Rc-o319 S utilizes a narrow range of Rhinolophus bat ACE2 orthologs as receptors

We used a S:ACE2 interaction mediated cell-cell fusion assay to identify ACE2 orthologs that can interact with Rc-o319 S-protein [37]. Effector cells expressing Rc-o319 S-proteins were tested against receptor cells expressing a panel of ACE2 orthologs for fusion activity (Fig 1A). In this assay, several Rhinolophus affinis (Ra) and Rhinolophus sinicus (Rs) ACE2 molecules, including Ra5538, Ra9479, Rs1446, Rs3359, and Rs3366, were found to mediate varying levels of cell-cell fusion, with Ra9479 showing the highest activity and Rs3366 the lowest. However, all were substantially weaker than the cognate bACE2_R.cor receptor. Rc-o319 S-protein is unable to utilize pangolin and human ACE2 (hACE2) orthologs for fusion. Receptor binding assays using biolayer interferometry (BLI) show that Rc-o319-RBD-Fc and Rc-o319-S-trimer bind bACE2_R.cor with moderate-to-high affinities (Fig 1B left panels, 76 nM for RBD-Fc and 27 nM for S-trimer, and S1 Table). Although Rc-o319 appears to somewhat use bACE2_Ra9479, showing weak fusion activity in the cell-cell fusion assay. BLI assays were unable to detect bACE2_Ra9479 binding by the Rc-o319 S-trimer and dimeric Rc-o319-RBD-Fc (Fig 1B, middle panels). Rc-o319-S-trimer binds bACE2_R.cor with relatively weak responses and an estimated affinity of 27 nM, tighter than that observed for the dimeric Rc-o319-RBD-Fc, indicating avidity-amplified receptor binding (Fig 1B, left panels). Consistent with the cell-cell fusion assay results, neither Rc-o319-RBD-Fc nor Rc-o319-S-trimer was able to bind hACE2 (Fig 1B, right panels). These results demonstrate that Rc-o319-S-protein has a restricted ACE2 specificity, with a clear preference towards its cognate bACE2_R.cor receptor (Fig 1A and 1B).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Rc-o319 S-protein utilizes a narrow range of Rhinolophus bat ACE2 orthologs as receptors and adopts two locked, 3-RBD-down conformations.

(A) Cell-cell fusion assays using receptor cells expressing bat, pangolin or human ACE2 molecules and effector cells expressing the Rc-o319 S-proteins. Fused cells show GFP signals. Representative images of cell-cell fusion captured at 6 hours post-cell-mixing are shown. Scale bar, 500 μm. (B) Binding of Rc-o319 S-trimer to immobilized dimeric bACE2_R.cor, hACE2 and bACE2_Ra9479 as assessed by BLI assays. (C) A schematic representation showing the eight functional domains or regions of Rc-o319 S-protein. (D) Cryo-EM density and associated molecular model showing the overall architecture of a Rc-o319 S-trimer in a locked-1 conformation. (E) Lipid binding pocket in the RBD of locked-1 Rc-o319 S-trimer. Modeled linoleic acid (magenta) interacts with the shown hydrophobic residues of the lipid binding pocket. Salt bridge and hydrogen bond are shown as dashed lines. (F) Structure of the Domain D region in the locked-1 Rc-o319 S-trimer. Consistent with a locked conformation, the fusion peptide proximal region (FPPR) is ordered. Residues 580-605_Rc-o319 form a large disordered loop in the locked-1 conformation. (G) Cryo-EM density and associated molecular model showing the overall architecture of a locked-2 Rc-o319 S-trimer. (H) The RBD lipid binding pocket is empty with Y339_Rc-o319 side-chain reorientated in the locked-2 conformation. (I) Residues 580-605_Rc-o319 refold and form an ordered structure. A stabilizing cation-π interaction (red dashed line) is formed between Domain D residue W598_Rc-o319 and Domain C-D junction residue R293_Rc-o319.

https://doi.org/10.1371/journal.ppat.1014245.g001

Structures of the Rc-o319 S-protein identify a new type of RBM

To understand the unusual restricted receptor specificity of Rc-o319 S-protein, purified Rc-o319 S-protein ectodomain was imaged by cryo-electron microscopy (Cryo-EM) (S2 Fig and S2 Table), revealing that Rc-o319 S-trimers adopt two distinct conformations (Figs 1D-1I, S2, S3, S4A, S4F and S4G). In both conformations, the 3 RBDs in each S-trimer are all in “down” positions (Fig 1D and 1G). Detailed structural analysis reveals tight trimer packing for both conformations. Based on their structural features, the two conformations differ primarily at domain D (Figs 2A, 2B, S3, S4F and S4G) and have been identified as “locked-1” and “locked-2” conformations previously identified for both SARS-CoV-1 and SARS-CoV-2 S-trimers [22,25,38–40].

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Rc-o319 S-trimer structures reveal an RBD featuring a novel RBM and an NTD with altered loops.

(A-B) Top-views of locked-1 and locked-2 Rc-o319 S-trimer structures. N-linked glycans are shown as light gray spheres. NTDs are colored blue, RBDs are colored orange (locked-1, A) or pink (locked-2, B). The lower panels show intra- or inter-protomer interactions involving RBD residues. K458_Rc-o319 forms an inter-protomer hydrogen bond with N344_Rc-o319 of an RBD of a neighboring protomer in both locked-1 and locked-2 S-trimers. In the locked-1 conformation, D468_Rc-o319 forms an intra-protomer hydrogen bond with Y473_Rc-o319; in the locked-2 conformation, D468_Rc-o319 instead mediates two inter-protomer hydrogen bonds with S469_Rc-o319 of an adjacent protomer. (C-D) Comparison of Rc-o319 (C) and SARS-CoV-2 (D) N-terminal domain (NTD) structures. The five highly variable NTD loops (N1-N5) are shown in the Rc-o319 and SARS-CoV-2 (PDB: 7XU4) NTD structures. A four-amino-acid residue insertion is highlighted in the Rc-o319 NTD in red. (E) The structure of the Rc-o319 RBD. The receptor binding motif (RBM) of the RBD is colored pink and the core of the RBD is colored gray. Disulfide bonds within the RBD are indicated. Featured structural elements, beta-Loop (BL), small loop (SL), Lamella (LM) sheets, β5 and β6, and the large RBM loop are indicated. (F-G) For comparison, the SARS-CoV-2 RBD (PDB: 7XU4) is shown (F) and superposed on the Rc-o319 RBD structure (G).

https://doi.org/10.1371/journal.ppat.1014245.g002

In the locked-1 Rc-o319 S-trimer, each RBD contains a density that can be modelled as a fatty acid with 18-carbon atoms, previous mass-spectrometry analysis identified it being either linoleic acid [41,42] or oleic acid [43] (Figs 1E and S4C). Consistent with previous observations, the aliphatic tail of the fatty acid is bound within the hydrophobic fatty-acid binding pocket of one RBD, while its carboxylate group forms salt-bridges with R382_Rc-o319 of a neighbouring RBD (Figs 1E and S4C); these fatty acids mediate inter-protomer interactions likely stabilizing the S-trimer in the observed locked-1 conformation. Each NTD also contains a density corresponding to a biliverdin molecule. The binding of biliverdin has been correlated with evasion of antibody immunity [44,45] (Figs 1D, 1G and S4D). Consistent with being in a locked conformation, the fusion peptide proximal region (FPPR) is ordered (Fig 1F). Finally, consistent with the locked-1 conformation observed for the S-proteins of SARS-CoV-1 [25] and SARS-CoV-2 [40], residues 580–605_Rc-o319 form a large disordered loop within Domain D of the Rc-o319 S-trimer (Figs 1D, 1F and S4F).

In the locked-2 conformation, no fatty acid density was observed within the RBD fatty-acid binding pocket (Fig 1H). This is accompanied with a restructuring of the fatty-acid binding pocket (Figs 1H and S4C) as observed when the fatty-acid binding pockets were vacated in other sarbecovirus S-trimers [41,46]. Each NTD still contains a density modelled as a biliverdin (Fig 1D and 1G). The FPPR is ordered (Fig 1I). Differently, the disordered Domain D loop in the locked-1 conformation refolds and forms a large ordered helical structure, packing against the core of the Domain D and the junction between Domains C and D with non-interupted cryo-EM density (Figs 1F, 1I and S4G) [25,40].

A structural change between locked-1 and locked-2 conformations in the Domain C-D junction region can be observed similar to previous observations on the S-proteins of SARS-CoV-1 and SARS-CoV-2 [25,40] (Figs 4F, 4G, and S4A). Near the 3-fold axis, to the top of the S-trimer, D468_Rc-o319 forms an intra-protomer hydrogen bond with Y473_Rc-o319 in the locked-1 conformation (Fig 2A). Whereas in locked-2 conformation, D468_Rc-o319 mediates two inter-protomer hydrogen bonds with S469_Rc-o319 of a neighbouring S-protomer (Fig 2B). Of Note, in both conformations, K458_Rc-o319, equivalent to Q493_SARS2, previously identified as an important determinant for ACE2 specificity [17,47,48], is found to form an inter-protomer hydrogen bond to the glycan attaching to N344_Rc-o319 of a neighbouring S-protomer (Fig 2A and 2B, the second panels). Inter-protomer hydrogen bonds likely contribute to S-trimer stability.

In both Rc-o319 S-trimer conformations, their NTDs are well resolved, revealing five loops that are markedly shorter than those in SARS-CoV-2 (Figs 2C, 2D and S4E). In addition, a four-amino-acid insertion (residues 189–192_Rc-o319) is identified in the Rc-o319 NTD, forming an “inserted loop” exposed to the exterior of the trimer (S4B Fig). It’s well-established that the SARS-CoV-2 S-protein NTD is under strong immune pressure [49,50]. The truncated and elongated loops observed in the Rc-o319 S-protein NTD may therefore confer a distinct antigenic profile, although whether these features reflect immune-driven selection in bats remains to be determined.

Finally, Rc-o319 S-trimer structures show that their RBDs form structures distinct from previous characterized sarbecovirus RBDs (Figs 2E-2G and S5). In contrast to type-3 RBDs, where 13- to 14-amino-acid deletions relative to SARS-CoV-2 eliminate the large-loop (LL) (S5C and S5F-S5G Fig), the Rc-o319 RBD retains an extension in a similar region as the LL of Type-1 RBDs, despite a nine-amino-acid deletion (Fig 2E and 2G). In the type-1 RBD, the LL is formed by two half-loops stabilized by an intra-loop disulfide bond (Fig 2F). By contrast, the equivalent extension in the Rc-o319 RBD comprises two small antiparallel β-sheets connected by a short loop (Figs 2E and 3). We designate this distinct extension as a beta-loop (BL) structure. Due to the nine-amino-acid deletion, the stabilizing disulfide bond found in the LL is lost. The shorter BL occupies a position roughly corresponding to the C-terminal half-loop of the LL in type-1 RBDs (Figs 2E-2G and 3D), albeit with notable conformational and tilting angle differences (Figs 2G and 3D). This distinct beta-loop (BL) structure, by comparison with the LL structure in type-1 RBDs (Fig 2F), suggests that the Rc-o319-RBD likely binds ACE2 with distinct interactions.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. Cryo-EM structure of the Rc-o319-RBD:bACE2_R.cor complex with a detailed comparison of RBD:ACE2 interactions versus those in the SARS-CoV-2-RBD:hACE2 complex.

(A) Molecular structure of a monomeric Rc-o319-RBD:bACE2_R.cor complex extracted from the dimeric Rc-o319-RBD:bACE2_R.cor complex structure resolved by cryo-EM (see S2 Fig). Detailed interactions formed between the Rc-o319-RBD and bACE2_R.cor are shown with hydrogen bonds and salt bridges shown as dashed lines. (B) Superposition of the Rc-o319-RBD:bACE2_R.cor and SARS-CoV-2-RBD:hACE2 complex structures. The interface areas between the RBD and ACE2 in the two structures are divided into four regions: beta-loop (BL in Rc-o319-RBD)/large-loop (LL in SARS-CoV-2-RBD) (purple box), lamella (LM, blue box), anchor-loop (AL, green box), and small-loop (SL, orange box). (C-D) Comparison of the Rc-o319 RBM-loop and beta-loop (BL) with the RBM-loop and large-loop (LL) of SARS-CoV-2. (E) An N-linked glycan (shown in grey stick representation overlayed with white cryo-EM density) attached to Asn38 of bACE2_R.cor is involved in the binding interface with Rc-o319-RBD, representing a unique structural feature. (F-I) Detailed interactions identified in the beta-loop+RBM-loop (BL, purple box), lamella (LM, blue box), anchor-loop (AL, green box), and small-loop (SL, orange box) regions of the interface between Rc-o319-RBD and bACE2_R.cor. (J-M) For comparison, detailed interactions are shown for the corresponding regions in the interface between SARS-CoV-2-RBD and hACE2. Hydrogen bonds, pi-pi (sugar-pi) interactions and cation-pi interactions are indicated by black, yellow (green) and red dashed lines, respectively.

https://doi.org/10.1371/journal.ppat.1014245.g003

Structure of the Rc-o319-RBD:bACE2_R.cor complex

To further understand receptor binding by the Rc-o319 S-protein, we determined a structure of Rc-o319-RBD in complex with its cognate bACE2_R.cor receptor (Figs 3 and S2B). This structure uncovers several previously unrecognized interactions at the Rc-o319-RBD:bACE2_R.cor interface, even though the molecular surfaces employed for binding largely mirror those in other sarbecovirus RBD:ACE2 complexes (Fig 3A and 3B). These interactions are mediated by the specialized RBD BL and RBM-loop and a glycan linked to Asn38 of bACE2_R.cor (Fig 3A, 3E and Fig 3H-3I) (interfacial residues in the RBD and ACE2 are denoted using single-letter and three-letter amino acid codes, respectively). The buried surface area (BSA) of the complex was calculated as 1005 Ų, which includes a contribution of 252 Ų from the binding of the Asn38-linked glycan. In line with previous interface analysis, the binding interface can be divided into four regions, namely the beta-loop (BL), lamella (LM), anchor loop (AL) and small loop (SL) regions (Fig 3F-3I).

The beta-loop (BL) of Rc-o319 RBD engages highly distinctive interactions (Fig 3A and 3F). Due to its distinct structure and protrusion from the RBM, the BL contacts the bACE2_R.cor near Lys27, forming a hydrogen bond with the side chain of Lys27 (Fig 3A and 3F). In contrast, LL in type-1 RBDs contacts ACE2 near residue 24, more towards the ACE2 N-terminus (Fig 3J). Distinctively, BL residue Y447_Rc-o319 forms a charged hydrogen bond with Asp31 of bACE2_R.cor (Fig 3A and 3F). Of note, the positively charged residue K454_Rc-o319, unique to the Rc-o319 RBM, extends from the BL to form a salt-bridge with Glu75_Rc-o319 of bACE2_R.cor (Fig 3A and 3F) and is positioned in close proximity to the negatively charged bACE2_R.cor residues Asp31 and Glu35 (Fig 3F and 3G). Changing this residue (K454Y_Rc-o319) to the corresponding residue Y in SARS-CoV-2 substantially reduced bACE2_R.cor binding (S9B Fig and S4 Table).

Distinct from SARS-CoV-2, the large Rc-o319 RBM-loop (Figs 2E, 3C, 3F and S7) adjacent and N-terminal to the BL contributes to bACE2_R.cor binding and contacts the BL structure with specialized interactions (compare Fig 3F and 3J): W430_Rc-o319 in the RBM-loop forms a cation-π interaction with Lys27 of bACE2_R.cor and likely stabilizes the BL via a π-π interaction with the BL residue Y447_Rc-o319 (Figs 3F and S7B and see below).

In the lamella (LM) region, K458_Rc-o319 forms a salt-bridge with bACE2_R.cor Glu35 (Fig 3G). Distinctively, in this region, the Asn38-glycan is part of the ACE2-RBD interface, forming hydrogen bonds with N459_Rc-o319 (Fig 3G). Also in this region, Lys353_Rc-o319 forms a hydrogen bond with the G461_Rc-o319 backbone (Fig 3G). In the anchor-loop (AL) region, a hydrogen bond network is observed among bACE2_R.cor residues Tyr41, Asn330, Glu335, Arg357 and the RBD residue S465_Rc-o319 (Fig 3H). Further, the glycan attached to Asn38 of bACE2_R.cor forms an inter-molecular hydrogen bond with bACE2_R.cor residue Gln42 and also hydrogen bonds with RBD residue Y463_Rc-o319 (Fig 3H). Finally, in the small-loop (SL) region, F423_Rc-o319 forms a sugar-π contact with the glycan attached to Asn38 (Fig 3I). These interactions, by comparison with those used by SARS-CoV-2 to engage ACE2 (Fig 3J-3M), show that Rc-o319 RBD engages its cognate bACE2_R.cor receptor with unique characteristics.

Molecular determinates underlying Rc-o319 specificity for bACE2_R.cor

To understand the specificity of the Rc-o319 S-protein towards the bACE2_R.cor receptor. We employed a mutagenesis approach to systematically change RBM regions to match those of SARS-CoV-2 (BL change = swapping of 443–456_Rc-o319 “AHYDYQVGTQFKSS” to 469–491_SARS2 “STEIYQAGSTPCNGVEGFNCYFP”; SL change = F423Y_Rc-o319; LM change = K458Q_Rc-o319; AL change = S465T_Rc-o319 + A466N_Rc-o319 + H470Y_Rc-o319; RBM-loop change = W430F_Rc-o319 + S432K_Rc-o319). These variants were used to assess hACE2 and bACE2_R.cor binding by BLI assays using RBD-Fc proteins, as well as receptor utilization by cell-cell fusion and pseudovirus entry assays using full-length S-proteins (Figs 4M and S13–S15). A similar approach has been previously used to elucidate how high-affinity hACE2 binding may be acquired by S-proteins with type-2 RBDs, as we defined previously [17], from clade 3 sarbecoviruses.

We found that changing AL and SL regions individually had very little effect in improving hACE2 binding (Fig 4D and 4E, the second panels, and S3 Table). The SL change retained very strong bACE2_R.cor utilization (Fig 4E, the third and fifth panels), whereas the AL change reduced bACE2_R.cor utilization somewhat in cell-cell fusion (Fig 4D, the third and fifth panels) and pseudovirus entry assays (Fig 4M). Interestingly, the AL change very slightly increased bACE2_R.cor binding when assayed using both Rc-o319-RBD-Fc and the Rc-o319 S-trimer (Figs 4D, the first panels, S10C and S14F, and S3, S5 and S9 Tables). The discrepancy between the increased BLI affinity and the reduced cellular activity may reflect differences in assay conditions. In the cellular context, phenotypes may also be influenced by S-protein-related factors beyond receptor binding, such as S-protein stability, efficiency of surface expression, incorporation into pseudovirus particles, and activation by cellular proteases, which could contribute to the observed discrepancy (see the Limitations section). Changing the LM region alone (the LM change) reduced bACE2_R.cor binding and utilization (Fig 4C, the first, third and fifth panels) but still failed to confer hACE2-mediated fusion activity and pseudovirus entry (Fig 4C, the fourth and fifth panels).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. ACE2 utilization by wild-type and variant Rc-o319 RBD-Fc in BLI and wild-type and variant Rc-o319 S-proteins in cell-cell fusion and Pseudovirus entry assays.

(A-L) Top two panels, binding of bACE2_R.cor and hACE2 by wild-type (WT) and variant Rc-o319-RBD-Fc proteins as assessed by BLI assays. Dimeric Rc-o319-RBD-Fc proteins were immobilized on the BLI sensors and tested binding against dimeric Rhinolophus cornutus bat ACE2 (bACE2_R.cor) or hACE2 protein as the analyte in solution. bACE2_R.cor and hACE2 were three-fold serial diluted from 3000 to 4.1 nM. K_D values estimated from each binding experiment are shown alongside the corresponding binding curves. Lower two panels, Representative images of cell-cell fusion captured at 12 hours post-cell-mixing are shown. Scale bar, 500 μm. Effector cells expressing WT or variant Rc-o319 S-proteins were tested against receptor cells expressing either bACE2_R.cor or hACE2. The bottom panels show the quantification of cell-cell fusion measuring the GFP+ area at 3, 6, 12 hours (h) post-cell-mixing. Data are presented as mean ± s.d., n = 2. The P values were determined by unpaired t-test comparing wild type (WT) with each Rc-o319 variant. *P < 0.05, **P< 0.01, ***P< 0.001; NS, not significant. (M) Pseudovirus assay. Entry of VSV particles pseudotyped with Rc-o319 S-protein variants into HEK293T cells transiently expressing hACE2 or bACE2_R.cor and co-expressing TMPRSS2. Luciferase activity was measured 24 h post-transduction. Data are presented as mean ± s.d. (n = 3). P values were calculated using unpaired t tests comparing wild-type (WT) with each Rc-o319 variant. * P< 0.05, ** P< 0.01, *** P < 0.001 and ****P< 0.0001.

https://doi.org/10.1371/journal.ppat.1014245.g004

Swapping of the Rc-o319 BL region to the LL sequence of SARS-CoV-2-RBD (the BL change), appears to confer Rc-o319 with the ability to utilize hACE2 for cell fusion activity at the expense of substantially reduced fusion activity mediated by its cognate bACE2_R.cor receptor (Fig 4B). Consistently, the BL change conferred hACE2 mediated pseudovirus entry, while pseudovirus entry mediated by the cognate bACE2_R.cor reduced by ~52-fold (Fig 4M). Despite conferring detectable cell-cell fusion activity and pseudovirus entry, hACE2 binding by the dimeric BL variant of the Rc-o319 RBD-Fc protein remained below the detection limit of the BLI assay, while bACE2_R.cor binding was markedly reduced and fell below the detection limit of the BLI assay (Fig 4B, the first panels). A subsequent BLI assay using the BL variant of the Rc-o319 S-trimer detected weak hACE2 binding (S10B Fig and S5 Table), likely made detectable by amplification through trivalent binding-mediated avidity, underscoring a dependence of assay sensitivity on binding valency. As cell-based assays involve highly multivalent interactions mediated by multiple S-trimers and receptors, avidity-enhanced binding renders them substantially more sensitive to weak interactions (also see the Limitations section).

Therefore, the above results show that replacing each RBM region singly with its SARS-CoV-2 counterpart in the Rc-o319 RBD-Fc failed to confer high-affinity binding to hACE2. This finding is consistent with our previous observations that high-affinity hACE2 binding requires a sarbecovirus RBD to establish optimal contacts across its all four RBM regions, including LL/BL, LM, AL, and SL [17]. A similar requirement for synergistic interactions across multiple RBM regions to achieve high-affinity receptor binding has also been observed in merbecoviruses [51].

However, unexpectedly, the BL + LM + AL + SL variant, designed to introduce optimal interactions across all four RBM regions, exhibited enhanced hACE2 utilization in cell-cell fusion and pseudovirus entry assays, yet its hACE2 binding remained too weak to be detected by BLI (Fig 4F, the second and fourth panels). The BL + LM + AL + SL variant of Rc-o319-RBD-Fc protein also showed no bACE2_R.cor binding by BLI, and consequently exhibited very little bACE2_R.cor mediated fusion activity (Fig 4F, the first and third panels). Therefore, in the case of Rc-o319 S-protein, determinants outside of the BL, LM, AL, and SL regions likely hinder high-affinity hACE2 binding.

Structural comparison shows that W430_Rc-o319 and S432_Rc-o319 in the large RBM-loop [amino acid (AA) 429–442_Rc-o319, corresponding to AA 455–468_SARS2], located N-terminal to the BL, are unique to the Rc-o319 RBD (Figs 3C, 3F, S5G and S7). In SARS-CoV-2-RBD, the corresponding residues F456_SARS2 and K458_SARS2, respectively interact with the LL residues Y473_SARS2 and E471_SARS2 (Figs 3J, S7A and S7D). These interactions likely stabilize the SARS-CoV-2 LL structure. We introduced these residues into the Rc-o319 BL + LM + AL + SL variant as the W430F_Rc-o319 + S432K_Rc-o319 change (collectively named as the “RBM-loop” change). BLI binding results show that the dimeric BL + SL + LM + AL + RBM-loop variant of Rc-o319-RBD-Fc is able to bind hACE2 with a much increased affinity of 31.7 nM (Fig 4G, the second panel). We also attempted the swap of Rc-o319 RBM for the SARS-CoV-2 RBM (Rc-o319-RBD-RBM_SARS2), as expected, the Rc-o319-RBD-Fc variant with the swapped RBM also showed high-affinity binding to hACE2 with only slightly higher affinity (12.3 nM) than the BL + LM + AL + SL + RBM-loop variant (compare Fig 4G and 4H, the second panels). Consistently, both BL + LM + AL + SL + RBM-loop and RBM swapped Rc-o319-RBD-RBM_SARS2 show effective hACE2 mediated cell-fusion activities (Fig 4G and 4H, the fourth and fifth panels). Therefore, our mutagenesis results identify residues in the RBM-loop of Rc-o319, N-terminal to the BL, that are incompatible with the LL structure of type-1 RBDs, thereby preventing the LL from mediating optimal ACE2 interactions (S7C, S7D and S8C Figs).

To further clarify the role of the RBM-loop in hACE2 recognition, we introduced the RBM-loop change alone into the Rc-o319 S-protein. This change did not confer hACE2-dependent cell-cell fusion (S8 Fig), consistent with structural analyses indicating that the RBM-loop does not make direct contacts with ACE2 (Fig 3). No detectable hACE2 binding was observed for the Rc-o319 BL-variant RBD-Fc (Fig 4B), and a similar lack of hACE2 binding was also observed for the Rc-o319 BL + LM-variant RBD-Fc in BLI assays (Fig 4I), despite the corresponding S-proteins mediating clear hACE2-dependent cell-cell fusion and pseudovirus entry (Fig 4B, 4I and 4M). When the RBM-loop change was introduced into the Rc-o319 BL- or BL + LM-variant RBD-Fc (i.e., BL + RBM-loop and BL + LM + RBM-loop variants), weak but detectable hACE2 binding could be detected in BLI assays (Fig 4B and 4I–4K). These results confirm that the RBM-loop plays a modulatory role, essential for mediating high-affintiy ACE2 interaction.

Utilization of bACE2_R.cor by different types of sarbecoviruses

Our structural and biochemical analyses show that the Rc-o319 S-protein has a highly restrictive specificity towards its cognate bACE2_R.cor receptor, we further characterized the potential of bACE2_R.cor to support diverse sarbecoviruses (Fig 5). A panel of sarbecovirus S-proteins was surface-expressed on effector cells and tested for cell-cell fusion activity against receptor cells expressing bACE2_R.cor (Fig 5A). The panel included S-proteins with type-1 RBDs [17] from clade 1a and 1b sarbeviruses, including SARS-CoV-1, LYRa11, SHC014, WIV1, RaTG13, BANAL-20-52, BANAL-20-236, and the SARS-CoV-2 ancestral strain; S-proteins with type-2 RBDs [17] from clade 3 sarbeviruses, including BtKY72, BM48-31, RfGB02, and RhGB07; and S-proteins with type-3 RBDs [17] from clade 2 sarbeviruses, including GX2013, HeB2013, YN2013, and RmYN02. The results show that bACE2_R.cor supported receptor binding and mediated cell fusion activities of several clade 1 sarbecovirus S-proteins with type-1 RBDs (Fig 5B and 5C). However, potent fusion activity was only observed for several S-proteins of viruses from the SARS-CoV-2 lineage (clade 1b). Although the SARS-CoV-1 S-protein exhibits only intermediate bACE2_R.cor-mediated cell-cell fusion activity, it is nevertheless the most potent among S-proteins from viruses within the SARS-CoV-1 lineage (clade 1a) (Fig 5A-5C). Among bACE2_R.cor utilizing S-proteins, only the BANAL-20-236 S-trimer was found to bind bACE2_R.cor in BLI assays, whereas S-trimers from BANAL-20-52, SARS-CoV-2, and SARS-CoV-1 exhibited no detectable binding to bACE2_R.cor despite displaying robust cell-cell fusion activity (Figs 5C, S11B, S11C, S11E and S11F, and S6 Table). These observations indicate that, in addition to receptor binding, complex cell-associated factors, potentially including S-protein expression level, protein stability, and susceptibility to activation by cellular proteases, can critically influence S-protein-mediated membrane fusion (also see the Limitations section).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 5. bACE2_R.cor utilization by different sarbecoviruses S-proteins.

(A) Different sarbecovirus S-proteins are sorted based on S-protein phylogenetics (clade 1-–3) and grouped by RBD types (I-IV, as defined by Si et al [28]. and 1–3, as defined by Wang et al [17].). Type I (type-1) RBD strains are shown in red (SARS-CoV-2 clade, Clade 1b) and blue (SARS-CoV-1 clade, clade 1a), type II (type-2) strains in orange, type IV (type-3) strains in brown and type III strains in green. Quantification of S-protein expression on the cell surface of effector cells by flow cytometry using an S2 antibody as the probe. Flow cytometry was repeated three times with representative data shown. (B) Effector cells shown in (A) were tested against receptor cells expressing hACE2 (gray line), bACE2_R.cor (red line) and bACE2_R.cor- ΔGLC₃₈ (green line, bACE2_R.cor with the Thr40Ala mutation to remove the Asn38-glycan). Cell-cell fusion was quantified by assessing GFP+ areas (y-axis) at 4, 18, and 24 hours (h, x-axis) post-cell-mixing (pt). Data are presented as mean  ±  s.d., n  =  2. P values were calculated using unpaired t-tests comparing fusion activities mediated by bACE2_R.cor and bACE2_R.cor-ΔGLC₃₈. *P < 0.05, **P <0.01, ***P < 0.001; NS, not significant. (C) Representative images of cell-cell fusion captured at 18 hours post-cell-mixing are shown for receptor cells expressing different ACE2. Scale bar, 500 μm.

https://doi.org/10.1371/journal.ppat.1014245.g005

The Asn38-linked glycan in bACE2_R.cor is involved in S-protein binding, a feature observed for the first time between an ACE2 and a sarbecovirus S-protein. To understand the effect of this glycan on S-protein function, we removed the glycan by introducing a Thr40Ala change (designated as ‘bACE2_R.cor-ΔGLC₃₈’). In cell-cell fusion assays, the bACE2_R.cor-ΔGLC₃₈ variant receptor showed a tendency to mediate enhanced fusion activities for all the sarbecovirus S-proteins capable of utilizing the WT bACE2_R.cor (Fig 5B and 5C). BLI assay results show that the removal of the Asn38-glycan was able to enhance bACE2_R.cor binding not only for Rc-o319 but also for the BANAL-20-52 and BANAL-20-236 S-trimers (S11A, 11C and 11E Fig and S6 Table). In particular, although unable to utilize WT bACE2_R.cor, BtKY72 S-protein, with a type-2 RBM, of a clade 3 sarbevirus, was able to utilize bACE2_R.cor-ΔGLC₃₈, albeit exhibiting only very weak cell fusion activity. Consistently, BLI binding results show that the BtKY72 S-trimer was unable to bind bACE2_R.cor, but gained bACE2_R.cor-ΔGLC₃₈ binding with an estimated affinity of 11.8 nM (S11H Fig, the second panel, and S6 Table).

To further assess the functional role of the Asn38-glycan, an N-linked glycosylation site was introduced into a Rhinolophus affinis bat ACE2 (bACE2_Ra9479), by substituting Asp38 with Asn, generating bACE2_Ra9479-GLC₃₈. In both cell-cell fusion and BLI assays, BtKY72 and SARS-CoV-1 S-proteins exhibited substantially reduced fusion activity utilizing bACE2_Ra9479-GLC₃₈ compared with the WT bACE2_Ra9479 receptor (S16B, S16F; S11J and S11K Figs, and S7 Table). In parallel, we swapped the RBMs of SARS-CoV-1 and BtKY72 S-proteins with that of Rc-o319, generating chimeric SARS1_Rc-o319RBM and BtKY72_Rc-o319RBM S-proteins. Although being expressed by the cell (S16Iand S16J Fig), the RBM swapping appeared to render the SARS1_Rc-o319RBM S-protein non-functional, unable to mediate cell-cell fusion mediated by either bACE2_Ra9479 or bACE2_R.cor receptor and their Asn38-glycan variants (S16C and S16D Fig). The BtKY72_Rc-o319RBM S-protein showed substantially reduced utilization of the bACE2_Ra9479 and bACE2_Ra9479-GLC₃₈ receptors (S16F and S16H Fig). In contrast, its utilization of both bACE2_R.cor and bACE2_R.cor-ΔGLC₃₈ was only marginally enhanced compared with the WT BtKY72 S protein, yet remained substantially weaker than that of the Rc-o319 S protein (Figs 5C, S16E and S16G). These results highlight compatibility constraints between different RBMs and S-protein backbones, indicating that RBM changes can adversely affect overall S-protein function.

Limitations of the study

A limitation of this study is that the biophysical BLI assays employed relatively low-valency analytes (monomeric RBD, dimeric ACE2, or trimeric S-protein), thereby limiting avidity-mediated binding amplification and sensitivity to weak interactions. In contrast, cell-cell fusion and pseudovirus entry assays (Figs 4, 5, and S16) involve highly multivalent interactions between multiple surface-displayed S-trimers and ACE2 molecules, resulting in substantially enhanced avidity and sensitivity. Accordingly, several interactions that were undetectable by BLI nonetheless supported detectable fusion or entry in cell-based assays (Figs 4B, 4F, 4G, 4I, 4K, 4L, and 5). A further limitation is that, in the cellular context, phenotypes can be influenced by S-protein-related factors beyond receptor binding, potentially including S-protein stability, efficiency of surface expression, incorporation into pseudovirus particles, and activation by cellular proteases. This is exemplified in Fig 5, where the SARS-CoV-2 S-protein exhibited the strongest fusion activity mediated by both WT bACE2_R.cor and the glycan mutant bACE2_R.cor-ΔGLC₃₈, despite no detectable binding to either receptor in BLI assays. An additional example is shown in S16 Fig, where a SARS-CoV-1 S-protein variant carrying the Rc-o319 RBM (SARS1_Rc-o319RBM) failed to mediate bACE2_R.cor-dependent fusion despite detectable expression, suggesting incompatibility between the swapped RBM and the S-protein backbone. While cell-based assay readouts often scale with large differences in receptor-binding affinity, they may not scale linearly when affinity differences are modest and the above-discussed S-protein-related factors are involved. Several RBM-modified S-proteins exhibited modestly increased ACE2 binding in BLI assays but reduced fusion or entry activity in cell-based assays (Fig 4A vs 4D and 4G vs 4H). We speculate that RBM changes may affect S-protein stability, surface presentation, or interactions with host factors, leading to nonlinear relationships between receptor-binding affinity and cellular phenotype. The effects observed in cell-based systems may not fully recapitulate authentic viral infection, where additional biological constraints and host factors come into play. Addressing this will require studies with authentic viruses and in vivo models to evaluate how receptor-affinity changes influence infection in a physiological context. Despite these limitations, our findings delineate key barriers to receptor adaptation and provide a conceptual framework for assessing cross-species transmission risk.

Protein constructs

S gene of Rc-o319 (Taxonomy ID: 694009) was codon-optimized and synthesized by Sangon Biotech (Shanghai, China). The transmembrane domain and C-terminal end of Rc-o319 (AA 1176-1235) were removed before the C-terminus of the coding sequence was fused to a T4 fibritin trimerization foldon, an HRV3C cleavage site, an Octo-His tag, and a double strep-tag to allow secretory expression and purification.

Rhinolophus cornutus (R.cor, Genebank: BCG67443.1), human (hACE2, Genebank: BAB40370.1), Rhinolophus ferrumequinum (R.ferr, Genbank: BAH02663.1) and greater mouse-eared bat Myotis lucifugus (M.luci, Genbank: XP_023609438.1) ACE genes were synthesized. Pangolin (Pangolin, Genbank: XP_017505752.1), Rhinolophus affinis 9479 (Ra9479, Genbank: QMQ39227), Rhinolophus affinis 787 (Ra5538, Genbank: QMQ39222), Rhinolophus sinicus WJ2 (Rs-WJ2, Genbank: QMQ39202.1), Rhinolophus sinicus 1434 (Rs1434, Genbank: QMQ39216.1), Rhinolophus sinicus 1438 (Rs1438, Genbank: QMQ39203.1), Rhinolophus sinicus 1446 (Rs1446 Genbank: MT394194.1), Rhinolophus sinicus 3357 (Rs3357 Genbank: AGZ48803.1), Rhinolophus sinicus 3359 (Rs3359 Genbank: QMQ39211.1), Rhinolophus sinicus 3366 (Rs3366, Genbank: QMQ39215.1) and Rhinolophus sinicus 5720 (Rs5720, Genbank: MT394182.1) ACE2 gene sequences were from a previous study [65]. The peptidase domain and the C-terminal collectrin-like domain of ACE2, bearing a carboxy-terminal Hexa-His tag, was generated for hACE2. Due to the presence of the C-terminal collectrin-like domain, these his-tagged ACE2 (AA 1-731) proteins form dimers. Additionally, the same length of ACE2s were fused to the Fc domain of human IgG with one thrombin site and a GGGG linker in between, resulting in Fc-tagged ACE2.

Rc-o319 RBD and variants (AA 293-506), SARS-CoV-2 RBD (AA 319-541), were fused with human IgG Fc domains to generate dimeric RBD-Fc proteins.

Protein expression and purification

Proteins were expressed by transient transfection in Expi293F cells, 1 mg of DNA was transfected into 1 L of cells using Linear Polyethylenimine (PEI) transfection reagent. After transfection, cells were cultured at 33 °C for 4 days for S-proteins, 5 days for RBD and ACE2 proteins. For his-tagged S-proteins and RBD proteins, the supernatants were harvested by centrifugation and 25 mM phosphate pH 8.0, 3 mM imidazole, 300 mM NaCl, and 0.5 mM PMSF were supplemented. Supernatant was recirculated onto a 5-ml Talon Cobalt column for about 3 times and the column was washed by 100 ml buffer A (25 mM phosphate pH 8.0, 5 mM imidazole, and 300 mM NaCl). Proteins were eluted using a linear gradient with increasing concentration of Buffer B (25 mM phosphate pH 8.0, 500 mM imidazole, and 300 mM NaCl). Eluted protein samples were quality checked by SDS-PAGE, concentrated and buffer-exchanged into PBS using a 100 kDa (for S-proteins) or 10 kDa (for RBD proteins) MWCO Amicon Ultra filtration device (Merck Millipore). The concentrated proteins were frozen in liquid nitrogen and stored at −80 °C.

For RBD-Fc and ACE2-Fc proteins, 5 days post-transfection, each culture supernatant was cleared by filtering through a 0.45 μm filter membrane before being loaded onto a 5 mL HiTrap Protein A column (Cytiva). The column was washed by PBS (10 mM Na₂HPO₄ and 1.8 mM KH₂PO₄ pH 7.4, 137 mM NaCl, and 2.7 mM KCl). Proteins were eluted using a linear gradient with increasing concentration of a citrate buffer (0.1 M citric acid, pH 3.0). Eluted protein fractions were immediately neutralized with equal volumes of 1.0 M Tris-HCl pH 8.0 buffer. RBD-Fc and ACE2-Fc proteins were concentrated, buffer-exchanged into PBS before being frozen and stored at −80 °C.

Biolayer Interferometry (BLI) binding analysis

Binding of Rc-o319 S-proteins or RBDs to ACE2-Fc proteins was assessed using biolayer interferometry (BLI) on an Octet RED 96 instrument (FortéBio, USA). All steps were performed at 25 °C with an orbital shaking speed of 1000 rpm. All reagents were formulated in PBS-TB buffer (PBS with 0.02% v/v Tween-20 and 0.1% w/v BSA). Before the experiments, all Protein A biosensors were pre-equilibrated in the PBS-TB buffer for 10 min. ACE2-Fc proteins (11 μg/ml) were immobilized onto Protein A biosensors (FortéBio, USA) to a response value of ~1.2 nm. After a 60 s baseline step in PBS-TB, the biosensors immobilized with ACE2-Fc proteins were submerged in the analyte (S-proteins or RBDs) solutions for 300 s to measure association, before being submerged in PBS-TB for 600 s to measure dissociation. Data were reference subtracted and analyzed in the FortéBio Data Analysis Software (FortéBio) by fitting to 1:1 or 2:1 binding models to determine k_on, k_off, and K_D values as previously described [66]. Raw data and fits were plotted using GraphPad Prism 8 (GraphPad Software).

Binding of dimeric his-tagged hACE2, pangolin ACE2, and bACE2s to Rc-o319 RBD-Fc protein was assessed using the BLI assay. The RBD-Fc proteins (11 μg/ml) were loaded onto Protein A biosensors (FortéBio, USA) to a level of ~1.2 nm. After a 60 s baseline step in PBS-TB buffer, the biosensors immobilized with RBD-Fc proteins were submerged in the ACE2 protein solutions for 300 s to measure association, before being submerged in PBS-TB for 600 s to measure dissociation. Data were reference subtracted and analyzed in the FortéBio Data Analysis Software.

Cryo-EM grid preparation and data acquisition

Cryo-EM grids were prepared with Rc-o319 S-trimer samples at concentration of 1.65 mg/ml. All grids were prepared by applying 3.0 μl of S-trimer solutions directly to glow discharged (15 mA, 30 s) holey carbon grids (Quantifoil Cu mesh 300 R1.2/R1.3) before blotting. Cryo-EM grids were blotted for 2.5 s with a blot force of 4, before being plunge-frozen into liquid ethane using a Vitrobot Mark IV (ThermoFisher) at 4 °C and 100% humidity.

For Rc-o319 S-proteins, micrographs were collected using the EPU software (Thermo Fisher) on a Titan Krios G3i electron microscope (Thermo Fisher) operated at 300 keV. Micrographs were recorded at a nominal magnification of 130,000× on a Falcon 4 direct electron detector coupled to a Selectris-X energy filter set with a slit width of 10 eV. For Rc-o319 S-protein, data were collected with a raw pixel size of 0.93 Å on the image plane. Image stacks were obtained at a dose rate of ~7.1 electrons per Å² per second with a defocus ranging from -0.8 to -2.0 μm. A total of 6255 movies were collected, each with 7.05 s total exposure time, resulting in a total dose of 50 electrons per Ų.

For complex formation, wild-type dimeric Rc-o319 RBD was mixed with dimeric Fc-tagged bACE2_R.cor at a 1.2:1 molar ratio, followed by incubation at 4 °C in the presence of 0.01% glutaraldehyde for 20 min. The glutaraldehyde was quenched by adding Tris-HCl to a final concentration of 50 mM. Gel filtration was performed to remove excess RBD on a Superose 6 10/300 GL column (Cytiva) equilibrated in PBS buffer. Complex formation was confirmed by SDS-PAGE, and the Rc-o319-RBD:bACE2_R.cor complex peak was collected and concentrated to prepare cryo-EM grids. 3 μL of 6.8 mg/mL the RBD-ACE2 complex was mixed with 8 mM 3-[(3-Cholamidopropyl)dimethylammonio]-2-hydroxy-1-propanesulfonate (CHAPSO) before being applied onto glow discharged (15 mA, 30 s) Ni-Au-300 mesh R1.2/1.3 grids. Grids were blotted for 2.5 s with a blot force of 5 before being plunge frozen into liquid ethane using a Vitrobot Mark IV (Thermo Fisher Scientific) at 4 °C and 100% humidity. Micrographs were collected on a Titan Krios electron microscope (Thermo Fisher) operated at 300 keV using the EPU software (Thermo Fisher). Micrographs were recorded at a nominal magnification of 130,000× on a K3 direct electron detector (Gatan) coupled to a Gatan BioQuantum energy filter set with a slit width of 20 eV (Thermo Fisher). For the complex, data were collected with a raw pixel size of 0.65 Å on the image plane. Image stacks were obtained at a dose rate of ~18.7 electrons per Ųper second with a defocus ranging from -0.8 to -2.0 μm. A total of 33954 movies were collected, each with 2.67 s total exposure time, resulting in a total dose of 50 electrons per Ų.

Image processing

For dataset processed by the CryoSPARC package [67], particles were picked before 2D classification was performed. Well featured particles from the 2D classification were subjected to ab initio 3D classification. Homogeneous Refinement and Local Refinement were applied to the set of particles with good features identified by the ab initio 3D classification. For Rc-o319 S-trimer, ab initio 3D classification was performed with the resolution limit restricted to 10.0 Å to minimize overfitting and emphasize large-scale conformational differences. Eight initial classes were obtained from this procedure. Three well-resolved classes were subsequently subjected to independent non-uniform refinement. The resulting refined maps were then rigid-body aligned and compared using pairwise map-map correlation analysis to quantitatively assess structural similarity among the classes. Classes yielding similar structures were combined and subjected to final refinement, resulting in two Rc-o319 S-trimer structures adopting distinct conformations.

For the Rc-o319-RBD:bACE2_R.cor complex dataset, data processing was performed mainly using cryoSPARC v4.6.2 [67]. After patch-motion correction and CTF estimation, micrographs with CTF fit resolution above 6 Å were picked by the curate exposures job. Particles were then picked using a blob picker to generate templates for the subsequent template picker. Picked particles were extracted at 3x binning and subjected to 2D classification, ab initio reconstruction, and heterogeneous refinement. Particles belongs to classes with high-resolution details were selected for a final round of Topaz picker by the Topaz cross validation job. Picked particles were extracted at 3x binning and subjected to 2D classification, ab initio reconstruction, and heterogeneous refinement again. Highly similar classes in heterogeneous refinement showing high-resolution details were combined and re-extracted at 1.5x binning and subjected to non-uniform refinement with optimization of per-group CTF parameter analysis. The CTF-refined particles and volumes were subjected to reference-based motion correction and an extra round of non-uniform refinement to get the final dimeric structure.

Due to conformational dynamics in the dimeric RBD-ACE2 complex structure, several local regions were resolved to lower resolution. Local refinement was applied to improve the local resolution at the RBD-ACE2 binding interface. Particles were symmetry expanded, the dynamic parts were masked, before the particles were subtracted through masking. The subtracted particles were imported to perform NU-Refine and local refinement jobs.

Model building and structure refinement

A previous determined SARS-CoV-2 S-trimer structure (PDB ID: 6ZP2) [22] was used as the starting models for the Rc-o319 S-trimer structures. Model building and adjustment were performed manually in Coot 0.98 [68]. Steric clash and sidechain rotamer conformations were improved using the Namdinator web server [69]. Final structures after additional manual adjustments were refined and validated in PHENIX-1.18.261 [70]. The SARS-CoV-2-RBD:hACE2-B⁰AT1 complex structure (PDB ID: 6M17) [27,71] was used as the starting model for the Rc-o319 RBD:bACE2_R.cor complex structure. The starting model was fitted to the cryo-EM density in Chimera 1.18.0 before manual adjustments were carried out in Coot. The model was refined by multiple rounds of refinement in Namdinator and PHENIX. Data collection and refinement statistics for all the reported structures are provided in S2 Table.

GFP-split fusion assay

The split-GFP system based cell-cell fusion assay mediated by the interaction between the S-protein and the ACE2 receptor was previously described [37]. Briefly, HEK293T cells transfected with full length S-proteins and pQCXIP-GFP1–10 were prepared as effector cells, while HEK293T cells transfected with bat ACE2 (R.cor, Rs-WJ2, Rs1446, Rs1434, Rs1438, Rs3357, Rs3366, Rs3359, Rs5720, Ra9479, Ra5538, R.ferr, and M.luci) or pangolin ACE2 or human ACE2 (hACE2) and pQCXIP-BSR-GFP11 served as receptor cells. 24 h after transfection, effector and receptor cells were washed and resuspended in DMEM containing 10% FBS without trypsin treatment, mixed at a 1:1 ratio, and plated at ~2 × 10⁵ cells per well in a 96-well plate. Fluorescence images were recorded at the indicated time points using a Nikon TS2-S-SM fluorescence microscope. The GFP area was quantified with ImageJ. S-protein expression was analyzed via flow cytometry using a SARS-CoV-2 S2 antibody (1:2500 dilution, Sino Biological, 40590-T62) and western blot using an anti-FLAG antibody (1:1000 dilution, Sino Biological, 109143-MM13). The GFP-positive area was normalized to the mean fluorescence intensity of surface S-proteins, and resulting values at different time points were plotted to calculate the area under the curve (AUC) for quantitative analysis.

VSV-Pseudovirus entry assay

VSV pseudotypes were generated according to a published protocol [72]. Briefly, HEK293T cells were transfected with 50 μg Rc-o319 WT or variant S plasmid into a 15 cm cell culture dish. After 24h, the cells were washed twice and inoculated with VSV*ΔG-Luc at a multiplicity of infection of 0.1 for 1h. After the inoculum was removed, the cells were washed 5 times using PBS with 2% FPS and further cultured with a DMEM culture medium for 24 h. The supernatant was harvested and centrifuged at 3000 rpm to remove cellular debris, filtered through a 0.45 μm syringe filter, 10-fold concentrated and stored at –80 °C in small aliquots.

To quantify entry of pseudovirons into cells, HEK293T cells stably co-expressing TMPRSS2 and ACE2 (2 × 10⁵ cells in 100 μL DMEM per well) were seeded in 96-well plates and inoculated with an equal volumes of pseudotype particles. 24 h post-transduction, culture supernatants were aspirated and cells lysed in 1x cell culture lysis reagent (Promega) for 20 min at room temperature. The lysates were then transferred to white 96-well plates and luciferase activity was measured using a luciferase assay system (Promega).

Western blot

To analyze S proteins and ACE2 expression in cells, HEK293T cells were transfected with expression vectors and incubated for an additional 24h. Then, cells were washed with ice-cold PBS, and soluble proteins were extracted with cell lysis buffer (100 mM Tris-HCl pH 8.0, 150 mM NaCl, 1% NP-40) containing 1 mM PMSF (Beyotime, ST505) and 2.5 U/mL BeyoZonase Super Nuclease (Beyotime, D7121). Non-reducing SDS-PAGE loading buffer (4×) was added to each sample immediately before boiling at 98 °C for 10 min and subjected to SDS-PAGE and immunoblotting.

To analyze S-protein incorporation into pseudotyped particles, 1 mL of  each VSV pseudotype preparation was subjected to ultracentrifugation (25 000 × g, 120 min, 4 °C). The supernatant (1 mL) was carefully removed. The remaining pellet was resuspended in 20 μL of 5x SDS-sample buffer, heated at 98 °C for 10 min and then analyzed by SDS-PAGE and immunoblotting.

Equal amounts of protein samples were separated by 4–10% SDS-PAGE and transferred to PVDF membranes (Millipore). After blocking with 5% (w/v) skimmed milk in TBST buffer, the membranes were incubated with primary antibodies, such as rabbit anti-SARS-CoV-2 S2 polyclonal antibody (1:2,500 dilution, Sino Biological, 40590-T62), mouse monoclonal antibody targeting FLAG tag (1:2,500 dilution, TransGen Biotech, HT201), beta-actin (1:5,000 dilution, TransGen Biotech, HC201) and mouse anti-VSV matrix protein (Kerafast, EB0011, 1:2,500). Horseradish peroxidase-conjugated (HRP) goat anti-mouse IgG antibody (1:10,000 dilution, Beyotime, A0216) and goat anti-rabbit IgG antibody (1:10,000 dilution, Beyotime, A0208) were used as secondary antibodies. The quantitative results of VSV-M were analyzed using ImageJ software.

Flow cytometry

To validate surface expression of S protein and its binding affinity for ACE2 following transfection, we performed flow cytometry using a BD Accuri C6 Plus flow cytometer, and the data were analyzed with FlowJo software. HEK293T cells transfected with S-protein expression plasmids were harvested 24 h post-transfection by centrifugation (3,000 rpm, 5 min, 4 °C), resuspended in ice-cold PBS containing 1% BSA, and incubated on ice for 1 h with rabbit anti-SARS-CoV-2 S2 polyclonal antibody (Sino Biological, Cat. #40590-T62, 1:500) and/or recombinant human ACE2-Fc fusion protein (10 μg/mL). Cells were then washed twice with PBS and incubated on ice for 45 min with Alexa Fluor 488-conjugated goat anti-rabbit IgG (H + L) secondary antibody (Beyotime, A0423, 1:500) and/or APC-conjugated anti-human IgG Fc antibody (BioLegend, 410711, 1:1000).

Statistical analysis

All statistical analyses were performed using GraphPad Prism software (Graphpad Version 10.6.1). The significance of differences between the two groups was determined with an unpaired, two-tailed Student’s t-test. For all analyses, only a probability (P) value of 0.05 or lower were considered statistically significant (P > 0.05 [n.s, not significant], P≤ 0.05 [*], P ≤ 0.01 [**], P ≤ 0.001 [***], P≤ 0.0001 [****]).