Intranasal trimeric sherpabody inhibits SARS-CoV-2 including recent immunoevasive Omicron subvariants

Figures

• 

  Fig. 1. Discovery of RBD-targeting sherpabodies. a, b Six and three residues in the RT- and n-Src-loops, respectively, of the human nephrocystin SH3 domain were replaced with hexapeptides (in red) comprising random combinations of the 20 natural amino acids to create a large semisynthetic M13 phage library displaying ~1011 individual sherpabodies with unique binding surfaces formed by the 12 randomized loop residues shown red in the illustration based on the structure pdb 1S1N (https://www.ncbi.nlm.nih.gov/Structure/pdb/1S1N)13. c Heat map showing relative binding in an ELISA assay of fifteen affinity-selected sherpabody-displaying phage clones targeted to RBD of SARS-CoV-2 spike protein (RBD-mFc), negative control (mFc), or an antibody evenly recognizing all displayed clones (anti-E-tag). d Semi-quantitative ELISA-based analysis of the RBD-binding affinity (30 nM) of the sherpabody clone Sb92. The experiments were repeated independently three times with similar results. A representative assay is shown. Source data are provided as a Source Data file.

• 

  Fig. 2. Neutralization of SARS-CoV-2 and SARS-CoV-1 by TriSb92. Half-maximal inhibitory concentration (IC50) of TriSb92 against different authentic SARS-CoV-2 variants or pseudoviruses using the indicated spike proteins were established based on automated counting of fluorescently labelled infected cells (see figure insert) or luciferase reporter gene expression, respectively. Standard deviation for the calculated IC50 values is shown. The pseudovirus neutralization experiments were repeated independently three or more times with similar results. Average and standard deviations of representative assays performed in duplicates is shown. Neutralization of authentic viruses were repeated with similar results three times for Wuhan, twice for Delta, BA.1, and BA.5 and once for Beta in experiments performed in quintuplicate and involved counting of approximately 25,000 cells per condition after immunostaining illustrated in the figure insert. The original data are provided as a Source Data file. An alignment of RBD sequences of spike variants used in this study is shown in Fig. S1.

• 

  Fig. 3. TriSb92 does not prevent virus attachment to target cells but prevents infection by inducing a conformational change in spike. a Spike-decorated pseudoviruses were treated with TriSb92, COVID-19 convalescent serum (Patient), or left untreated (None) before incubation with ACE2-expressing HEK293 cells. The amount of cell-associated pseudoviral capsid protein (p24) after washing of the cultures was determined by Western blotting. b–e Spike-pseudotyped virions treated (+) or not treated (−) with TriSb92 were lysed in SDS-PAGE loading buffer and warmed (37 °C) or boiled (100 °C) before analysis by Western blotting using antibodies specific for the p24 capsid protein (b), spike S1 subunit (c), spike S2 subunit (d), or COVID-19 convalescent serum that can recognize S1 and S2 (e). The experiment 3a-e was repeated independently three times with similar results. Uncropped original scans of all blots presented in panels a–e are provided as Source Data files.

• 

  Fig. 4. Prophylactic and early therapeutic efficacy of TriSb92 in mice. a Subgenomic viral RNA was quantified in lung tissues of Balb/c mice 2 days after challenge with 2 × 105 PFU of SARS-CoV-2 B.1.351 in three independent infection experiments. The infected mice were treated with intranasal administration TriSb92 at the indicated doses and times before or after the infection. b SARS-CoV-2 nucleoprotein immunostaining in the nasal cavity and lung tissues 2 days after infection in an untreated mouse (left) and in a mouse pretreated with 25 μg of TriSb92 per nostril (right). The untreated animal exhibits viral antigen expression in the nasal respiratory epithelium (RE) and olfactory epithelium (OE). In the lung, several bronchioles (B) exhibit viral antigen in a variable number of epithelial cells that are often degenerate (arrowheads) or sloughed off (arrows). In infected alveoli, both type I pneumocytes (arrowhead) and type II pneumocytes (arrow) are positive. The bronchial lymph node (top inset) exhibits several positive cells (macrophages or dendritic cells). The animal that received TriSb92 does not exhibit any evidence of viral antigen expression in nasal epithelium, lung and…

• 

  Fig. 5. TriSb92 can protect from a lethal challenge with mouse-adapted SARS-CoV-2. Body weight development of Balb/c mice that were infected with 1 × 106 or 5 × 106 PFU of the pathogenic SARS-CoV-2 strain maVie16 following intranasal administration TriSb92 (25 μg of per nostril; red lines and symbols) or the same volume of saline buffer (grey lines and symbols) 1 h earlier, as indicated. Four control mice received similar mock treatment but no TriSb92 or virus (blue lines and symbols). The experiment was terminated at day 4 post-infection when all infected mice that had not received TriSb92 prophylaxis had been euthanized because of losing 15% or more of their weight (crossed empty symbols). This experiment was performed once. Original data on mouse weight development are provided as Source Data files.

• 

  Fig. 6. Cryo-EM of SARS-CoV-2 S in complex with TriSb92 reveals the highly conserved TriSb92 binding site. a Side and top views of the cryoEM reconstruction of spike trimer in complex with TriSb92. b Locally refined cryoEM reconstruction of the RBD region, with fitted models of Sb92 (generated with I-TASSER) and the RBD (PDB 7KMS; https://www.ncbi.nlm.nih.gov/Structure/pdb/7KMS). c Surface rendering of the RBD with the ACE2 binding site and the TriSb92 epitope highlighted (upper panel), and the location of amino acid changes in VOCs relative to the TriSb92 epitope highlighted (lower panel). d Pairwise sequence alignment of SARS-CoV-1 and SARS-CoV-2. The binding sites of TriSb92 and ACE2 are highlighted in orange and in purple, respectively. Circles above the RBD residues indicate amino acid changes in Omicron, Delta, and/or Beta variants, and are coloured as follows: white is found in Beta or Delta and later (Omicron) variants; mint green is found in Omicron BA.1 and later subvariants; medium lime green is found in Omicron BA.2, BA.4/5; dark green is found in the most recent Omicron subvariants. The mapped amino acid changes and associated variant information are presented in Fig. S1.