key: cord-0707369-1n0qw2a5
authors: Hoffmann, Markus; Kleine-Weber, Hannah; Pöhlmann, Stefan
title: A Multibasic Cleavage Site in the Spike Protein of SARS-CoV-2 Is Essential for Infection of Human Lung Cells
date: 2020-05-01
journal: Mol Cell
DOI: 10.1016/j.molcel.2020.04.022
sha: ee56b7782fa23e1977458001ef662bf21b829a35
doc_id: 707369
cord_uid: 1n0qw2a5

Summary The pandemic coronavirus SARS-CoV-2 threatens public health worldwide. The viral spike protein mediates SARS-CoV-2 entry into host cells and harbors a S1/S2 cleavage site containing multiple arginine residues (multibasic) not found in closely related animal coronaviruses. However, the role of this multibasic cleavage site in SARS-CoV-2 infection is unknown. Here, we report that the cellular protease furin cleaves the spike protein at the S1/S2 site and that cleavage is essential for S-protein-mediated cell-cell fusion and entry into human lung cells. Moreover, optimizing the S1/S2 site increased cell-cell, but not virus-cell, fusion, suggesting that the corresponding viral variants might exhibit increased cell-cell spread and potentially altered virulence. Our results suggest that acquisition of a S1/S2 multibasic cleavage site was essential for SARS-CoV-2 infection of humans and identify furin as a potential target for therapeutic intervention.

Coronavirus spike proteins are activated by host cell proteases. Hoffmann and colleagues show that the pandemic SARS-CoV-2 harbors a highly cleavable S1/S2 cleavage site not found in closely related coronaviruses. Cleavage at this site is mediated by furin and is required for viral entry into human lung cells.

It is believed that the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2, previously termed nCoV-2019) was introduced into the human population from a poorly characterized animal reservoir in late 2019 (Ge et al., 2013; Wang et al., 2020; Zhou et al., 2020b; Zhu et al., 2020) . The epicenter of the subsequent SARS-CoV-2 spread was Wuhan, Hubei province, China, with more than 65,000 cases occurring in this area (WHO, 2020a) . However, infections have now been detected in more than 110 countries and massive outbreaks are currently ongoing in the United States, Italy, and Spain (WHO, 2020a (WHO, , 2020b . Understanding which features of SARS-CoV-2 are essential for infection of human cells should provide insights into viral transmissibility and pathogenesis and might reveal targets for intervention.

The spike protein of coronaviruses is incorporated into the viral envelope and facilitates viral entry into target cells. For this, the surface unit S1 binds to a cellular receptor while the transmembrane unit S2 facilitates fusion of the viral membrane with a cellular membrane Hulswit et al., 2016; Millet and Whittaker, 2018) . Membrane fusion depends on S protein cleavage by host cell proteases at the S1/S2 and the S2 0 site ( Figure 1A ), which results in S protein activation Hulswit et al., 2016; Millet and Whittaker, 2018) . Cleavage of the S protein can occur in the constitutive secretory pathway of infected cells or during viral entry into target cells and is essen-tial for viral infectivity. Therefore, the responsible enzymes constitute potential targets for antiviral intervention.

Our previous work revealed that the activity of the cellular serine protease TMPRSS2, which activates several coronaviruses Gierer et al., 2013; Glowacka et al., 2011; Matsuyama et al., 2010; Shirato et al., 2013 Shirato et al., , 2016 Shulla et al., 2011) , is also required for robust SARS-CoV-2 infection of human lung cells (Hoffmann et al., 2020) . However, it is conceivable that the activity of other cellular proteases is also necessary. Thus, the Middle East respiratory syndrome coronavirus spike protein (MERS-S) is activated by a two-step process: MERS-S is first cleaved by furin at the S1/S2 site in infected cells, which is required for subsequent TMPRSS2-mediated cleavage at the S2 0 site ( Figure 1A ) during viral entry into lung cells (Kleine-Weber et al., 2018; Park et al., 2016; Millet and Whittaker, 2014) . A cathepsin B/L-dependent auxiliary activation pathway is operative in many TMPRSS2 À cell lines but seems not to be available in viral target cells in the lung because TMPRSS2dependent activation of the S protein is essential for robust MERS-CoV and SARS-CoV spread and pathogenesis in the infected host (Iwata-Yoshikawa et al., 2019; Simmons et al., 2005; Zhou et al., 2015) .

The S1/S2 site in SARS-CoV-2 forms an exposed loop ( Figure 1B ) that harbors multiple arginine residues (multibasic) (Walls et al., 2020; Wrapp et al., 2020) that are not found in SARS-CoV-related coronaviruses (SARSr-CoV) but are present in the human coronaviruses OC43, HKU1, and MERS-CoV ( Figure 1C ). However, the contribution of this multibasic cleavage site to SARS-CoV-2 infection of human cells is unknown and was in the focus of the present study.

The Multibasic S1/S2 Site in the Spike Protein of SARS-CoV-2 Is Required for Efficient Proteolytic Cleavage of the Spike Protein In order to address the role of the multibasic S1/S2 cleavage site in SARS-CoV-2 infection, we generated S protein mutants with (B) Protein models for SARS-S and SARS-2-S based on the PDB: 5X5B structure as a template. Colored in red are the S1/S2 and S2 0 cleavage sites. Further, the S1 subunit (blue), including the RBD (purple), and the S2 subunit (gray) are depicted. (C and D) Amino acid sequence alignment of residues around the S1/S2 and S2 0 cleavage sites of group 2b betacoronaviruses found in humans, civet cats, raccoon dog, pangolin, and bats (C) or coronaviruses that are able to infect humans (D). Basic amino acid residues are highlighted in red, while gray boxes mark the presence of multibasic motifs. Numbers refer to amino acid residues (n/a, no information available). The symbol ''*'' refers to amino acid residues that are conserved among all tested sequences, while the symbols '':'' and ''.'' indicate positions with heterogeneous amino acid residues that share highly similar or similar biochemical properties. altered S1/S2 cleavage sites ( Figure 2A ). In particular, we exchanged the multibasic cleavage site against its monobasic counterparts present in SARS-S or RaTG13-S (Figure 2A , RaTG13 is a bat coronavirus closely related to SARS-CoV-2 [Zhou et al., 2020b] ). This resulted in mutants SARS-2-S (SARS) and SARS-2-S (RaTG). Moreover, we either deleted all arginines in the S1/S2 site of SARS-2-S or inserted an additional arginine residue (jointly with an alanine to lysine exchange), giving rise to mutants SARS-2-S (delta) and SARS-2-S (opt), respectively. Finally, we introduced the S1/S2 sites of SARS-2-S and RaTG13-S into the background of SARS-S (Figure 2A ), which yielded the mutants SARS-S (SARS-2) and SARS-S (RaTG).

The effects of the above described S1/ S2 mutations on viral entry were examined using vesicular stomatitis virus (VSV) particles bearing S proteins because these particles are safe and adequately reflect coronavirus entry into target cells. Immunoblot of VSV particles bearing S proteins with a C-terminal antigenic tag revealed that all S proteins were readily incorporated into VSV particles. SARS-2-S WT (wild type) was efficiently cleaved at the S1/S2 site ( Figure 2B ), in keeping with published data (Hoffmann et al., 2020; Walls et al., 2020) . Exchange of the S1/S2 site of SARS-2-S against those of SARS-S and RaTG13-S abrogated cleavage, and this effect was also seen when the multibasic motif was deleted ll Short Article ( Figure 2B ). Moreover, insertion of an additional arginine residue jointly with an alanine-to-lysine exchange at the S1/S2 site did not appreciably increase cleavability. Finally, insertion of the S1/S2 site of SARS-2-S into SARS-S increased S protein cleavability while insertion of the RaTG13 S1/S2 site did not (Figure 2B) . These results indicate that the presence of several arginine residues at the S1/S2 site is required for efficient SARS-2-S proteolytic processing in human cells and also confers high cleavability to SARS-S.

Furin Cleaves the SARS-CoV-2 Spike Protein at the S1/ S2 Site, and Cleavage Is Required for Efficient Cell-Cell Fusion We next investigated which protease is required for S protein processing at the S1/S2 site. The S1/S2 motif matches the minimal furin sequence RXXR and is closely related to the furin consensus sequence RX[K/R]R. Therefore, we analyzed whether decanoyl-RVKR-CMK, a furin inhibitor, blocks SARS-2-S pro-cessing at the S1/S2 site. Decanoyl-RVKR-CMK inhibited processing of MERS-S, which is known to depend on furin (Gierer et al., 2015; Millet and Whittaker, 2014) , in a concentrationdependent manner and had no effect on SARS-S expression ( Figure 2C ), as expected. Processing of SARS-2-S was also inhibited, indicating that furin cleaves SARS-2-S at the S1/S2 site. In order to determine whether cleavage at the S1/S2 site is required for SARS-2-S-driven cell-cell fusion, we studied S-protein-dependent formation of multinucleated giant cells (syncytia). No syncytia were observed in the absence of S protein expression while MERS-S WT expression resulted in syncytium formation, which was increased upon addition of trypsin or expression of TMPRSS2 ( Figure 2D ). Expression of SARS-S WT or SARS-S harboring the S1/S2 site of RaTG13-S did not induce syncytium formation in the absence of protease, but modest multikaryon formation was detected in the presence of trypsin or TMPRSS2. In contrast, SARS-S harboring the SARS-2-S S1/S2 site induced syncytia in the absence of Transduction efficiency is shown relative to that measured for particles not bearing a viral glycoprotein. Error bars indicate the standard error of the mean. Statistical significance was tested by one-way analysis of variance with Dunnett's post test (p > 0.05, ns; ***p % 0.001).

Molecular Cell 78, 1-6, May 21, 2020 3 protease, and syncytium formation was markedly increased by trypsin and, particularly, TMPRSS2. SARS-2-S expression triggered syncytium formation that was strongly increased by trypsin and TMPRSS2. Syncytium formation was clearly less prominent and required the presence of trypsin or TMPRSS2 when the SARS-2-S S1/S2 site was replaced by that of SARS-S or RaTG13-S. Moreover, deletion of the multibasic motif resulted in a spike protein that was no longer able to induce syncytium formation even in the presence of trypsin or TMPRSS2. Finally, the addition of an arginine residue to the S1/S2 site of SARS-2-S jointly with alanine-to-lysine exchange strongly increased syncytium formation, indicating that viral variants with optimized S1/S2 sites might show augmented cell-cell spread and potentially altered pathogenicity. Thus, the S1/S2 site of SARS-2-S is required for cell-cell fusion, and this process can be augmented by adding basic residues to the S1/S2 site.

We finally examined the importance of the S1/S2 site for S-protein-mediated virus-cell fusion. Blockade of SARS-2-S cleavage at the S1/S2 site (mutants SARS-2-S (SARS), SARS-2-S (RaTG), and SARS-2-S (delta)) abrogated entry into the TMPRSS2 + human lung cell line Calu-3 ( Figure 2E ), in which the cathepsin B/ L-dependent S protein activation pathway is not sufficiently available (Park et al., 2016) . In contrast, entry into TMPRSS2 À Vero cells, which is known to be cathepsin B/L dependent, was not affected by these mutations (Figure 1E ), in keeping with results reported by Walls and colleagues (Walls et al., 2020) . Optimization of the S1/S2 site did not increase entry into the cell lines tested; it slightly decreased entry into both Vero and Calu-3 cells, for, at present, unclear reasons. Finally, alterations of the S1/S2 site of SARS-S did not augment entry efficiency. Collectively, these results demonstrate that a multibasic S1/S2 site is essential for SARS-2-S-driven entry into human lung cells while a monobasic site is sufficient for SARS-S.

Our results reveal commonalities between the proteolytic activation of SARS-CoV-2 and MERS-CoV. Both viruses depend on furin-mediated pre-cleavage of their S proteins at the S1/S2 site for subsequent S protein activation by TMPRSS2 in lung cells, which fail to express robust levels of cathepsin L (Park et al., 2016) . Thus, inhibitors of furin and TMPRSS2 might be considered as a treatment option for COVID-19, and a TMPRSS2 inhibitor that blocks SARS-CoV-2 infection has recently been described (Hoffmann et al., 2020) . Regarding furin inhibition, it must be taken into account that furin, unlike TMPRSS2, is required for normal development (Roebroek et al., 1998) . Blockade of this enzyme for prolonged time periods might thus be associated with unwanted toxic effects. In contrast, a brief treatment might be well tolerated and still associated with a therapeutic benefit (Sarac et al., 2002 (Sarac et al., , 2004 .

For avian influenza A viruses, a multibasic cleavage site in the viral hemagglutinin protein is a central virulence factor (Luczo et al., 2015) . Thus, viruses with a monobasic cleavage site are activated by TMPRSS2 or related proteases with an expression profile confined to the aerodigestive tract. As a consequence, viral replication is limited to these organs and does not result in severe disease. In contrast, viruses with a multibasic cleavage site are activated by ubiquitously expressed proprotein convertases, including furin, and can thus spread systemically and cause massive disease. In the context of coronavirus infection, S protein cleavability has been identified as a determinant of zoonotic potential (Menachery et al., 2020; Yang et al., 2014) . The presence of a highly cleavable S1/S2 site in SARS-2-S may therefore not have been unexpected. However, it is noteworthy that all SARS-CoV-2-related coronaviruses of bats and pangolins identified today harbor a monobasic cleavage site (Lam et al., 2020; Li et al., 2020; Zhang et al., 2020) . It will thus be interesting to determine how the multibasic motif was acquired by SARS-CoV-2, and a recent study suggested that a recombination event might have been responsible Zhou et al., 2020a) .

Our results demonstrate that the multibasic S1/S2 cleavage site is essential for SARS-2-S-driven entry into TMPRSS2 + lung cells. It will be interesting to extend these studies to primary human respiratory epithelial cells and to authentic SARS-CoV-2, which requires a reverse genetics system not available to the present study.

Detailed methods are provided in the online version of this paper and include the following: 

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Stefan Pö hlmann (spoehlmann@dpz.eu).

All unique/stable reagents generated in this study are available from the Lead Contact with a completed Materials Transfer Agreement.

The study did not generate unique datasets or code.

Cell cultures 293T (human, kidney) and Vero (African green monkey, kidney) cells were cultivated in Dulbecco's Modified Eagle Medium (PAN-Biotech) supplemented with 10% fetal bovine serum (Biochrom), 100 U/mL of penicillin and 0.1 mg/mL of streptomycin (PAN-Biotech). Vero cells that stably express human TMPRSS2 have been described previously (Hoffmann et al., 2020) and were cultivated in the presence of 10 mg/mL blasticidin (Invivogen). Calu-3 (human, lung; kindly provided by Stephan Ludwig, Westf€ alische Wilhelms-Universit€ at, Muenster/Germany) cells were cultivated in Minimum Essential Medium (Thermo Fisher Scientific) supplemented with 10% fetal bovine serum (Biochrom), 100 U/mL of penicillin and 0.1 mg/mL of streptomycin (PAN-Biotech), 1x non-essential amino acid solution (from 100x stock, PAA) and 10 mM sodium pyruvate (Thermo Fisher Scientific). All cell lines were incubated at 37 C and 5% CO 2 in a humidified atmosphere.

A vesicular stomatitis virus replicon-based bioassay for the rapid and sensitive determination of multi-species type I interferon

The glycoprotein of vesicular stomatitis virus promotes release of virus-like particles from tetherin-positive cells

Isolation and characterization of a bat SARS-like coronavirus that uses the ACE2 receptor

The spike protein of the emerging betacoronavirus EMC uses a novel coronavirus receptor for entry, can be activated by TMPRSS2, and is targeted by neutralizing antibodies

Priming Time: How Cellular Proteases Arm Coronavirus Spike Proteins

Differential sensitivity of bat cells to infection by enveloped RNA viruses: coronaviruses, paramyxoviruses, filoviruses, and influenza viruses

SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor

Coronavirus Spike Protein and Tropism Changes

TMPRSS2 Contributes to Virus Spread and Immunopathology in the Airways of Murine Models after Coronavirus Infection

Functional analysis of potential cleavage sites in the MERS-coronavirus spike protein

Mutations in the Spike Protein of Middle East Respiratory Syndrome Coronavirus Transmitted in Korea Increase Resistance to Antibody-Mediated Neutralization

YASARA View -molecular graphics for all devices -from smartphones to workstations

Identifying SARS-CoV-2 related coronaviruses in Malayan pangolins

Evolutionary history, potential intermediate animal host, and cross-species analyses of SARS-CoV-2

Molecular pathogenesis of H5 highly pathogenic avian influenza: the role of the haemagglutinin cleavage site motif

The EMBL-EBI search and sequence analysis tools APIs in 2019

Efficient activation of the severe acute respiratory syndrome coronavirus spike protein by the transmembrane protease TMPRSS2

Trypsin Treatment Unlocks Barrier for Zoonotic Bat Coronaviruses Infection

Host cell entry of Middle East respiratory syndrome coronavirus after two-step, furin-mediated activation of the spike protein

Physiological and molecular triggers for SARS-CoV membrane fusion and entry into host cells

Proteolytic processing of Middle East respiratory syndrome coronavirus spikes expands virus tropism

Failure of ventral closure and axial rotation in embryos lacking the proprotein convertase Furin

The furin inhibitor hexa-Darginine blocks the activation of Pseudomonas aeruginosa exotoxin A in vivo

Protection against anthrax toxemia by hexa-D-arginine in vitro and in vivo

Comparison of vesicular stomatitis virus pseudotyped with the S proteins from a porcine and a human coronavirus

Middle East respiratory syndrome coronavirus infection mediated by the transmembrane serine protease TMPRSS2

Clinical Isolates of Human Coronavirus 229E Bypass the Endosome for Cell Entry

Inhibitors of cathepsin L prevent severe acute respiratory syndrome coronavirus entry

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

A novel coronavirus outbreak of global health concern

Novel Coronavirus(2019-nCoV) Situation Report 52

Coronavirus disease 2019 (COVID-19) Situation Report -81

Cryo-EM structure of the 2019-nCoV spike in the prefusion conformation

Receptor usage and cell entry of bat coronavirus HKU4 provide insight into bat-to-human transmission of MERS coronavirus

Cryo-EM structures of MERS-CoV and SARS-CoV spike glycoproteins reveal the dynamic receptor binding domains

Probable Pangolin Origin of SARS-CoV-2 Associated with the COVID-19 Outbreak

Protease inhibitors targeting coronavirus and filovirus entry

A novel bat coronavirus reveals natural insertions at the S1/S2 cleavage site of the Spike protein and a possible recombinant origin of HCoV-19

A pneumonia outbreak associated with a new coronavirus of probable bat origin

A Novel Coronavirus from Patients with Pneumonia in China

Based on the SARS-S and SARS-2-S expression plasmids we cloned mutated versions with alterations at the S1/S2 cleavage site: We generated SARS-S containing the cleavage site of SARS-2-S, SARS-S (SARS-2), or BetaCoV/bat/Yunnan/RaTG13/2013 (RaTG; GISAID: EPI_ISL_402131), SARS-S (RaTG). Further, we generated SARS-2-S harboring the S1/S2 cleavage site of SARS-S, SARS-2-S (SARS) or RaTG-S, SARS-2-S (RaTG). Finally, we constructed SARS-2-S variants in which either the multibasic motif was deleted, SARS-2-S (delta), or in which the proline residue preceding the multibasic motif was mutated to arginine and the alanine residue within the minimal furin motif was changed to lysine in order to increase the basic environment at the S1/S2 site, SARS-2-S (opt)

At 16 h posttransfection, the cells were inoculated with VSV*DG-fLuc (kindly provided by Gert Zimmer, Institute of Virology and Immunology, Mit-telh€ ausern/Switzerland), a replication-deficient VSV vector that lacks the genetic information for VSV-G and encodes for eGFP and firefly luciferase (fLuc), at a multiplicity of infection of 3. After 1 h of incubation, the inoculum was removed and cells were washed with phosphate-buffered saline (PBS) before medium containing anti

Cells were further incubated for 16 h, before the VSVpp containing supernatants were harvested, freed from cellular debris by centrifugation and used for experiments. For transduction, target cells were grown in 96-well plates until they reached 50%-80% confluency. The culture supernatant was removed by aspiration and 100 ml/well of the respective pseudotype were added (quadruplicate samples)

Next, we removed 1 mL of supernatant, added 50 ml of 2x SDS-sample buffer and incubated the samples for 15 min at 96 C. Thereafter, the samples were subjected to SDS-PAGE and protein transfer to nitrocellulose membranes by western blot. The membranes were subsequently blocked in 5% skim milk solution (PBS containing 0.05% Tween-20 [PBS-T] and 5% skim milk powder) for 1 h at room temperature. The blots were then incubated over night at 4 C with primary antibody solution (all antibodies were diluted in PBS-T containing 5% skim milk; mouse anti-HA tag [Sigma-Aldrich, H3663, 1:2,500], mouse anti-V5 tag

GenBank: KJ473816.1), bat SARSr-CoV Rs/HuB2013 (GenBank: KJ473814.1), bat SARSr-CoV Rs/GX2013 (GenBank: KJ473815.1), bat SARSr-CoV Rf/SX2013 (GenBank: KJ473813.1), bat SARSr-CoV Rf/JL2012 (GenBank: KJ473811.1), bat SARSr-CoV Rf/HeB2013 (GenBank: KJ473812.1), bat SARSr-CoV YNLF/34C (GenBank: KP886809.1), bat SARSr-CoV YNLF/31C (GenBank: KP886808.1), bat SARSr-CoV Rs672

QUANTIFICATION AND STATISTICAL ANALYSIS If not stated otherwise, statistical significance was tested by one-way analysis of variance with Dunnet's posttest (GraphPad Prism 7.03)

We thank Inga Nehlmeier for technical assistance. We gratefully acknowledge the authors and the originating and submitting laboratories for their sequence and metadata shared through GISAID, on which this research is based. This work was supported by the Bundesministerium f€ ur Bildung und Forschung (BMBF, RAPID Consortium, 01KI1723D to S.P.). We further like to thank Andrea Maisner and Stephan Ludwig for providing the Vero cells and Calu-3 cells, respectively.