Structure of the ACE2-RBD Complex
Previous studies showed that the extracellular Peptidase Domain (PD) of ACE2, which adopts a claw-like morphology, is the interaction site of SARS-CoV RBD (Towler et al., 2004; Li, 2015; Kirchdoerfer et al., 2018; Song et al., 2018). More specifically, ACE2 engages SARS-CoV RBD by establishing contacts with the 424–494 residue domain, which is referred to as Receptor Binding Motif (RBM) (Li et al., 2005).
In spite of the recent emergence of SARS-CoV-2, multiple authors have already highlighted the structure of ACE2/SARS-CoV-2 RBD complex by either X-ray (Lan et al., 2020; Shang et al., 2020b; Wang et al., 2020) or cryo-EM (Walls et al., 2020; Wrapp et al., 2020; Yan et al., 2020). Pleasantly, all data converged to a consistent tridimensional arrangement of the receptor (Wang et al., 2020). In the “closed” conformation of S protein, the RBD is buried at the interface between protomers (Walls et al., 2020). Only in the “open” S conformation, RBD engages PD of ACE2 (Wrapp et al., 2020), and the complex may involve a dimeric ACE2 that accommodates two S protein trimers (Yan et al., 2020). In keeping with their sequence similarity, strong structural homology was found between ACE2/SARS-CoV RBD and ACE2/SARS-CoV-2 RBD (Lan et al., 2020; Shang et al., 2020b; Wang et al., 2020). SARS-CoV-2 RBM spans from residue 438–506 of S sequence and, likewise SARS-CoV RBM, it approaches the outer surface of ACE2 by a gently concave surface with a ridge on one side (Figure 8). The concave surface is made up by the two short β5 and β6 sheets of the external RBD subdomain, whereas the ridge contains the β5/β6 loop (loop 1: residues 474–489). A second smaller loop (loop 2: residues 498–505) is visible on the other side of the concave surface. Inspection of the complex structure and molecular dynamics (MD) highlighted that the motifs 453–456 (in β5), 484–489 (in the loop 1), and 500–505 (in the loop 2) are at the basis of the largest differences between SARS-CoV-2 and SARS-CoV RBM interactions with ACE2. SARS-CoV RBM ridge contains a Pro-Pro-Ala motif that is replaced by Gly-Val-Glu-Gly in SARS-CoV-2 (residues 482–485), yielding a more compact loop able to engage more interactions with proximal ACE2 residues (e.g., Ser19 and Gln24) (Shang et al., 2020b). Additionally, Phe486 of SARS-CoV-2 RBM (which replaces Ile of SARS) inserts into a hydrophobic pocket on the receptor surface, establishing strong aromatic interactions with Tyr83 of ACE2 (Wang et al., 2020). Asn501 in loop 2 further engages recognized hotspots on the ACE2 surface (Shang et al., 2020b). Consistently, MD studies confirmed that loop 1 and 2 are much more rigid in RBM-ACE2 complex of SARS-CoV-2 with respect to SARS (Brielle et al., 2020). These subtle structural differences probably account for the higher affinity of SARS-CoV-2 for ACE2 (Wang et al., 2020). Interestingly, MD simulations suggest that the difference in affinity is largely due to the solvation energy, emphasizing the relevant role of hydrophobic patches in RBM/ACE2 binding surface (He et al., 2020).
It is worth noting that the RBD-receptor engagement is the crucial effector of viral-host interaction, which eventually determines viral host range, and in tandem with the host proteases is responsible for virus tropism and pathogenicity (Millet and Whittaker, 2015). Structure-guided sequence analysis has suggested that several mammals, including pets such as cats and dogs, host ACE2 receptors that could bind effectively to SARS-CoV-2 S protein and propagate COVID-19 infection (Luan et al., 2020). Yet, no correlation between genetic distance and the S/ACE2 interaction was found.
In this context, a third human CoV, hCoV-NL63, has been previously found to use ACE2 for cell entry (Hofmann et al., 2005), although its S1 sequence is rather dissimilar from SARS (23.4%) and SARS-CoV-2 (29.2%). In spite of this, the structures of hCoV-NL63 (Wu et al., 2009) and SARS-CoV RBD (Li et al., 2005) were found to engage some sterically overlapping sites in ACE (Li, 2015). This homology can be transitively extended to SARS-CoV-2, suggesting these three CoVs have evolved to recognize a “hotspot” region in ACE2 for receptor binding. This might represent a critical feature for the appearance of novel CoV able to infect humans in the future, as it is known that at least three more bat CoVs bind ACE2 (Hoffmann et al., 2020b). Indeed, bat RaTG13 CoV binds ACE2 and contains a similar four-residue motif in the ACE2 binding ridge of RBM, suggesting that SARS-CoV-2 could have evolved from RaTG13 or a yet-unknown related bat CoV (Shang et al., 2020b).