1. Biochemistry and Chemical Biology
  2. Structural Biology and Molecular Biophysics

SARS-CoV-2 S protein:ACE2 interaction reveals novel allosteric targets

  1. Palur V Raghuvamsi
  2. Nikhil K Tulsian
  3. Firdaus Samsudin
  4. Xinlei Qian
  5. Kiren Purushotorman
  6. Gu Yue
  7. Mary M Kozma
  8. Wong Y Hwa
  9. Julien Lescar
  10. Peter J Bond  Is a corresponding author
  11. Paul A MacAry  Is a corresponding author
  12. Ganesh S Anand  Is a corresponding author
  1. Department of Biological Sciences, National University of Singapore, Singapore
  2. Bioinformatics Institute, Agency for Science, Technology, and Research (A*STAR), Singapore
  3. Centre for Life Sciences, Department of Biochemistry, National University of Singapore, Singapore
  4. Life Sciences Institute, Centre for Life Sciences, National University of Singapore, Singapore
  5. School of Biological Sciences, Nanyang Technological University, Singapore
  6. Current address: Department of Chemistry, Department of Biochemistry and Molecular Biology, Center for Infectious Disease Dynamics -Huck Institute of the Life Sciences, The Pennsylvania State University, United States
https://doi.org/10.7554/eLife.63646
Share this article
Cite this article
  1. Palur V Raghuvamsi
  2. Nikhil K Tulsian
  3. Firdaus Samsudin
  4. Xinlei Qian
  5. Kiren Purushotorman
  6. Gu Yue
  7. Mary M Kozma
  8. Wong Y Hwa
  9. Julien Lescar
  10. Peter J Bond
  11. Paul A MacAry
  12. Ganesh S Anand
(2021)
SARS-CoV-2 S protein:ACE2 interaction reveals novel allosteric targets
eLife 10:e63646.
https://doi.org/10.7554/eLife.63646
  2. Download BibTeX
  3. Download .RIS

Abstract

The spike (S) protein is the main handle for SARS-CoV-2 to enter host cells via surface angiotensin-converting enzyme 2 (ACE2) receptors. How ACE2 binding activates proteolysis of S protein is unknown. Here, using amide hydrogen–deuterium exchange mass spectrometry and molecular dynamics simulations, we have mapped the S:ACE2 interaction interface and uncovered long-range allosteric propagation of ACE2 binding to sites necessary for host-mediated proteolysis of S protein, critical for viral host entry. Unexpectedly, ACE2 binding enhances dynamics at a distal S1/S2 cleavage site and flanking protease docking site ~27 Å away while dampening dynamics of the stalk hinge (central helix and heptad repeat [HR]) regions ~130 Å away. This highlights that the stalk and proteolysis sites of the S protein are dynamic hotspots in the prefusion state. Our findings provide a dynamics map of the S:ACE2 interface in solution and also offer mechanistic insights into how ACE2 binding is allosterically coupled to distal proteolytic processing sites and viral–host membrane fusion. Thus, protease docking sites flanking the S1/S2 cleavage site represent alternate allosteric hotspot targets for potential therapeutic development.

Introduction

The COVID-19 pandemic caused by the SARS-CoV-2 virus has sparked extensive efforts to map molecular details of its life cycle to drive vaccine and therapeutic discovery (Bar-Zeev and Inglesby, 2020). SARS-CoV-2 belongs to the family of Coronaviridae, which includes other human pathogens including common cold-causing viruses (hCoV-OC43, HKU, and 229E), SARS, and MERS-CoV (Corman et al., 2018; St-Jean et al., 2004; Lau et al., 2006; Warnes et al., 2015). SARS-CoV-2 has an ~30 kb RNA (positive stranded) genome with 14 open reading frames, encoding four structural proteins – spike (S) protein, membrane (M) protein, envelope (E) protein, and nucleo-protein; 16 non-structural proteins, and 9 accessory proteins (Su et al., 2016; Andersen et al., 2020; Tan et al., 2005) An intact SARS-CoV-2 virion consists of a nucleocapsid core containing genomic RNA within a lipid–protein envelope forming a spherical structure of diameter ~100 nm (Ke et al., 2020). The viral envelope is decorated with S, M, and E proteins (Ke et al., 2020). The prefusion S protein is a club-shaped homotrimeric class I viral fusion protein that has distinctive ‘head’ and ‘stalk’ regions (Figure 1A).

Figure 1
Download asset Open asset
Structure and domain organization of trimeric spike (S) protein showing steps in the virus–host entry initiated by S recognition and binding to angiotensin-converting enzyme 2 (ACE2) receptor.

(A) Prefusion S protein trimer in closed conformational state, with monomers shown in yellow, green, and pink. S protein construct (1–1245) used in this study showing head, stalk, and transmembrane (TM) segments as generated by integrative modeling. The S1/S2 and S2′ cleavage sites are in red. Proteolytic processing (furin) of S protein generates S1 and S2 subunits. (B) Schematic of viral entry into host cell mediated by S:ACE2 interactions as previously outlined (Shang et al., 2020): (i) Intrinsic dynamics of prefusion S protein trimer decorating SARS-CoV-2 and host ACE2 dimeric structure showing sweeping motions of S protein and ACE2 to facilitate S:ACE2 recognition. (ii) In the open conformation (Sopen), receptor binding domain adopts an ‘up’ orientation to recognize and bind the host membrane-bound ACE2 receptor (Protein Data Bank [PDB] ID: 1R42). ACE2 binding induces conformational changes promoting Furin* (red) proteolysis at the S1/S2 cleavage site (red arrows), leading to dissociation of S1 and S2 subunits, the mechanism of which is unknown. *Furin here also denotes relevant related proteases. (iii) The residual ACE2-bound S1 subunit becomes stably bound to ACE2 and S2 subunits dissociate. (iv) Conformational changes in the separated S2 subunit promote formation of an extended helical fusogenic intermediate (PDB ID: 6M3W) (Fan et al., 2020) for fusion into the host cell membrane, membrane fusion, and viral entry into the host cell (Hoffmann et al., 2020).

A characteristic feature of SARS-CoV-2 is that upon host entry, its prefusion S protein is proteolyzed by host proteases into constituent S1 and S2 subunits. The S1 subunit comprises an N-terminal domain (NTD) and a receptor binding domain (RBD) that interacts with the host receptor angiotensin-converting enzyme-2 (ACE2) (Lan et al., 2020; Hoffmann et al., 2020) to initiate viral entry into the target cell (Yan et al., 2020). The defining virus–host interaction for entry is therefore that mediated by the viral S protein with the host ACE2 receptor (Lan et al., 2020). Binding to ACE2 primes the S protein for proteolysis by host furin proteases at the S1/S2 cleavage site (Walls et al., 2017; Vankadari, 2020). The S2 subunit consists of six constituent domains harboring the membrane fusion machinery of the virus. These comprise the fusion peptide (FP), heptad repeat (HR1), central helix (CH), heptad repeat 2 (HR2), connector domain (CD), transmembrane domain (TM), and cytoplasmic tail (CT) domain (Walls et al., 2020; Wrapp et al., 2020). Extensive structural studies (Ke et al., 2020; Walls et al., 2020; Fan et al., 2020; Turoňová et al., 2020) have captured S protein of coronaviruses in distinct open (PDB: 6VXX) (Walls et al., 2020) and closed (PDB: 6VYB) (Walls et al., 2020) conformational states relative to the orientation of the RBD. These structures additionally reveal distinct orientations of the ectodomain (ECD) in pre- and postfusion states and highlight the intrinsic ensemble nature of the S protein in solution. The S2 subunit promotes host–viral membrane fusion and viral entry (Figure 1B).

Despite extensive cryo-Electron Microscopy (cryo-EM) studies, a map of the S:ACE2 interface in solution and how ACE2 binding to the RBD primes enhanced proteolytic processing at the S1/S2 site is entirely unknown. Amide hydrogen/deuterium exchange mass spectrometry (HDXMS), together with molecular dynamics (MD) simulations, offers a powerful combined approach for describing virus protein conformational dynamics and breathing (Lim et al., 2017a) and mapping protein–protein interactions for host receptor–virus interactions (Lim et al., 2017b). Here, we describe dynamics of free S protein and S:ACE2 complex, which reveal allosteric effects of ACE2 binding-induced conformational changes at distal stalk and protease docking sites flanking the S1/S2 cleavage sites. Our studies uncover distal ‘hotspots’ critical for the first step of the SARS-CoV-2 infection and thereby represent novel targets beyond the RBD for therapeutic intervention.

Results and discussion

Subunit-specific dynamics and domain motions of S protein trimer

Structural snapshots of the ACE2 binding interface with the SARS-CoV-2 S protein have previously been described for the RBD alone (Lan et al., 2020; Wrapp et al., 2020; Ali and Vijayan, 2020; Chan et al., 2020; Wang et al., 2020a). In this study, we have expanded this to map interactions and dynamics of ACE2 binding with a larger S protein construct, S (1–1208), lacking only the C-terminal membrane spanning helices. Mutations at the S1/S2 cleavage site (PRRAS motif substituted by PGSAS motif) and 986–987 (KV substituted PP) were engineered (Wrapp et al., 2020) to block host cell-mediated S protein proteolysis during expression and purification (Figure 2—figure supplement 1). S (1–1208), ACE2, and RBD eluted as trimers, dimers, and monomers, respectively, on size-exclusion chromatography (Figure 2—figure supplement 1, Figure 3—figure supplement 1, and Figure 5—figure supplement 1). S protein hereafter in the text denotes S (1–1208). Isolated RBD constructs showed high-affinity binding to ACE2 (Figure 3—figure supplement 1, Figure 5—figure supplement 1).

HDXMS of S protein alone was next carried out as described in 'Materials and methods'. Pepsin proteolysis of the S protein generated 317 peptides with high signal to noise ratios, yielding a primary sequence coverage of ~87% (Figure 2—figure supplement 2). S protein is highly glycosylated (at least 22 sites have been predicted and characterized on S protein) (Watanabe et al., 2020). Of these, 20 sites are predicted to be N-linked glycosylation modifications. We obtained peptides spanning 12 of the 20 predicted glycosylation sites. None of these peptides were glycosylated, making deuterium exchange of non-glycosylated peptides the focus of this study.

HDXMS results were overlaid onto integrative models of the full-length S protein trimer built using experimental structures of prefusion S ECD in the open conformation (PDB ID: 6VSB) (Wrapp et al., 2020) and HR2 domain from SARS S protein as templates. A deuterium exchange heat map (t = 1 and 10 min) revealed the stalk region to show the greatest relative deuterium exchange (Figure 2A). This is consistent with earlier studies showing at least 60° sweeping motions of the three identified hinge regions of the stalk (Turoňová et al., 2020). This was further verified via all-atom MD simulations of the S protein model embedded in a viral model membrane, which showed significant motions of the S protein ECD resulting from the high flexibility of the stalk region (Figure 2B), combined with large atomic fluctuations around the HR2 domain, compared to the rest of the protein (Figure 2—figure supplement 3, Figure 2—figure supplement 4).

Figure 2 with 5 supplements see all
Download asset Open asset
Deuterium exchange heat map and molecular dynamics simulations reveal domain-specific conformational dynamics of prefusion spike (S) protein trimer.

(A) Deuterium exchange at t = 1 min deuterium exchange mapped onto the structure of S protein (shades of blue [low exchange] and red [high exchange]). (B) Per-residue root mean square fluctuation (RMSF) of the S protein mapped onto the surface of the S trimer. Deuterium exchange-based dynamics across N-terminal domain (NTD) (C), receptor binding domain (RBD) (D), and the S2 subunit (E). (i) Relative fractional deuterium uptake (RFU) plots of NTD, RBD, and the S2 subunit at 1 min (orange) and 10 min (black) deuterium exchange times, with pepsin digest fragments displayed from N to C-terminus (X-axis). Peptides are grouped into clusters indicated by brackets (X-axis) for ease of display. Individual peptides within each cluster are identifiable from the Supplementary Excel file, which lists clusters and each peptide within each cluster (Supplementary file 1: Table S1). (Also see Figure 2—figure supplement 2). (ii) Deuterium exchange maps on close-up of the structures of NTD (21–303), RBD (318–552), and the S2 subunit (821–1197). Peptides spanning NTD–RBD interaction sites (166–182, 213–223, 294–303, 318–325, 375–387, and 442–449) showing relatively high deuterium exchange at t = 1 min are highlighted. (iii) The first principal motion and RMSF values of backbone atoms on the NTD, RBD, and the S2 subunit. Residues with high RMSF are labeled. Different domains (fusion peptide [FP], heptad repeat 1 [HR1], central helix [CH], connector domain [CD], heptad repeat 1 [HR2]) showing domain-specific RFU changes are labeled. RFU values are tabulated in Figure 2—source data 1.

Figure 2—source data 1

Relative fractional deuterium uptake values for spike (S) protein peptides at indicated labeling times.

https://cdn.elifesciences.org/articles/63646/elife-63646-fig2-data1-v2.xlsx

Interestingly, the deuterium exchange heat map also showed highest relative exchange in the S2 subunit (Figure 2—figure supplement 3) and helical segments of the stalk, while peptides spanning the FP showed relatively lower deuterium exchange overall. Individually, S1 and S2 subunits showed different intrinsic deuterium exchange kinetics, where average relative fractional deuterium uptake (RFU) at early deuterium exchange time points of S1 subunit (~0.25) was lower than the average RFU (~0.35) for the S2 subunit (Figure 2—figure supplement 3, source data – Figure 2—source data 1). Furthermore, peptides connecting the RBD to the rest of the S protein showed greater deuterium exchange, suggesting a ‘hinge’ role for this segment to facilitate RBD adopting an ensemble of open and closed conformational states (Figure 2C). Indeed, in our simulations of the S protein (Figure 2B), the RBD oriented initially in an ‘up’ conformation and exhibited spontaneous motion toward the ‘down’ conformation relative to the hinge region (Figure 2D, Figure 2—figure supplement 4A). Interestingly, a part of the receptor binding motif, specifically residues 476–486, exhibited a higher degree of flexibility based on its average atomic fluctuations (Figures 2B and 3B), suggesting a role for the ACE2 receptor in stabilizing S protein dynamics and priming it for host furin proteolysis.

Figure 3 with 2 supplements see all
Download asset Open asset
Map of receptor binding domain (RBD)isolated:angiotensin-converting enzyme 2 (ACE2) interactions.

(A) Relative fractional deuterium uptake values at t = 1 min for RBD (314–547) of spike (S) protein (RBDS) mapped onto the structure of RBD extracted from S protein model (see Supplementary file 1: Table S2). High and low exchanging regions are represented as shown in key, and regions with no coverage are shown in black. (B) The root mean square fluctuation (RMSF) values of backbone atoms on the RBD showing residues with high RMSF (476–486) as per key. Differences in deuterium exchanged between RBDisolated:ACE2 complex and free RBDisolated (C) and RBDS:ACE with free RBDS (D) at 1 min of deuterium labeling are mapped onto the structure of RBD. Protection from deuterium uptake and increases in exchange are indicated in blue and red, respectively. Regions with no peptide coverage are in black. RFU: relative fractional deuterium uptake.

The NTD of the S protein showed low overall RFU (~0.2), consistent with its well-structured arrangement of β-sheets connected by loops (Figures 1B and 2C). Importantly, certain regions showed significantly higher deuterium exchange (~0.4), of which two loci (136–143, 243–265) span the dynamic interdomain interactions with the RBD. This is supported by the high per-residue root mean square fluctuations (RMSFs) and large principal motions observed for residues 249–259 during simulations (Figure 2C, Figure 2—figure supplement 4C). One locus (291–303) at the C-terminal end of the NTD connecting to the RBD showed high deuterium exchange, indicating high relative motions of the two domains. The RBD (Figure 2D) showed an overall higher deuterium exchange (RFU ~0.35), with the peptides spanning the hinge regions (318–336) showing greatest deuterium exchange (~0.6). Peptides spanning residues 351–375 and 432–452 showed significantly increased deuterium uptake, and these correspond to the NTD interdomain interaction sites. Interestingly, certain loci of the RBD at the ACE2 interface (453–467, 491–510) showed higher intrinsic exchange.

Overall, the S2 subunit showed variable deuterium exchange across the constituent domains (Figure 2E, Figure 2—figure supplement 3). Interestingly, peptides spanning the region directly C-terminal to the S1/S2 cleavage site showed the greatest deuterium exchange (0.6). Congruently, our MD simulations revealed the unstructured loop housing the S1/S2 cleavage site (residues 677–689) to be highly dynamic (Figure 2—figure supplement 4), with RMSFs reaching >1.0 nm. It is important to note that the S1/S2 cleavage site has been abrogated in the construct of the S protein used in this study to block proteolytic processing into S1 and S2 subunits during expression in host cells. We observed lower deuterium exchange (and lower RMSF values) at peptides forming the CH and CD, suggesting their function as the central stable core of prefusion S. In contrast, peptides spanning hinge segments and heptad repeats (HR1 and HR2) showed high exchange and RMSF values, reflecting the S protein’s ensemble properties encompassing prefusion, fusion, and postfusion conformations in solution.

Domain-specific and global effects of ACE2 binding to the RBD

Comparative HDXMS of the S protein and S:ACE2 complex showed large-scale changes in S protein upon ACE2 binding. The RBD forms the main interaction site on S protein for ACE2. We therefore set out to comparatively map HDXMS of ACE2:RBD interface of an isolated MBP fusion construct of the RBD (‘RBDisolated’) (Figure 3C, Figure 3—figure supplement 2—source data 1 Supplementary file 1: Table S2) with S:ACE2 complex (Figure 4A, B). A list of peptides common to RBDisolated and S protein (‘RBDS’) showed differences in deuterium exchange only at interdomain interfaces within individual monomers and trimer interaction sites in the S protein (Supplementary file 1: Table S3). Several RBDS peptides showed decreased exchange upon complexation with ACE2 (Figure 3). These include peptides 340–359, 400–420, 432–452, and 487–502 in the RBDS:ACE2 complex (Figure 4). Sites showing deuterium exchange protection are consistent with the RBD:ACE2 interface described by X-ray crystallography (PDB: 6M0J) (Lan et al., 2020). Further, HDXMS revealed the core of this interface to be contributed by peptides 340–359, 400–420, 432–452, and 491–510 (Figure 4A, D, Figure 2—figure supplement 3). Interestingly, loci showing large-magnitude differences in deuterium exchange correlate to certain mutational hotspots (Wang et al., 2020b).

Figure 4
Download asset Open asset
Angiotensin-converting enzyme 2 (ACE2) interaction induces large-scale allosteric changes across spike (S) protein.

(A) Differences in deuterium exchange (ΔDex) (t = 1 min) in S protein upon binding ACE2 showing decreased (blue) and increased (red) deuterium exchange, mapped onto the structure of S protein. Deuterium exchange differences (X-axis) for peptides from (B) receptor binding domain (RBD)S and S2 subunit (C). Peptides are grouped into clusters indicated by brackets (X-axis) for ease of display. Individual peptides within each cluster are identifiable from the source data (Figure 4—source data 1). Difference cutoff ±0.3 D (Houde et al., 2011) is the deuterium exchange significance threshold indicated by pink shaded box with standard error values in gray. Positive differences (>0.3 D) denote increased deuterium exchange, and negative differences (<−0.3 D) denote decreased deuterium exchange in S protein bound to ACE2. (B) Peptides spanning residues interacting with ACE2 are in purple. (C) Peptides spanning S1/S2 cleavage site, fusion peptide (FP) and heptad repeat 1 (HR1) are highlighted in pink boxes, while peptides spanning central helix and heptad repeat 2 (HR2) are in blue. (D) Stacked mass spectra with isotopic envelopes after deuterium exchange (t = 1, 10 min) for select peptides from (i) RBD (residues 476–498), (ii) S1/S2 cleavage site (residues 672–690), (iii) FP (residues 823–832), and (iv) HR2 (residues 1175–1188) are shown for the S protein and S:ACE2 complex. Mass spectra of the equivalent undeuterated peptide are shown for reference. The centroid masses are indicated by red arrowheads.

Figure 4—source data 1

Differences in deuterium exchange values between spike (S):angiotensin-converting enzyme 2 (ACE2) complex and free S protein at indicated labeling times.

https://cdn.elifesciences.org/articles/63646/elife-63646-fig4-data1-v2.xlsx

A closer examination of the RBDisolated:ACE2 interface by HDXMS also revealed decreased exchange in peptides spanning these regions (Figure 3). However, the magnitude of deuterium exchange protection was significantly more in RBDisolated than in RBDS, potentially reflecting the higher flexibility in the full-length S trimer relative to free RBD, interfering with ACE2 binding. High-resolution structures of RBD:ACE2 reveal the core of the RBD interface to be formed by amino acids Y449, Y453, N487, Y489, G496, T500, G502, Y505, L455, F456, F486, Q493, Q498, and N501 (Wang et al., 2020a). These correspond to peptide 448–501 from S protein and RBDisolated in our HDXMS study.

Cryo-EM studies have shown that each RBD in the trimeric S protein can adopt an open conformation irrespective of other RBDs, indicating an absence of cooperativity between the three RBDs within a trimer (Ke et al., 2020). Therefore, we compared the deuterium exchange profiles of RBDisolated with RBDS and observed differences in dynamics imposed by quaternary contacts (Figure 3). Overall, the loci with high and low deuterium exchange profiles were similar when compared between RBDisolated and RBDS, both at the disordered ACE2 receptor binding region and the folded regions at the N- and C-termini. In solution, RBDS toggles between open and closed conformations, resulting in an average readout of deuterium exchange measurements.

ACE2 binding to RBDisolated and RBDS resulted in similar effects, where we observed deuterium exchange protection at the peptide regions spanning the known binding interface of RBD. Notably, increased deuterium exchange was observed at the hinge region (Figure 3D), indicating allosteric conformational changes, associated with restricting the open and closed states interconversion. Therefore, the destabilization/local unfolding observed at the hinge region as a result of ACE2 binding enables RBD to maintain the open conformation. It therefore seems likely that small molecules and biologics targeting the hinge region to lock RBD in the closed state would be of potential high therapeutic value.

ACE2 binding to RBD is allosterically propagated to the S1/S2 cleavage site and HR

Unexpectedly, ACE2 binding at the RBD induced large-scale changes in deuterium exchange in distal regions of the S protein. Some of the peptides in the stalk of S protein showed decreased exchange in the S:ACE2 complex (Figure 4C,D). This indicates that ACE2 receptor interactions stabilized the hinge dynamics in the S protein. Decreased exchange was also seen in the distal sites in the S2 subunit, localized at the FP locus and CH. Interestingly, increased exchange was seen in multiple peptides flanking the S1/S2 cleavage site, HR1 domain, and critically at the S1/S2 cleavage site (Figure 4D). Even though the protease cleavage site is abrogated in the construct used in this study, we still observed increased dynamics as inferred by the higher relative deuterium exchange at the S1/S2 locus. Furthermore, this region exhibited high RMSF values during simulations (Figure 2—figure supplement 4B). These results clearly indicate that ACE2 binding induces allosteric enhancement of dynamics at this locus, providing mechanistic insights into the conformational switch from the prefusion to fusogenic intermediate. Differences in deuterium exchange between free S protein and the S:ACE2 complex show stabilization at the ACE2 interacting site and local destabilization at peptides juxtaposed to the S1/S2 cleavage site and HR1 ( peptides 931–938). This suggests that ACE2 binding allosterically primes HR1 and other high exchanging regions flanking the S1/S2 cleavage site for enhanced furin protease binding and cleavage. Importantly, these results suggest that the S1/S2 cleavage site is a critical hotspot for S protein dynamic transitions for facilitating SARS-CoV-2’s entry into the host, and therefore represents a new target for inhibitory therapeutics against the virus.

Dynamics of RBD:ACE2 and S:ACE2 protein interactions provides insights for viral–host entry

Considering the indispensable role of ACE2 binding in SARS-CoV-2 infection, it is crucial to assess the effects of S protein and RBD binding on ACE2 dynamics (Figure 5, Figure 5—figure supplements 1–3, Supplementary file 1: Table S4). We therefore mapped the corresponding binding sites of RBD, both isolated and within the S protein, onto ACE2. The S:ACE2 complex represents the prefusion pre-cleavage state wherein full-length S protein is bound to the ACE2 receptor (Figure 1B, ii), while the RBDisolated:ACE2 complex represents the post-furin cleavage product formed by the S1 subunit and ACE2 (Figure 1B, iii). Previous studies have shown that 14 key amino acids of RBD interact with ACE2, wherein mutations at six sites resulted in higher binding affinity of SARS-CoV-2 (Li et al., 2005). SARS-CoV-2 adopted a different binding mode to ACE2 as a superior strategy for infection compared to SARS-CoV-1. A crystal structure of RBDisolated:ACE2 complex has identified 24 key ACE2 residues, spanning across peptides 16–45, 79–83, 325–330, 350–357, and R393 (Towler et al., 2004). While most of these residues are conserved in binding to both SARS-CoV-1 and SARS-CoV-2, R393 and residues 325–330 are unique to SARS-CoV-1 interaction (Wang et al., 2020b). Interestingly, we observed increased deuterium exchange at these residues in the S:ACE2 complex compared to ACE2 alone (Figure 5C). Identifying the intrinsic dynamics and allosteric changes upon binding could guide development of therapeutic antibodies and small molecule drugs.

Figure 5 with 3 supplements see all
Download asset Open asset
Effect of receptor binding domain (RBD)isolated and RBDS complexes on angiotensin-converting enzyme 2 (ACE2) dynamics.

(A) Structure of extracellular domain of ACE2 receptor (PDB ID: 1R42) depicting the relative fractional deuterium uptake (RFU) at t = 1 min. (B) Differences in deuterium exchange of RBDisolated:ACE2 complex and free ACE2 at t = 1 min mapped onto the structure of ACE2, predominantly showing decreased deuterium exchange in ACE2 (shades of blue). (C) Heat map of differences in deuterium exchange (t = 1 min) of S:ACE2 complex and free ACE2. (D) Plot showing differences in deuterium exchange between ACE2 and complexes with RBDisolated (i) and S (ii) at different labeling times. Peptides are grouped into clusters for ease of display and listed in source data (Figure 5—source data 1). Cutoff ± 0.3 D is the deuterium exchange significance threshold, indicated by pink shaded box, and standard errors are in gray. Positive differences denote increased deuterium exchange in (i) RBDisolated:ACE2 or (ii) S:ACE2 compared to free ACE2, while negative differences denote decreased deuterium exchange. Peptides spanning the sites of interaction with RBD and two distal sites (278–292, 574–585) are highlighted. (E) Stacked mass spectra showing isotopic distribution for select peptides spanning the binding sites (Ali and Vijayan, 2020; Chan et al., 2020; Wang et al., 2020a; Watanabe et al., 2020; Cai et al., 2020; Wang et al., 2020b; Li et al., 2005; Towler et al., 2004; Hamuro et al., 2008; Hoofnagle et al., 2003; Houde et al., 2011; Šali and Blundell, 1993; Hakansson-McReynolds et al., 2006; Dev et al., 2016; Eramian et al., 2006; Ramachandran et al., 1963; Petit et al., 2007; van Meer, 1998; Krijnse-Locker et al., 1994) and a distal allosteric site (575–586) for ACE2, S:ACE2, and RBDisolated:ACE2 are shown at 1 min deuterium labeling time. Centroids indicated by red arrowheads.

Figure 5—source data 1

Differences in deuterium exchange between receptor binding domain (RBD)isolated:angiotensin-converting enzyme 2 (ACE2) and spike (S):ACE2 complexes with free ACE2 at indicated labeling times.

https://cdn.elifesciences.org/articles/63646/elife-63646-fig5-data1-v2.xlsx

Simulations of the ACE2 dimer complexed with the B0AT1 amino acid transporter (PDB: 6M1D) (Yan et al., 2020) in a model epithelial membrane revealed a large motion of the peptidase domain, which recognizes the S protein RBD, with respect to the transmembrane and juxtamembrane domains (Figure 5—figure supplement 3). This large motion is reminiscent of the flexible tilting displayed by the S protein ECD itself, suggesting that both S protein and ACE2 have adaptable hinges that allow for orientational freedom of the domains involved in recognition. To understand how S protein binding affects ACE2 dynamics, we performed HDXMS experiments of monomeric ACE2 alone, S:ACE2 and RBD:ACE2 complexes (Figure 5, Figure 5—figure supplement 2) and mapped the deuterium exchange values onto a deletion construct of ACE2 (PDB: 1R42) (Towler et al., 2004; Figure 5, Figure 5—figure supplement 2). We observed a reduction in deuterium exchange across both RBDisolated:ACE2 and larger S:ACE2 complexes compared to free ACE2 (Figure S8B and S8C). Differences in deuterium exchange between RBDisolated:ACE2 complex and free ACE2 showed that RBD binding stabilizes ACE2 globally, specifically large differences at the binding site (peptides 21–29, 30–39, and 75–92), and also at distal regions (peptides 121–146, 278–292, and 575–586) from the RBD binding site of ACE2 (Figure 5E). Cryo-EM studies have shown that a dimeric full-length ACE2 receptor can stably bind to one trimer of the S protein (Yan et al., 2020).

Conclusions

Here, a combination of HDXMS and MD simulations provides a close-up of S protein dynamics in the prefusion, ACE2-bound, and other associated conformations. Our results reveal the energetics of the S:ACE2 complex interface. ACE2 binding to the isolated RBD and S protein alike leads to binding and stabilization. Interestingly, ACE2 binding to the RBD induces global conformational changes across the entire S trimer. Importantly, the stalk region undergoes dampening of conformational motions while showing increased deuterium exchange at the proteolytic processing sites. This study may help in explaining how mutations in emerging strains in the ongoing COVID-19 outbreak might alter dynamics and allostery of ACE2 binding and offer a mechanistic basis for altered infectivities observed in emerging strains. Sites on S protein showing altered deuterium exchange describe allosteric propagation of ACE2 binding and represent novel cryptic targets for therapeutic small molecule inhibitor/antibody discovery.

Materials and methods

Key resources table
Reagent type
(species) or resource		DesignationSource or referenceIdentifiersAdditional information
Gene (SARS-CoV-2)pTT5 expression vectorGenBankQHD43416.1For recombinant S protein
Gene (ACE2)pHL-sec expression vectorGenBankAB046569.1For recombinant ACE2 protein
Cell line (Homo sapiens)Human embryonic kidney (HEK293-6E)NRC, CanadaRRID:CVCL_HF20
Cell line (Homo sapiens)Expi293FThermo Fisher ScientificRRID:CVCL_D615
AntibodyAnti-human IgG Fc HRHorseradish Peroxidase (HRP)(goat polyclonal)Thermo ScientificRRID:AB_2536544WB (1:5000)
Recombinant DNA reagentpHLmMBP-10 (plasmid)Addgene, USARRID:Addgene_72348For recombinant RBD protein
Recombinant DNA reagentpTT5 expression vector (plasmid)Addgene, USARRID:Addgene_52367
Recombinant DNA reagentpHL-sec expression vector (plasmid)Addgene, USARRID:Addgene_99845recombinant DNA reagent
Chemical compound, drug3,3′,5,5′-TetramethylbenzidineSigma AldrichRRID:AB_2336758
Chemical compound, drugDeuterium oxide (chemical)Cambridge Isotope LaboratoriesCAS# 7789-20-0Deuterium exchange experiments
Software, algorithmDynamXWaters Corporation (Milford, MA)Version 3.0
Software, algorithmProteinLynx Global Server (PLGS)Waters Corporation (Milford, MA)Version 3.0.1
Software, algorithmGraphPad Prism softwareGraphPad Prism (https://graphpad.com)RRID:SCR_002798Version 5.0.0
Software, algorithmModeller1989–2020 Andrej SaliRRID:SCR_008395Version 9.21
Software, algorithmVisual molecular dynamicsUniversity of Illinois at Urbana-ChampaignRRID:SCR_001820Version 1.9.3

Materials

Mass spectrometry grade acetonitrile, formic acid, and water were from Fisher Scientific (Waltham, MA); deuterium oxide was from Cambridge Isotope Laboratories (Tewksbury, MA). All reagents and chemicals were research grade or higher and obtained from Merck-Sigma-Aldrich (St. Louis, MO).

Methods

Transient expression and purification of recombinant SARS-CoV-2 spike, RBD, and ACE2 receptor

Request a detailed protocol

A near-full-length S protein of SARS-CoV-2 (1–1208; Wuhan-Hu-1; GenBank: QHD43416.1), excluding TD and CT, was codon optimized for mammalian cell expression and cloned into pTT5 expression vector with a twin strep tag at the C-terminus (Twist Biosciences, Singapore). Mutations were introduced into this construct at two sites: (i) RRAR motif at the S1/S2 cleavage site (682–685) was substituted by GSAS and (ii) KV motif (986–987) was substituted with two prolines. A gene encoding SARS-CoV-2-RBD (319–591 of SARS-CoV-2 spike) (BioBasic, Singapore) was cloned into the expression vector pHLmMBP-10 as a fusion protein with N-terminal mMBP and C-terminal hexahistidine tags. A gene encoding human ACE2 (residues 21–597) fused to a C-terminal Fc-tag (BioBasic, Singapore) was cloned into vector pHL-sec between the signal peptide and C-terminal 6xHis tag. S (1–1208) was expressed in HEK293-6E using polyethylenimine as the transfection reagent while the isolated RBD (‘RBDisolated’) and ACE2 constructs were expressed in Expi293F using the Expi293 System. Culture supernatant was harvested on day 7 for HEK293-6E expression and day 5 for Expi293F expression. S protein was affinity purified using Strep-TactinXT column (IBA), RBD protein was affinity purified using cOmplete His-Tag Purification column (Merck, Darmstadt, Germany), and ACE2 receptor was affinity purified using HiTrap MabSelect SuRe column (GE Healthcare, Chicago, IL,USA). Purified proteins were concentrated and buffer exchanged into phosphate buffered saline (PBS) using VivaSpin, and the purity was assessed by denaturing polyacrylamide gel electrophoresis (Figure 2—figure supplement 1A, Figure 3—figure supplement 1A, and Figure 5—figure supplement 1A). Cell lines obtained commercially are listed in key resources table and were tested for contamination by Mycoplasma species.

Characterization of RBD:ACE2 receptor binding

Request a detailed protocol

Interactions between recombinant purified MBP-RBD and ACE2 receptor (Figure 3—figure supplement 1A and Figure 5—figure supplement 1A) were confirmed by enzyme-linked immunosorbent assay. To test binding activity of ACE2, 96-well maxisorp plates were coated with 100 µL of 27.2 nM MBP-RBD diluted in PBS at 4°C for 16 hr and blocked with 350 µL of 4% skimmed milk in PBST (0.05% Tween 20 in PBS) at room temperature for 1.5 hr. This was followed by 1 hr incubation with ACE2 (100 µL) at varying concentrations and detection with 100 µL of goat-anti-human IgG Fc HRP diluted at 1:5000 in 2% skimmed milk in PBST for 1 hr. Plates were washed three times in PBST after each incubation step above. After 5 min incubation with 100 µL of 3,3′,5,5′-tetramethylbenzidine, reaction was stopped with 100 µL of 1 M H2SO4 and absorbance at 450 nm (A450) was recorded. A similar protocol was adopted for the quality testing of MBP-RBD – it was coated at variable concentrations in PBS at 4°C for 16 hr and blocked at room temperature for 1.5 hr. This was followed by 1 hr incubation with 10.4 nM ACE2 (100 µL) diluted in blocking buffer. Detection, plate washing, and color development steps were performed in the same manner as described above. Data represents an average of three replicates, along with their error bars and plotted using GraphPad Prism 5 (San Diego, CA).

Deuterium exchange

Request a detailed protocol

S protein (8 µM), ACE2 (52 µM), and RBD (67 µM) solubilized in PBS (pH 7.4) were incubated at 37°C in PBS buffer reconstituted in D2O (99.90%), resulting in a final D2O concentration of 90%. S:ACE2 and RBDisolated:ACE2 complexes (KD of ~15 and ~150 nM, respectively) (Wrapp et al., 2020) were pre-incubated at 37°C for 30 min in a 1:1 molar ratio to achieve >90% binding prior to each hydrogen–deuterium exchange reaction. Deuterium labeling was performed for 1, 10, and 100 min for isolated construct of RBD, free ACE2, and RBDisolated:ACE2 complex. For isolated S protein and S:ACE2 complex, 1 and 10 min labeling timescales were used. Pre-chilled quench solution 1.5 M GnHCl and 0.25 M Tris(2-carboxyethyl) phosphine-hydrochloride was added to deuterium exchange reaction mixture to lower the pHread to ~2.5 and lower the temperature to ~4°C. Next, the quenched reaction was incubated at 4°C on ice for 1 min followed by online pepsin digestion.

Mass spectrometry and peptide identification

Request a detailed protocol

Approximately 100 pmol quenched samples were injected onto chilled nanoUPLC HDX sample manager (Waters, Milford, MA). The injected samples were subjected to online digestion using immobilized Enzymate BEH pepsin column (2.1 × 30 mm) (Waters, Milford, MA) in 0.1% aqueous formic acid at 100 μL/min. Simultaneously, the proteolyzed peptides were trapped in a 2.1 × 5 mm C18 trap (ACQUITY BEH C18 VanGuard Pre-column, 1.7 μm, Waters, Milford, MA). Following pepsin digestion, the proteolyzed peptides were eluted using acetonitrile gradient of 8–40% in 0.1% formic acid at a flow rate of 40 µL/min into reverse phase column (ACQUITY UPLC BEH C18 Column, 1.0 × 100 mm, 1.7 μM, Waters, Milford, MA) pumped by nanoACQUITY Binary Solvent Manager (Waters, Milford, MA). Electrospray ionization mode was used to ionize peptides sprayed onto SYNAPT G2-Si mass spectrometer (Waters, Milford, MA) acquired in HDMSE mode. A flow rate of 5 µL/min was used to continually inject 200 fmol μL−1 of [Glu1]-fibrinopeptide B ([Glu1]-Fib) as lockspray reference mass.

For identification of the resolved and eluted peptides, HDMSE method was used with ion-mobility settings 600 m/s wave velocity and 197 m/s transfer wave velocity. Low collision energies of 4 and 2 V were used for trap and transfer, respectively, while high collision energy was ramped from 20 to 45 V. A constant 25 V cone voltage was used, and mass spectra within 50–2000 Da were acquired for 10 min with mass spectrometer operated in positive ion mode.

Undeuterated protein samples were used to identify sequences from mass spectra data (in HDMSE mode) using ProteinLynx Global Server (PLGS) v3.0. Peptide identification search was performed against a separate sequence database of each protein sequence, along with its respective affinity purification tag sequences. PLGS search parameters selected for peptide sequence identification were (i) no specific protease and (ii) variable N-linked glycosylation modification. Additional cutoff filters applied included (i) minimum intensity = 2500, (ii) minimum products per amino acids = 0.2, and (iii) a precursor ion mass tolerance of <10 ppm in DynamX v.3.0 (Waters, Milford, MA) and confirmed for pepsin cleavage specificity as described (Hamuro et al., 2008). Peptides independently identified under the specified condition and present in at least in two out of three undeuterated sample replicates were retained for HDXMS analysis. S protein contains 22 variable glycosylation sites (Watanabe et al., 2020), out of which we identified peptides spanning 12 glycosylation sites in our sample (Figure 2—figure supplement 2). However, none of these peptides showed glycosylation. For ACE2, we obtained peptides overlapping four glycosylation sites (Figure 5—figure supplement 3).

RFU is the ratio of number of deuterons exchanged to the total number of exchangeable amides of the peptide. Centroid masses of undeuterated reference spectra were subtracted from equivalent spectra of deuterium exchanged peptides to calculate the average deuterons exchanged for each peptide. Deuterium exchange plots, relative deuterium exchange, and difference plots were generated by DynamX v.3.0. The N-terminus and all prolines in each peptide were excluded for estimation of exchangeable amides per peptide (Hoofnagle et al., 2003). Deuterium exchange experiments for two biological replicates and technical triplicates of S protein and the S:ACE2 complex were carried out. Average deuterium exchange measurements between the two biological replicates were within ±0.3 Da (Supplementary file 1: Table S5, S6) (Houde et al., 2011). While deuterium exchange values are not corrected for back exchange, fully deuterated S protein samples were used to measure deuterium back exchange. A list of peptides with back exchange values is shown in Supplementary file 1: Table S7. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE [1] partner repository with the dataset identifier PXD023138.

Modeling and MD simulations

Request a detailed protocol

An integrative model of full-length SARS-CoV-2 S protein was built using Modeller v.9.21 (Šali and Blundell, 1993). The cryo-EM structure of prefusion S ECD in the open conformation (PDB: 6VSB) (Wrapp et al., 2020) was used as the template for the ECD with missing loops on the NTD and the C-terminus of the ECD modeled based on the cryo-EM structure of S ECD in the closed conformation resolved at a higher resolution (PDB: 6XR8) (Cai et al., 2020). The Nuclear Magnetic Resonance (NMR) structure of the SARS S HR2 domain (PDB: 2FXP) (Hakansson-McReynolds et al., 2006) was used as the template for the HR2 domain, while the TM domain was modeled using the NMR structure of the HIV-1 gp-41 TM domain (PDB: 5JYN) (Dev et al., 2016). Ten models were built and subjected to stereochemical assessment using the discreet optimized protein energy (DOPE) score (Eramian et al., 2006) and Ramachandran analysis (Ramachandran et al., 1963). The model with the lowest DOPE score and the smallest number of Ramachandran outliers was chosen. Palmitoylation was performed at three cysteine residues (C1236, C1240, and C1243) on the CT domain based on a study showing its importance in SARS S protein function (Petit et al., 2007). The S protein model was then embedded into a model membrane representing the endoplasmic reticulum–Golgi intermediate compartment (ERGIC) (van Meer, 1998), where coronaviruses are known to assemble in a bud form (Krijnse-Locker et al., 1994; Klumperman et al., 1994). The ERGIC model membrane was built using CHARMM-GUI Membrane Builder (Lee et al., 2019).

All-atom MD simulation was performed for 200 ns using GROMACS (University of Groningen, Netherlands) 2018 (Abraham et al., 2015) and the CHARMM36 force field (Huang and MacKerell, 2013). The system was solvated with 590,742 TIP3P water molecules and 0.15 M NaCl salt, achieved by adding 3235 Na+ and 2103 Cl- ions. Minimization and equilibration were performed following standard CHARMM-GUI protocols (Lee et al., 2016). This includes six steps of equilibration; the first two steps used a 1 fs integration time step for 125 ps, while the last four used 2 fs time step for 250 ps. With each step, the magnitude of positional and dihedral restraints imposed on the protein and lipid molecules was gradually reduced by lowering the force constants from 1000 (step 1) to 0 kJ mol−1 nm−2 (step 6). Temperature and pressure were maintained at 310 K and one atm, respectively, using the Berendsen thermostat and barostat during equilibration. This was then followed by the production run, whereby the temperature was maintained using the Nosé–Hoover thermostat (Nosé, 1984; Hoover, 1985) and the pressure was maintained via semi-isotropic coupling to the Parrinello–Rahman barostat (Parrinello and Rahman, 1981). Electrostatics were calculated using the smooth particle mesh Ewald method (Essmann et al., 1995) with a real space cutoff of 1.2 nm and the van der Waals interactions were truncated at 1.2 nm with force switch smoothing between 1.0 and 1.2 nm. Constraints were applied to covalent bonds with hydrogen atoms using the LINCS algorithm (Hess et al., 1997) and a 2 fs integration time step was employed. Snapshots of the trajectory were saved every 100 ps. To assess whether the system was properly equilibrated, we calculated domain-specific root mean square deviations (RMSDs) of the Cα atoms following least-squares fitting (Figure 2—figure supplement 5). For all three domains tested (NTD, RBD, and HR2) in all three chains of the S protein, the RMSD reached a plateau after around 50 ns. Additionally, we also calculated RMSF profiles using 20 ns trajectory windows along the simulations. Similarly, the per-residue RMSF values for all three domains converged after the first three windows (60 ns).

For simulations of the ACE2 receptor, the cryo-EM structure of the ACE2-B0AT1 complex in the open conformation (PDB: 6M1D) (Yan et al., 2020) was used. The ACE2-B0AT1 complex was embedded into a model membrane representing the epithelial cell membrane (Jia et al., 2005; Sampaio et al., 2011). The system was solvated with 314,442 TIP3P water molecules and 0.15 M NaCl salt (1868 Na+ and 1300 Cl- ions). Minimization, equilibration, and production runs were performed using the protocols described above. Principal component analysis and RMSF analyses were performed using GROMACS, and simulations were visualized in VMD (University of Illinois at Urbana-Champaign, USA) (Humphrey et al., 1996).

Data availability

All data generated or analysed during this study are included in the manuscript and supporting files. Source data files have been provided for Figures 2, 3, 4 and 5.

References

Decision letter

  1. Donald Hamelberg
    Reviewing Editor; Georgia State University, United States
  2. Olga Boudker
    Senior Editor; Weill Cornell Medicine, United States
  3. Donald Hamelberg
    Reviewer; Georgia State University, United States
  4. Elizabeth Komives
    Reviewer; UCSD, United States

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

Understanding the mechanism of action of the SARS-Cov-2 virus, which is responsible for the COVID-19 pandemic, at the molecular level is key to developing therapeutics. This work is a timely and interesting exploration of the interaction between the Spike protein on the surface of the SARS-Cov-2 virus and the ACE2 receptor on the surface of human cells. This Spike-ACE2 interaction is an important step in the initial stages of viral infection. The results suggest that the Spike-ACE2 interaction induces extremely long-range allosteric effects on the Spike protein that could help trigger proteolysis of the Spike protein. It is anticipated that the results of this work would have implications for the development of small molecule inhibitors to prevent SARS-Cov-2 infection.

Decision letter after peer review:

Thank you for submitting your article "SARS-CoV-2 S protein ACE2 interaction reveals novel allosteric targets" for consideration by eLife. Your article has been reviewed by three peer reviewers, including Donald Hamelberg as the Reviewing Editor and Reviewer #1, and the evaluation has been overseen by Olga Boudker as the Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: Elizabeth Komives (Reviewer #2).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

We would like to draw your attention to changes in our revision policy that we have made in response to COVID-19 (https://elifesciences.org/articles/57162). Specifically, we are asking editors to accept without delay manuscripts, like yours, that they judge can stand as eLife papers without additional data, even if they feel that they would make the manuscript stronger. Thus the revisions requested below only address clarity and presentation.

Summary:

This is a timely and interesting exploration of the interaction between the Spike protein of SARS-CoV-2, the virus responsible for the COVID-19 pandemic, and the ACE2 receptor using hydrogen deuterium exchange mass spectrometry and molecular dynamics simulations. The Spike protein consists of two sub-domains S1 and S2 with the S1 needing to be cleaved-off so the S2 can become the fusion protein responsible for getting the SARS-CoV-2 into the cell. Structures are available but they do not shed light on how the protease furin can access the cleavage site between S1 and S2 in order to begin the process of fusion. The results suggest that the Spike-ACE2 interaction induces extremely long-range allosteric effects on the Spike protein that could trigger proteolysis of the Spike protein. Specifically, when ACE2 binds to the Spike protein, a conformational change occurs near the S1/S2 cleavage site, exposing it and likely making it more susceptible to furin cleavage. The binding also dampens exchange in the stalk region of the Spike protein. The authors refer to these regions as "dynamic hotspots in the pre-fusion state". The results of this work have implications for the development of small molecule inhibitors.

In general, the work is timely, and the results will be of interest to many in the field. The major conclusions of the work are generally supported by the results. Below are suggestions to improve the manuscript and strengthen the conclusions of the work.

Essential revisions:

1) The manuscript appears to have been hastily written, it would benefit from a scientific editor making it more readable. For example, "Average deuterium exchange at these 91 reporter peptides was monitored for comparative deuterium exchange analysis of S protein, ACE2 receptor and S:ACE2 complex, along with a specific ACE2 complex with the isolated RBD." Presumably "reporter peptides" refers to the 321 peptides mentioned two sentences earlier…Why is the ACE2 complex with the isolated RBD qualified as "specific" while none of the others are? Then the article continues with more information about glycosylation…

2) Subsection “Localizing subunit specific dynamics and domain motions of S protein trimer” and Figure 2—figure supplement 2: A bit more should be said about the glycosylation sites. If only non-glycosylated peptides being observed in the pepsin digestion, the coverage map (Figure 2—figure supplement 2), shows expected lack of coverage for only a few sites (17, 122, 149, 165, 234, 282, 709, 1134) whereas many other sites are covered by peptides. Does this indicate that these sites are mostly not glycosylated?

3) Figure 2—figure supplement 3 legend seems to indicate that uptake of each peptide is plotted, whereas uptake per residue should be plotted because overlapping peptides make these data misleading. The peptides are shown in the other relative uptake graphs, but then there is more than one data point per peptide. Can the authors explain a bit more in the legend how they got the data in these figures?

4) Figure 2—figure supplement 4 seems to indicate that the cleavage site is already very dynamic. Can the authors explain this?

5) "… Mapping the relative deuterium exchange across all peptides onto this S protein model showed the greatest deuterium exchange at the stalk region" seems to contradict "The deuterium exchange heat map showed the highest relative exchange in the S2 subunit (Figure 2—figure supplement 3) and helical segments," Please clarify.

6) Figure 2A and B look like the same molecular structure (nice that they are in the same orientation) but the domains are labeled differently. Yet a third domain listing is used in panel E. Comparing panels A and B, it's a little strange that some of the least dynamic spots in the Head/ECD are the highest exchanging, do the authors want to comment on this?

7) If it wouldn't slow things down too much, it would be great if the RFU data were calculated after back exchange correction. Even an imperfect correction (such as a global correction for the back exchange during analysis) would make the data more meaningful.

8) Figure 3C and D look remarkably different considering that they both are reflecting the RBD:ACE2 interaction. Did the authors attempt to find a convergent set of peptides to do this analysis? Perhaps if the binding site were labeled it would help make the differences look less important (overall the top part of the molecule is blue and the bottom more-or-less has some red and if that's all we are supposed to get out of this figure then it is ok).

9) Figure 4. The authors state that the significance cut-off for difference in deuterium exchange is 0.3 D but there is no explanation of how the value were derived.

10) Figure S1: The SDS-PAGE showed around 90 kDa for the molecular weight of RBDisolated, which should be around 25 kDa based on its sequence (318-547). Please check and clarify.

11) It is confusing about the existing forms of the S protein and ACE2 and their binding stoichiometry, regarding the statements such as "we measured dynamics of a trimer of this near-full length S protein…", "we performed HDXMS experiments of monomeric ACE2…", "…were pre-incubated at 37°C for 30 min in a molar ratio of 1:1 to achieve >90% binding…". Please confirm whether the expressed ACE2 is dimeric and S protein is trimeric or not, and their binding stoichiometry is 1:1 or 2:3. Please also provide the concentration and calculation details for ensuring the >90% binding. If only one ACE2 in the ACE dimer and one S protein in the S protein trimer are involved in the binding, how sensitive and accurate could the HDX-MS results reflect the binding, since no HDX difference would be observed for the other ACE2 and other 2 S proteins?

12) Abstract: Other studies (e.g., Hoffman et al., 2020) have shown that ACE2 binding can enhance S1/S2 cleavage by furin and S1/S2 cleavage site could be possible targets for small molecule inhibitor/antibody development. It would be helpful if further evidence could be provided to support that the stalk hinge regions could also be the targets for that.

13) The authors should provide more technical details of the molecular dynamics simulations in the supplementary materials. Could the authors provide more details on the equilibration protocol? Was there any analysis done or metric used to assess whether the system was properly equilibrated? How often were snapshots of the trajectory saved for analysis? How many Na+ and Cl- ions were added to achieve 0.15 M of salt concentration? Also, how many water molecules were added? These details are relevant to the non-casual readers.

14) Figure 2—figure supplement 1: Could the authors elaborate on Figure 2—figure supplement 1B in the figure legend? Is (i) measuring the binding of ace2 to the S protein? Is (ii) measuring the binding of RBD to the ace2 protein? The distinction between (i) and (ii) is not made in the figure legend.

https://doi.org/10.7554/eLife.63646.sa1

Author response

Essential revisions:

1) The manuscript appears to have been hastily written, it would benefit from a scientific editor making it more readable. For example, "Average deuterium exchange at these 91 reporter peptides was monitored for comparative deuterium exchange analysis of S protein, ACE2 receptor and S:ACE2 complex, along with a specific ACE2 complex with the isolated RBD." Presumably "reporter peptides" refers to the 321 peptides mentioned two sentences earlier…Why is the ACE2 complex with the isolated RBD qualified as "specific" while none of the others are? Then the article continues with more information about glycosylation…

We accept the reviewer’s suggestions and feedback and have revised and rewritten the manuscript to make it more accessible to the readers and address reviewers’ concerns.

For the example cited above, we have significantly amended the text. The reporter peptides indeed refer to the 321 pepsin fragment peptides (317 in the revised manuscript) used in the HDXMS analysis. We have made it much clearer in the revised manuscript for a broader readership new to HDXMS (subsection “Subunit specific dynamics and domain motions of S protein trimer”). We have amended and removed any qualifier denoting RBD:ACE2 complex as a “specific” complex. We thank the reviewer for their feedback.

2) Subsection “Localizing subunit specific dynamics and domain motions of S protein trimer” and Figure 2—figure supplement 2: A bit more should be said about the glycosylation sites. If only non-glycosylated peptides being observed in the pepsin digestion, the coverage map (Figure 2—figure supplement 2), shows expected lack of coverage for only a few sites (17, 122, 149, 165, 234, 282, 709, 1134) whereas many other sites are covered by peptides. Does this indicate that these sites are mostly not glycosylated?

We thank the reviewers for this query. Spike protein has been predicted to be glycosylated at 22 sites, 20 of which are predicted to be N-linked glycosylations. Spectra with good signal to noise ratios were observed only for 12 glycosylation sites (61, 74, 331, 343, 603, 656, 717, 801 1074, 1098, 1173, 1194) and included in this study. Peptides spanning other 8 (16, 122, 149, 165, 234, 282, 616 709) glycosylation sites were of poor quality and not considered. As pointed out by the reviewers, a peptide identification search with glycosylation filter did not show glycosylation at the 12 analyzed peptides. This is amended in the subsection “Subunit specific dynamics and domain motions of S protein trimer”.

3) Figure 2—figure supplement 33 legend seems to indicate that uptake of each peptide is plotted, whereas uptake per residue should be plotted because overlapping peptides make these data misleading.

We thank the reviewer for pointing this out and have amended the figure legend for Figure 2—figure supplement 3 to explain this better. This figure now shows relative fractional uptake of deuterium (RFU), not absolute deuterium uptake as erroneously reported in our first submission. We have divided peptides (denoted by a dot) into clusters, indicated by brace brackets on the X-axis, for ease of display. Individual peptides within each cluster are identifiable from their source data and Supplementary file 1 which list each cluster and each peptide within that cluster.

We acknowledge the reviewer’s point that the distribution of pepsin fragments as generated in a modified mirror plot by DynamX appear uniform but are in fact a function of propensity of pepsin cleavage, which is in turn a function of protein sequence. Consequently, some clusters have more constituent peptides, some of which may be overlapping and are nonuniform. Not all peptides within each cluster are overlapping and therefore the plot describes deuterium exchange per peptide. We have amended legends for respective figures.

The peptides are shown in the other relative uptake graphs, but then there is more than one data point per peptide. Can the authors explain a bit more in the legend how they got the data in these figures?

We thank the reviewers for their comment and acknowledge lack of clarity in our figure legends surrounding our grouping peptides into clusters. Each point denotes a single peptide but due to the lack of space and for ease of display, we have grouped peptides into clusters. We have now amended this for each figure legend in the main text.

4) Figure 2—figure supplement 4 seems to indicate that the cleavage site is already very dynamic. Can the authors explain this?

In free S protein HDXMS and MD simulations, a single S1/S2 cleavage site spanning peptide 672690 (Figure 4D) includes a loop region with high intrinsic deuterium exchange and high simulated RMSF values (Figure 2—figure supplement 4B). Importantly, upon ACE2 binding, this peptide showed even greater exchange in the S:ACE2 complex. This forms the primary allosteric response site which we propose to enhance furin proteolytic processing of the S:ACE2 complex. We have included an explanation in the subsection “ACE2 binding to RBD is allosterically propagated to the S1/S2 cleavage site and Heptad Repeat”.

5) "… Mapping the relative deuterium exchange across all peptides onto this S protein model showed the greatest deuterium exchange at the stalk region" seems to contradict "The deuterium exchange heat map showed the highest relative exchange in the S2 subunit (Figure 2—figure supplement 3) and helical segments," Please clarify.

We seek to clarify that the stalk region spanning (central helix and heptad repeat) is part of the S2 subunit showing the greatest deuterium exchange. Notably, within the S2 subunit, domain-specific deuterium exchange was observed, of which the stalk region (central helix and heptad repeats) showed greatest magnitude deuterium exchange. In Figure 2—figure supplement 3, average RFU values (indicated by solid red line) of the S1 and S2 subunits are represented in separate plots. These highlight that the S2 subunit shows higher overall deuterium exchange and by extension dynamics relative to the S1 subunit.

6) Figure 2A and B look like the same molecular structure (nice that they are in the same orientation) but the domains are labeled differently. Yet a third domain listing is used in panel E. Comparing panels A and B, it's a little strange that some of the least dynamic spots in the Head/ECD are the highest exchanging, do the authors want to comment on this?

We thank the reviewer for this insight. We have amended the figure in the revised manuscript. Panel E lists the subdomains of the S2 subunit, for which we have added the residue numbers for the peptides. We would like to clarify that 2A and 2B only show exchange and RMSF values of surface exposed peptides and not the interior structures due to surface representation.

7) If it wouldn't slow things down too much, it would be great if the RFU data were calculated after back exchange correction. Even an imperfect correction (such as a global correction for the back exchange during analysis) would make the data more meaningful.

We appreciate the reviewers’ suggestion to include details of deuterium back-exchange. We have carried out deuterium back-exchange control experiments with fully deuterated S protein (subsection “Mass Spectrometry and peptide identification”). A list of peptides with their back exchange values is provided in Table S7 in Supplementary file 1. Back exchange showed a range between 14.7 to 51%. However the median back exchange for multiple peptides is ~ 34%. This does not change the deuterium exchange heat map significantly. The stalk and S1/S2 proteolysis sites continue to show highest exchange. We have therefore not factored this and applied deuterium exchange corrections. We believe that upon back exchange correction, the differences in deuterium exchange values would be more pronounced. We welcome the reviewer’s suggestion to show readers our back exchange for different peptides.

We have included biological replicate measurements of HDXMS (S protein) and report the new data in Tables S5 and S6 in Supplementary file 1. 4 peptides that showed lower signal to noise were excluded from the final list of 317 peptides. This did not change the sequence coverage in our study. Differences in deuterium exchange values between the two biological replicates are within the low standard errors for measurements (± 0.3 Da).

8) Figure 3C and D look remarkably different considering that they both are reflecting the RBD:ACE2 interaction. Did the authors attempt to find a convergent set of peptides to do this analysis? Perhaps if the binding site were labeled it would help make the differences look less important (overall the top part of the molecule is blue and the bottom more-or-less has some red and if that's all we are supposed to get out of this figure then it is ok).

We thank the reviewers for this suggestion. Figure 3C shows the effect of ACE2 binding on the isolated RBD, while Figure 3D is from the RBD from the larger S protein construct. These were used for comparison. We were also interested in comparing ACE2 interactions of a monomer (isolated RBD) with that of a trimer (full-length S protein). Importantly, we found no differences in the interface mapped by HDXMS for these two constructs. Our results suggest that trimeric S protein bound ACE2 marginally less tightly than isolated RBD. This might be due to the intrinsic dynamics of the S protein trimer interfering with ACE2 binding.

We thank the reviewer for their suggestion. We include a table comparing the common peptides between RBDisolated and RBD of S protein (Table S3 in supplementary file 1), which reveal this marginal difference between RBD and S protein. We have included a sentence to describe this in the subsection “Domain-specific and global effects of ACE2 binding to the RBD”.

9) Figure 4. The authors state that the significance cut-off for difference in deuterium exchange is 0.3 D but there is no explanation of how the value were derived.

The significance cut-off for differences in deuterium exchange were experimentally determined from standard deviations of the average deuterium uptake values from technical and biological replicates, which were obtained from DynamX software. All peptides showed a standard deviation within ± 0.3 D. This was the basis for this value to represent the significance threshold for measurements and is consistent with findings from a previous study. (Houde, Berkowitz and Engen. 2011). We have added a sentence to describe the basis for the deuterium exchange significance cutoff to the Figure 4 legend.

10) Figure 2—figure supplement 1: The SDS-PAGE showed around 90 kDa for the molecular weight of RBDisolated, which should be around 25 kDa based on its sequence (318-547). Please check and clarify.

We thank the reviewer for their query. As described in the Materials and methods, we used a MBP fusion construct of RBD in this study. The total molecular weight for our fusion protein adds up to 90 kDa. Figure 2—figure supplement 1 legend has been amended to describe this.

11) It is confusing about the existing forms of the S protein and ACE2 and their binding stoichiometry, regarding the statements such as "we measured dynamics of a trimer of this near-full length S protein…", "we performed HDXMS experiments of monomeric ACE2…", "……were pre-incubated at 37°C for 30 min in a molar ratio of 1:1 to achieve >90% binding……".

We thank the reviewer. We have included size exclusion chromatography (SEC) profiles of MBPRBD, Fc-ACE2, S protein in Figure 2—figure supplement 1, Figure 3—figure supplement 1 and Figure 5—figure supplement 1 respectively, which clearly show their elution profiles as monomers, dimers and trimers respectively. Multiple Cryo-EM studies have also confirmed that this Spike double mutant construct used in the current study is trimeric, while Fc tagged-ACE2 is dimeric in solution (Fc forms dimers).

Please confirm whether the expressed ACE2 is dimeric and S protein is trimeric or not, and their binding stoichiometry is 1:1 or 2:3. Please also provide the concentration and calculation details for ensuring the >90% binding. If only one ACE2 in the ACE dimer and one S protein in the S protein trimer are involved in the binding, how sensitive and accurate could the HDX-MS results reflect the binding, since no HDX difference would be observed for the other ACE2 and other 2 S proteins?

In the Materials and methods section, stoichiometries and calculations of the protein concentrations used have been provided. Earlier studies have reported ACE2 binding affinities of 15 nM and 150 nM to S protein and RBD. We used 1:1 ratio of RBDisolated:ACE2 in their monomeric forms. We determined that under our experimental conditions with saturating concentrations of RBD isolated and ACE2, nearly all RBD and ACE2 (>99% bound as per the program % bound; University of California San Diego), indicative of their complexation without any predominant free forms in solution.

For S:ACE2 complex, the open state of RBD alone binds to ACE2. Benton et al. (Nature, 2020, https://doi.org/10.1038/s41586-020-2772-0) have shown that RBD can exist in combination of open and closed state in solution i.e. all closed, one open-, two closed or two open, one closed or three open states making it difficult to predict the exact binding stoichiometries to ACE2. However, we estimate complete saturation binding of ACE2 with S protein at concentrations used in our HDXMS experiments (550nM ACE2:550nM S protein) during deuterium exchange conditions. In our HDXMS reaction mixture we have used 1:1 ratios of S protein (monomer): Fc-ACE2 (monomer) and equilibrated for 30 min prior to performing the deuterium labeling to obtain saturated binding.

We observed no bimodal deuterium exchange kinetics for any S protein or ACE2 peptides which would have been observed for substoichiometric complexation of S with ACE2. Deuterium exchange protection was also only marginally lower for S:ACE2 compared to RBDisolated:ACE2, saturation binding under HDXMS experimental conditions.

12) Abstract: Other studies (e.g., Hoffman et al., 2020) have shown that ACE2 binding can enhance S1/S2 cleavage by furin and S1/S2 cleavage site could be possible targets for small molecule inhibitor/antibody development. It would be helpful if further evidence could be provided to support that the stalk hinge regions could also be the targets for that.

We thank the reviewer for this suggestion. The S1/S2 cleavage site for host furin proteases is essential for the S protein to dissociate into individual S1 and S2 subunits for the host-entry process. We envisage this represents a primary hotspot for targeting small molecule inhibitors against SARS-CoV-2.

The stalk provides flexibility to the S protein for binding to ACE2. However, parts of the stalk regions showed contrasting decreased deuterium exchange upon ACE2 binding. Tomogram studies of SARS-CoV-2 virus have shown that S protein binds to ACE2 even at various angles/conformations of the stalk. We don’t have evidence for the efficacy of targeting the stalk hinge regions for better neutralization but rank the S1/S2 cleavage site better as a drug target.

13) The authors should provide more technical details of the molecular dynamics simulations in the supplementary materials. Could the authors provide more details on the equilibration protocol? Was there any analysis done or metric used to assess whether the system was properly equilibrated? How often were snapshots of the trajectory saved for analysis? How many Na+ and Cl- ions were added to achieve 0.15 M of salt concentration? Also, how many water molecules were added? These details are relevant to the non-casual readers.

We thank the reviewer for their query. We have now updated our simulations, using an improved S protein model in which missing loops located in the NTD and the C-terminus of the ECD have been modelled based on a higher resolution structure (PDB: 6XR8). The new model nevertheless exhibited comparable dynamics to our previous model.

As requested by the reviewer, we have now added more technical details of the molecular dynamics simulations in the Materials and methods section, including the equilibration protocol, the frequency of saving frames from trajectories, the number of water molecules as well as Na+ and Cl- ions to the systems. To assess proper system equilibration, we calculated RMSD curves for subdomains of the S protein and found that they converged after around 50 ns. Likewise, we made similar assessments on the basis of calculation of block-wise RMSF profiles. This new data has been added to Figure 2—figure supplement 5.

14) Figure 2—figure supplement 1: Could the authors elaborate on Figure 2—figure supplement 1B in the figure legend? Is (i) measuring the binding of ace2 to the S protein? Is (ii) measuring the binding of RBD to the ace2 protein? The distinction between (i) and (ii) is not made in the figure legend.

We thank the reviewer for this point. We have provided experimental details of the ELISA performed in the Materials and methods section of the revised manuscript.

https://doi.org/10.7554/eLife.63646.sa2

Article and author information

Author details

  1. Palur V Raghuvamsi

    1. Department of Biological Sciences, National University of Singapore, Singapore, Singapore
    2. Bioinformatics Institute, Agency for Science, Technology, and Research (A*STAR), Singapore, Singapore
    Contribution
    Conceptualization, Resources, Data curation, Formal analysis, Visualization, Methodology, Writing - original draft, Writing - review and editing
    Contributed equally with
    Nikhil K Tulsian
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-0897-6935

  2. Nikhil K Tulsian

    1. Department of Biological Sciences, National University of Singapore, Singapore, Singapore
    2. Centre for Life Sciences, Department of Biochemistry, National University of Singapore, Singapore, Singapore
    Contribution
    Conceptualization, Resources, Data curation, Formal analysis, Visualization, Methodology, Writing - original draft, Writing - review and editing
    Contributed equally with
    Palur V Raghuvamsi
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-6577-6748

  3. Firdaus Samsudin

    Bioinformatics Institute, Agency for Science, Technology, and Research (A*STAR), Singapore, Singapore
    Contribution
    Conceptualization, Data curation, Formal analysis, Visualization, Methodology, Writing - original draft
    Competing interests
    No competing interests declared

  4. Xinlei Qian

    Life Sciences Institute, Centre for Life Sciences, National University of Singapore, Singapore, Singapore
    Contribution
    Resources, Data curation, Methodology
    Competing interests
    No competing interests declared

  5. Kiren Purushotorman

    Life Sciences Institute, Centre for Life Sciences, National University of Singapore, Singapore, Singapore
    Contribution
    Resources, Data curation, Methodology
    Competing interests
    No competing interests declared

  6. Gu Yue

    Life Sciences Institute, Centre for Life Sciences, National University of Singapore, Singapore, Singapore
    Contribution
    Resources, Data curation, Methodology
    Competing interests
    No competing interests declared

  7. Mary M Kozma

    Life Sciences Institute, Centre for Life Sciences, National University of Singapore, Singapore, Singapore
    Contribution
    Resources, Data curation, Methodology
    Competing interests
    No competing interests declared

  8. Wong Y Hwa

    School of Biological Sciences, Nanyang Technological University, Singapore, Singapore
    Contribution
    Resources
    Competing interests
    No competing interests declared

  9. Julien Lescar

    School of Biological Sciences, Nanyang Technological University, Singapore, Singapore
    Contribution
    Methodology
    Competing interests
    No competing interests declared

  10. Peter J Bond

    1. Department of Biological Sciences, National University of Singapore, Singapore, Singapore
    2. Bioinformatics Institute, Agency for Science, Technology, and Research (A*STAR), Singapore, Singapore
    Contribution
    Conceptualization, Resources, Data curation, Software, Formal analysis, Supervision, Validation, Investigation, Visualization, Methodology, Writing - original draft, Project administration, Writing - review and editing
    For correspondence
    peterjb@bii.a-star.edu.sg
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-2900-098X

  11. Paul A MacAry

    Life Sciences Institute, Centre for Life Sciences, National University of Singapore, Singapore, Singapore
    Contribution
    Conceptualization, Resources, Data curation, Supervision, Funding acquisition, Validation, Investigation, Project administration, Writing - review and editing
    For correspondence
    micpam@nus.edu.sg
    Competing interests
    No competing interests declared

  12. Ganesh S Anand

    1. Department of Biological Sciences, National University of Singapore, Singapore, Singapore
    2. Current address: Department of Chemistry, Department of Biochemistry and Molecular Biology, Center for Infectious Disease Dynamics -Huck Institute of the Life Sciences, The Pennsylvania State University, University Park, United States
    Contribution
    Conceptualization, Formal analysis, Supervision, Validation, Investigation, Writing - original draft, Project administration, Writing - review and editing
    For correspondence
    gsa5089@psu.edu
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-8995-3067

Funding

National Medical Research Council (WBS#R-571-000-081-213 Establishment of assays for drug screening and virus characterization of the newly emerged novel coronavirus (2019-nCoV) which is also known as the Wuhan coronavirus)

  • Paul A MacAry

A*STAR Bioinformatics Institute

  • Peter J Bond

National University of Singapore

  • Palur V Raghuvamsi

Ministry of Education - Singapore (MOE2017-T2-A40-112)

  • Nikhil K Tulsian
  • Ganesh S Anand

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

We thank Dr. Lu Gan, Dept. of Biological Sciences, National University of Singapore, Sean Braet, Theresa Buckley and Varun Venkatakrishnan, Dept. of Chemistry, the Pennsylvania State University for helpful discussions. Additionally, we thank reviewers and a reader for their feedback. We thank Protein Production Platform of Nanyang Technological University for their help in making the RBD and ACE2 expression constructs and small-scale protein expression tests. HDXMS experiments were carried out as a fee for service at the Singapore National Laboratory for Mass Spectrometry (SingMass) funded by NRF, Singapore. PVR was supported by research scholarship from National University of Singapore, Singapore. NKT was supported by research grant from Ministry of Education, Singapore awarded to GSA (MOE2017-T2-A40-112). This work was supported by BII of A*STAR. Simulations were performed on the petascale computer cluster ASPIRE-1 at the National Supercomputing Centre of Singapore (NSCC) and the A*STAR Computational Resource Centre (A*CRC).

Senior Editor

  1. Olga Boudker, Weill Cornell Medicine, United States

Reviewing Editor

  1. Donald Hamelberg, Georgia State University, United States

Reviewers

  1. Donald Hamelberg, Georgia State University, United States
  2. Elizabeth Komives, UCSD, United States

Publication history

  1. Received: October 1, 2020
  2. Accepted: February 5, 2021
  3. Accepted Manuscript published: February 8, 2021 (version 1)
  4. Version of Record published: March 4, 2021 (version 2)

Copyright

© 2021, Raghuvamsi et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 8,223
    Page views
  • 1,037
    Downloads
  • 46
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, Scopus, PubMed Central.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Palur V Raghuvamsi
  2. Nikhil K Tulsian
  3. Firdaus Samsudin
  4. Xinlei Qian
  5. Kiren Purushotorman
  6. Gu Yue
  7. Mary M Kozma
  8. Wong Y Hwa
  9. Julien Lescar
  10. Peter J Bond
  11. Paul A MacAry
  12. Ganesh S Anand
(2021)
SARS-CoV-2 S protein:ACE2 interaction reveals novel allosteric targets
eLife 10:e63646.
https://doi.org/10.7554/eLife.63646

Categories and tags

Further reading

    1. Epidemiology and Global Health
    2. Medicine
    3. Microbiology and Infectious Disease

    COVID-19: A Collection of Articles

    Edited by Diane M Harper et al.
    Collection

    eLife has published the following articles on SARS-CoV-2 and COVID-19.

    1. Biochemistry and Chemical Biology
    2. Structural Biology and Molecular Biophysics

    Timeline of changes in spike conformational dynamics in emergent SARS-CoV-2 variants reveal progressive stabilization of trimer stalk with altered NTD dynamics

    Sean M Braet, Theresa SC Buckley ... Ganesh S Anand
    Research Article Updated

    SARS-CoV-2 emergent variants are characterized by increased viral fitness and each shows multiple mutations predominantly localized to the spike (S) protein. Here, amide hydrogen/deuterium exchange mass spectrometry has been applied to track changes in S dynamics from multiple SARS-CoV-2 variants. Our results highlight large differences across variants at two loci with impacts on S dynamics and stability. A significant enhancement in stabilization first occurred with the emergence of D614G S followed by smaller, progressive stabilization in subsequent variants. Stabilization preceded altered dynamics in the N-terminal domain, wherein Omicron BA.1 S showed the largest magnitude increases relative to other preceding variants. Changes in stabilization and dynamics resulting from S mutations detail the evolutionary trajectory of S in emerging variants. These carry major implications for SARS-CoV-2 viral fitness and offer new insights into variant-specific therapeutic development.

    1. Biochemistry and Chemical Biology

    Reserpine maintains photoreceptor survival in retinal ciliopathy by resolving proteostasis imbalance and ciliogenesis defects

    Holly Y Chen, Manju Swaroop ... Anand Swaroop
    Research Article

    Ciliopathies manifest from sensory abnormalities to syndromic disorders with multi-organ pathologies, with retinal degeneration a highly penetrant phenotype. Photoreceptor cell death is a major cause of incurable blindness in retinal ciliopathies. To identify drug candidates to maintain photoreceptor survival, we performed an unbiased, high-throughput screening of over 6,000 bioactive small molecules using retinal organoids differentiated from induced pluripotent stem cells (iPSC) of rd16 mouse, which is a model of Leber congenital amaurosis (LCA) type 10 caused by mutations in the cilia-centrosomal gene CEP290. We identified five non-toxic positive hits, including the lead molecule reserpine, which maintained photoreceptor development and survival in rd16 organoids. Reserpine also improved photoreceptors in retinal organoids derived from induced pluripotent stem cells of LCA10 patients and in rd16 mouse retina in vivo. Reserpine-treated patient organoids revealed modulation of signaling pathways related to cell survival/death, metabolism, and proteostasis. Further investigation uncovered dysregulation of autophagy associated with compromised primary cilium biogenesis in patient organoids and rd16 mouse retina. Reserpine partially restored the balance between autophagy and the ubiquitin-proteasome system at least in part by increasing the cargo adaptor p62, resulting in improved primary cilium assembly. Our study identifies effective drug candidates in preclinical studies of CEP290 retinal ciliopathies through cross-species drug discovery using iPSC-derived organoids, highlights the impact of proteostasis in the pathogenesis of ciliopathies, and provides new insights for treatments of retinal neurodegeneration.