The Spike (S) protein of the SARS-CoV-2 virus operates much like a mechanical crane, starting in a folded 'down' position before extending and swinging into an 'up' position to latch onto ACE2 receptors on human cells.
Peter J. Bond, Senior Principal Investigator at A*STAR’s Bioinformatics Institute (BII), describes the S protein as a sophisticated molecular machine, adding that most antibody-based COVID-19 therapeutics and vaccines are designed to block its binding dynamics.
Throughout the pandemic, mutations in the S protein have enabled the virus to evade antibodies, diminishing the effectiveness of vaccines and therapies and allowing the virus to continue spreading.
"The central motivation for this study was to characterise convalescent patient-derived antibodies against the ancestral variant of SARS-CoV-2," explained Bond. This strategy would shed light on how mutations in the Delta and Omicron variants contribute to immune escape.
In collaboration with Paul MacAry and a team of researchers from National University of Singapore; Nanyang Technological University, Singapore; A*STAR’s Singapore Immunology Network (SIgN); and BII; Bond’s team characterised the neutralisation mechanisms of antibodies from 15 recovered COVID-19 patients.
The team leveraged hydrogen-deuterium exchange mass spectrometry (HDXMS)—a technique that replaces hydrogen atoms in the protein with heavier deuterium atoms and measures how quickly this swap occurs—to study protein structure and movement, and interpreted this data using cutting-edge computational modelling approaches based on molecular simulations.
“One of the crucial insights we gained from the study was the unique dynamics that a given antibody binding to a specific site induces upon the S protein,” said Bond.
Some antibodies were found to bind to a hidden site on the receptor-binding domain (RBD), destabilising the S protein trimer and preventing the virus from effectively latching on to host cells. Others stabilised the trimer by binding to the receptor-binding motif (RBM), blocking ACE2 interaction.
Interestingly, some antibodies that are less effective against Delta and Omicron variants still bound to S protein sites in the ancestral strain but altered the protein dynamics, hinting at vaccine escape mechanisms. Moreover, antibody cocktails targeting different sites were found more effective at neutralising the virus than individual antibodies.
“This approach could potentially be used to inform clinicians on the choice of antibodies that target mutually exclusive sites and could thereby be coupled in combinatorial antibody cocktails,” suggested Bond.
In the post-vaccination era, SARS-CoV-2 infections have become endemic, with the S protein constantly mutating due to various pressures. Funded by the Programme for Research in Epidemic Preparedness and Response (PREPARE), Bond’s team will continue to collaboratively monitor new variants and develop broad-spectrum S protein inhibitors for future coronavirus outbreaks.
The A*STAR-affiliated researchers contributing to this research are from the Bioinformatics Institute (BII) and Singapore Immunology Network (SIgN).
