Loss of FCoV-23 spike domain 0 enhances fusogenicity and entry kinetics

Fig. 1. Architecture and evolution of the FCoV-23 S glycoprotein. a,b, Ribbon diagrams of the cryo-EM structures of prefusion FCoV-23 S-long in two distinct conformations defined by the positioning of D0: the ‘swung out’ conformation (a) and the ‘mixed’ structure (b). The structure with mixed D0 conformations (b) is rendered from a composite model of the global S and local D0 refinements. The positions of the S1 and S2 subunits are indicated. c, Superimposition of FCoV-23 (blue) and CCoV-HuPn-2018 (yellow) S1 subunits with D0 swung out. d, Superposition of the FCoV-23 and CCoV-HuPn-2018 S2 subunits. e, Structure-guided phylogenetic classification of alphacoronavirus and deltacoronavirus RBDs. Conserved APN-interacting residues are shown in red (except for the FCoV1683 RBD, which is an AlphaFold3-predicted structure62). f, Zoomed-in view of FCoV-23 S domain A and RBDs (domain B) showing the conformational masking and glycan shielding of the receptor-binding loops (labelled loop 1 and loop 2) in the context of the S trimer (only part of two protomers are shown for clarity). g, Amino acid sequence conservation of the residues spanning the S1–S2 subunit junction and the S2′ cleavage site for…

Fig. 2. APN is an entry receptor for FCoV-23. a, BLI binding analysis of dimeric human APN-Fc, human APN-Fc(R741T) (glycan knock-in), FcAPN-Fc, FcAPN-Fc(T742R) (glycan knockout), CfAPN-Fc and GgAPN-Fc ectodomains at a concentration of 1 µM to immobilized biotinylated FCoV-23 RBD. b,c, BLI binding kinetic analysis of FcAPN-Fc (b) and CfAPN-Fc (c) to immobilized biotinylated FCoV-23 RBD. Global fit (1:1 model) to the data is shown in black and affinities are reported as Kd,app owing to avidity resulting from the dimeric nature of APN. A representative experiment is shown out of two biological replicates. d,e, Entry of VSV particles pseudotyped with FCoV-23 S-long (d) or FCoV-23 S-short (e) into HEK293T cells transiently transfected with membrane-anchored FcAPN, FcAPN(T742R), CfAPN, SsAPN, GgAPN, human APN or human APN(R741T) orthologues. Each data point represents a biological replicate performed with technical duplicates. Bars represent the mean of 2–3 biological replicates. The colour key is shown in a. RLU, relative luciferase units. f,g, Concentration-dependent inhibition of FCoV-23 S-long (f) and FCoV-23 S-short (g) VSV pseudovirus entry into HEK293T cells transiently transfected with…

Fig. 3. Evaluation of cross-neutralization of FCoV-23 by alphacoronavirus-elicited antibodies. a,b, Dose-dependent neutralization of FCoV-23 S-short (a) and FCoV-23 S-long (b) pseudotyped VSV particles in the presence of various concentrations of the 1AF10 neutralizing monoclonal Fab fragment using HEK293T cells transiently transfected with membrane-anchored FcAPN. Each curve represents a biological replicate (n = 2, using distinct pseudovirus batches) each performed with 2 technical replicates. Error bars represent the s.e.m. of the technical duplicates. c–e, Inhibition of FCoV-23 S-short (c) CCoV-HuPn-2018 S (d) and HCoV-229E S (P100E isolate) (e) VSV pseudovirus entry into HEK293T cells transiently transfected with membrane-anchored FcAPN (FCoV-23, c), CfAPN (CCoV-HuPn-2018, d) or human APN (HCoV-229E, e) in the presence of a 1:10 dilution of HCoV-229E (P100E isolate) S-elicited mouse sera. Each data point represents a biological replicate (n = 2, using distinct pseudovirus batches) performed with technical duplicates. Bars represent the mean of two independent biological replicates. Source data

Fig. 4. Loss of S D0 increases fusogenicity and entry kinetics. a–e, Left, kinetics of cell–cell fusion (GFP+ area) between BHK-21–GFP1–10 cells transiently transfected with FCoV-23 S-long or S-short (a–c) or CCoV-HuPn-2018 S-long or S-short (d,e) with BHK-21–GFP11 target cells transiently transfected with membrane-anchored FcAPN (a) or CfAPN (d), cat macrophage-like Fcwf-CU cells (b), cat epithelial kidney CRFK cells (c) or canine fibroblast A-72 cells (e) endogenously expressing FcAPN or CfAPN and transiently transfected with GFP11. Middle, representative fields of view obtained 18 h (a,c,d,e) or 5.5 h (b) after S-expressing cells (untreated or pretreated with trypsin) were added to target cells are shown (time points are indicated by arrows on the corresponding left panels). Scale bars, 1,000 µm. Right, western blot quantification of membrane-anchored S expression in BHK-21–GFP1–10 cells detected using the 76E1 monoclonal antibody63. Data shown are one representative of 2–4 biological replicates. Each dot in the bar graphs represents a biological replicate (n = 3 with 2–4 technical replicates each for a, n = 2 with 5 technical replicates each for b, n = 3 with 5–10 technical…

Extended Data Fig. 1. Sequence alignment of the FCoV-23 S-long and S-short constructs used in this study. The S-short construct corresponds to the C10-DA isolate13.

Extended Data Fig. 2. Data processing and validation of the FCoV-23 S-long glycoprotein (with D0) dataset. a,b, Representative electron micrograph out of 13,099 (a) and 2D class averages (b) of the prefusion FCoV-23 S-long ectodomain trimer embedded in vitreous ice. Scale bars of the micrograph and the 2D class averages are 100 nm and 220 Å, respectively. c, Gold-standard Fourier shell correlation (FSC) curves for the FCoV-23 S-long trimer with D0 in the “swung out” conformation (black line) or with mixed D0 conformations (gray line), and for the local refinements of D0 in swung out conformation (dashed black line) and in the proximal conformation (dashed gray line). The 0.143 cutoff is indicated by a horizontal dotted line. d, Cryo-EM data processing flowchart. e,f, Unsharpened maps of the reconstructions of the FCoV-23 S-long trimer with mixed D0 conformations (e) and the locally refined proximal D0 (f) colored according to local resolution determined using cryoSPARC. g,h, Unsharpened maps of the reconstructions of the FCoV-23 S-long trimer with D0 in swung out conformation (g) and the locally refined swung out D0 (h) colored according to local resolution determined using cryoSPARC. Angular…

Extended Data Fig. 3. Data processing and validation of the FCoV-23 S-short glycoprotein (without D0) dataset. a,b, Representative electron micrograph out of 11,208 (a) and 2D class averages (b) of the prefusion FCoV-23 S-short (without D0) ectodomain trimer embedded in vitreous ice. Scale bars of the micrograph and the 2D class averages are 100 nm and 180 Å, respectively. c, Gold-standard Fourier shell correlation curve for the FCoV-23 S-short trimer reconstruction. The 0.143 cutoff is indicated by a horizontal dotted line. d, Cryo-EM data processing flowchart. e, Unsharpened map of the reconstruction of the FCoV-23 S-short ectodomain trimer colored according to local resolution determined using cryoSPARC. f, Angular distribution plot of all the particles that contributed to the map. g, Ribbon representation of FCoV-23 S-short highlighting the S1 and S2 subunits. N-linked glycans are shown in surface representation colored dark blue. CTF: contrast transfer function. NUR: non uniform refinement. Source data

Extended Data Fig. 4. S1 subunit architectures of one representative S glycoprotein from the alpha, beta, gamma and delta coronavirus genera. a-d, Ribbon diagrams of the FCoV-23 S-short (a), SARS-CoV-2, PDB 6VXX (b), IBV, PDB 6CV0 (c) and PDCoV, PDB 6BFU (d) S1 subunits. A, B, C and D domains are labeled in each panel.

Extended Data Fig. 5. Sensitivity of FCoV-23 S to host proteases. a-f, Quantification of the fluorogenic FCoV-23 S2’ peptide cleavage by the indicated panel of proteases. The S2’ peptide was coupled to a fluorescence resonance energy transfer pair of fluorophores: 7-methoxycoumarin-4-yl acetyl (MCA) at the N-terminus and N-2,4-dinitrophenyl (DNP) at the C-terminus yielding MCA-S958KRKYRSAIE967-DNP. Data points represent nine technical replicates from n = 1 biological replicate. Means and SEM are shown as bars and lines, respectively. RFU: relative fluorescence units. g-i, Western blot evaluation of proteolytic sensitivity of FCoV-23 S-short and S-long anchored in the membrane of VSV pseudoviruses upon incubation with trypsin alone (g), with the recombinant CfAPN ectodomain dimer (h) or anchored in the membrane of VSV pseudoparticles (i) followed by trypsin and proteinase K digestion. Anti-HA tag polyclonal antibody and Alexa Fluor 680-conjugated goat anti-rabbit were used as primary and secondary antibodies, respectively. Each western blot corresponds to one representative experiment out of two biological replicates. Source data

Extended Data Fig. 6. FCoV-23 S D0 does not hemagglutinate erythrocytes. I53-50 nanoparticles (NP) were prepared by incubation of D0-I53-50A (FCoV-23 S-D0), domain A-I53-50A (FCoV-23 S-DA), or domain B-I53-50A (FCoV-23 S-DB) trimers with pentameric I53-50B at a 1:1 molar ratio, for at least 30 min at room temperature. Assembly of the CCoV-HuPn-2018 D0-I53-50 NP (CCoV-HuPn-2018 S-D0) required mixing domain D0-I53-50A and bare I53-50A before adding pentameric I53-50B at 6 (1:5):1 molar ratio. a, Hemagglutination assay. Wells positive for hemagglutination are circled. b, Representative SDS-PAGE analysis of purified nanoparticles out of 2 biological replicates (defined as distinct batches of purified proteins). c-f, Representative electron micrographs of negatively stained FCoV-23 S D0- (c), DA- (d), DB-I53-50 NPs (e), and CCoV-HuPn-2018 S D0-I53-50 NP (f). Scale bars: 200 nm. One hemagglutination assay corresponding to one preparation of nanoparticles is shown out of n = 2 biological replicates. g, Structure-based sequence alignment between FCoV-23 S D0 and CCoV-HuPn-2018 S D0.

Extended Data Fig. 7. Characterization of receptor binding and entry of FCoV-23, CCoV-HuPn-2018 and HCoV-229E. a,b, BLI binding kinetic analysis of dimeric FcAPN- (a) and CfAPN- (b) Fc ectodomains to immobilized biotinylated CCoV-HuPn-2018 RBD. Global fit (1:1 model) to the data is shown in black and reported affinities are expressed as apparent KD (KD, app) due to avidity resulting from the dimeric nature of APN. A representative experiment is shown out of two biological replicates. c-d, Coomassie-stained SDS-PAGE analysis of CfAPN-Fc pull-down experiments using either FCoV-23 S-short (c) or S-long (d) immobilized on NTA-cobalt based magnetic beads. Asterisks indicate the positions of the pulled down APN/S complexes. UB1: unbound fraction after incubating with S or buffer, UB2: unbound fraction after incubation with CfAPN-Fc or buffer; W1: final wash after incubating with S or buffer, W2: final wash after incubating with CfAPN-Fc; B, bound. Purified S coupled to beads and not incubated with CfAPN-Fc and CfAPN-Fc incubated with uncoupled beads are shown on the gels at the left-hand side of panels c and d. One pull down experiment corresponding to one preparation of proteins is shown out of two…

Extended Data Fig. 8. Data processing and validation of the FCoV-23 S RBD in complex with F.cat APN. a,b, Representative electron micrograph out of 13,540 (a) and 2D class averages (b) of the FCoV-23 S RBD in complex with dimeric FcAPN embedded in vitreous ice. Scale bars of the micrograph and the 2D class averages are 100 nm and 150 Å, respectively. c, Gold-standard Fourier shell correlation curve for the FcAPN-bound FCoV-23 RBD (solid black line). The 0.143 cutoff is indicated by a horizontal dotted line. d, Cryo-EM data processing flowchart. e, Unsharpened map of the FcAPN-bound FCoV-23 RBD colored according to local resolution determined using cryoSPARC. f, Angular distribution plots with all the particles contributing to the final map. CTF: contrast transfer function. NUR: non uniform refinement. Source data