The coronavirus disease 2019 (COVID-19) pandemic disrupted the normal functioning of society, business, and education while taking a grievous toll on human health. The pathogen, the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has undergone several rounds of mutations to emerge as a host of variants.
Study: Site Density Functional Theory and Structural Bioinformatics Analysis of the SARS-CoV Spike Protein and hACE2 Complex. Image Credit: Juan Gaertner/Shutterstock
Given the serious threat posed by successive coronaviruses, scientists have been investigating the binding between the angiotensin-converting enzyme 2 (ACE2) molecule, which acts as the virus receptor on the target host cell, with the virus receptor-binding domain (RBD) on its spike protein. A new paper in the journal Molecules describes this in molecular terms, comparing SARS-CoV-2 with an earlier highly pathogenic coronavirus, SARS-CoV.
The spike protein consists of a homotrimer in the natural state, each protomer having an S1 and an S2 subunit. The S1 subunit contains the RBD that binds the ACE2 peptidase domain, helping to stabilize the spike in its prefusion conformation. To understand this process better, the scientists in the current paper used a method based on density functional theory (DFT).
The application of DFT to biological macromolecules is a complicated affair because of the dimensional nature of these molecules, which means six-dimensional density profiles are required to examine rigid molecules with DFT. One potential solution is site density representation, which refers to the density of each atom within a solvent molecule, based on which the site density functional theory (SDFT) was developed.
This is suitable for any examination that requires atomic-level structural analysis, as it looks at the density of individual atoms and chemical bonds. However, chemical bonds are localized through strong interactions, in contrast to the equally important long-range but weaker intermolecular forces.
It is difficult to apply SDFT to both types of forces simultaneously, which is a huge challenge in the use of this method for molecular liquids. The three-dimensional reference interaction site model (3DRISM) is a simplified SDFT version that is a partial answer to this problem, by representing such liquids as an “atomic” mixture, in effect, thus opening the way for the use of multiple simple fluid theory techniques.
With coronaviruses, some other issues arise, like the large size of the virus and the low level of flexibility of the virus structure. To reduce the cost of evaluating such structures using conventional mechanics with 3DRISM, the researchers in this study used structural bioinformatics instead. This includes techniques like molecular docking, affinity and binding constraint studies, constrained geometric simulations (CGS) to sample different conformations, and normal mode analysis (NMA).
The use of 3DRISM allowed the scientists to examine the interactions between the protein and its ligands, in addition to environmental effects due to electrostatic interactions, hydrogen bond formation, steric clashes, and van der Waals bonds.
The 3DRISM seems to be a suitable methodological choice to reveal a role of electrostatic and short-ranged protein-solvent interactions in solvation thermodynamics, as the theory can be used to detect water-mediated contacts in protein.”
What did the study show?
The scientists examined the spike protein protomers of SARS-CoV and SARS-CoV-2 in isolation as well as the RBD-ACE2 formed in both cases. An important difference between the two is the RRAR insertion at the furin recognition site in the SARS-CoV-2 spike, at the S1/S2 interface, compared to the single arginine in SARS-CoV.
The RBD moves like a hinge with this change in conformational motion before recognizing the host receptor. For either virus, at least one RBD has to be in the “up” conformation for ACE2 binding because otherwise, the ACE2 peptidase clashes with the spike in the three-down conformation.
With SARS-CoV-2, however, the area covered by the RBD is greater due to a conformational change. The binding ridge is more compact, and virus-binding hotspots at the interface are more stable because of changes in the amino acids at the interface. This allows for greater interactions between the molecules.
The molecular interactions occur via loops and beta-sheets in the RBD vs. one alpha-helix of ACE2, indicating inherent instability. Most interactions occur at the two ends and the middle. Computational modeling shows comparable binding free energy for both RBD-ACE2 complexes, but experimental evidence indicates that the SARS-CoV-2 RBD-ACE2 binding free energy is lower than the other by about 1.1 kcal/mol.
While the chemical shifts between the two complexes are insignificant, the relative absolute differences at the interface are marked, especially near residues forming hydrogen bonds. When water-mediated interactions are also considered, using water molecules only at the most likely positions, strong interactions are seen with both RBD and ACE2. These were predicted to be stronger for the SARS-CoV-2-RBD-ACE2 complex.
Protein-solvent interaction energies confirmed this shift, with a large fall in the electrostatic energy leading to a larger improvement in the total interaction energy for the CoV-2-ACE2 complex. This is attributable to a higher degree of polarization of the interfacial water with this complex and the formation of two more water bridges than with CoV-1-ACE2. The water bridges are also stronger with the former complex.
The bridging water molecules form hydrogen bonds with both the RBD and the ACE2 simultaneously in the former complex, but this is both much weaker and more diffuse with the CoV-1-ACE2 complex. “Therefore, we conclude that bridge water molecules play a significant and, perhaps, crucial role in stabilizing CoV-2-hACE2 complex.”
The binding energies were also predicted to follow the same trend. Still, the difference is always less than expected from the experimental data, probably because of the ionic bonds that perturb the electrostatic and relaxation-induced entropy changes. The results of each method vary significantly.
Moreover, multiple factors such as the conformation state and conformation angle of the molecules in the complex contribute to the binding energy and binding affinity, indicating the need for further studies to validate the 3DRISM results.
Other computational modeling methods suggest that the bound and unbound RBD shows large conformational differences for the two viruses. Following this up with NMA, the scientists observed that pairs of residues are coupled. Both RBDs show large degrees of movement in the unbound state but less for the S2 subunit of the spike.
With ACE2 binding, anti-phase dynamic movements of the extended RBD region disappear, indicating a stabilization of the molecule with slightly faster RBD movements with SARS-CoV-2.
Again, the SARS-CoV-2 molecule is more rigid than SARS-CoV in the RBD “up” conformation, with the latter showing a greater shift with increasing temperatures. With the latter, two-up and three-up RBDs are common in the unbound state but are rare with the former. Such findings may underlie the increased binding of RBD-ACE2 with the former.
Overall, the increased strength of interfacial contacts accounts for the stability of the bound RBD with SARS-CoV-2. An analysis of the rigidity of the structure also confirms that as the hydrogen-bond energy cut-off increases, SARS-CoV loses rigidity but not SARS-CoV-2. This is because the RBD for SARS-CoV becomes a large conformational ensemble with increasing cut-off H bond energy.
As a result, it loses binding affinity for ACE2. In contrast, the other RBD is mostly a single rigid cluster, which remains rigid on ACE2 binding and favors the proteolytic splitting that leads to membrane fusion.
What are the implications?
The results show how important water is in the RBD-ACE2 interactions, stabilizing it and rendering the CoV-2-ACE2 complex more stable than the CoV-ACE2 complex. The dissociation constant of the latter is one order of magnitude lower than the former.
The lower conformational stability of the SARS-CoV RBD compared to SARS-CoV-2 contributes to the lower binding affinity when coupled with, the larger surface area of the latter and the large local conformational shifts in the residues. This leads to more RBD-ACE2 interactions, including more hydrogen bonding.
The increased rigidity of the CoV-ACE2 complex compared to that formed by SARS-CoV is due to stronger bridging interactions via water molecules, along with two added polar interactions. The binding of the receptor predisposes the molecule to proteolysis and membrane fusion, favoring virus entry.
Understanding prefusion conformational dynamics as well as its binding mechanism to the receptor among closely related species is critical for designing vaccine and inhibitors to strop viral entry.”