Cell lines
The following cell lines used in this study were obtained from the American Type Culture Collection (ATCC): human epithelial embryo cells (HEK293T, CRL-3216), human lung carcinoma epithelial cells (A549, CRM-CCL-185), human lung adenocarcinoma epithelial cells (Calu3, HTB-55), African green monkey kidney cells (VeroE6, CRL-1586), rhesus monkey kidney cells (LLC-MK2, CCL-7), Syrian golden hamster kidney fibroblast cells (BHK-21, CCL-10), rat lung epithelial cells (L2, CCL-149), chicken embryo fibroblast cells (DF-1, CRL-3586), feline epithelial kidney cells (CRFK, CCL-94), feline airway epithelial cells (AK-D, CCL-150), canine tumour fibroblast cells (A-72, CRL-1542), canine epithelial kidney cells (MDCK, CCL-34), pig epithelial kidney cells (LLC-PK1, CL-101) and bovine epithelial kidney cells (MDBK, CCL-22). Human bronchial epithelial cells (HBEC cells) were provided by R. Cerione’s laboratory at Cornell University. Ovine kidney cells (OV KID) and ovine fetal turbinate (OFT) cells were provided by D. Diel at Cornell University Animal Health Diagnostic Center. Feline macrophage-like cells, Fcwf-4, were initially obtained from the ATCC (CRL-2787), and the progeny (Fcwf-CU) that was significantly better at propagating feline coronavirus was selected by E. Dubovi and G.R.W. at Cornell University to be significantly better at propagating feline coronavirus64. VeroE6 cells expressing transmembrane serine protease TMPRSS2 (VeroE6-TMPRSS2, JCRB1819) were obtained from the Japanese Collection of Research Bioresources (JCRB Cell Bank). The cell lines ExpiCHO and Expi293F were obtained from Thermo Fisher. Cells were cultured at 37 °C in the medium recommended by the manufacturer in an atmosphere of 5% CO2. Cell lines were not routinely tested for mycoplasma contamination nor were they authenticated.
Plasmids
Genes used in this study were synthesized by GenScript, codon optimized for expression in mammalian cells and cloned into pcDNA3.1(+) between KpnI and XhoI, in-frame with a Kozak’s sequence to direct translation. Ectodomains used the signal peptide derived from µ-phosphatase (MGILPSPGMPALLSLVSLLSVLLMGCVAETGT) and a C-terminal Avi-tag and octa-histidine tag. FCoV-23 S-long and S-short sequences were obtained from a previous study13 and are described in Extended Data Fig. 1. To stabilize the prefusion FCoV-23 S ectodomain constructs, residues E1146 and L1147 were mutated65 to proline, and a foldon trimerization domain was added. The HCoV-229E S ectodomain used for mouse immunizations was synthetized by GeneArt (Thermo Fisher) and comprised residues 1–1128, with a C-terminal truncation of 60 residues, and was stabilized by mutations of residues T871 and I872 to proline and by the addition of a foldon trimerization domain.
To pseudotype VSV(ΔG-luc)66, membrane-anchored wild-type S glycoproteins with their native signal peptide encoding genes from HCoV-229E (P100E isolate, 2001, AAK32191.1) residues 1–1,155, TGEV (ABG8935.1) residues 1–1,425, CCoV-HuPn-2018 (QVL91811.1) residues 1–1,425, FCoV-23 S-long and S-short, CCoV-HuPn-2018 lacking D0 (residues 259–1,425) (QVL91811.1), all with C-terminal truncations of 23 residues, except for HCoV-229E with a 18 C-terminal truncation, were codon optimized for expression in mammalian cells and cloned into pcDNA3.1(+) as indicated above. FCoV-23 S-long, FCoV-23 S-short and CCoV-HuPn-2018 S lacking D0 also had a C-terminal GGSYPYDVPDYA sequence (HA-tag in bold).
FCoV-23 S and CCoV-HuPn-2018 S B domains (also referred as RBDs) matching with the full-length sequences indicated above included residues 529–677 and 523–671, respectively, and were fused to a C-terminal Avi-tag and His-tag for biotinylation and affinity purification.
Full-length APN constructs have been previously described9. In brief, APN from human (NCBI GenBank identifier NP_001141.2), F. catus (NP_001009252.2), C. familiaris (NP_001139506.1), S. scrofa (AGX93258.1) and G. gallus (ACZ95799.1) comprise residues 1–967, 1–967, 1–975, 1–963 and 1–967, respectively. APN ectodomains from human wild-type and R741T mutant (both comprising residues 66–967), F. catus (residues 64–967), C. familiaris (residues 71–975), S. scrofa (residues 62–963) and G. gallus (residues 69–967) from the same sequence codes shown above were fused to a thrombin-cleavage site followed by a human Fc fragment at the C-terminal end for affinity purification.
The 1AF10 Fab constructs for expression in mammalian cells of the heavy and light chains have been previously described9. For NP blockembly, FCoV-23 D0 (residues 19–264), domain A (residues 265–504) and domain B (residues 529–677) were fused to the N terminus of the trimeric I53-50A NP component67 using a 16-residue-long glycine and serine linker. CCoV-HuPn-2018 D0 (residues 17–259) fused to the N terminus of the trimeric I53-50A NP component has been previously described9. Heavy chain and light chain genes encoding 76E1 full IgG were inserted each into a different pcDNA3.1(+) plasmid, fused to the N-terminal CD5 signal peptide MPMGSLQPLATLYLLGMLVASVLA.
The plasmid pQCXIP-BSR-GFP11 to express GFP11 for fusion blockays was obtained from Addgene (https://www.addgene.org/68716/).
Mutagenesis
The membrane-anchored wild-type FcAPN-encoding plasmid was used as a template to knockout the glycan at position N740 (NWT to NWR) using the following nonoverlapping primers: 5′-aagaattggagggaccaccccc-3′ (the codon change T to R is indicated in bold) and 5′-tgtcacgcgctcaaagtggttg-3′ using fusion DNA polymerase (Thermo Fisher). After treating the PCR products with DpnI (New England Biolabs) for 1 h at 37 °C, amplified plasmids were purified using a PCR & DNA cleanup kit (Monarch) treated with T4 polynucleotide kinase (New England Biolabs) for 1 h at 37 °C and ligated using T4 DNA ligase (New England Biolabs) at 25 °C overnight before being used for the transformation of One Shot MAX Efficiency DH10B chemically competent cells (Invitrogen). Introduction of the desired mutations was verified by sequencing purified plasmids by Plasmidsaurus. Plasmids with the desired mutation were amplified and purified using an EndoFree mega kit (Qiagen) so that they were suitable for transfection into mammalian cells.
Protein expression and purification
To produce stabilized FCoV-23 S-long and S-short, 200 ml Expi293F cells grown to a density of 3 × 106 cells per ml and at 37 °C were transfected with 640 µl Expifectamine reagent (Thermo Fisher) and 200 µg of the corresponding plasmids following the manufacturer’s recommendations. The day after transfection, feed and enhancer were added to the cells. At 4–5 days after transfection, supernatants were clarified by centrifugation at 400g for 15 min and resuspended in 20 mM Tris-HCl pH 8.0, 20 mM imidazole and 300 mM NaCl. Supernatants were further centrifuged at 14,000g for 30 min and pblocked through a 1 ml Histrap Excel column (Cytiva) previously equilibrated with binding buffer (25 mM Tris-HCl pH 7.4 and 350 mM NaCl). FCoV-23 S was eluted using a linear gradient of 500 mM imidazole. To produce stabilized HCoV-229E S 2P, 500 ml of Expi293F cells at 3 × 106 cells per ml were transiently transfected using polyethylenimine linear (PEI). In brief, 500 µg DNA (1 µg ml–1 final concentration) was mixed with Opti-MEM and 1.5 mg PEI (3 µg ml–1 final concentration) was mixed with Opti-MEM. The PEI mixture was added to the DNA mixture and gently mixed. After 15 min of incubation at room temperature, the PEI–DNA mixture (25 ml) was added dropwise to cells. On day 4 after transfection, stabilized HCoV-229E S 2P was purified using the protocol described above.
To express B domains from FCoV-23 S and CCoV-HuPn-2018 S C-terminally fused to an Avi-tag and a histidine tag, 100 ml of Expi293F cells at 3 × 106 cells per ml were transiently transfected with 320 µl Expifectamine and 100 µg of the respective plasmids following the manufacturer’s instructions. Four days after transfection, supernatants were clarified by centrifugation at 800g for 10 min, supplemented with 20 mM imidazole, 300 mM NaCl and 25 mM Tris-HCl pH 8.0, further centrifuged at 14,000g for 30 min and pblocked through a 1 ml His trap HP column (Cytiva) previously equilibrated with binding buffer (25 mM Tris-HCl pH 7.4 and 150 mM NaCl). B domains were eluted using 25 mM Tris-HCl pH 7.4, 150 mM NaCl and a linear gradient of 500 mM imidazole. A similar protocol was used to express and purify the Fab domain of 1AF10 fused to an 8-residue histidine tag, except that 50 ml of Expi293F cells were transfected with a mixture containing 50 µg of each individual plasmid encoding the Fab light and heavy chain and 160 µl Expifectamine reagent (Thermo Fisher).
To express the human APN, FcAPN, CfAPN, SsAPN and GgAPN ectodomain orthologues fused to the Fc fragment of human IgG, Expi293F cells were transiently transfected with the respective plasmids following the manufacturer’s protocol. In brief, 50 ml of Expi293F cells at 3 × 106 cells per ml were transfected using 160 µl Expifectamine and 50 µg APN plasmid. Four days after transfection, supernatants were clarified by centrifugation at 800g for 10 min, supplemented with 300 mM NaCl and 25 mM Tris-HCl pH 8.0, further centrifuged at 14,000g for 30 min and pblocked through a 1 ml HiTrap Protein A HP column (Cytiva). Proteins were eluted using 0.1 M citric acid pH 3.0 in individual tubes containing 200 µl of 1 M Tris-HCl pH 9.0 to immediately neutralize the low pH needed for elution. APN samples were further purified by size-exclusion chromatography (SEC) on a Superdex 200 column 10/300 GL (GE Life Sciences) previously equilibrated in 25 mM Tris-HCl pH 8.0 and 150 mM NaCl. Fractions containing the proteins were pooled and buffer exchanged to 25 mM Tris-HCl pH 8.0 and 150 mM NaCl.
To produce FcAPN and CfAPN without the Fc fusion for cryo-EM and for protease digestion experiments, respectively, the Fc fragments were removed using thrombin (Millipore Sigma) in a reaction mixture containing 3 µg thrombin per mg of FcAPN-Fc, 25 mM Tris-HCl pH 8.0, 150 mM NaCl and 2.5 mM CaCl2 and incubated overnight at room temperature. The reaction mixture was loaded to a protein A column to remove uncleaved FcAPN-Fc and the Fc tag, and purified FcAPN was buffer exchanged to 25 mM Tris-HCl pH 8.0 and 150 mM NaCl.
To express the 76E1 full IgG, 200 ml of Expi293F cells at 3 × 106 cells per ml were transiently transfected using Expifectamine and a mix containing 200 µg each of plasmids encoding the IgG heavy chain and light chain following the manufacturer’s protocol. Four days after transfection, Expi293 cell supernatant was clarified by centrifugation at 4,121g for 30 min, supplemented with 20 mM phosphate buffer pH 8.0 and 0.5 mM phenylmethylsulfonyl fluoride. Supernatant was pblocked through a protein A affinity column (Cytiva) twice before being washed with 20 mM phosphate buffer pH 8.0. Antibody 76E1 was eluted with 100 mM citric acid pH 3.0 into tubes containing 1 M Tris-HCl pH 9.0 for immediate neutralization of the low pH used for elution. Eluted 76E1 antibody was buffer exchanged into 20 mM phosphate buffer pH 8.0 and 100 mM NaCl, and purity was evaluated by SDS–PAGE.
Protein biotinylation
B domains of FCoV-23 S and CCoV-HuPn-2018 S were biotinylated using a BirA biotin-protein ligase standard reaction kit (Avidity) following the manufacturer’s protocol. In a typical reaction, 40 μM B domains was incubated overnight at 4 °C with 2.5 μg BirA enzyme in reaction mixtures containing 1× BiomixB, 1× BiomixA and 40 µM BIO200. The B domains were further separated from BirA by SEC using Superdex 75 increase 10/300 GL (GE LifeSciences) and concentrated using 10 kDa filters (Amicon).
BLI binding blockays
Biotinylated B domains from FCoV-23 S and CCoV-HuPn-2018 S were immobilized at 2 µg ml–1 in 10X kinetics buffer (Sartorius) to 0.5 nm shift to streptavidin (SA) biosensors (Sartorius) previously hydrated in 10X kinetics buffer for at least 10 min. Loaded tips were dipped into a solution containing 1 µM APN orthologues from F. catus, C. familiaris, S. scrofa, G. gallus, human, and the human APN(R741T) mutant in 10X kinetics buffer. For Kd,app determination, loaded tips were dipped into various concentrations of FcAPN-Fc or CfAPN-Fc for 300 s followed by a 300-s dissociation phase in 10X kinetics buffer. Data were baseline subtracted, and the plots were fitted using Sartorius blockysis software (v.11.1). Data were plotted in Graphpad Prism 10. These experiments were done side-by-side with two different batches of B domain and APN orthologue preparations.
Pull-down blockays
To further characterize the interaction between CfAPN-Fc and FCoV-23 S-short and S-long, we performed pull-down blockays. In brief, 100 µl (4 mg) magnetic beads, Dynabeads His-tag (Thermo Fisher) were washed twice with 200 µl 1× TBS (20 mM Tris-HCl pH 8.0 and 150 mM NaCl) using a magnetic stand (Thermo Fisher). Before the last wash, beads were divided into eight aliquots, after which the buffer was removed. Beads were coupled or not with 10 µg FCoV-23 S-short or S-long in TBS buffer. The bead–protein mixture was incubated at room temperature with gentle rotation. After 30 min, unbound proteins were removed and beads were washed three times with 30 µl TBS each time before resuspension in a solution containing 15 µg purified CfAPN-FC. After incubation for 1 h at room temperature with gentle rotation, beads were washed three times with TBS. The following samples were collected and characterized by SDS–PAGE: unbound proteins after coupling the spike proteins (UB1) and after incubation with CfAPN (UB2); the last washes after coupling the spike proteins (W1) and after incubation with CfAPN (W2); and bound proteins. Negative controls consisting of beads not coupled to FCoV-23 S-short or S-long were also blockysed by SDS–PAGE.
Bacterial protein expression and purification of NP components
The I53-50A and I53-50B proteins were expressed as previously described67. In brief, transformed Lemo21(DE3) (NEB) in LB broth (10 g tryptone, 5 g yeast extract and 10 g NaCl) were grown at 37 °C to an OD600 of 0.8 with constant agitation. Protein expression was induced by adding 1 mM isopropil-β-d-1-thiogalactopyranoside and the temperature was reduced to 18 °C. Cells were collected after 16 h and lysed by microfluidization using a Microfluidics M110P at 18,000 p.s.i. in 50 mM Tris-HCl pH 8.0, 500 mM NaCl, 30 mM imidazole, 1 mM phenylmethylsulfonyl fluoride and 0.75% 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS). Lysates were clarified by centrifugation at 24,000g for 30 min and applied to a Ni Sepharose 6 fast flow column (Cytiva) for purification by immobilized metal affinity chromatography on an AKTA Avant150 FPLC system (Cytiva). Proteins were eluted with a linear gradient of 30 mM to 500 mM imidazole in 50 mM Tris-HCl pH 8.0, 500 mM NaCl and 0.75% CHAPS buffer. Peak fractions were pooled, concentrated in 10 K MWCO centrifugal filters (Millipore), sterile filtered (0.22 mm) and applied to either a Superdex 200 Increase 10/300 or HiLoad S200 pg GL SEC column (Cytiva) previously equilibrated in 50 mM Tris-HCl pH 8.0, 500 mM NaCl and 0.75% CHAPS buffer.
In vitro NP blockembly
Concentration of purified individual NP components was determined by measuring the absorbance at 280 nm and the corresponding calculated extinction coefficients. NPs were prepared by incubating D0-A-I53-50A, domain A-I53-50A or domain B-I53-50A trimers with pentameric I53-50B at molar ratios of 1.1:1, respectively, in 50 mM Tris-HCl pH 8.0 and 500 mM NaCl. Formation of CCoV-HuPn-2018 D0 NPs required mixing D0-I53-50A with I53-50A at a molar ratio of 1:6 with pentameric I53-50B. All in vitro blockemblies were incubated at room temperature with gentle rocking for at least 20 min before subsequent purification by SEC on a Superose 6 column to remove residual unblockembled components. Fractions were blockysed by negative-stain electron microscopy and by SDS–PAGE. Assembled NPs eluted in the void volume of a Superose 6 column and were pooled and stored at 4 °C until use in haemagglutination blockays.
Negative-stain electron microscopy
NPs diluted to 0.01 mg ml–1 in 50 mM Tris-HCl pH 8.0 and 150 mM NaCl were adsorbed to glow-discharged home-made carbon-coated copper grids for 30 s. The excess liquid was blotted away with filter paper (Whatman 1) and 3 µl of 2% w/v uranyl formate was applied to the grids. Finally, the stain was blotted away, and the grids were allowed to air dry for 1 min. Grids were imaged on a 120 kV FEI Tecnai G2 Spirit with a Gatan Ultrascan 4000 4k × 4k CCD camera at ×67,000 nominal magnification using a defocus ranging between 1.0 and 2.0 mm and a pixel size of 1.6 Å.
Haemagglutination blockay
The haemagglutination blockay was done according to standard procedures. In brief, 25 µl CCoV-HuPn-2018 S D0, FCoV-23 S D0, S domain A and S domain B NPs at 125 µg ml–1 were incubated with 25 µl 1% cat (Cornell University), chicken (Lampire), human (Rockland Immunochemicals), cow (Lampire) and turkey (Lampire) erythrocytes diluted in DPBS (Gibco) in V-bottom, 96-well plates (Greiner Bio-One) for 30 min at room temperature. Plates were then photographed and haemagglutination was blockysed. CCoV-HuPn-2018 D0 NPs were used as a positive control.
VSV pseudotyped virus production
FCoV-23, CCoV-HuPn, HCoV-229E and TGEV S pseudotyped VSV particles were generated as previously described9. In brief, HEK293T cells in DMEM supplemented with 10% FBS and 1% penicillin–streptomycin and seeded in poly-d-lysine-coated 10-cm dishes were transfected with a mixture of 24 µg of the corresponding plasmid encoding CCoV-HuPn S, TGEV S or HCoV-229 S using 60 µl Lipofectamine 2000 (Life Technologies) in 3 ml Opti-MEM following the manufacturer’s instructions. After 5 h of incubation at 37 °C, DMEM supplemented with 20% FBS and 2% penicillin–streptomycin was added. The next day, cells were washed three times with DMEM and were transduced with VSVΔG-luc66. After 2 h, the virus inoculum was removed and cells were washed five times with DMEM before the addition of DMEM supplemented with anti-VSV-G antibody (Il-mouse hybridoma supernatant diluted 1:25 (v/v), from CRL-2700, ATCC) to minimize parental background. After 18–24 h, supernatants containing pseudotyped VSV were collected, centrifuged at 2,000g for 5 min to remove cellular debris, filtered through a 0.45 µm membrane, concentrated ten times using a 30 kDa cut-off membrane (Amicon), aliquoted and frozen at −80 °C.
VSV pseudotyped virus infection and neutralization
For pseudotyped VSV infection and neutralization, HEK293T cells were transfected with plasmids encoding the different full-length APN orthologues following a previously described protocol9. In brief, HEK293T cells at 90% confluency and seeded in poly-d-lysine-coated 10-cm dishes were transfected with a mixture of Opti-MEM containing 8–10 µg of the corresponding plasmid encoding full-length APN orthologues using 30 µl Lipofectamine 2000 (Life Technologies) according to the manufacturer’s instructions. After 5 h of incubation at 37 °C, cells were trypsinized, seeded into poly-d-lysine-coated clear-bottom white-walled 96-well plates at 50,000 cells per well and cultured overnight at 37 °C. For infections, 20 µl of the corresponding pseudotyped VSV diluted 1:20 was mixed with 20 µl DMEM, and the mixture was added to cells that were previously washed twice with DMEM. After 2 h of incubation at 37 °C, 40 µl DMEM was added and cells were further incubated overnight at 37 °C.
For neutralization blockays, a single dilution of 1:10 for mouse serum samples or 11 3-fold serial dilutions of Fab 1AF10 or APN ectodomains were prepared in DMEM. Next, 20 µl FCoV-23 S, CCoV-HuPn-2018 S, HCoV-229E S or TGEV S pseudotyped VSV was added 1:1 (v/v) to each Fab 1AF10 or APN ectodomains, and mixtures were incubated for 45–60 min at 37 °C. After removing medium, transfected HEK293T cells were washed three times with DMEM, and 40 μl of the mixture containing pseudotyped VSV and Fab or APN ectodomains and sera were added. One hour later, 40 μl DMEM was added to the cells. After 17–24 h, 60 μl One-Glo-EX substrate (Promega) was added to each well and plates were placed in a shaker in the dark. After 5–15 min of incubation, plates were read on a Biotek plate reader. Relative luciferase units were plotted and normalized in Graphpad Prism 10. Cells alone without pseudotyped virus were defined as 0% infection, and cells with virus only (no sera) were defined as 100% infection.
VSV pseudotyped virus entry blockay with protease inhibitors
Camostat mesylate (100 mM), nafamostat mesylate (100 mM), E-64d (at 10 mM) and bafilomycin A1 (100 µM) (Sigma) inhibitors were dissolved in DMSO and frozen in aliquots. Fcwf-CU cells and CRFK were seeded into white-walled, clear-bottom 96-well plates (Corning) at a density of 40,000 cells and grown overnight at 37 °C and 5% CO2. The next day, the growth medium was removed and cells were washed twice with DMEM (for CRFK cells) or EMEM–1% HEPES (for Fcwf-CU cells) before adding 50 µl of the respective medium containing 25 µM camostat mesylate, nafamostat mesylate or E-64d, 1 µM bafilomycin A1 or DMSO. After 2 h of incubation at 37 °C, protease inhibitors were removed, except for bafilomycin, which was kept at 1 µM throughout the whole experiment, and 25 µl VSV-pseudotyped viruses with FCoV-23 S-long or S-short diluted 1:20 were added to the cells. After 2 h, an equal volume of DMEM supplemented with 20% FBS (for CRFK cells) or EMEM–1% HEPES supplemented with 40% FBS (for Fcwf-CU cells) was added. After 20–24 h, ONE-Glo EX (Promega) was added to each well, and the cells were incubated for 5 min at 37 °C. Luminescence values were measured using a BioTek Synergy Neo2 plate reader. Luminescence readings were blockysed using (GraphPad Prism 10). The relative luminescence units values recorded from cells infected in the presence of DMSO and values from cells infected in the presence of protease inhibitors were plotted and compared. Two biological replicates (n = 2) were each done with a different batch of VSV-pseudotyped virus and 10–12 technical replicates were done for each inhibitor.
VSV pseudotyped virus entry kinetics
For pseudotyped VSV virus entry kinetics, FCoV-23 S-long and S-short were quantified in VSV pseudotyped viruses by western blot blockysis to ensure infection with the same amount of S from both pseudotypes. Fcwf-CU and CRFK cells were seeded into white-walled, clear-bottom 96-well plates (Corning) at a density of 40,000 cells and grown overnight at 37 °C and 5% CO2. The next day, the growth medium was removed and cells were washed twice with DMEM (for CRFK cells) or EMEM–1% HEPES (for Fcwf-CU cells) before adding 50 µl VSV pseudotyped viruses with FCoV-23 S-long or S-short. Luminescence values were measured at the indicated times using a BioTek Synergy Neo2 plate reader, and the readings were blockysed using GraphPad Prism 10. The relative luminescence values recorded from cells infected with VSV pseudotyped with FCoV-23 S-long and from cells infected with VSV pseudotyped with FCoV-23 S-short were plotted and compared. Four biological independent experiments (n = 4) were each done with a different batch of VSV-pseudotyped virus and 12 technical replicates. Comparison of entry values between S-long and S-short for each time point was performed using multiple ratio paired t-tests.
MLV pseudotyped virus production
MLV pseudotyped particles were produced as previously described68. In brief, HEK293T cells were seeded at 5 × 105 cells per ml in 6-well plates (Costar) a day before transfection. At about 60% confluency, cells were transfected with 800 ng pCMV-MLV gag-pol, 600 ng pTG-Luciferase and 600 ng pCDNA3.1(+) FCoV-23 S-short or pCDNA3.1(+) FCoV-23 S-long, pCAGGS empty vector as a negative control or pCAGGS-VSV-G as a positive control and incubated at 34 °C. Supernatants were collected 48 h after transfection and cell debris was removed by centrifugation at 1,000g. The supernatant-containing particles were filtered through a 0.45 µm membrane, aliquoted and frozen at –80 °C.
Pseudotyped MLV infection and entry kinetics
Fcwf-CU, CRFK, AK-D, A-72, MDCK, LLC-MK2, VeroE6, VeroE6-TMPRSS2, A549, Calu3, HBEC, LLC-PK1, MDBK, OFT, OVKID, L2, BHK 21 and DF-1 cells were seeded at 3–5 × 105 cells per ml in a 24-well plate (Costar) and cultured at 37 °C. At about 90–100% confluency, cells were infected with 200 µl MLV particles and incubated on a rocker at 37 °C. After 1.5 h, 300 µl of the corresponding complete medium was added to cells. After 72 h (Extended Data Fig. 8a) or 8 h (Extended Data Fig. 8c), cells were lysed with 100 µl of 1× luciferase cell culture lysis reagent (25 mM Tris-phosphate pH 7.8, 2 mM dithiothreitol (DTT), 2 mM EDTA, 10% glycerol and 1% Triton X-100, Promega), and luminescence was measured using a Luciferase Assay system (Promega). Luciferase activity was measured by adding 20 µl luciferin substrate to 10 µl cell lysate and read using a GloMax 20/20 luminometer (Promega). MLV infection blockays were done in duplicate or triplicate for each of the n = 2 biological replicates. For the entry kinetics experiments, FCWF-CU, AK-D and A72 cells were seeded at 3 × 105 cells per ml in a 48-well plate (Costar) and cultured at 37 °C. At 100% confluency, cells were infected with 120 μl MLV particles and incubated on a rocker at 37 °C. Cells were lysed with 100 μl of 1× luciferase cell culture lysis reagent, and luminescence was measured by adding 20 μl luciferase substrate to 10 μl cell lysate at 8 h and 72 h after infection. The kinetic blockay was performed with four biological independent experiments, each with three to four technical replicates. Comparison of entry values between S-long and S-short for each time point was performed using ratio paired t-tests. We note that for MLV pseudoviruses, expression of the luciferase reporter gene entails reverse transcription, genomic integration and forward transcription and is therefore more complex than for VSV pseudoviruses.
Fluorogenic peptide cleavage blockay
The FCoV-23 S2′ fluorogenic peptide, 958SKRKYRSAIE967, site was synthesized by Biomatik, with a 7-methoxycoumarin-4-yl acetyl group covalently bound at the N terminus end and a 2,4-dinitrophenyl group at the C-terminal end used as a substrate in a fluorogenic peptide cleavage blockay performed as previously described69. In brief, FCoV-23 S2′ peptide was resuspended in water to 1 mM and cleavage was performed in a black, flat-bottom 96-well plate (Costar) in a total volume of 100 µl per well as follows: 94.5 µl enzyme-specific buffer and 5 µl S2′ fluorogenic peptide was added to each well (50 µM per well) with 0.5 µl of protease. TPCK trypsin (Thermo Fisher) was used at 4.3 nM per well in 67 mM NaH2PO4 pH 7.6. Furin (NEB) was used at 1 unit per well in 20 mM HEPES–NaOH pH 7.5, 0.2 mM CaCl2 and 1 mM 2-mercaptoethanol. PC1 (NEB) was used at 1 U per well in 100 mM HEPES–NaOH pH 6.0, 1 mM CaCl2 and 1 mM 2-mercaptoethanol. Xa (NEB) was used at 1 U per well in 20 mM Tris-HCl pH 8.0, 100 mM NaCl and 2 mM CaCl2. Plasmin (Innovative Research) was used at 0.5 µg ml–1 per well in 100 mM Tris-HCl pH 7.4, 0.1% Tween 20 and 0.1 mM EDTA. Cathepsin L (R&D) was used at 0.5 µg ml–1 per well in 50 mM 2-(N-morpholino) ethanesulfonic acid and NaOH pH 6.0, 5 mM dithiothreitol, 1 mM EDTA and 0.005% Brij-35. After buffer, protease and peptide were mixed, fluorescence was measured every 60 s for 2 h at 37 °C using a SpectraMax Gemini XPS (Molecular Devices) at an excitation wavelength of 330 nm and an emission wavelength of 390 nm. Data were plotted as relative fluorescent units on the y axis and time on the x axis.
Protease digestion experiments of VSV pseudotyped viruses
VSV pseudotyped with FCoV-23 S-short or S-long were thawed and with or without trypsin at 10 mg ml–1 for 15 min at 37 °C. To blockyse the influence of APN on trypsin digestion patterns, VSV pseudotyped with FCoV-23 S-short or S-long were incubated with either 6.2 µM recombinant CfAPN ectodomain or membrane-anchored CfAPN for 30 min at room temperature before the addition or not of trypsin at 10 mg ml–1 for 15 min at 37 °C followed by treatment (or not) with proteinase K at 0.6 mg ml–1 for 15 min at room temperature. 4× SDS–PAGE loading buffer was then added to all samples before boiling at 95 °C for 10 min. Samples were run on a 4–15% gradient Tris-glycine gel (Bio-Rad) and transferred to PVDF membranes. HA-tag polyclonal antibodies at 1:1,000 dilution (Proteintech) and Alexa Fluor 680-conjugated goat anti-rabbit at 1:50,000 dilution (Jackson Immuno Research) were used as primary and secondary antibodies, respectively. A LI-COR processor was used to develop the western blots.
S-mediated cell–cell fusion blockay
Membrane fusion kinetics experiments were conducted at 37 °C in an atmosphere of 5% CO2, as previously described70,71 but with some modifications. On day 1, BHK-21–GFP1–10 cells were seeded into 6-well plates at a density of 4 × 105 cells per well and BHK-21–GFP11 cells were seeded into poly-d-lysine-coated 10-cm dishes at 4–5 × 106 cells. On day 2, BHK-21–GFP1–10 cells were washed once with DMEM, placed in DMEM containing 10% FBS and 1% penicillin–streptomycin and transfected using Lipofectamine 2000 (Life Technologies) with 4 µg per well of plasmid encoding membrane-anchored FCoV-23 S-long or CCoV-HuPn-2018 S-long, or 1 µg per well of plasmid encoding for membrane-anchored FCoV-23 S-short or CCoV-HuPn-2018 S-short (to produce comparable expression levels between S-long and S-short). The same day, BHK-21–GFP11 cells were transfected with plasmid encoding the indicated membrane-anchored APN orthologue using a mixture of 8 µg DNA and 30 µl Lipofectamine 2000 (Life Technologies) per 10-cm dish according to the manufacturer’s instructions. Alternatively, if A-72, CRFK or Fcwf-CU cells were used as target cells (instead of BHK-21–GFP11 cells), they were trypsinized before seeding at 5–6 × 106 cells per poly-d-lysine-coated 10-cm dish. After 2–3 h of incubation, A-72, CRFK and Fcwf-CU cells were transfected with a mixture of 24 µg DNA plasmid encoding pQCXIP-BSR-GFP11 and 60 µl Lipofectamine 2000 (Life Technologies) per 10-cm dish for the transient expression of GFP11. After 4–5 h of incubation, all target cells were trypsinized and seeded into a 96-well-glblock bottom, black-walled plates (CellVis) at a density of 36,000 cells per well and cultured overnight. On day 3, S-expressing BHK-21–GFP1–10 cells were washed three times using FluoroBrite DMEM (Thermo Fisher), detached using an enzyme-free cell dissociation buffer (Gibco) and pblocked through a cell strainer to remove aggregates. BHK-21–GFP1–10 cells were subsequently counted, diluted to 60,000 cells per ml, 40,000 cells per ml or 120,000 cells per ml to seed on top of BHK-21-GFP11, Fcwf-Cu, and A-72 and CRFK cells, respectively, which were previously washed three times with FluoroBrite. Trypsin pretreatment of S-expressing cells was performed when applicable using 5–10 µg trypsin (Sigma) during 15 min of incubation at 37 °C, after which 10–15 µg soybean trypsin inhibitor (Sigma) was added before adding them on target cells. Cells were incubated at 37 °C and 5% CO2 in a Cytation 7 plate Imager (Biotek), and both bright-field and GFP images were collected every 30 min for 18 h. Fusogenicity was blockessed by quantifying the fraction of the imaged area with GFP fluorescence for each image using Gen5 Image Prime (v.3.11) software. AUCs were calculated using GraphPad Prism 10, and data were collected for 18 h. For Fcwf-CU cells, comparisons between S-long-mediated and S-short-mediated fusion was done until 5.5 h after adding S-expressing cells (after this length of time, FcwF-CU cells were detaching or clumping, which hindered fluorescence measurements). Statistical blockysis was performed using the AUC of the GFP+ area between the biological replicates with two-way ANOVA or one-way ANOVA with Tukey’s multiple comparison test, as indicated in the relevant figure legends. S-transfected cells added to BHK-21–GFP11 target cells not expressing APN (no APN) or untransfected BHK-21–GFP1–10 cells (no S) added to target cells expressing GFP11 and APN were used as negative controls.
Western blotting
For VSV pseudotypes characterization and quantification, 15 µl of each pseudotype was mixed with 4× SDS–PAGE loading buffer, run on a 4–15% gradient Tris-glycine gel (Bio-Rad) and transferred to a PVDF membrane using the protocol mix molecular weight of a Trans-Blot Turbo system (Bio-Rad). The membrane was blocked with 5% milk in TBS-T (20 mM Tris-HCl pH 8.0 and 150 mM NaCl) supplemented with 0.05% Tween-20 at room temperature and with agitation. After 1 h, the fusion-peptide-specific 76E1 monoclonal antibody was added at 2 µg ml–1 and incubated overnight at 4 °C with agitation. The next day, the membrane was washed three times with TBS-T and an Alexa Fluor 680-conjugated goat anti-human secondary antibody (1:50,000 dilution, Jackson Immuno Research) was added and incubated for 1 h at room temperature. Alternatively, HA-tag polyclonal antibody at 1:1,000 (Proteintech) and anti-VSV M (23H12) monoclonal antibody (Kerafast) were used as the primary antibodies in combination with Alexa Fluor 680-conjugated goat anti-rabbit and Alexa Fluor 680-conjugated goat anti-mouse (each at 1:50,000 dilution, Jackson Immuno Research) as secondary antibodies, using the same protocol as described above. Membranes were washed three times with TBS-T, after which a LI-COR processor was used to develop the western blots.
For quantification of membrane-anchored S expressed in BHK-21–GFP1–10 for fusion blockays, we used 76E1 monoclonal antibody and Alexa Fluor 680-conjugated goat anti-human as the primary and secondary antibodies, respectively.
For MLV pseudotypes quantification, 1.5 ml MLV particles were centrifuged using a SW55Ti rotor (Beckman) with an Optima XPN ultracentrifuge (Beckman) at 110,000g for 1 h and 30 min. Pellets were resuspended in ice-cold PBS (Corning) and centrifuged again at 110,000g for 1 h. Pellets were resuspended in 20 µl PBS (Corning), and NuPage LDS sample buffer (Thermo Fisher) and NuPage sample reducing agent (Thermo Fisher) were added to the resuspended sample. After boiling at 70 °C for 10 min, samples were run on a NuPage 4–12% gradient Bis-Tris gel (Thermo Fisher) and transferred to a PVDF membrane. The membrane was blocked for 30 min with 5% BSA (VWR) in TBS-T (20 mM Tris-HCl, pH 8.0 and 150 mM NaCl) supplemented with 0.05% Tween-20. After 30 min, HA monoclonal antibody (1:1,000 dilution, Thermo Fisher, 26183) and MLV p30 (4B2) monoclonal antibody (1:2,000 dilution, Abcam, ab130757) were added, and the membrane was incubated overnight at 4 °C with agitation. The membrane was washed three times with TBS-T and incubated with Alexa Fluor 488-conjugated goat anti-mouse secondary antibody (1:50,000 dilution, Thermo Fisher, A-11029) for 1 h at room temperature with agitation. The membrane was washed three times and imaged using ChemiDoc Imaging software (Bio-Rad). Spike and MLV p30 band intensity was calculated using Bio-Rad Image Lab software.
Cryo-EM sample preparation, data collection and data processing for FCoV-23 S-short and S-long
A volume of 3 µl FCoV-23 S-short (without D0) at approximately 0.1–0.3 mg ml–1 was loaded three times onto freshly glow-discharged NiTi grids covered with a thin layer of home-made continuous carbon before plunge-freezing using a vitrobot MarkIV (Thermo Fisher) with a blot force of –1 and 4.5 s blot time at 100% humidity and 21 °C. For FCoV-23 S-long (with D0), 3 µl of the sample at 0.15 mg ml–1 was loaded onto lacey grids freshly glow-discharged and covered with a thin layer of home-made continuous carbon following the same protocol as for FCoV-23 S-short.
For FCoV-23 S-short and S-long, 11,208 and 13,099 videos were collected, respectively, with a defocus range between –0.8 and –2.0 μm. Both datasets were acquired using a FEI Titan Krios transmission electron microscope operated at 300 kV equipped with a Gatan K3 direct detector and a Gatan Quantum GIF energy filter, operated with a slit width of 20 eV. Automated data collection was carried out using Leginon software (v.3.5)72 at a nominal magnification of ×105,000 corresponding to a pixel size of 0.843 Å. The dose rate was adjusted to 9 counts per pixel per s, and each video was acquired in counting mode fractionated in 100 frames of 40 ms. Video frame alignment, estimation of the microscope contrast-transfer function parameters, particle picking and extraction were carried out using Warp73. One round of reference-free 2D clblockification was performed using cryoSPARC (v.4.4.1)74 with binned particles to select well-defined particle images. To further improve particle picking, we trained Topaz picker75 on the Warp-picked particles on the selected clblockes after 2D clblockification. Topaz-picked particles were extracted and 2D clblockified using cryoSPARC (v.4.4.1). Topaz-duplicated picked particles were removed using 60 Å (for FCoV-23 S-short) or 90 Å (for FCoV-23 S-long) as a minimum distance cut-off. Initial model generation was done using ab initio reconstruction in cryoSPARC (v.4.4.1) and used as a reference for a non-uniform refinement76 (NUR) (for FCoV-23 S-short) or heterogenous refinement followed by NUR (for FCoV-23 S with D0) in cryoSPARC (v.4.4.1), which enabled the identification of two conformations for the FCoV-23 S-long dataset. Particles were transferred from cryoSPARC (v.4.4.1) to Relion (v.5.0b)77 using pyem78 (https://github.com/asarnow/pyem) to be subjected to one round of 3D clblockification with 50 iterations, using the NUR map as a reference model (angular sampling 7.5° for 25 iterations and 1.8° with local search for 25 iterations) and without imposing symmetry. Selected particles were subjected to a NUR using cryoSPARC (v.4.4.1). Particles were subjected to Bayesian polishing79 using Relion (v.5.0b), during which they were re-extracted with a box size of 512 pixels at a pixel size of 1 Å. Another round of 2D clblockification was performed in cryoSPARC (v.4.4.1), which was followed by a NUR that included per-particle defocus refinement.
To improve the density of D0 from the FCoV-23 S-long and S-short maps, particles were symmetry-expanded (for the FCoV-23 S-long map) or not (for the FCoV-23 S-short map) and subjected to Relion (v.5.0b) 3D focus clblockification without refining angles and shifts using soft masks encompblocking D0 in the swung-out conformation from the FCoV-23 S-long map or D0 in the proximal conformation from the FCoV-23 S-short map. Local refinement and local resolution estimation were carried out using cryoSPARC. Reported resolutions are based on the gold-standard Fourier shell correlation using the 0.143 criterion80, and Fourier shell correlation curves were corrected for the effects of soft masking by high-resolution noise substitution81.
Cryo-EM sample preparation, data collection and data processing for FCoV-23 RBD in complex with FcAPN
Purified FcAPN was incubated overnight at 4 °C with a molar excess of purified FCoV-23 RBD of 1:3 and the complex was purified by SEC. Next, 2.7 µl FcAPN–FCoV-23 RBD at 4 or 6 mg ml–1 was applied for 15 s onto freshly glow-discharged UltrAuFoil R 2/2 grids with 0.3 µl of 30 or 60 mM of CHAPSO, respectively, before plunge-freezing using a vitrobot MarkIV (Thermo Fisher) with a blot force of –0 and 6 s blot time at 100% humidity and 21 °C. A total of 13,540 micrographs were collected using Leginon (v.3.5)72, with a defocus range between –0.8 and –2.0 μm on the same microscope set up described for the FCoV-23 S glycoprotein ectodomain. Data processing was similar as described for the FCoV-23 S glycoprotein ectodomain, except that two cycles of Topaz picker75 on the Warp-picked particles on the selected clblockes after 2D clblockification were performed.
Cryo-EM model building and blockysis
Model Angelo82 was used to generate an initial model, and UCSF Chimera (v.1.8)83 and Coot (v.0.9.8.8)84 were used to manually build the model. The model was refined and rebuilt into the maps using Coot (v.0.9.8.8), Phenix (v.1.21)85 and Rosetta (v.2021.07.61567)86,87. Model validation was done using Molprobity88 and Privateer89 from the CCP4i2 suite90. Figures were generated using UCSF ChimeraX91. Palmitoleic acids resolved in our cryo-EM maps were built in the final models based on previous findings for PEDV55. We note that no density was present for palmitoleic acid molecules interacting with domain A and domain D of the protomer with D0 in the proximal conformation.
Phylogenetic blockysis
Structure-guided phylogenetic clblockification of alphacoronavirus and deltacoronavirus RBDs was carried out using FoldTree92.
In vivo immunogenicity study
Female BALB/cAnNHsd mice were purchased from Envigo (order code 047) at 7 weeks of age and were maintained in a specific-pathogen-free facility in the Department of Comparative Medicine at the University of Washington accredited by the Association for Assessment and Accreditation of Laboratory Animal Care. Animal experiments were conducted in accordance with the University of Washington’s Institutional Animal Care and Use Committee under approved protocol 4470-01. For each immunization, low-endotoxin HCoV-229E S ectodomain was diluted to 100 µg ml–1 in buffer and mixed 1:1 (v/v) with AddaVax adjuvant (InvivoGen vac-adx-10) to obtain a final dose of 5 µg of HCoV-229E S ectodomain per animal per injection. At 8 weeks of age, 10 mice per group were subcutaneously injected in the inguinal region with 100 μl immunogen at weeks 0, 3 and 6. Animals were bled by submental venous puncture at weeks 2 and 5, and terminal blood was collected at week 8. Whole blood was collected in BD microtainer collection tubes and rested at room temperature for 30 min for coagulation. Tubes were then centrifuged for 10 min at 2,000g and serum was collected and stored at −80 °C until use.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
