Why Are Some Coronavirus Strains More Infectious Than Others?

In March 2003, a strain of coronavirus, severe acute respiratory syndrome coronavirus (SARS, or SARS-CoV-1) was recognized as a global threat, causing 774 deaths. By July of the same year, no new case were reported and the World Health Organization declared that the outbreak was over.

SARS-CoV-2, the coronavirus that causes COVID-19, has devastated the human population, infecting over 110 million people and causing almost 2.5 million deaths to date. SARS-CoV-1 and SARS-CoV-2 both belong to the same coronavirus family. A key question is: What makes one more infectious than the other?

The lab of or Dr Mahmoud Moradi at the University of Arkansas has used molecular simulations, performed at the Texas Advanced Computing Center and the Pittsburgh Supercomputing Center, to investigate this question.

The team aimed to explore the precise dynamics of how the spike proteins of SARS-CoV-1 and SARS-CoV-2 move from one position to another, "active" to "inactive", which is a key step required for viral entry into host cells.

Their findings, presented at the 65^th Annual Meeting of the Biophysical Society, suggest that, once activated, the spike protein of SARS-CoV-2 has a higher tendency to stay active compared to the spike protein of SARS-CoV-1. The simulations also suggest that there are less well studied parts of the spike protein that contribute to its stability and warrant further research.

Technology Networks interviewed Moradi to learn more about the study, the utility of computer simulations in this context and the potential implications of their findings.

Molly Campbell (MC): Please can you briefly summarize why the spike protein is a crucial component required for coronavirus infection of a human cell?

Mahmoud Moradi (MM): Coronavirus spike protein is not only responsible for recognizing the host cell, but also mediates viral entry by triggering the fusion of the cellular and viral membranes. The gene encoding the spike protein is associated with the most variable part of the coronavirus genome, resulting in distinct behaviors in various coronaviruses such as SARS-CoV-1 and SARS-CoV-2 as well as different variants of SARS-CoV-2.

MC: Can you describe your study rationale and method?

MM: The rationale of this study is that if we characterize the structural dynamics of spike proteins of SARS-CoV-1 and SARS-CoV-2 and identify their similarities and differences mechanistically, we may not only capture what has caused the two viruses to have substantially different transmissibility but also predict what mutations in the SARS-CoV-2 could potentially be consequential in terms of its activation dynamics and perhaps infectivity. We have used large-scale simulations of spike proteins of both viruses including all atoms of the protein and its immediate environment such as water and salt. Using this method, known as all-atom molecular dynamics, we needed to solve the Newton’s equation of motion for close to a million atoms simultaneously – for millions of simulation steps – to be able to only partially construct the structural dynamics of these proteins.

MC: For anyone that may be unfamiliar, can you explain what steered MD (SMD) simulations are?

MM: The majority of our simulations do not use any sort of bias. However, our unbiased simulations only provide a partial picture of the structural dynamics of spike proteins due to the limited capability of the computers and the complexity of the problem. An alternative method used in some of the simulations was the steered molecular dynamics. In this method we trigger the activation or deactivation process of spike protein by applying forces to the system to steer it from one state to another. Since we do this similarly for both SARS-CoV-1 and SARS-CoV-2 spike proteins, the comparison still provides a reliable way of determining the differential behavior of the two viruses.

MC: How accurate are molecular simulations in this context?

MM: We use state-of-the-art simulation techniques that have been used recently to study other similar systems successfully. These are certainly the most extensive set of simulations performed so far to reveal the similarities and differences of the spike proteins of SARS-CoV-1 and 2. However, there are always assumptions and approximations involved in not only simulations but also experiments, and it is always necessary to use complementary computational and experimental techniques to validate the results coming from a particular method. In this case, the best way to determine the accuracy of the calculations is to perform experiments to validate them.

MC: You state that "Our results based on nearly 50 microseconds of equilibrium and nonequilibrium MD simulations indicate that the active form of the SARS-CoV-2 spike protein is considerably more stable than the active SARS-CoV-1 spike protein", can you please explain this finding in lay terms?

MM: SARS-CoV-1 and 2 spike proteins are not always capable of recognizing and binding to the human cell. They only can do this when they are in their active form. Our results show that the SARS-CoV-2 spike protein has a higher tendency to stay active when it is activated, while the SARS-CoV-1, even when activated, does not stay active long enough to infect the human cell as quickly and effectively as the SARS-CoV-2 can.

MC: Your hypothesis is that mutations in other parts of protein that are not directly involved in binding may lead to changes in the effective binding of coronaviruses. Can you explain this hypothesis further?

MM: The part of protein that is directly involved in binding to the human cell, known as the receptor binding domain (RBD), has been the focus of much of the previous studies. However, spike protein needs to be activated and remain active prior to binding to the human cell. Based on our simulations, there are other parts of the protein far from the RBD that are involved in the activation and deactivation process of the protein. Certain mutations in these apparently unrelated parts of the protein could potentially alter the activation/deactivation dynamics of the spike protein and perhaps lead to higher or lower levels of infectivity.

MC: What implications might your data have in the development of therapeutics/preventives for SARS-CoV-2?

MM: Our data provides a new framework for developing vaccines and therapeutics. In addition to receptor binding and other processes that have been targeted previously, we are suggesting that the activation of SARS-CoV-2 spike protein could be inhibited and/or its deactivation promoted by designing therapeutics that target various hot spots involved in the activation/deactivation process, not only around the RBDs but also other parts of protein previously unnoticed. In terms of the spike-based vaccine development, an important implication is that while the pre- vs post-fusion conformations of spike protein has justifiably been taken into account, the active vs inactive conformations should also be considered.

MC: You have begun studying the new SARS-CoV-2 variant, B.1.1.7. Can you talk to us about your findings thus far?

MM: Unfortunately, we will not be able to provide much information prior to a thorough analysis of the results.

Mahmoud Moradi was speaking to Molly Campbell, Science Writer for Technology Networks.