of bright monomeric orange 551 and red fluorescent proteins

Three Amino Acid Changes in Avian Coronavirus Spike Protein 556

560 Leginon: a system for fully automated acquisition of 1000 electron 561 micrographs a day

Appion: An integrated, database-driven pipeline to facilitate EM image 565 processing

DoG 568 Picker and TiltPicker: Software tools to facilitate particle selection in single 569 particle electron microscopy

5μg of protein was subjected without or with PNGase F or EndoH for 1hr and 726 subjected to SDS-PAGE and western blot analyzes

Supplemental figure 2. Binding of RBD proteins to cell lines 733 (A) Protein binding of RBD proteins observed on VERO E6 cells

Proteins were applied 50μg/ml and where indicated pre-incubated with 735

Spike proteins were detected using anti-strep and 736 goat-anti-mouse antibodies

Binding of full-length SARS-CoV-2 ectodomain, IBV-M41, antibodies 738 only, and NTD spike proteins to A549 cells. Proteins were applied 50μg/ml 739 and detected using anti-strep and goat-anti-mouse antibodies

C) Non-binding of RBD trimers to MDCK cells

Binding of RBD proteins to tissues 747 (A) Binding of RBD proteins fused to mOrange2 to ferret lung tissues 748

Proteins were applied 50μg/ml and detected using anti-strep and goat-anti-749 mouse antibodies

Binding of RBD proteins fused to sfGFP proteins to ferret lung tissues, 751 using HRP as a readout

Identical experiment to (A) but using an HRP readout using anti-strep and goat-753 anti-mouse antibodies

Control staining on ferret lung tissues using HRP as readout

Proteins were 756 applied 50μg/ml and detected using anti-strep and goat-anti-mouse antibodies

Scalebar is 100μm

Lack of NTD binding to ferret lung tissue using fluorescence 759

Proteins were applied 50μg/ml and detected using anti-strep and goat-anti-760 mouse antibodies

Control stainings to Syrian hamster tissues, antibodies only and M41

Proteins were applied 50μg/ml and where indicated pre-incubated with 763 recombinant ACE2 protein