In a paradigm shift paper currently on the bioRxiv* Preprint servers, the researchers have drastically expanded the possibilities of computer microscopy in order to better understand both the structure and the dynamics of breath aerosols that are loaded with the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2).
Although droplets were seen as the main route of transmission for SARS-CoV-2 – a causative agent of the ongoing coronavirus pandemic (COVID-19) – it quickly became clear that airborne transmission also plays a crucial role in the spread of the disease.
However, the greatest challenge to understanding airborne transmission has been the inability to reliably study the structure and dynamics of viruses once they become part of airway aerosol particles.
Correctly defined, aerosols are less than five microns in diameter, can travel great distances and float in the air for hours (similar to cigarette smoke), and are relatively easy to inhale.
In addition to the different sizes and compositions of aerosols, viruses can also be modified to reflect new variants of concern, with different mutations affecting the interactions with certain species in the aerosol and subsequently changing their viability and / or structural dynamics.
It is for this reason that a recent paper from the Amaro group at the University of California San Diego has completely reworked contemporary models of airborne viral transmission by providing an unprecedented understanding of SARS-CoV-2 at the atomic level in a breathable aerosol.
Overall scheme showing the structure and the multi-scale simulations of Delta SARS-CoV-2 in a breath aerosol. (Note: the size of the divalent cations has been increased for visibility.)
Development of a “computer microscope”
The approach of these researchers was based on the use of molecular dynamics simulations of all atoms as a multi-scale “computer microscope”. These types of simulations can result in multiple types of biological data (e.g., multi-resolution structural data sets) in cohesive and biologically precise structural models.
Individual protein components of the SARS-CoV-2 delta virion. The tip is highlighted with the surface in cyan and with deltas mutated residues and deletion sites in pink and yellow, respectively. Glycans attached to the thorn are shown in blue. The E protein is yellow and the M protein is shown in silver and white. Visualized with VMD.
And once it has been created, the model can then be approximated to its many atoms in order to develop trajectories of its time-dependent dynamics under aerosol-like conditions. In addition, such simulations can also provide specific parameters that can be incorporated into physical models of aerosols.
“Our approach to simulating the entire aerosol follows a composite framework in which each of the individual molecular parts is refined and simulated for itself before it is integrated into the composite model,” the study authors explain their methodological approach further.
First views at the atomic level of virus-laden aerosols
This study demonstrated an extensible multi-scale computational framework that encompasses time and length scales from electronic structure to entire aerosol particle dynamics and morphology. In addition, the researchers also developed all-atom simulations of respiratory mucins to understand the structural basis of the interaction with the SARS-CoV-2 spike glycoprotein.
The SMA system uses multi-scale modeling to capture everything from classic MD to AI-capable quantum mechanics. For all panels: S-protein in cyan, S-glycans in blue, m1 / m2 in red, ALB in orange, Ca2 + in yellow spheres, virus membrane in purple. A) Interactions between mucins and S, facilitated by glycans and Ca2 +. B) Snapshot from SMA simulations. C) Example of a Ca2 + binding site from SMA simulations (1800 sites, 1000+ atoms each) used for AI-assisted quantum mechanical estimates by OrbNet Sky. D) Quantification of the contacts between S and mucin from SMA simulations. E) OrbNet Sky energies vs. CHARMM36m energies for each under-selected system, colored according to the total number of atoms. Performance of OrbNet Sky vs. DFT in subplot B97x-D3 / def-TZVP, R2 = 0.99, for 17 systems of peptides that chelate Ca2 + (Hu et al., 2021)). Visualized with VMD.
As a result, this paper presents an all-atom simulation that captures the massive biological and chemical complexity within a breathing aerosol. In addition, a simulation described provides the first atomic views of virus-laden aerosols, which serve as the basis for the development of a large number of verifiable hypotheses.
Changes in pH observed in the aerosol environment can alter the dynamics and communication pathways in crucial functional regions of the spike glycoprotein of SARS-CoV-2 (and especially for the Delta strain), while its dramatically open state affects binding to the human host can facilitate cells.
Finally, the study showed how high-performance computing and cloud resources can accelerate scientific endeavors and facilitate complex collaborations. Artificial intelligence can also be coupled with high-performance computing on several levels to improve effective performance.
Implications for Understanding Airborne Transmission
This work expands the possibilities of using multiscale computed microscopy to answer important questions about our understanding of aerosols at the atomic and molecular level that are currently hindering our knowledge of airborne transmission.
“We are demonstrating how our integrated data-driven platform offers a new way of researching the composition, structure and dynamics of aerosols and aerosolized viruses and at the same time driving the development of simulation methods along several important axes,” the study authors say bioRxiv Paper.
Nonetheless, many technological advances are required to study even more complex biologically complex systems, ranging in size from nanometers to a micrometer and on longer time scales (ie, microseconds to seconds). This is a challenge that needs to be addressed through further study in the area.
bioRxiv publishes preliminary scientific reports that are not peer-reviewed and therefore are not considered conclusive, guide clinical practice / health-related behavior, or should be treated as established information.