key: cord-0296113-6qqq64l2
authors: Maffei, Mariano; Montemiglio, Linda Celeste; Vitagliano, Grazia; Fedele, Luigi; Sellathurai, Shaila; Bucci, Federica; Compagnone, Mirco; Chiarini, Valerio; Exertier, Cécile; Muzi, Alessia; Roscilli, Giuseppe; Vallone, Beatrice; Marra, Emanuele
title: The nuts and bolts of SARS-CoV-2 Spike Receptor Binding Domain heterologous expression
date: 2021-09-17
journal: bioRxiv
DOI: 10.1101/2021.09.17.460782
sha: f7b1b941d6b0871c88cc59d512c83e6cc1b9cce4
doc_id: 296113
cord_uid: 6qqq64l2

COVID-19 is a highly infectious disease caused by a newly emerged coronavirus (SARS-CoV-2) that has rapidly progressed into a pandemic. This unprecedent emergency has stressed the significance of developing effective therapeutics to fight current and future outbreaks. The receptor-binding domain (RBD) of the SARS-CoV-2 surface Spike protein is the main target for vaccines and represents a helpful “tool” to produce neutralizing antibodies or diagnostic kits. In this work, we provide a detailed characterization of the native RBD produced in three major model systems: Escherichia coli, insect and HEK-293 cells. Circular dichroism, gel filtration chromatography and thermal denaturation experiments indicated that recombinant SARS-CoV-2 RBD proteins are stable and correctly folded. In addition, their functionality and receptor-binding ability were further evaluated through ELISA, flow cytometry assays and bio-layer interferometry.

At the end of 2019, a novel respiratory pathogen responsible for the COVID-19 39 disease, namely severe acute respiratory syndrome-related coronavirus (SARS-CoV-2), 40 emerged in Wuhan, China 1 . Only three months later, the virus spread worldwide causing 41 one of largest outbreak of the century that rapidly progressed into pandemic with more than 42 222 million of confirmed cases and 4,59 million deaths by September 9 th 2 . In response to 43 this exceptional situation, an enormous effort has been made by the scientific community to 44 study and characterize the pathogen and to quickly develop safe and effective prophylactic 45 and therapeutic drugs. 46 The SARS-CoV-2 is an enveloped virus whose surface is decorated with an integral 47 membrane protein (M), an envelope protein (E), a surface spike protein (S), and an 48 additional unexposed structural nucleocapsid protein (N) 3,4 . Among those, the Spike protein 49 is critical to recognize the host-cell receptors and for mediating viral entry, therefore it 50 represents the most studied viral component and the best candidate for drug target 5,6 . The 51 140 kDa SARS-CoV-2 S protein is organized into two major subunits (S1 and S2) connected 52 by a furin-cleavage site 7 . The S1 subunit contains the receptor-binding domain (RBD; aa 53 319-541), a 25 kDa domain that is directly involved in the interaction with the Angiotensin-54 converting enzyme 2 (ACE2) 8, 9 . RBD contains nine cysteines, including eight that form 55 disulfide-bridges involved in the RBD fold. In addition, the domain displays two N-56 glycosylation sites (Asn331 and Asn343) known to participate to folding, stability, and 57 function [10] [11] [12] . Mutations occurring within this domain are constantly monitored to predict the 58 emergence of novel variants that could be naturally selected and quickly spread, such as Rhamnose (data not shown). The target protein was totally recovered from inclusion bodies 104 with yields representing 5.2% (Star) and 8.1% (Lemo21) of the total protein extract ( Figure   105 1(b), Suppl. Figure 1 (a) ). Protein purification was carried out in the presence of denaturing 106 agents (6 M urea) followed by a slow refolding-process through an over-night dialysis 107 against buffer containing the redox pair of oxidized and reduced glutathione to induce proper 108 disulfide bond formation (Figure 1(b) , Suppl. Figure 1(b) ). As shown in Figure 1 (c), the 109 purified E. coli-RBD protein shows a high degree of purity (>90%) and migrated as a single-110 smeared band at the expected height on 4-12% SDS-PAGE (theoretical mass, 26052 Da).

Moreover, Western blot analysis indicated that the protein was efficiently recognized by anti-

His and anti-SARS-CoV-2 Spike S1 subunit antibodies (Figure 1(c) ). Approximately 1.25 mg 113 of purified RBD was obtained starting from 0.5 L of bacterial culture (final yield ~2.5 mg/L).

Among distinct batches, concentration ranged from 0.1 mg/mL (3.8 µM) to 0.3 mg/mL (11.5 115 µM). Concentrations higher than 0.3 mg/mL led to protein precipitation. Finally, the 116 molecular weight and primary aminoacidic sequence of SARS-CoV-2 RBD purified from E. detected by immunoblotting (anti-His and anti-SARS-CoV-2 Spike S1 subunit) (Figure 1(d) ).

At a laboratory-scale, final yields were around 6.5 mg RBD per Liter of insect cells with batch 130 concentrations ranging from 0.25 mg/mL (9.5 µM) to 0.5 mg/mL (19 µM). The experimental 131 mass (28936 Da) of the recombinant insect-RBD determined by MALDI mass-spectrometry 132 analysis (Suppl. Figure 2 Figure   143 2(e),(f)). Additionally, the eluted protein was efficiently detected by anti-His and anti-S1 were observed between plates, suggesting that PBS buffer is the most efficient buffer for 217 coating (data not shown).

Subsequently, plates were coated with increasing RBD protein concentrations 219 (ranging between 1 and 5 µg/mL) and serum from rats previously immunized with COVID-220 eVax 29 vaccine was applied to each plate for RBD protein binding. Of note, independently 221 from protein concentration (1, 3 and 5 μg/mL) both insect-RBD and HEK-293-RBD were 222 efficiently recognized by rat IgG, whereas rat IgG-E. coli-RBD interaction was much lower 223 (Figure 3(a) ). The observed differences between RBD produced in E. coli and in insect or 224 mammalian counterparts are probably due to the major affinity of the latter ones to the IgG 225 produced in rats. 226 We also monitored the interaction between RBD and a commercial antibody against 227 the S1 subunit of SARS-CoV-2 Spike. As shown in Figure 3 (b), the observed OD signal of 228 insect-RBD was not markedly different from that of HEK-293-RBD although at 229 concentrations < 1 μg/mL, RBD from insect showed a slightly higher binding ability 230 compared to HEK-293-RBD (Suppl. Figure 3(a),(b) ). In contrast, E. coli-RBD showed a lower 231 binding compared to HEK-293 and insect RBDs. We hypothesized that this lower binding 232 ability of RBD produced in E. coli may, again, be due to the presence of a sub-population of 233 the protein that failed refolding. Figure 4 shows that all the three studied RBDs were 251 able to efficiently bind Vero E6 cells, while no signal was observed when cells were 252 incubated only with antibodies (Suppl. Figure 4) . This result suggests that recombinant RBD 253 proteins are efficient in recognizing ACE2. insect-RBD gave a slightly better signal than HEK-293-RBD. We also investigated the 312 capability of isolated RBDs to bind the ACE2 receptor. All the RBDs produced in this work 313 efficiently bind to Vero E6 cells as confirmed by FACS assay. ACE2-RBD binding was further 314 confirmed and quantified by bio-layer interferometry, with the bacterial-RBD displaying again 315 the lowest binding efficiency. Our data suggest that the absence of glycosylations could 316 partially affect ACE-2 binding in-vitro as also previously observed 10 . In addition to this, we 317 must consider that the presence of a sub-population of protein that failed refolding, as 318 indicated by circular dichroism and ELISA assays, might also contribute to the observed 319 lower binding efficiency of the bacterial RBD.

To summarize, this work offers a technical and practical overview of RBD production 321 using the three most widely used expression systems, highlighting the main advantages and 322 drawbacks, reported in Table 1 . RBD obtained from both eukaryotic systems resulted in a 323 high-quality final bioproduct potentially eligible for diverse downstream applications (vaccine 324 design, diagnostic kits, drug screening etc.). However, the high costs (resources), the time-325 consuming production, the requirement of specific equipment, and access to dedicated 326 facilities could be a limitation for many laboratories or for the industrial production. By 327 contrast, the bacterial-derived RBD offers a low production cost, a broader availability, and 328 easy handling as main advantages, which make it more accessible. However, limitations in 329 the quality of the produced sample include the absence of glycosylation that partially affects (1):

where ∆GD-N is the free energy of the unfolding process, Tm is the melting temperature that 503 corresponds to midpoint of the thermal denaturation, ∆Hm is the enthalpy of denaturation at 504 the transition midpoint, and ∆Cp is the change of heat capacity of denaturation. The latter 505 parameter is related to the amount of hydrophobic area that becomes exposed to solvent Binding studies were carried out using the Octet Red system (Forte Bio Biosensors were soaked for 10 min in 1x kinetic buffer followed by a baseline signal 530 measurement for 60 s and then loaded with ACE2-hFc recombinant protein (10 µg/mL) for 531 300 s (until the biosensor was fully saturated). After a wash step in 1× kinetic buffer for 120 532 s, the ACE2-Fc-captured biosensor tips were then submerged for 300 s in wells containing 533 different concentrations of antigen (RBD E. coli, insect, and HEK-293) to evaluate 534 association curves, followed by 900 s of dissociation time in kinetic buffer. The ACE2-hFc 535 captured biosensor tips were also dipped in wells containing kinetic buffer to allow single 536 reference subtraction to compensate for the natural dissociation of captured ACE2-hFc.

Biosensor tips were used without regeneration.

The binding curve data were collected and then analysed using data analysis 

Severe acute respiratory syndrome 567 coronavirus 2 (SARS-CoV-2) and coronavirus disease-2019 (COVID-19): The epidemic and 568 the challenges

Characterization of 573 the receptor-binding domain (RBD) of 2019 novel coronavirus: implication for development 574 of RBD protein as a viral attachment inhibitor and vaccine

SARS-CoV-2 spike protein: a key target for eliciting persistent 580 neutralizing antibodies

Potential drug targets of SARS-CoV-2: From genomics to therapeutics

Furin Cleavage Site Is Key to SARS-CoV-2

Biomechanical 592 characterization of SARS-CoV-2 spike RBD and human ACE2 protein-protein interaction

Structural basis of receptor recognition by SARS-CoV-2

Nanoluciferase complementation-based 600 bioreporter reveals the importance of N-linked glycosylation of SARS-CoV-2 S for viral entry

Systemic effects of missense 603 mutations on SARS-CoV-2 spike glycoprotein stability and receptor-binding affinity

Beyond shielding: The roles of glycans in 607 the SARS-CoV-2 spike protein

Design of Specific Primer Sets for the Detection of B

SARS-CoV-2 Variants using Deep Learning

WorldPop : Preliminary risk analysis of the international 614 spread of new COVID-19 variants

Structure-function analyses of new sars-cov-2 variants 618 b.1.1.7, b.1.351 and b.1.1.28.1: Clinical, diagnostic, therapeutic and public health 619 implications

Application of newly developed SARS-CoV2 serology test along with real-time PCR for early 625 detection in health care workers and on-time plasma donation

Robust and 628 low-cost ELISA based on IgG-Fc tagged recombinant proteins to screen for anti-SARS-CoV-629 2 antibodies

SARS-CoV-2 spike protein: An optimal immunological target for vaccines

A serological assay to detect 636

SARS-CoV-2 seroconversion in humans

AA (2020) Structural and functional comparison of SARS-CoV-2-spike 639 receptor binding domain produced in Pichia pastoris and mammalian cells

CoV-2 Receptor Binding Domain Is Important for Functional Expression in Plants

Production and purification of recombinant proteins from Escherichia 646 coli

648 (2020) SARS-CoV-2 spike produced in insect cells elicits high neutralization titres in non-649 human primates

Deducing the N-and O-652 glycosylation profile of the spike protein of novel coronavirus SARS-CoV-2. Glycobiology 653 [Internet] 30:981

Site-specific glycan 655 analysis of the SARS-CoV-2 spike. Science (80-. )

The 2019 coronavirus (SARS-CoV-2) surface protein (Spike)

Receptor Binding Domain undergoes conformational change upon heparin binding

661 (2020) Structure of the SARS-CoV-2 spike receptor-binding domain bound to the ACE2 662 receptor

COVID-eVax, an electroporated plasmid DNA vaccine 666 candidate encoding the SARS-CoV-2 Receptor Binding Domain, elicits protective immune 667 responses in animal models of COVID-19

SARS-coronavirus-2 671 replication in Vero E6 cells: Replication kinetics, rapid adaptation and cytopathology

Analysis of ACE2 in polarized epithelial cells: 676 Surface expression and function as receptor for severe acute respiratory syndrome-677 associated coronavirus

SARS-CoV-2 Vaccines

An overview of current COVID-19 vaccine platforms

Anon COVID-19 Studies from the World Health Organization Database -684

Receptor-686 binding domain of SARS-Cov spike protein: Soluble expression in E.coli, purification and 687 functional characterization

Bacterial expression and purification of functional 690 recombinant SARS-CoV-2 spike receptor binding domain

Accurate 694 secondary structure prediction and fold recognition for circular dichroism spectroscopy

Proc. Natl. Acad. Sci

BeStSel: a web server for accurate protein secondary structure prediction 699 and fold recognition from the circular dichroism spectra

Monitoring protein aggregation during 703 thermal unfolding in circular dichroism experiments

Structure and mechanism in protein science. A guide to enzyme 706 catalysis and protein folding, by A. Fersht

Denaturant m values and heat capacity changes: 710 relation to changes in accessible surface areas of protein unfolding

Additional supporting information may be found online in the Supporting Information section

We acknowledge the assistance of L. Salvini and L.Tinti of the Toscana Life Science 562 facilities for the support with mass spectrometry measurements and analysis of proteins.