Long-range allosteric influences on SARS-COV-2-ACE2 stability and detachment

By Dr. Liji Thomas, MDDec 3 2020

The COVID-19 pandemic is caused by SARS-CoV-2, a novel beta-coronavirus which has numerous spike proteins on its membrane. These are responsible for its engagement with the angiotensin-converting enzyme 2 (ACE2), the host cell receptor.

Image Credit: https://www.biorxiv.org/content/10.1101/2020.11.30.405340v1.full.pdf

*Important notice: bioRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.

The spike tip exists in both ‘up’ and ‘down’ conformations. In the former, the homotrimeric spike protein exposes the receptor-binding domain (RBD) so that it can attach to the host. The structures of the spike protein, RBD, and ACE2 have all been elucidated.

The researchers used deep mutagenesis scanning, a method to examine each residue at the ACE2-RBD interface and find out what would happen to the level of expression of, and the binding of, the RBD if another residue is substituted. The study was published in the preprint server bioRxiv* in December 2020.

Allosteric Effects on Receptor-Virus Binding

Changing the amino acid sequence of the RBD interface could alter the binding affinity of the domain to the receptor via the resulting disruption of local interactions, such as an electrostatic bond, or because of long-range effects that may cause large structural changes. The latter seems to be involved in the transition of the spike protein from ‘up’ to ‘down’ and vice versa. The RBD-ACE2 binding could also be allosteric in nature.

These allosteric interactions may involve multiple linked residues throughout the complex as seen in other similar large complexes. The researchers used a computational method to predict the long-range effects of such transitions, to round out the picture provided from the earlier structural and sequencing studies. Called the Structural Perturbation Method (SPM), it can determine the Allostery Wiring Diagram (AWD) that shows the network of residues responsible for large amplitudes of movement.

Identifying Key Residues

The first part of this analysis is setting up an Elastic Network Model (ENM) to represent the complex. This first sets out the system at a macroscopic scale, with the components showing interactions mediated by harmonic potentials. This representation of the energy of the system underlies its transitions, which are then represented by several normal modes.

The AWD is drawn on the ground of the direct correlation between the important amino acids that mediate the change in conformation in the different allosteric states and the largest change in the small number of low-frequency soft modes that define the large-scale movement.

The responses to changes in the interactions at each residue due to deliberate perturbations are examined, and the residues with the largest response are identified as hotspots. The AWD is simply the network of these hotspot residues.

The AWD is similar for both viruses, but mutations and some smaller changes in the structure have led to some differences, mostly as outliers at the part of the complex where detachment begins.

The SPM in this case was analyzed based on the complexes formed by the SARS-CoV or SARS-CoV-2 RBDs with the metallopeptidase (PD) domain of the human ACE2 receptor. The RBD sequences are 81% similar, but at the interface, only 58% are identical.

Allosteric Effects on Complex Stability

The ENM showed that the disruption of the complex could be best described by a single dominant mode, similar for both viruses. This mode had a subset of interfacial interactions hinging around the G502 residue on the RBD (in SARS-CoV, it is residue G488), which break apart as the RBD rotates in a counter-clockwise direction, moving farther away from the ACE2.

The researchers also found that the ACE2 was allosterically linked to the RBD-surface of the receptor. The interface between them was covered by three different regions of interaction. Some of the important residues that were linked to an altered binding affinity for the receptor for both viruses were also identified.

The researchers carried out Molecular Dynamics (MD) simulations to evaluate how these residues affected the stability of the RBD-ACE2 complex and to explain the findings at the atomic level. They found that an oscillatory movement along with the dominant mode, triggered by the detachment, would explain how the two ends of the ACE2-RBD interface break their contacts more often than the central part because the interaction at the two ends are predicted to weaken alternately.

The SPM also predicts that the key residues in the AWD are located around G502 (G488). If so, the MD simulations of the G502P (proline-substituted) mutant should reflect this. Indeed, this mutation led to marked destabilization of the complex. Again, when any other residue of G502 was mutated, the RBD-ACE2 interaction showed markedly poorer affinity.

This is due to the thermodynamic unfavorability of such substitutions, mediated by several changes including the loss of the backbone hydrogen bond connecting this residue with K353. Ultimately, they observed, this mutation impairs the fit of the mutant RBD on the ACE2 surface, showing how important steric hindrance and complementarity of shape is to binding. Other mutations were also explored for their effects on binding affinity and strength.

The SPM also successfully predicted the allosteric connection between RBD detachment and the PD of the ACE2 molecule, where the ACE2 subdomains opened up like jaws in a manner opposite to that seen when in the presence of an ACE2 inhibitor.

The RBD detachment triggers the AWD to reach the receptor’s zinc-binding site, indicating that ACE2-mediated catalysis could be linked to its formation of a complex with the RBD. More evidence is required for this allosteric effect, however.

Thus, the researchers found that the stability of the complex depends upon the allosteric linkage of residues that are located at some distance from it, and these long-range transitions are critical in driving the conformational change from ‘up’ to ‘down’ state.

What are the Implications?

The AWD is scattered through the complex, which may indicate that electrostatic forces, as well as the allosteric interactions inherent in the stiffness of the protein folding, mediate the switch to the ‘up’ conformation. This underlies the ability of the RBD to bind to the host cell receptor.

The study shows that apart from the interactions between the interface residues themselves, long-range interactions also play a part in stabilizing binding, and thus facilitate fusion of viral and host cell membranes.

Thus, both for global mechanical movements through allosteric communication and for stability, it is critical to determine the AWD. The current study provides a sound computational framework to investigate the AWD activated transitions upon dissolution of the complex.”

Future studies may focus on the role of the glycosylation of both ACE2 and the RBD in stabilizing the complex, taking forward current knowledge gained from MD simulations and experimental observations.

*Important notice: bioRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.