Essential Envelope Spike Motifs for Cell Entry of Transmissible Gastroenteritis Virus and Its Evolution in Coronavirus

Abstract

Background: Transmissible gastroenteritis virus (TGEV), a coronavirus (CoV) infecting pigs, uses its spike (S) glycoprotein to bind porcine aminopeptidase N (pAPN) for cell entry. Although structural studies have identified receptor-binding motifs (RBMs) within the receptor-binding domain (RBD) of the S protein, the functional relevance of individual residues for TGEV receptor recognition, cell entry, and infection remain unclear. Methods: In this study, we performed structure-guided mutagenesis of the TGEV RBD to evaluate the contribution of specific residues to receptor binding and viral infectivity. Results: Using soluble RBD proteins, we found that most of the RBD residues within the pAPN-binding interface contribute to the binding interaction. Nonetheless, TGEV reverse genetics experiments revealed that just three RBD residues (Gly527, Tyr528, and Trp571) were indispensable for viral cell entry. Mutations at these positions, which are conserved among group 1 alpha-CoVs abolished infectivity, highlighting their central role in the virus–receptor interface. Conclusions: Our findings provide a detailed functional map of the TGEV RBD and offer insights into the evolution of receptor recognition across CoV.

1. Introduction

Coronaviruses (CoVs) are a large group of single-stranded, positive-sense RNA viruses that infect animals. They are classified into four genera: alpha-, beta-, gamma-, and delta-CoVs [1,2]. Transmissible gastroenteritis virus (TGEV), a representative member of the alpha-CoV genus, infects pigs and exhibits both enteric and respiratory tropism [3,4]. The determinants of CoV tropism reside in the envelope spike (S) glycoprotein, a type I membrane protein anchored to the viral envelope [1]. The N-terminal S1 subunit (600–800 residues) of the S protein mediates attachment to host cell surface molecules, initiating viral entry. Receptor-binding domains locate at the N or C-terminal portion of the S1, and certain alpha-CoV bear an additional N-terminal domain (termed D0) that recognizes host cell-surface carbohydrates [5]. In TGEV, D0 is responsible for its enteric tropism [3,6], whereas in feline CoV-23 (FCoV23) modulates the viral fusogenic activity mediated by the S2 subunit of the S protein [7].

The canonical receptor-binding domain (RBD or CTD) of TGEV and other CoVs is located in the C-terminal portion of the S1 subunit [8,9,10]. In alpha-CoV, the RBD adopts a β-barrel fold, with the receptor-binding motif (RBM) formed by loops at the tip of the barrel, distant from the domain’s N- and C-terminal connections to the S protein [10,11,12]. This β-barrel fold is conserved in delta-CoVs and maintains a remnant β-barrel topology in gamma-CoVs [13,14], but diverges in beta-CoVs [15,16], where the RBM forms a distinct subdomain. Structural studies of CoV S proteins have revealed their domain architecture, showing that the RBD is positioned at the most distal region from the viral membrane [6,17,18,19]. The RBD exhibits dynamic conformations, alternating between receptor-binding active and inactive states, with the RBM either exposed or hidden [6,19,20].

For cell entry, CoV RBDs recognize ectoenzymes such as aminopeptidase N (APN), angiotensin-converting enzyme 2 (ACE2), or dipeptidyl peptidase 4 (DPP4) [21,22,23]. In contrast, CoV N-terminal domains (NTDs) do not bind ectoenzymes [24]. Most alpha-CoVs, including TGEV [25], human HCoV-229E [26], canine CoV–human pneumonia (CCoV-HuPn-2018) [5], FCoV23 [7], and porcine epidemic diarrhea virus (PEDV) [27] recognize and use APN for cell entry. APN is a zinc-dependent metallopeptidase involved in diverse biological processes, such as blood pressure regulation and angiogenesis, and is overexpressed in certain tumors [28]. The APN ectodomain comprises four domains (DI-DIV) and forms dimers on the cell surface [10,29]. It can adopt multiple conformations representing different functional states [30]. TGEV specifically recognizes the open, catalytically inactive form of APN. Inhibiting the transition from the closed to the open APN conformation blocks TGEV infection [30].

The crystal structure of pAPN in complex with the RBD of porcine respiratory CoV (PRCV), which shares 97% sequence identity with the TGEV RBD, revealed the mechanism by which porcine and related CoVs recognize APN [10]. The PRCV/TGEV binding footprint on APN is similar to that of other group 1 alpha-CoVs (alpha1-CoVs) [5,7], but it differs from that of porcine delta-CoV (PdCoV) and HCoV-229E [12,14]. The TGEV and alpha1-CoV RBDs adopt a β-barrel fold with an APN-binding edge characterized by two protruding loops containing exposed aromatic residues that insert into small cavities on APN. One loop features a solvent-exposed tyrosine that fits between an APN α-helix and an N-linked glycan on domain DIV, while the other loop contains a tryptophan that docks into a cavity formed by DII and DIV [10]. The size of this cavity in the open APN conformation determines TGEV’s specific binding [30]. Neutralizing antibodies target the APN-binding motifs in the TGEV RBD, thereby preventing infection [10]. It has been proposed that immune pressure on alpha-CoV receptor-binding motifs has shaped their conformation, leading to shifts in receptor recognition among alpha-CoVs [10,12].

Although previous structural studies have identified motifs critical for APN recognition by TGEV and other alpha-CoV, the role of individual residues in viral cell entry remained undefined. In this study, we performed systematic amino acid substitutions in the APN-binding region of TGEV and assessed their impact on receptor binding and viral infectivity using soluble proteins and recombinant viruses, respectively. We identified three key viral residues at the virus–receptor interface that are conserved in alpha1-CoV and discuss their evolutionary significance across CoVs.

2. Materials and Methods

2.1. RBD Binding to pAPN and Determination of the RBD Mutants’ Binding Activity

Flow cytometry was used to characterize the interaction of the TGEV RBD with the pAPN expressed on the surface of BHK-21 cells (BHK-pAPN), as previously described [9]. An RBD-Fc fusion protein (CoSA3-Fc) comprising the residues 505 to 653 of the matured TGEV S protein (virus strains SC11, GenBank accession number AJ271965) and containing an HA peptide at the N-terminus of the RBD and the IgG1 Fc region at the C-terminus was engineered [9]. RBD-Fc mutants were generated by site-directed mutagenesis following the quick-change protocol (Stratagene, La Jolla, CA, USA) using PfuI Turbo polymerase (Agilent Technologies, Santa Clara, CA, USA) and DpnI (NEB, Ipswich, MA, USA), and the mutations were verified by sequencing. HEK293T cells cultured in DMEM supplemented with 10% FCS were transiently transfected to produce RBD-Fc proteins secreted into the cell supernatants. The RBD-Fc proteins concentrations were quantified by sandwich ELISA using antibodies against both the HA tag and the Fc region, with an Fc-fusion protein of known concentration as standard.

A range of RBD-Fc protein concentrations (30 to 0.1 μg/mL) was tested for binding to pAPN on BHK-pAPN cells by flow cytometry. Briefly, cells were washed with PBS, detached with 0.02% EDTA solution, sedimented at 1400 rpm, resuspended in binding buffer (1% BSA in PBS), counted and plated into conical wells of a 96-well plate at a density of 5 × 10⁴ cells/well. Binding was performed at 4 °C. The RBD-Fc proteins and a control Fc-fusion protein were added in 20 µL of buffer for binding to pAPN for 30 min, followed by three washes with the binding buffer. Subsequently, cells were stained with anti-human IgG-FITC antibody (Invitrogen, Carlsbad, CA, USA) and incubated on ice for 30 min, followed by 3 washes before analysis by flow cytometry. The percentage of stained cells was determined for each RBD-Fc protein and corrected by the background staining with a control Fc-fusion protein. Specific binding was plotted, and the effective concentration at 50% binding (BC₅₀) was determined using the GraphPad 5.0b software.

2.2. Generation of TGEV Mutants, Virus Rescue, and Titration

Recombinant TGEV (PUR46MAD) mutants carrying substitutions in the RBD region of the S protein were generated following established BAC-based reverse genetics methods [31,32]. First, a recombinant cloning vector (pGEM-T) containing a copy of the full-length cDNA of the TGEV S gene was used as a template to introduce mutations. Mutations were confirmed by DNA sequencing. Then, using the PacI and MluI restriction enzymes, the mutated S fragment (4.514 kb) was cloned into an intermediate recombinant BAC plasmid to generate the pBAC-TGEV^∆ClaI construct. This BAC plasmid carried an engineered cDNA copy of the full-length TGEV genome except for a ClaI fragment that is toxic to bacterial cultures [31]. This ClaI fragment was maintained separately in another BAC construct (pBAC-TGEV^ClaI) and re-introduced in the intermediate BAC plasmid at the final cloning step, yielding the full-length infectious BAC construct, pBAC-TGEV^FL. Mutations were confirmed by sequencing of the resultant BAC plasmids, and restriction digestion (EcoRI-XhoI) was used to verify correct orientation and plasmid stability. Because BACs are maintained at low copy number (1–2 per cell), plasmid DNA was prepared using Qiagen (Venlo, The Netherlands) midiprep kits with BAC-specific recommendations. For BAC and insert DNA purification, agarose gel extraction was performed with the QIAEX II Gel Extraction Kit (Qiagen, Venlo, The Netherlands). This process was repeated for each mutant. To generate recombinant viruses, highly pure preparations of pBAC-TGEV^FL DNA (wild-type and mutants) were transfected into BHK-pAPN cell monolayers using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA), following established protocols [32]. After 6 h, transfected cells were briefly trypsinized and layered onto fresh swine testis (ST) cell monolayers for virus rescue. Cytopathic effect (CPE) was observed at ~72 h post-infection, at which point cells and supernatants were harvested. Viral RNA rescued from ST cells was sequenced to confirm the presence of engineered mutations. Virus titers were determined by plaque-forming assays in ST cell monolayers. In brief, for titration, ST cells were seeded in 24-well plates and inoculated with serial dilutions of supernatants in DMEM supplemented with 2% FCS. After 90 min absorption, monolayers were washed with media and overlaid with semisolid medium (0.7% agar in DMEM). Plates were incubated for 2 days, followed by crystal violet staining for plaque visualization and counting. All infectivity assays were performed using virus stocks after a single amplification passage in ST cells (P1). Infectious units were quantified for wild-type and mutant viruses.

2.3. Structure Analysis and Representation

Virus–receptor binding interfaces were determined with the PISA server (https://www.ebi.ac.uk/msd-srv/prot_int/cgi-bin/piserver, accessed on 23 October 2025). Figures were prepared with pymol (https://www.pymol.org/, accessed on 23 October 2025).

3. Results

3.1. Identification of TGEV RBD Residues Important for Binding to Cell Surface pAPN

The crystal structure of the PRCV RBD-pAPN complex revealed the receptor-binding region and the viral residues that interact with the pAPN protein [10]. This region includes two protruding loops at the RBD tip containing aromatic residues, which function as receptor-binding motifs (RBMs) (Figure 1A). The loop connecting the β1 and β2 strands (β1–β2) defines RBM1 (magenta in Figure 1A), which contains a conserved tyrosine residue among alpha1-CoV (Tyr528 in TGEV), located at a β-turn in the RBD tip (Figure 1A,B). In the RBD-pAPN interface, the side chain of Tyr528 is buried between an α-helix and an N-linked glycan in DIV of the pAPN ectodomain [10]. This RBD residue, along with others in RBM1, mediates a network of hydrophilic and hydrophobic interactions with the pAPN DIV.

RBM2 also contains a β-turn located at the beginning of the β3-β4 loop (Figure 1A,B), and it includes a tryptophan residue (Trp571 in TGEV) (red in Figure 1A), which is conserved in other alpha1-CoVs (Figure 1B). Trp571 docks into a cavity formed by DII and DIV in the open pAPN conformation [10,30]. A third loop connecting β5 and β6 strands (RBM3, orange in Figure 1A), located near RBM1, also contacts the receptor, although its contribution to the binding interface is less significant [10].

Consistent with the structural data, mutation of Tyr528 or Trp571 in TGEV abolishes RBD binding to pAPN [10]. To assess the contribution of additional RBD residues buried at the virus–receptor interface to pAPN recognition, we performed structure-guided mutagenesis of the RBMs in the TGEV RBD (Figure 1C). We analyzed RBD residues Arg525 to Ile532 in RBM1 and Leu570 to Trp571 in RBM2, as well as several residues in RBM3. Wild-type and mutant RBD-Fc fusion proteins were expressed, and their binding to BHK cells expressing pAPN was assessed by flow cytometry (see Materials and Methods). Binding was quantified as the percentage of cells stained by RBD-Fc proteins across a range of concentrations (Supplementary Figure S1), and the binding activity of RBD mutants was expressed as BC₅₀ values (Figure 1C). In RBM1, all tested mutants exhibited very low binding activity compared to the wild-type (WT) RBD (Figure 1C). In RBM2, Trp571 proved more critical than Leu570 (Figure 1C). These results highlighted the essential role of the RBM1 and Trp571 in binding of TGEV to pAPN. In line with the lower contribution of RBM3 observed in the structural data, β5–β6 loop mutants retained pAPN binding, although N632A and Q634A substitutions reduced the RBD binding activity (Figure 1C). Because the RBMs are conserved in alpha1-CoV (see Figure 1B), these findings can likely be extrapolated to other CoVs in this group.

3.2. Identification of TGEV RBD Residues Essential for Cell Entry and Infection

After confirming the role of the RBMs identified by the crystal structure of the RBD-pAPN complex in pAPN recognition, we examined their contribution to TGEV entry and infection. Recombinant TGEV mutants carrying single-residue substitutions in the RBMs were generated using a BAC-based reverse genetics system [31] and rescued in BHK-pAPN cells, as illustrated in Figure 2A and detailed in Materials and Methods. Infectivity of the TGEV mutants was quantified by plaque assays in ST cells (Figure 2B).

Among the TGEV mutants tested, only three exhibited no or negligible infectivity. These included two mutations in the RBM1 (G527D or Y528A) and one in the RBM2 (W571A) (Figure 2B). Tyr528 was identified as the most critical APN-binding residue, as its substitution completely prevented TGEV cell infection. Similarly, the G527D mutation, which introduces a side chain adjacent to Tyr528, was deleterious, indicating that Gly527 is essential for proper RBM1 docking into pAPN DIV. In contrast, the G529D mutation, located on the distal side of the β-turn (Figure 1B), retained infectivity (Figure 2B). This observation is consistent with the previously reported TGEV G529D mutant that escapes antibody neutralization [34,35]. The other RBM1mutations, Q530A and I532A, reduced the TGEV infectivity by 1 and 2 logs (Figure 2B), respectively. This is consistent with their involvement in pAPN recognition (Figure 1B), but they did not impair viral infection.

In RBM2, the W571A substitution reduced TGEV infectivity by approximately 7 logs (Figure 2B), indicating that this second aromatic residue at the RBD tip (Figure 1) is also essential for TGEV cell entry and infection. In contrast, consistent with their low contribution to pAPN recognition (Figure 1C), substitutions L570A in RBM2, and N632A, Q634A, and the double mutant N632Q634AA in RBM3 did not substantially affect TGEV infectivity (Figure 2B).

Together, these results identify three critical residues, Gly527, Tyr528, and Trp571, as indispensable for TGEV cell entry and infection, whereas other residues at its receptor-binding interface appear to modulate the binding affinity and infectivity without being individually essential.

3.3. Evolution of CoV RBD Structure and RBMs

To place our findings in an evolutionary context, we next compared RBD structures across CoVs from the four genera (alpha-, delta-, gamma-, and beta-CoV), highlighting how structural differences at the RBM relate to receptor specificity and evolutionary divergence. The sequence and conformation of the RBD varies considerably among CoV, most likely because of the immune pressure on this main antigenic site [10,12]. In alpha-CoV, the RBD adopts a β-barrel fold with two highly twisted β-sheets, three disulfide bonds, and N-linked glycans on one side of the β-barrel (Figure 1 and Figure 3A). While alpha- and delta-CoVs preserve a β-barrel fold, gamma- and especially beta-CoVs diverge markedly (Figure 3A). These structural differences translate into distinct receptor recognition strategies across genera.

Alpha1-CoV (e.g., TGEV, PRCV, CCoV-HuPn-2018, FCoV23, and CCoV) all display two protruding loops in RBM1 and RBM2, each bearing exposed Tyr and Trp residues (Figure 3A) [7]. The Tyr residue in RBM1 and the preceding Gly (Gly527), which form a β-turn in the RBD of alpha1-CoV (Figure 1B), are indispensable for TGEV infection, whereas Gly529 tolerates substitutions without abolishing infection. The Trp in RBM2 lies in a β-turn at the start of a long loop connecting β-strands 3 and 4, disulfide-bridged to strand 3 (Figure 1B), restricting flexibility. These residues fit into small APN cavities: The Tyr side chain stacks between an α-helix and an N-linked glycan attached to DIV of pAPN, while the Trp inserts between DII and DIV [5,7,10]. Despite its conserved overall architecture, the RBD tip adopts distinct conformations among other alpha-CoVs that also use APN, such as PEDV and HCoV-229E (Figure 3B), particularly in RBM1 and RBM2 (Figure 3C). PEDV, as the alpha1-CoV, contains two prominent RBM1 and RBM2 with a Leu and a Tyr, respectively (Figure 3B,C), supporting APN recognition [27], although the precise receptor-binding interface remains undefined. In HCoV-229E, the binding region preserves a long loop in RBM1 with a Pro at the tip (substituted by Val or Gly in some variants [12]), which fits into an APN cavity containing an N-linked glycan [12]. Nonetheless, RBM2 and RBM3 form a flatter receptor-binding surface in HCoV-229E that interacts with a distinct APN site at low affinity (buried surface ~525 Ų). HCoV-NL63 diverges further: Its receptor-binding region lacks prominent RBM features and instead forms a concave surface (Figure 3B), enabling recognition of ACE2 rather than APN [36]. Aromatic residues are not exposed but cluster centrally, contacting the N-terminal region of ACE2 [11]. These observations suggest that APN specificity in alpha-CoVs depends on hydrophobic residues protruding from the β-barrel tip, which we demonstrated are essential for cell entry.

PdCoV RBD retains the β-barrel fold and presents two prominent loops with aromatic residues at the receptor-binding moiety, resembling alpha1-CoV (Figure 3A). Like alpha1-CoV, PdCoV uses APN as its receptor, but engages a distinct binding site [14]. The Phe residue in RBM1, which also contacts a helix and a glycan in pAPN, and the Trp residue in RBM3 both fit into small cavities on the receptor surface. Thus, PdCoV exhibits similar receptor-binding features and modes to alpha1-CoVs. However, PdCoV’s receptor-binding surface is substantially larger (~925 Ų), covering most of the glycan-free side of the β-barrel, compared to ~700 Ų in TGEV or PRCV, which is restricted to the RBD tip (Figure 3A). This expanded interface may reflect evolutionary adaptation to broaden receptor engagement.

In gamma-CoVs such as infectious bronchitis virus (IBV), the RBD adopts an irregular β-barrel fold with short β-strands and long loops (Figure 3A) [13]. This architecture diverges from the β-barrel fold preserved in alpha- and delta-CoVs, reflecting a distinct receptor recognition mode and underscoring evolutionary divergence within CoV genera.

Beta-CoVs, including Middle East respiratory syndrome CoV (MERS-CoV), severe acute respiratory syndrome CoV (SARS-CoV), and SARS-CoV-2, have lost the β-barrel fold entirely. Their RBD consists of two subdomains: a core with five-stranded β-sheet and an external RBM subdomain connected to two adjacent β-strands of the core (Figure 3A) [15,16]. The RBM is structurally variable among beta-CoVs, accommodating multiple mutations that alter receptor affinity and enable antibody escape [37]. This variability is exemplified by the SARS-CoV-2 Omicron variant [38,39].

To date, no single indispensable residues for SARS-CoV-2 entry have been identified, unlike the essential Gly527, Tyr528, and Trp571, we report for TGEV. This suggests that beta-CoV receptor recognition relies on multiple distributed interactions rather than in a few critical contacts, enabling efficient immune evasion. Although the β-barrel fold is preserved in alpha- and delta-CoV, divergences in RBMs (Figure 3) result in marked differences in receptor recognition. Alpha1-CoVs depend on a conserved architecture centered at two aromatic residues for APN binding. Delta-CoVs share this strategy but with expanded interfaces. Gamma-CoVs diverged structurally, while beta-CoVs evolved a distinct two-subdomain RBD fold, relying on distributed interactions for receptor engagement and immune escape. These genus specific differences highlight how immune pressure and receptor availability have shaped CoV RBD evolution and specificity.

4. Discussion

In this study, we leveraged a reverse genetics system to identify key residues within the RBD of TGEV that are critical for cell entry and infection. Whereas binding studies reported here and elsewhere have shown that numerous amino acids at the TGEV RBD-APN interface contribute to receptor recognition, our findings pinpoint just three essential RBD residues (Gly527, Tyr528, and Trp571) that are indispensable for viral entry. Together, they form a conserved functional core in the RBM of TGEV and alpha1-CoV, and maintaining its integrity is required for infection. Structural variations within this framework lead to differences in receptor recognition modes among alpha-CoVs, as well as with delta-CoV, which share a similar RBD fold.

Our results confirm and extend previous structural analyses of TGEV and other alpha-CoVs in complex with receptors, which highlighted the importance of prominent loops at the RBD tip in APN recognition [5,7,10]. APN-binding CoVs bear at least one motif that projects onto the receptor-binding surfaces (Figure 3), and the loss of this feature results in a receptor-binding switch. In TGEV, two receptor-binding motifs (RBM1 and RBM2) within the RBD contain aromatic residues that are solvent-exposed in the unbound state but become largely buried (~80% of their surface area) upon APN receptor engagement; this suggests a substantial contribution to the binding energy. Furthermore, the high shape complementary between these viral motifs and the pAPN, together with the polar interactions involving Tyr528 and Trp571, demonstrate their critical role in receptor-binding specificity [10]. The complete loss of infectivity following substitution of these residues underscores their central role in establishing the virus–receptor interface. Notably, the adjacent Gly527 residue appears to be structurally critical for maintaining the β-turn conformation that positions Tyr528 for optimal interaction. This is supported by the deleterious effect of the G527D mutation, which likely introduces steric hindrance or disrupts local backbone flexibility.

Interestingly, other residues within the RBMs that contribute to pAPN binding, such as Gln530, Ile532, and Leu570, exhibited reduced but not abolished infectivity when mutated. This suggests that while these residues enhance binding affinity, they are not individually essential for viral entry. The reduced binding of the mutant is likely compensated by the multivalent attachment of the viral particle to APN on the cell surface (avidity effect). Gln530 forms a network of polar interactions with an pAPN-linked glycan, potentially contributing to the binding specificity rather than to the binding energy [10]. Ile532 and Leu570 stablish van de Waals interaction with pAPN; however, binding assays and infectivity for I532A and L570A mutants (Figure 1 and Figure 2) indicated that Ile532 plays a greater role in the binding energy, likely because it becomes almost fully buried at the interface, whereas L570A is more solvent-exposed. Nonetheless, only substitutions at core anchoring residues (Gly527, Tyr528, Trp571) disrupted essential contacts required for productive receptor engagement and spike activation, leading to a disproportionate collapse in infectivity. This distinction between indispensable anchoring contacts and affinity-modulating residues clarifies the observed phenotypic differences and highlights the hierarchical contribution of individual residues to receptor recognition and viral entry. Such a modular architecture, where a few residues are indispensable while others modulate binding strength, may represent an evolutionary balance between efficient receptor engagement and immune evasion. The presence of escape TGEV mutants at position Gly529 that retain infectivity [34,35] and receptor-binding loop diversity in HCoV-229E [12] further support the idea that certain RBM residues are more tolerant to variation, potentially facilitating antigenic drift.

From an evolutionary perspective, the conservation of Tyr and Trp residues exposed at the RBD tip among alpha1-CoVs that use APN [7] (e.g., TGEV, CCoV-HPn-2018, FCoV23) suggested a shared receptor-binding strategy. Structural comparisons reveal that these residues are consistently positioned within β-turns at the apex of the β-barrel fold, enabling specific insertion into APN cavities. In contrast, other alpha-CoV such as HCoV-229E and HCoV-NL63, which use APN or ACE2, respectively, exhibit divergent RBM conformations and receptor engagement modes. HCoV-229E interacts with a distinct site on APN using a flatter RBD surface but retains APN-binding specificity, likely through the prominent RBM1 (Figure 3A), which inserts into a glycan-containing cavity similar to RBM1 in alpha1-CoVs. Glycan flexibility likely enables accommodation of variable RBMs, such as those described for HCoV-229E [12], emphasizing its essential role in modulating CoV–receptor interactions [21].

HCoV-NL63 lacks prominent features at the RBD tip and, instead of APN, engages ACE2 via a concave surface. These differences highlight the plasticity of CoV RBDs and suggest that shifts in receptor usage are accompanied by structural remodeling of the RBMs. The evolutionary divergence of RBD architecture is even more pronounced in beta-CoVs, where the RBD lost the β-barrel fold and contains a structurally distinct RBM subdomain. Unlike alpha-CoVs, beta-CoVs such as SARS-CoV and SARS-CoV-2 rely on a broader interface involving multiple contact points with ACE2, which may confer greater mutational flexibility [40]. This is exemplified by the emergence of SARS-CoV-2 variants with altered RBMs that escape neutralizing antibodies while retaining receptor affinity [38,41]. In contrast, the reliance of TGEV and alpha1-CoV on a few critical residues for receptor binding may impose evolutionary constraints, reducing their ability to escape immune pressure without compromising infectivity. Notably, anti-TGEV antibodies (6A.C3) targeting RBM1 and RBM2 [10] effectively prevent the appearance of neutralization-resistant mutants [34,42].

Our findings also have important implications for vaccine and therapeutic development. Identifying essential RBM residues provides strategic targets for neutralizing antibodies or small molecules designed to block virus-binding cavities in the APN and prevent receptor engagement. Given the structural conservation of the RBD among alpha1-CoVs, such interventions may offer cross-protective potential. Moreover, functional mapping of the TGEV RBD can inform surveillance efforts by highlighting mutations likely to affect viral fitness or host range, thereby guiding early detection of variants with epidemiological significance.

5. Conclusions

In conclusion, this study delineates the minimal set of residues required for TGEV and alpha1-CoV cell entry and uncovers evolutionary principles shaping CoV–receptor recognition. The strict reliance on a few key contacts for cell entry receptor engagement contrasts sharply with the distributed interaction networks observed in beta-CoVs, revealing lineage-specific strategies for host adaptation and immune evasion. These insights not only clarify the structural constraints influencing cross-species transmission in alpha-CoVs, but also provide a rational basis for vaccine design by pinpointing conserved, functionally indispensable motifs that represent promising targets for broadly neutralizing antibodies.