key: cord-0746916-n4u72n0a authors: Golcuk, Mert; Yildiz, Ahmet; Gur, Mert title: Omicron BA.1 and BA.2 Variants Increase the Interactions of SARS-CoV-2 Spike Glycoprotein with ACE2 date: 2022-04-01 journal: bioRxiv DOI: 10.1101/2021.12.06.471377 sha: dcf886266a32ae7d3088902c9ddb9cf1f4ac014c doc_id: 746916 cord_uid: n4u72n0a SARS-CoV-2 infection is initiated by binding of the receptor-binding domain (RBD) of its spike glycoprotein to the peptidase domain (PD) of angiotensin-converting enzyme 2 (ACE2) receptors in host cells. Recently detected Omicron variant of SARS-CoV-2 (B.1.1.529) is heavily mutated on RBD. Currently, the most common Omicron variants are the original BA.1 Omicron strain and the BA.2 variant, which became more prevalent since it first appeared. To investigate how these mutations affect RBD-PD interactions, we performed all-atom molecular dynamics simulations of the BA.1 and BA.2 RBD-PD in the presence of full-length glycans, explicit water and ions. Simulations revealed that RBDs of BA.1 and BA.2 variants exhibit a more dispersed interaction network and make an increased number of salt bridges and hydrophobic interactions with PD compared to wild-type RBD. Although BA.1 and BA.2 differ in two residues at the RBD-ACE2 interface, no major difference in RBD-PD interactions and binding strengths were observed between these variants. Using the conformations sampled in each trajectory, the Molecular Mechanics Poisson-Boltzmann Surface Area (MMPBSA) method estimated ~34% and ~51% stronger binding free energies for BA.1 and BA.2 RBD, respectively, than wild-type RBD, which may result in higher binding efficiency of the Omicron variant to infect host cells. The recent appearance and the rapid rate of infection of a heavily mutated B.1.1.529 variant of SARS-CoV-2, named Omicron, have raised concerns around the world, with many countries temporarily limiting their international travel. World Health Organization has designated the omicron variant as a variant of concern (VOC) 1 . Currently, the omicron variant has three major sub-lineages, namely BA.1, BA.2, and BA.3. 2 BA.2 RBD (RBDBA.2) are surface-exposed and being targeted by various antibodies ( Figure S1 ) and nanobodies. In addition, for BA.1, 11 of these 15 mutations are located on the ACE2 binding interface, while for BA.2 nine of these are located on the ACE2 binding interface (Figure 1 ). For both BA.1 and BA.2 four hydrophilic residues mutated to positively charged residues (N440K, T478K, Q493R, and Q498R), one negatively charged residue mutated to hydrophobic residue (E484A), one positively charged residue mutated to hydrophilic residue (K417N), and for hydrophilic residues are mutated to again hydrophilic residues (S477N, N501Y, and Y505H) at RBD's PD binding interface. In addition, to these mutations, two neutral residues mutated to hydrophilic residues (G446S and G496S) in BA.1. Thus, both RBDBA.1's and RBDBA.2's PD binding interface are more positively charged than RBDWT. Furthermore, the PD binding interface of RBDBA.1 comprises more hydrophilic residues than RBDBA.2. Our previous all-atom Molecular Dynamics (MD) simulations 4 showed that 5 of these mutated residues form pairwise interactions between wild-type (WT) S and ACE2 (salt bridges between K417-D30 and E484-K31, and hydrogen bonding between Q493-E35, Q498-Q42, Q498-K353, and Y505-E37). It is still unclear how BA.2 omicron mutations affect the binding strength of RBD to ACE2 and the ability of existing SARS-CoV-2 antibodies to neutralize this interaction. Furthermore, the difference in binding characteristics and strength of omicron BA.1 and BA.2 variants remains to be explored. We performed all-atom MD simulations of the RBDBA.1-PD in the presence of full-length glycans on both S RBD and ACE2 PD, 5,6 explicit water and ions (~200k atoms in total). Four sets of MD simulations each of 300 ns in length were performed using the parameters of our previous RBD-PD simulations for the WT, 4 alpha, and beta variants. 7 These four sets of simulations were combined into a single 1,200 ns long trajectory to investigate the RBDBA.1-PD interactions. Simulations revealed a more extensive interaction network for RBDBA.1-PD with PD compared to RBDWT. We detected five salt bridges between RBDBA.1 and PD; one of them (K440-E329) medium and four (R403-E37, R493-E35, R493-D38, and R498-D38) with high frequency (Figure 2 ). In comparison only 2 high frequency salt bridges existed between RBDWT and PD 4 ( Figure 2 and Table S1 ) and both of those disappeared in the BA.1 variant ( Figure 2 and Table S2 ). The RBDBA.1 forms all of the ten high frequency hydrophobic interactions that were observed for RBDWT-PD and an additional high frequency hydrophobic interaction between Y501-Y41. Compared to eight hydrogen binding between RBDWT and PD (3 high and five medium frequency), six hydrogen bonds were observed between RBDBA.1 and PD (3 high and 3 medium frequency). Only two of these interactions were also observed for the WT, while other four are newly formed ( Figure 2) . Collectively, the total number of salt bridges, hydrophobic interactions, hydrogen bonds at the S-ACE2 interface changed by 150%, 10% and -25%, respectively. Our simulations also revealed a change in the spatial distribution of RBD-PD interactions along the interaction surface due to the mutations in the BA.1 variant, which are mostly consistent with recently reported RBDBA.1-ACE2 structures. [8] [9] [10] [11] Between RBDWT and PD, salt bridges are concentrated at the interface of contact region 1 (CR1) and CR2, while hydrogen bonding and hydrophobic interactions are concentrated in CR3 and CR1, respectively ( Figure 2A ). 4 In comparison, RBDBA.1 exhibits a more dispersed interaction network along the RBD-PD interaction surface This may result in an altered binding mechanism and negatively impact the current inhibition mechanism by neutralizing antibodies and nanobodies. for all of these antibody classes. For example, we expect E484A mutation to eliminate E484-R52 salt bridge and E484-S57 hydrogen bonds in H11-H4 and H11-D4 nanobodies, and E484-N56, and E484-Y335 hydrogen bonds in Ty1 nanobody. 7, 23, 24 Additionally, Q493R mutation would eliminate the hydrogen bonds Q493-Y104 and Q493-S104 in H11-H4, and H11-D4, respectively. Furthermore, mutations are expected to eliminate salt bridge E484-R96 and hydrogen bonds E484-H35, Q493-R97, and Q493-S99 between RBD and a class 2 antibody C002 ( Figure 5 ). Consistent with this view, point mutation Q493R mutations was reported to decrease the RBD dissociation constant of C002 from 11 nM to 596 nM. 22, 25, 26 Four sets of simulations each of 300ns length were performed for the RBDBA.2-PD complex in the presence of explicit water molecules, ions and also full length glycans (MD7-10). Prior to these production simulations, each system was minimized for 10,000 steps and then equilibrated for 2 ns by keeping the protein fixed. Subsequently, system was minimized for an additional 10,000 steps without fixing the protein, which is followed by 4 ns of equilibration with harmonic constraints applied on Cα atoms. All constraints were removed from the system and an additional 4 ns of MD simulations were performed; finalizing the minimization and equilibration steps prior to production runs. MD simulations were performed in NAMD-2.14 28 , for MMPBSA calculations and system minimizations and equilibrations, and NAMD3 28 for all production simulations under N, P, T conditions. CHARMM36 29 force field and a time step of 2 fs was used in the simulations. Pressure was kept at 1 atm using the Langevin Nosé-Hoover method with an oscillation period of 100 fs and a damping time scale of 50 fs. Temperature was maintained at 310 K using Langevin dynamics with a damping coefficient of 1 ps -1 . Periodic boundary conditions were applied in simulations and Particle-mesh Ewald method was used for long-range electrostatic interactions. 12 Å cutoff distance was used for van der Waals interactions. To determine salt bridge formation in MD simulations, a cutoff distance of 6 Å between the basic nitrogen and acidic oxygen was used. 30 , while for hydrophobic interactions, a cutoff distance of 8 Å between the side chain carbon atoms was used. 31-33 A cutoff distance of 3.5 Å between hydrogen bond donor and acceptor, and a 30° angle between the hydrogen atom, the donor heavy atom and the acceptor heavy atom was used to determine hydrogen bond formation. 34 Among those interaction pairs that satisfied the hydrogen bonding distance criterion but did not satisfy the angle criterion, were classified as electrostatic interactions. As was performed in our previous studies, 4,35 observation frequencies of interactions sampled from MD simulations were classified as high and moderate for interactions that occur in 49% and above and between 15 and 48% of the total trajectory, respectively. Pairwise interactions with observation frequencies below 15% were excluded from further analysis. For each set of simulation, 3000 snapshots each separated by 0.1 ns were selected from the simulations. The binding free energies were predicted for the RBD-PD complexes using the MMPBSA method 13, 14 which was conducted via VMD 15 plugin CaFE 16 as described by Liu and Tingjun. 16 Entropy change during binding was neglected in calculations, consistent with previous MMPBSA calculations for RBD-PD interactions. 36, 37 Default parameters were used in CaFE 16 calculations. WHO. Classification of Omicron (B.1.1.529): SARS-CoV-2 Variant of Concern Rapid epidemic expansion of the SARS-CoV-2 Omicron variant in southern Africa Nextstrain: real-time tracking of pathogen evolution Critical Interactions Between the SARS-CoV-2 Spike Glycoprotein and the Human ACE2 Receptor Beyond Shielding: The Roles of Glycans in the SARS-CoV-2 Spike Protein The Flexibility of ACE2 in the Context of SARS-CoV-2 Infection Binding Mechanism of Neutralizing Nanobodies Targeting SARS-CoV-2 Spike Glycoprotein SARS-CoV-2 Omicron variant: Antibody evasion and cryo-EM structure of spike protein– Structural basis of SARS-CoV-2 Omicron immune evasion and receptor engagement Receptor binding and complex structures of human ACE2 to spike RBD from omicron and delta SARS-CoV-2 Structures of the Omicron spike trimer with ACE2 and an anti-Omicron antibody Enhanced Receptor Binding of SARS-CoV-2 Through Networks of Hydrogen-Bonding and Hydrophobic Interactions Calculating Structures and Free Energies of Complex Molecules: Combining Molecular Mechanics and Continuum Models Continuum Solvent Studies of the Stability of DNA, RNA, and Phosphoramidate−DNA Helices VMD: Visual Molecular Dynamics CaFE: a Tool for Binding Affinity Prediction Using End-Point Free Energy Methods Mutations on RBD of SARS-CoV-2 Omicron variant result in stronger binding to human ACE2 receptor Scanning the RBD-ACE2 molecular interactions in Omicron variant SARS-CoV-2 Omicron variant: Antibody evasion and cryo-EM structure of spike protein–ACE2 complex Broadly neutralizing antibodies overcome SARS-CoV-2 Omicron antigenic shift The SARS-CoV-2 variant, Omicron, shows rapid replication in human primary nasal epithelial cultures and efficiently uses the endosomal route of entry. bioRxiv SARS-CoV-2 Neutralizing Antibody Structures Inform Therapeutic Strategies Neutralizing Nanobodies Bind SARS-CoV-2 Spike RBD and Block Interaction with ACE2 An Alpaca Nanobody Neutralizes SARS-CoV-2 by Blocking Receptor Interaction Antibody evasion properties of SARS-CoV-2 Omicron sublineages Striking antibody evasion manifested by the Omicron variant of SARS-CoV-2 Structure of the SARS-CoV-2 Spike Receptor-Binding Domain Bound to the ACE2 Receptor Scalable molecular dynamics on CPU and GPU architectures with NAMD Optimization of the additive CHARMM all-atom protein force field targeting improved sampling of the backbone ϕ, ψ and side-chain χ1 and χ2 dihedral angles Zipping and unzipping of adenylate kinase: atomistic insights into the ensemble of open↔ closed transitions Direct and Quantitative AFM Measurements of the Concentration and Temperature Dependence of the Hydrophobic Force Law at Nanoscopic Contacts A Study of the Preferred Environment of Amino Acid Residues in Globular Proteins Molecular Dynamics Simulation of Antimicrobial Peptide Arenicin-2: b-Hairpin Stabilization by Noncovalent Interactions HBonanza: a computer algorithm for molecular-dynamics-trajectory hydrogen-bond analysis Binding mechanism of neutralizing Nanobodies targeting SARS-CoV-2 Spike Glycoprotein Critical sequence hotspots for binding of novel coronavirus to angiotensin converter enzyme as evaluated by molecular simulations Mechanistic insights into the effects of key mutations on SARS-CoV-2 RBD-ACE2 binding The authors declare no competing interests.