AI-Driven Multiscale Simulations Illuminate Mechanisms of SARS-CoV-2 Spike Dynamics

Abstract

We develop a generalizable AI-driven workflow that leverages heterogeneous HPC resources to explore the time-dependent dynamics of molecular systems. We use this workflow to investigate the mechanisms of infectivity of the SARS-CoV-2 spike protein, the main viral infection machinery. Our workflow enables more efficient investigation of spike dynamics in a variety of complex environments, including within a complete SARS-CoV-2 viral envelope simulation, which contains 305 million atoms and shows strong scaling on ORNL Summit using NAMD. We present several novel scientific discoveries, including the elucidation of the spike’s full glycan shield, the role of spike glycans in modulating the infectivity of the virus, and the characterization of the flexible interactions between the spike and the human ACE2 receptor. We also demonstrate how AI can accelerate conformational sampling across different systems and pave the way for the future application of such methods to additional studies in SARS-CoV-2 and other molecular systems.ACM Reference FormatLorenzo Casalino1†, Abigail Dommer1†, Zied Gaieb1†, Emilia P. Barros1, Terra Sztain1, Surl-Hee Ahn1, Anda Trifan2,3, Alexander Brace2, Anthony Bogetti4, Heng Ma2, Hyungro Lee5, Matteo Turilli5, Syma Khalid6, Lillian Chong4, Carlos Simmerling7, David J. Hardy3, Julio D. C. Maia3, James C. Phillips3, Thorsten Kurth8, Abraham Stern8, Lei Huang9, John McCalpin9, Mahidhar Tatineni10, Tom Gibbs8, John E. Stone3, Shantenu Jha5, Arvind Ramanathan2∗, Rommie E. Amaro1∗. 2020. AI-Driven Multiscale Simulations Illuminate Mechanisms of SARS-CoV-2 Spike Dynamics. In Supercomputing ’20: International Conference for High Performance Computing, Networking, Storage, and Analysis. ACM, New York, NY, USA, 14 pages. https://doi.org/finalDOI

Potent neutralizing nanobodies resist convergent circulating variants of SARS-CoV-2 by targeting novel and conserved epitopes

“…Nevertheless, the unique RBD binding mode of Nb17 and its ease of bioengineering may facilitate the development of robust biosensors to detect specific conformations of the spike. Currently, this process can only be inferred computationally [36][37][38][39] or by single-molecule techniques which require extensive bioengineering to label the spike with fluorescence dyes 17 . Finally, since Nb17 binds extremely tightly to the RBD-up conformations, potentially, it could be used as a novel Nb "adjuvant" for vaccines to facilitate exposure of conserved yet cryptic RBD epitopes that could help elicit broadly neutralizing activities insensitive to the evolving variants.…”

“…The receptor-binding domain (RBD) of S1 is critical for interacting with the host receptor angiotensin-converting enzyme 2 (ACE2). In the prefusion state, the RBD is undergoing highly dynamic switching between closed ("down") and open ("up") conformations on the distal tip of the spike trimer [15][16][17] . In the post-fusion state, S1 shedding triggers a large conformational change of S2 to facilitate virus binding to the host membrane for infection 18 .…”

Dynamic Network Modeling of Allosteric Interactions and Communication Pathways in the SARS-CoV-2 Spike Trimer Mutants: Differential Modulation of Conformational Landscapes and Signal Transmission via Cascades of Regulatory Switches

“… 56 The development of a “bottom-up” coarse-grained model of the SARS-CoV-2 virion was recently presented in which cryo-EM, X-ray crystallography, and computational predictions were integrated to build a multiscale model of structural SARS-CoV-2 proteins and a complete virion model. 57 , 58 Another large-scale MD study of a 4.1 million atom system containing a patch of viral membrane with four full-length, fully glycosylated S proteins was reported, where structural dynamics and the analysis of steric accessibility were used for epitope mapping, revealing signatures of the flexible glycan coat that can shield a larger surface area than can be derived from static structures. 59 MD simulations of the SARS-CoV-2 spike glycoprotein identified differences in flexibility of functional regions that may be important for modulating the equilibrium changes and binding to the ACE2 host receptor.…”

Neuropilin-1 Assists SARS-CoV-2 Infection by Stimulating the Separation of Spike Protein Domains S1 and S2

“…Intriguingly, in our models Nrp1 domains do not interact with the RBD's motif for binding ACE2 (a sub-region comprising residues 437-538), thus the ACE2 association with the Spike protein should be unperturbed unless there are allosteric effects. [25][26][27] The Nrp1 a2-b1-b2 domains also do not interact extensively with S2 (just the very N-terminus in case of models 3,5 and 6) as it would make the Nrp1 domains too distant from the membrane (see below & ). In Fig.…”