Conformational dynamics and allosteric modulation of the SARS-CoV-2 spike

Curation statements for this article:
  • Curated by eLife

    eLife logo

    Evaluation Summary:

    This manuscript describes timely work on the structural dynamics of the SARS-CoV-2 spike protein that will be of importance to a broad range of scientists with interests in the biology of SARS-CoV-2 and COVID-19 and the function of anti-COVID-19 vaccines and antibodies as well as to molecular biophysicists generally interested in single-molecule imaging, protein dynamics, allostery, and molecular mechanisms. The experiments were very well-designed, controlled, and executed, and the data are of very high quality. Nonetheless, although the conclusions seem to be generally supported by the data and consistent with expectations based on previous findings, there are some concerns regarding the modeling and error analysis of some of the data.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #2 agreed to share their name with the authors.)

This article has been Reviewed by the following groups

Read the full article See related articles

Abstract

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infects cells through binding to angiotensin-converting enzyme 2 (ACE2). This interaction is mediated by the receptor-binding domain (RBD) of the viral spike (S) glycoprotein. Structural and dynamic data have shown that S can adopt multiple conformations, which controls the exposure of the ACE2-binding site in the RBD. Here, using single-molecule Förster resonance energy transfer (smFRET) imaging, we report the effects of ACE2 and antibody binding on the conformational dynamics of S from the Wuhan-1 strain and in the presence of the D614G mutation. We find that D614G modulates the energetics of the RBD position in a manner similar to ACE2 binding. We also find that antibodies that target diverse epitopes, including those distal to the RBD, stabilize the RBD in a position competent for ACE2 binding. Parallel solution-based binding experiments using fluorescence correlation spectroscopy (FCS) indicate antibody-mediated enhancement of ACE2 binding. These findings inform on novel strategies for therapeutic antibody cocktails.

Article activity feed

  1. Evaluation Summary:

    This manuscript describes timely work on the structural dynamics of the SARS-CoV-2 spike protein that will be of importance to a broad range of scientists with interests in the biology of SARS-CoV-2 and COVID-19 and the function of anti-COVID-19 vaccines and antibodies as well as to molecular biophysicists generally interested in single-molecule imaging, protein dynamics, allostery, and molecular mechanisms. The experiments were very well-designed, controlled, and executed, and the data are of very high quality. Nonetheless, although the conclusions seem to be generally supported by the data and consistent with expectations based on previous findings, there are some concerns regarding the modeling and error analysis of some of the data.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #2 agreed to share their name with the authors.)

  2. Reviewer #1 (Public Review):

    This manuscript reports a very timely and interesting single-molecule fluorescence resonance energy transfer (smFRET) study of SARS-CoV-2 spike (S) proteins derived from the ancestral, Wuhan-1 strain (carrying an aspartic acid at amino acid position 614; D614) and the B.1 variant strain (carrying an aspartic acid-to-glycine mutation at amino acid position 614; D614G). The aim of the study was to characterize the dynamics with which the angiotensin-converting enzyme 2 (ACE2) receptor-binding domain (RBD) of the D614 variant of S undergoes transitions between its "down" conformation, in which the receptor-binding motif (RBM) is occluded, and its "up" conformation, in which the RBM is accessible and to assess the effects that the D614G mutation has on these dynamics. Moreover, the authors sought to elucidate if and how the binding of ACE2 or each of a set of antibodies that target different S epitopes to the D614 and D614G variants of S alter these dynamics. Finally, the authors attempt to determine whether the dynamic effects imparted by each of the antibodies interfere with or enhance ACE2 binding to the D614 and D614G variants of S. Interestingly, the authors find that the D614 and D614G variants of S intrinsically fluctuate between the up and down conformations, with the D614G mutation shifting the conformational equilibrium towards the up state through a mechanism in which the mutation stabilizes S in the up state while having little to no effect on the down state. In a very elegant set of results, the authors show that binding of ACE2 also shifts the conformational equilibrium towards the up state through a very similar mechanism, which, in the case of the D614G variant of S, pushes the conformational equilibrium even further towards the up state. The authors then show that binding of each of the antibodies tested to the D614 variant of S again shifts the conformational equilibrium of S towards the up state through a very similar mechanism. One of the most interesting observations in this regard is that even the antibodies targeting epitopes distal to the RBD allosterically exert this conformational effect on the RBD. The authors further observe that the effects of antibody binding to the D614G variant of S are smaller than those of antibody binding to the D614 variant of S. Finally, the authors beautifully show that, together with consideration of whether binding of a particular antibody would be expected to sterically obstruct the RBM, the effects of antibody binding on the conformational equilibrium of the RBD can regulate ACE2 binding.

    Major strengths of this study are the carefully designed and biophysically and biochemically validated S constructs. Careful controls demonstrate that fluorescent labeling of the D614 and D614G variants of S does not seriously impair the ability of ACE2 to bind to S. Moreover, the representative smFRET trace shown in Fig 1 indicates that the smFRET data are of very high quality. In addition, the supporting and complementary fluorescence correlation spectroscopy and ELISA data are also of extremely high quality. Additional strengths of this work are the wide-ranging and unique exploration of the effects that the D614G mutation, ACE2 binding, and antibody interactions have on the dynamics of S as well as the very relevant investigation of if, and to what extent, the dynamic effects imparted by antibody-binding to S interfere with or enhance ACE2 binding to S.

    The study does have a few weaknesses. One concern regards how the authors determined that the smFRET traces were best analyzed using a two-state conformational model. The representative smFRET state shown in Fig 1 suggests the possible existence of more than two FRET states and, correspondingly, more than two conformational states. Nonetheless, there is no discussion in the manuscript regarding how the authors arrived at the conclusion that the data are best described by a two-state model. Another concern regards how the authors determined if, and to what extent, some of the ligand-binding interactions had reached equilibrium/saturation. The methods report that S was incubated with ligands for periods of 60-90 min at room temperature prior to performing the experiments and that the experiments were performed in the continued presence of the ligands, but some of the interactions are quite high affinity and the study does not report if and how the authors determined whether the 60-90 min, room temperature incubations were enough to achieve equilibrium/saturation. Moreover, the number of smFRET traces obtained and analyzed at each experimental condition is lower than what is reported in typical smFRET studies. The relatively low number of smFRET traces exacerbates a somewhat related concern regarding the reproducibility and error analysis of some of the experiments. Specifically, it is not entirely clear whether the smFRET experiments were independently repeated or whether the various types of error analyses applied to the histograms, occupancies, and rates extracted from the smFRET experiments are fully justified.

  3. Reviewer #2 (Public Review):

    Diaz-Salinas et al. use single molecule FRET to probe the conformational changes in a recombinant viral spike glycoprotein from the SARS-CoV-2 virus. The authors introduced enzyme-labeling sites spanning the receptor binding domain (RBD) that report on the conformational transitions in the presence of a soluble ACE2 receptor along with a series of neutralizing antibodies that target different regions of the spike glycoprotein. The work includes important controls to verify that labeling does not destroy native function and antigenicity. A key result is identifying allosteric antibodies that can modulate the conformation even when they bind outside the RBD. Based on kinetic analysis of FRET transitions, the authors conclude that ACE2 binds through a mechanism of conformational selection rather than induced fit and that neutralizing antibodies affect the dynamic timescales of these interconversions. This allosteric modulation of spike protein affects the interaction with the ACE2 receptor with some antibodies inhibiting ACE binding while others enhanced binding. Based on this data, the authors conclude that the point mutant, ACE2 and some antibodies act by lowering the free energy of the bound state while antibodies targeting S2 lower the transition barrier height to facilitate the conformational change. The model is well supported by the kinetic data. Previous smFRET and structural studies identified this conformational switch linked to receptor interactions. The use of recombinant proteins provides an accessible assay for examining this important protein. The interpretation of FRET efficiency values is qualitative buts provides an accessible metric for the critical conformational change. The work provides important information about a naturally occurring point mutant and the effect of interactions with an array of antibodies.

  4. SciScore for 10.1101/2021.10.29.466470: (What is this?)

    Please note, not all rigor criteria are appropriate for all manuscripts.

    Table 1: Rigor

    Ethicsnot detected.
    Sex as a biological variablenot detected.
    Randomizationnot detected.
    Blindingnot detected.
    Power Analysisnot detected.

    Table 2: Resources

    Antibodies
    SentencesResources
    Antibodies: Monoclonal antibodies MAb362 isotypes IgG1 and IgA1 has been described before(32).
    IgA1
    suggested: None
    REGN10987, S309 and CR3022 antibodies heavy and light variable region sequences(33, 34, 58) were synthesized and cloned into pcDNA3.1 vector (Invitrogen™, Thermo Fisher Scientific, Waltham, MA, USA) in-frame with human IgG heavy or light chain Fc fragment.
    CR3022
    suggested: None
    . 2G12 monoclonal antibody was expressed in ExpiCHO-S™ cells through co-transfection of plasmids encoding light and IgG heavy chains(59), using the ExpiFectamine™ CHO transfection kit (Gibco™, Thermo Fisher Scientific, Waltham, MA, USA) according to manufacturer instructions.
    2G12
    suggested: None
    Monoclonal antibodies 4A8 and 1A9 were purchased from BioVision (Milpitas, CA, USA) and GeneTex (Irvine, CA, USA), respectively.
    1A9
    suggested: None
    Anti-6x-His-tag polyclonal antibody, and both HRP-conjugated anti-mouse IgG Fc and anti-human IgG Fc were purchased from Invitrogen™ (Waltham, MA, USA)
    Anti-6x-His-tag
    suggested: None
    We used a rabbit anti-6X-His antibody (Invitrogen™, Waltham, MA, USA) to detect histidine-tagged proteins or mouse 1A9 antibody (GeneTex, Irvine, CA, USA) for specific detection of SARS-CoV-2 SΔTM.
    anti-6X-His
    suggested: None
    Membranes were washed three times with PBS-T and then incubated with secondary HRP-conjugated anti-rabbit IgG (Abcam, Cambridge, UK) or anti-mouse IgG (Invitrogen™, Waltham, MA, USA) antibodies diluted in 0.5% (w/v) skim milk/PBS-T and incubated for one hour at room temperature.
    anti-rabbit IgG
    suggested: None
    anti-mouse IgG
    suggested: None
    As secondary antibodies, HRP-conjugated anti-human kappa antibody (SouthernBiotech, Birmingham, AL, USA) diluted 1:4000 in PBS was used in wells treated with MAb362, CR3022 and S309 antibodies, while HRP-conjugated anti-human IgG Fc (Invitrogen™, Waltham, MA, USA) diluted 1:10,000 in PBS was used in wells treated with REGN10987, 4A8 and 2G12 antibodies.
    anti-human kappa
    suggested: (Abgent Cat# AT4000a, RRID:AB_1554597)
    S309
    suggested: None
    anti-human IgG
    suggested: None
    S1). smFRET imaging: Labeled SΔTM spikes (100-200 nM) were incubated in the absence or presence of unlabeled ACE2 or the indicated antibody at a monomer:ACE2 or monomer:antibody ratio of 1:3 for 90 minutes at room temperature.
    ACE2
    suggested: None
    Two-tailed nonparametric Spearman test with 95% confidence was performed to evaluate the correlation level between the occupancy of SΔTM in the open conformation due to allosteric antibody binding and ACE2 binding (Figs. 4 and 5).
    ACE2 binding (Figs. 4
    suggested: None
    Recombinant DNA
    SentencesResources
    glycoprotein ectodomain (SΔTM) (residues Q14–K1211) with SGAG substitution at the furin cleavage site (R682 to R685), and proline substitutions at K986 and V987, was synthesized by GenScript® (Piscataway, NJ, USA) and inserted into pcDNA3.1(−).
    pcDNA3.1
    suggested: RRID:Addgene_79663)
    SΔTM hetero-trimers for smFRET experiments were expressed by co-transfection with both the untagged SΔTM (D614 or D614G) construct and the corresponding 161/345A4-tagged SΔTM plasmid at a 2:1 molar ratio.
    SΔTM
    suggested: None
    Software and Algorithms
    SentencesResources
    SΔTM concentration was also estimated by densitometric analysis of protein bands on immunoblots with the monoclonal antibody 1A9 as described below, and using ImageJ software v1.52q (NIH, USA).
    ImageJ
    suggested: (ImageJ, RRID:SCR_003070)
    All smFRET data were processed and analyzed using the SPARTAN software (www.scottcblanchardlab.com/software) in Matlab (Mathworks, Natick, MA, USA)(65).
    Matlab
    suggested: (MATLAB, RRID:SCR_001622)
    smFRET trajectories were idealized to a 3-state hidden Markov model and the transition rates were optimized using the maximum point likelihood algorithm(66), implemented in SPARTAN.
    SPARTAN
    suggested: (SPARTAN, RRID:SCR_014901)
    Dissociation constants (KD) were determined using GraphPad Prism version 9.2.0 (GraphPad Software, San Diego, CA, USA)
    GraphPad Prism
    suggested: (GraphPad Prism, RRID:SCR_002798)
    GraphPad
    suggested: (GraphPad Prism, RRID:SCR_002798)
    Structural analysis: Protein structures from RCSB PDB were visualized and analyzed using PyMOL™ software version 2.0.7 (The PyMOL Molecular Graphic System,
    PyMOL™
    suggested: (PyMOL, RRID:SCR_000305)
    PyMOL
    suggested: (PyMOL, RRID:SCR_000305)

    Results from OddPub: We did not detect open data. We also did not detect open code. Researchers are encouraged to share open data when possible (see Nature blog).


    Results from LimitationRecognizer: An explicit section about the limitations of the techniques employed in this study was not found. We encourage authors to address study limitations.

    Results from TrialIdentifier: No clinical trial numbers were referenced.


    Results from Barzooka: We did not find any issues relating to the usage of bar graphs.


    Results from JetFighter: Please consider improving the rainbow (“jet”) colormap(s) used on page 21. At least one figure is not accessible to readers with colorblindness and/or is not true to the data, i.e. not perceptually uniform.


    Results from rtransparent:
    • Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
    • No funding statement was detected.
    • No protocol registration statement was detected.

    Results from scite Reference Check: We found no unreliable references.


    About SciScore

    SciScore is an automated tool that is designed to assist expert reviewers by finding and presenting formulaic information scattered throughout a paper in a standard, easy to digest format. SciScore checks for the presence and correctness of RRIDs (research resource identifiers), and for rigor criteria such as sex and investigator blinding. For details on the theoretical underpinning of rigor criteria and the tools shown here, including references cited, please follow this link.