Peptide Antidotes to SARS-CoV-2 (COVID-19)

  1. Andre Watson
  2. Leonardo Ferreira
  3. Peter Hwang
  4. Jinbo Xu
  5. Robert Stroud

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

The design of an immunogenic scaffold that serves a role in treating a pathogen, and can be rapidly and predictively modeled, has remained an elusive feat. Here, we demonstrate that SARS-BLOCK™ synthetic peptide scaffolds act as antidotes to SARS-CoV-2 spike protein-mediated infection of human ACE2-expressing cells . Critically, SARS-BLOCK™ peptides are able to potently and competitively inhibit SARS-CoV-2 S1 spike protein receptor binding domain (RBD) binding to ACE2, the main cellular entry pathway for SARS-CoV-2, while also binding to neutralizing antibodies against SARS-CoV-2. In order to create this potential therapeutic antidote-vaccine, we designed, simulated, synthesized, modeled epitopes, predicted peptide folding, and characterized behavior of a novel set of synthetic peptides. The biomimetic technology is modeled off the receptor binding motif of the SARS-CoV-2 coronavirus, and modified to provide enhanced stability and folding versus the truncated wildtype sequence. These novel peptides attain single-micromolar binding affinities for ACE2 and a neutralizing antibody against the SARS-CoV-2 receptor binding domain (RBD), and demonstrate significant reduction of infection in nanomolar doses. We also demonstrate that soluble ACE2 abrogates binding of RBD to neutralizing antibodies, which we posit is an essential immune-evasive mechanism of the virus. SARS-BLOCK™ is designed to “uncloak” the viral ACE2 coating mechanism, while also binding to neutralizing antibodies with the intention of stimulating a specific neutralizing antibody response. Our peptide scaffolds demonstrate promise for future studies evaluating specificity and sensitivity of immune responses to our antidote-vaccine. In summary, SARS-BLOCK™ peptides are a promising COVID-19 antidote designed to combine the benefits of a therapeutic and vaccine, effectively creating a new generation of prophylactic and reactive antiviral therapeutics whereby immune responses can be enhanced rather than blunted.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2020.08.06.238915: (What is this?)

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

    Table 1: Rigor

    NIH rigor criteria are not applicable to paper type.

    Table 2: Resources

    Antibodies
    SentencesResources
    Furthermore, ACE2-his (Sino Biological) and RBD-his (Sino Biological) exhibited extremely weak binding to HIS1K tips, so we opted to utilize dimeric-ACE2-biotin (UCSF) and RBD-biotin (UCSF) on SA tips, as well as neutralizing monoclonal IgG antibody against the SARS-CoV-2 spike glycoprotein (CR3022, antibodies-online) on AHC tips for all studies.
    RBD-biotin
    suggested: None
    A neutralizing monoclonal IgG antibody against the SARS-CoV-2 spike glycoprotein (CR3022, antibodies-online), ACE2 (Sino Biological)
    SARS-CoV-2 spike glycoprotein ( CR3022 , antibodies-online)
    suggested: None
    ACE2
    suggested: None
    Software and Algorithms
    SentencesResources
    SWISS-MODEL was utilized to generate a SARS-CoV-2 spike protein structure prior to the availability of Cryo-EM or X-ray crystallography data in February of 2020.64,65,66,67,68,69 The SARS-CoV-2 spike protein structure was aligned with SARS-CoV-1 spike protein bound to ACE2 (PDB ID 6CS2) using PyMOL (The PyMOL Molecular Graphics System, Version 2.3.5 Schrödinger, LLC.).
    PyMOL
    suggested: (PyMOL, RRID:SCR_000305)
    Other epitopes may be flexibility included within these modular stretches of Peptide 5 and homologues, representing 9- and 13-mer amino acid sequences corresponding to MHC-I and MHC-II binding, respectively.77,78 PEPTIDE SYNTHESIS: The best performing in silico peptides were synthesized via microwave-assisted solid-phase peptide synthesis (SPSS) by a commercial manufacturer, sb-PEPTIDE (France)
    SPSS
    suggested: (SPSS, RRID:SCR_002865)

    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: We detected the following sentences addressing limitations in the study:
    The synthetic nature, in silico screening, and precise conformation of these peptides allows for rapid synthesis without traditional limitations of recombinant, live-attenuated, gene delivery system, viral vector, and/or inactivated viral vaccine approaches. Due to the click chemistry nature of SARS-BLOCK™, it may also serve as a drug and gene delivery carrier by modifications with electrostatic sequences, or by click chemistry onto lipidic or other nanoparticles. We envision SARS-BLOCK™ and future permutations of these compounds vastly facilitating the design, development and scale-up of precise antidotes and vaccines against a variety of infectious agents as part of broader biodefense initiatives. SARS-BLOCK™ peptides are designed to overcome many limitations associated with antibody therapies, ACE2-Fc therapies, and other antiviral therapeutics. Though neutralizing antibodies may be used as “stopgap” therapeutics to prevent the progression of disease, the transient nature of administered antibodies leaves the organism susceptible to reinfection. Furthermore, as we demonstrate in this study, ACE2 is a potent inhibitor of neutralizing antibody binding to the SARS-CoV-2 spike protein receptor binding domain, and this may relate to why short-lived antibody responses are a hallmark component of SARS-CoV-1 infections in many patients, whereby SARS-CoV-2 exhibits as much as 10-15x stronger binding to ACE2 than SARS-CoV-1 and would likely exhibit even stronger antibody maturation ...

    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: We did not find any issues relating to colormaps.

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

    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.

  2. ScreenIT

    SciScore for 10.1101/2020.08.06.238915: (What is this?)

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

    Table 1: Rigor

    Institutional Review Board Statementnot detected.Randomizationnot detected.Blindingnot detected.Power Analysisnot detected.Sex as a biological variablenot detected.

    Table 2: Resources

    Antibodies
    SentencesResources
    3) Neutralizing IgG antibody was immobilized on AHC tips (~1nm capture). a) Peptides 1, 4, 5 and 6 were introduced to immobilized ACE2 in concentrations of 0.37, 1.11, 3.33 and 10μM (Figure 4a - 4d). b) RBD-his (Sino Biological) was introduced to immobilized neutralizing antibody (CR3022, antibodies-online) in concentrations of 1, 3, 9, 27 and 81μM. c) 117nM RBD-his (Sino Biological) was mixed with ACE2-his in concentrations of 0 (RBD-only), 2.88, 8.63, 25.9, and 77.7μM, and then introduced to immobilized neutralizing antibody (CR3022, antibodies-online).
    Neutralizing IgG
    suggested: None
    CR3022
    suggested: (Imported from the IEDB Cat# CR3022, RRID:AB_2848080)
    antibodies-online
    suggested: (Antibodies-Online Cat# ABIN335322, RRID:AB_10783489)
    77.7μM
    suggested: None
    antibodies-online).
    suggested: None
    Software and Algorithms
    SentencesResources
    SWISS-MODEL was utilized to generate a SARS-CoV-2 spike protein structure prior to the availability of Cryo-EM or X-ray crystallography data in February of 2020.​64,65,66,67,68,69 The SARS-CoV-2 spike protein structure was aligned with SARS-CoV-1 spike protein bound to ACE2 (PDB ID 6CS2) using PyMOL (The PyMOL Molecular Graphics System, Version 2.3.5 Schrödinger, LLC.).
    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: We detected the following sentences addressing limitations in the study:

    The synthetic nature, in silico screening, and precise conformation of these peptides allows for rapid synthesis without traditional limitations of recombinant, live-attenuated, gene delivery system, viral vector, and/or inactivated viral vaccine approaches. Due to the click chemistry nature of SARS-BLOCK™, it may also serve as a drug and gene delivery carrier by modifications with electrostatic sequences, or by click chemistry onto lipidic or other nanoparticles. We envision SARS-BLOCK™ and future permutations of these compounds vastly facilitating the design, development and scale-up of precise antidotes and vaccines against a variety of infectious agents as part of broader biodefense initiatives. SARS-BLOCK™ peptides are designed to overcome many limitations associated with antibody therapies, ACE2-Fc therapies, and other antiviral therapeutics. Though neutralizing antibodies may be used as “stopgap” therapeutics to prevent the progression of disease, the transient nature of administered antibodies leaves the organism susceptible to reinfection. Furthermore, as we demonstrate in this study, ACE2 is a potent inhibitor of neutralizing antibody binding to the SARS-CoV-2 spike protein receptor binding domain, and this may relate to why short-lived antibody responses are a hallmark component of SARS-CoV-1 infections in many patients, whereby SARS-CoV-2 exhibits as much as 10-15x stronger binding to ACE2 than SARS-CoV-1 and would likely exhibit even stronger antibody maturation issues according to this hypothesis.​40,41,42,43 Therapeutics that mimic ACE2 and shield this key epitope are likely to bias antibody formation towards off-target sites, which could contribute to antibody-dependent enhancement (ADE), vaccine-associated enhanced respiratory distress (VAERD), and a host of other immunological issues upon repeat viral challenge.​44,45,46,47,48,49,50,51,52,52 With SARS-CoV-1, a marked lack of peripheral memory B cell responses was observed in patients 6 years following infection.​53 These key issues are also important to consider in vaccine development, as there is precedent for enhanced respiratory disease in vaccinated animals with SARS-CoV-1.​54 Thus, any approach that promotes a specific and neutralizing immune response, whether freestanding or in conjunction with another vaccine approach or infection, should be considered as an alternative to immunosuppressant and potentially off-target antibody forming approaches. In particular, any approaches that have potential to limit endogenous antibody formation should be carefully reconsidered, due to the viral immune-evasive techniques already spanning a gamut of mechanisms, including but not limited to the S1 spike protein switching between “open” and “closed” conformations, heavy glycosylation limiting accessible regions, and also the presentation of T cell evasion due to MHC downregulation on infected cells and potential MHC-II binding of the SARS-COV-2 spike protein limiting CD4+ T cell responses, which all may be factors in contributing to T cell exhaustion and ineffective and/or transient antibody and memory B cell responses in infected patients.​55,56,57,58 Indeed, severe and critically ill patients exhibit extreme B cell activation and, presumably, antibody responses.​59,60 Yet, poor clinical outcomes are seen, suggesting that immune evasion and/or off-target antibody formation is dominant. The extent to which various factors individually play in contributing to these phenomena remains poorly understood. Surely, COVID-19 presents itself as a multifactorial disease with a cascade of deleterious effects. Also, the potential for reinfection across cohorts of varying disease severity remains to be fully elucidated, though numerous clinical and anecdotal reports indicate that immunity to coronaviruses is markedly short-lived, with seasonal variations in susceptibility to reinfection with alpha- and betacoronaviruses being frequently observed, and some antibody responses lasting for no longer than 3 months.​61 ​With SARS-CoV-2 in particular, patients developing moderate antibody responses are seen to have undetectable antibodies in as little as 50 days.​62 Additionally, one study on 149 recovered individuals reported that 33% of study participants did not generate detectable neutralizing antibodies 39 days following symptom onset, and that the majority of the cohort did not have high neutralizing antibody activity.​63 ​In sum, any strategy that risks further contributing to these immune-evasive properties warrants caution. An ideal therapeutic strategy should enhance neutralizing antibody formation, not blunt it, while also preventing the virus from entering cells and replicating. In our study, we offer clues as to the role of soluble ACE2 in abrogation neutralizing antibody binding to the viral spike protein RBD, and a potential therapeutic solution to this immune evasion. Importantly, we believe that SARS-BLOCK​™ ​peptides will be useful as both antidotes and vaccines, due to the presence of key epitopes for antibody formation, and the performance of Peptide 5 (which exhibits one MHC-I and one MHC-II epitope) within these experiments suggests future utility for our multifunctional scaffolds in eliciting T cell responses against dominant viral epitopes, including ones that are not on the spike protein. The MHC-I and MHC-II domains can be flexibly substituted to match HLA types in various populations, or pooled across panels of peptides exhibiting multiple domains. Because SARS-BLOCK​™ mimics the virus, rather than binding to it, and also due its its ability to displace ACE2 from cloaking the virus spike protein, we believe that these multifunctional peptides will prove to be an effective immune-enhancing strategy in infected patients, with additional potential to serve as a prophylactic vaccine. Therefore, SARS-BLOCK​™ ​peptides are an elegant solution towards preventing viral association with ACE2 and infection, while also contributing to a decrease in soluble ACE2 shielding of the virus. The thermodynamically favorable interaction of an antibody with the virus (~6nM kD with the neutralizing antibody we studied) versus our peptides (~1uM kD) suggests that the peptides can dissociate ACE2, promote antibody formation against the virus during infection, and preferentially train the immune system to eliminate the virus. Additionally, SARS-BLOCK™ effectively blocks infection by viruses displaying SARS-CoV-2 spike proteins. Our future studies will examine the improvements and therapeutic/vaccine utility of these peptides when displayed multivalently versus as free peptides, as well as refinements to structure and binding to further improve binding affinity and physiological behavior. MATERIALS AND METHODS SIMULATION AND DOCKING OF SARS-COV-2 SPIKE PROTEIN IN THE ABSENCE OF STRUCTURAL DATA In order to elucidate the binding motif of the receptor binding domain, in the absence of structural data, it was necessary to utilize the results of prior crystallography experiments on SARS-CoV-1 with ACE2. SWISS-MODEL was utilized to generate a SARS-CoV-2 spike protein structure prior to the availability of Cryo-EM or X-ray crystallography data in February of 2020.​64,65,66,67,68,69 The SARS-CoV-2 spike protein structure was aligned with SARS-CoV-1 spike protein bound to ACE2 (PDB ID 6CS2) using PyMOL (The PyMOL Molecular Graphics System, Version 2.3.5 Schrödinger, LLC.). This structure was then run through PDBePISA to determine the Gibbs free energy (∆G) and predicted amino acid interactions between the SARS-CoV-2 spike protein and the ACE2 receptor.​70 Upon the availability of structural data, this approach was compared and determined to have correctly identified the stretches of amino acids necessary for binding to ACE2. DESIGN AND SIMULATION OF SYNTHETIC PEPTIDES MIMICKING SPIKE PROTEIN RECEPTOR BINDING MOTIF Following the simulation of structure and binding of the SARS-CoV-2 spike protein receptor binding domain (RBD), a truncated receptor binding motif (RBM) was gathered. This sequence was designed to recreate the structure of the larger protein in this motif, with key modifications performed to facilitate beta sheet formation. These modified peptide sequences were then simulated using RaptorX. RaptorX is an efficient and accurate protein structure prediction software package, building upon a powerful deep learning technique.​71,72 ​Given a sequence, RaptorX runs a homology search tool HHblits to find its sequence homologs and build a multiple sequence alignment (MSA), and then derives sequence profile and inter-residue coevolution information.​73 Afterwards, RaptorX feeds sequence profile and coevolution information to a very deep convolutional residual neural network (of ~100 convolution layers) to predict inter-atom distance (i.e., Ca-Ca, Cb-Cb and N-O distance) and inter-residue orientation distribution of the protein under prediction. To predict inter-atom distance distribution, RaptorX discretizes the Euclidean distance between two atoms into 47 intervals: 0-2, 2-2.4, 2.4-2.8, 2.8-3.2,..., 19.6-20, and > 20A. To predict inter-residue orientation distribution, RaptorX discretizes the orientation angles defined in into bins of 10 degrees.​74 Finally, RaptorX derives distance and orientation potential from the predicted distribution and builds 3D models of the protein by minimizing the potential. Experimental validation indicates that such a deep learning technique is able to predict correct folds for many more proteins than ever before and outperforms comparative modeling unless proteins under prediction have very close homologs in PDB (Protein Data Bank). For our studies, the peptides in their 9 possible folded states were overlaid with the SARS-CoV-2 receptor binding domain docked to ACE2 using PyMOL align commands to approximate binding, and then exported to PDBePISA to estimate binding pockets before synthesis and analytical characterization. IMMUNE EPITOPE MAPPING AND INCLUSION WITHIN PEPTIDES IEDB is a predictive epitope discovery tool for determining possible MHC-I, MHC-II, and non-classical MHC restricted binding epitopes across various HLA genotypes.​75 We utilized IEDB to predict key epitopes prior to clinical data emerging on various T cell receptor (TCR) responses across populations with various HLA alleles, and compared epitopes on SARS-CoV-1 with known immunogenicity to predicted and conserved epitopes for MHC-I and MHC-II response on SARS-CoV-2.​76 A key region of SARS-CoV-1 (LPDPLKPTKRSFIED​LLFNKVTLA​DAGFMKQYG) was defined based on known immunogenicity of the monovalent peptide in terms of its ability to elicit an MHC-I response and antibody response, whereas many other peptides were only immunogenic while present multivalently. Next, a known MHC-II domain from SARS-CoV-1 was also defined (ASANLAAT​KMSECVLGQSKRV​DFCGKGYH). These two peptides were compared to the stretches in SARS-CoV-2, and shown to have a high degree of homology in SARS-CoV-2, with sequences QILPDPSKPSKRSFIEDLLFNKVTLADAGFIK (804-835) and ASANLAATKMSECVLGQSKRVDFCGKGY (1020-1047). IEDB determined that sequences KMSECVLGQSKRV and LLFNKVLTA of SARS-CoV-2, representing MHC-II and MHC-I binding domains for HLA-A*02:01, respectively, would be immunogenic with percentile ranks of 0.9 and 1.2, respectively. These sequences were included within a non-interfacing loop structure of Peptide 5, whereas Peptide 4 included a GSGSG linker and Peptide 6 included the wildtype receptor binding motif (RBM) sequence for the loop. Other epitopes may be flexibility included within these modular stretches of Peptide 5 and homologues, representing 9- and 13-mer amino acid sequences corresponding to MHC-I and MHC-II binding, respectively.​77,78 PEPTIDE SYNTHESIS The best performing ​in silico p ​ eptides were synthesized via microwave-assisted solid-phase peptide synthesis (SPSS) by a commercial manufacturer, sb-PEPTIDE (France). Mass spectrometry was utilized to confirm the appropriate peptide molecular weights. BIOLAYER INTERFEROMETRY Biolayer interferometry is a label-free method for measuring the wavelength shift of incident white light following loading of a ligand upon a sensor tip surface, and/or binding of soluble analytes to that ligand or the sensor tip surface.​79 The wavelength shift corresponds to the amount of analyte present, and can be used to determine dissociation constants and competition between multiple analytes and the immobilized ligand. An Octet® RED384 biolayer interferometer (Fortebio) was utilized with sensor tips displaying anti-human IgG Fc (AHC), streptavidin (SA), nickel-charged tris-nitriloacetic acid (NTA), or anti-penta-his (HIS1K) in 96-well plates. For streptavidin tips, we utilized 1mM biotin to block the surface after saturation with a given immobilized ligand. After protocol optimization with his-tagged vs. biotin-tagged variants of ACE2 and RBD, we determined that peptide analytes in solution exhibited nonspecific binding to the sensor tip surface with NTA and HIS1K tips, whereas biotinylated surfaces minimized this non-specific binding. Furthermore, ACE2-his (Sino Biological) and RBD-his (Sino Biological) exhibited extremely weak binding to HIS1K tips, so we opted to utilize dimeric-ACE2-biotin (UCSF) and RBD-biotin (UCSF) on SA tips, as well as neutralizing monoclonal IgG antibody against the SARS-CoV-2 spike glycoprotein (CR3022, antibodies-online) on AHC tips for all studies. Non-specific binding was still observed with Peptide 5 binding to a neutralizing antibody on AHC tips, which complicated efforts of determining the Kd of the Peptide 5 analyte vs. the neutralizing antibody ligand. All stock solutions were prepared in 1X PBS containing 0.2% BSA and 0.02% Tween20. Next, the following ligands and analytes were studied: 1) Dimeric ACE2-biotin was immobilized on SA tips (~2.5nm capture). a) Peptides 1, 4, 5 and 6 were introduced to immobilized ACE2 in concentrations of 1, 3 and 10μM (Figure 3a - 3d). b) Sensor tips were removed from peptide solutions and introduced to 35μM RBD-his (Sino Biological) (Figure 3e - 3h). 2) RBD-biotin was immobilized on SA tips (~5nm capture). a) ACE2-his (Sino Biological) was introduced to immobilized RBD in concentrations of 1.3, 3.9, 11.7, 35 and 105μM (Figure 3i). 3) Neutralizing IgG antibody was immobilized on AHC tips (~1nm capture). a) Peptides 1, 4, 5 and 6 were introduced to immobilized ACE2 in concentrations of 0.37, 1.11, 3.33 and 10μM (Figure 4a - 4d). b) RBD-his (Sino Biological) was introduced to immobilized neutralizing antibody (CR3022, antibodies-online) in concentrations of 1, 3, 9, 27 and 81μM. c) 117nM RBD-his (Sino Biological) was mixed with ACE2-his in concentrations of 0 (RBD-only), 2.88, 8.63, 25.9, and 77.7μM, and then introduced to immobilized neutralizing antibody (CR3022, antibodies-online). INFECTION OF ACE2-HEK293 WITH SARS-COV-2 SPIKE PROTEIN PSEUDOTYPED LENTIVIRUS ACE2-HEK293s (BPS Bioscience) were cultured and transduced in opaque 96-well white plates (Corning®) with pseudotyped lentivirions displaying the SARS-CoV-2 spike glycoprotein (BPS Bioscience). A neutralizing monoclonal IgG antibody against the SARS-CoV-2 spike glycoprotein (CR3022, antibodies-online), ACE2 (Sino Biological), receptor-binding domain (RBD) of spike glycoprotein (Sino Biological), and SARS-BLOCK​™ ​peptides (Ligandal) were used as inhibitors of infection. Infection was quantitated via bioluminescence, and toxicity was characterized via a trypan blue absorbance assay utilizing a Synergy™ H1 BioTek spectrophotometer 60h following viral transduction. The cells were stained with trypan blue for 15 minutes, washed 3x with ice-cold calcium- and magnesium-containing 1x PBS, and lysed before transfer to a clear-bottom plate for absorbance measurements as described elsewhere.​80

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

    Results from JetFighter: We did not find any issues relating to colormaps.

    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.

  3. Version published to 10.1101/2020.08.06.238915v1 on bioRxiv
  4. Version published to 10.1101/2020.08.06.238915v2 on bioRxiv