key: cord-1005418-f0o926sz
authors: Faisal, H. M. Nasrullah; Katti, Kalpana S.; Katti, Dinesh R.
title: Binding of SARS-COV-2 (COVID-19) and SARS-COV to human ACE2: Identifying binding sites and consequences on ACE2 stiffness
date: 2021-09-06
journal: Chem Phys
DOI: 10.1016/j.chemphys.2021.111353
sha: 3c8a9574085419d99954d3f6d67505daf68aa9ea
doc_id: 1005418
cord_uid: f0o926sz

The SARS-CoV-2 coronavirus (COVID-19) that is causing the massive global pandemic exhibits similar human cell invasion mechanism as the coronavirus SARS-CoV, which had significantly lower fatalities. The cell membrane protein Angiotensin-converting-enzyme 2 (ACE2) is the initiation point for both the coronavirus infections in humans. Here, we model the molecular interactions and mechanical properties of ACE2 with both SARS-CoV and COVID-19 spike protein receptor-binding domains (RBD). We report that the COVID-19 spike RBD interacts with ACE2 more strongly and at only two protein residues, as compared to multi-residue interaction of the SARS-CoV. Although both coronaviruses stiffen the ACE2, the impact of COVID-19 is six times larger, which points towards differences in the severity of the reported respiratory distress. The recognition of specific residues of ACE2 attachments to coronaviruses is important as the residues suggest potential sites of intervention to inhibit attachment and subsequent entry of the COVID-19 into human host cells.

Coronaviruses have been posing mild to serious health concerns for the public since their discovery in 1965 [1] . These large positive-stranded RNA viruses were named due to their crown-like appearance observed using electron microscopy [2] . About 200 different coronaviruses have been discovered to date, that infects different creatures, including bats, birds, cattle, dogs, pigs, rodents, monkeys, humans, etc. [3] [4] [5] [6] . The seven coronaviruses found among humans are HCoV-229E, HCoV-NL63, HCoV-OC43, HKU1, SARS-CoV, MERS-CoV, and SARS-CoV-2 [7, 8] . Among the mentioned coronaviruses, the first four are commonly found and generally cause symptoms of common cold, while the uncommonly found latter three can be much more deadly, causing severe pneumonia.

In 2002-03, the SARS-CoV (Severe Acute Respiratory Syndrome) infected around 8000 people around the globe and caused 774 deaths [9] . Since its first emergence in 2012, the MERS-CoV (Middle East Respiratory Syndrome) has infected 2494 people with a fatality of 858 in 27 countries [10] . The SARS-CoV-2, also known as COVID-19, the recently emerged pandemic, was first reported in Wuhan city, China, in late December 2019 [11, 12] . According to the World Health Organization (WHO), the number of globally confirmed cases of COVID-19 is 209,201,939 with 4,390,467 fatalities in 216 countries (as of August 20, 2021) [13] .

The SARS-CoV and SARS-CoV-2 are closely related coronaviruses that are classified as beta-coronaviruses, and both have originated in bats [8, 14] . The SARS-CoV-2 genome exhibits an 80% identity match with the SARS-CoV genome [15] . Another remarkable similarity between them is their host cell entry mechanism. Both coronaviruses utilize spike glycoproteins (S) to enter host cells by binding with cell surface Angiotensin-Converting Enzyme 2 (ACE2) receptors though their spike (S) genes share only 75% sequence similarity [8, [15] [16] [17] [18] . Spike glycoprotein (S), one of the four structural proteins of coronaviruses, is a class I virus membrane fusion protein [19] . The large ectodomain of spike protein comprises of receptor-binding domain S1 and membrane fusion domain S2 [20] [21] [22] . Both the N-terminal domain (NTD) and the C-terminal domain (CTD) of the S1 subunit are attributed to viral host receptor attachment of different coronaviruses [23] [24] [25] .

The S2 subunit, the most conserved region of the spike protein, carries the fusion peptide (FP) along with two heptad repeats (HR1 and HR2) for performing viral and host membrane fusion [19, 26] .

The ACE2, an essential carboxypeptidase of the renin-angiotensin system (RAS), plays a crucial role in maintaining cardiovascular homeostasis [27] . This Type I membrane protein is primarily expressed in the heart, kidneys, intestine, and lungs [28, 29] . As a homolog of ACE, it negatively regulates the RAS system by cleaving AngI into Ang1-9

and AngII into Ang1-7 [28, 30] . Inside a healthy human lung, alveolar epithelial Type II cells are characterized by abundant expression of ACE2 [31] . Downregulation of ACE2 in these cells causes severe lung injury that may be associated with acute respiratory distress syndrome (ARDS) occurring from alveolar collapse due to increased surface tension [32] [33] [34] . Both SARS-CoV and SARS-CoV-2 infection have been shown to cause ARDS in severely ill patients [35, 36] . The introduction of host cell infection by these coronaviruses is marked by the molecular interaction of the spike glycoprotein (S) receptor-binding domain (RBD) with ACE2 cell receptor [37] . This interaction ultimately leads to the invasion of the host cell by the virus replicating machinery.

The in vivo folding behavior of proteins contributes to their effective functioning [38, 39] , and variation in temperature and pH impacts the folded conformation [40, 41] . Cellular motion-induced mechanical stretching in the extracellular matrix, muscle, and cell receptors also result in protein unfolding [42] [43] [44] . As downregulation of ACE2 cell receptors with cyclic stretching of human lung epithelial cells may be associated with ARDS in case of coronavirus infections (both SARS-CoV and SARS-CoV-2), the molecular interactions and unfolding pathway of ACE2 with and without the presence of spike receptor-binding domain can highlight the deviation of ACE2 behavior due to viral infections [32, 45, 46] . This change in behavior can be modeled through pairwise nonbonded interactions and mechanical response to external forces. Molecular dynamics (MD) simulation is a computational technique that predicts the time-dependent behavior of a molecular system in terms of energy (bonded and non-bonded) and conformation.

MD simulations have been employed to investigate different material systems i.e., oil shale [47] [48] [49] , swelling clays [50, 51] , and proteins [52] . The interactions within coronaviral RNA dependent RNA polymerase (RdRp) have also been analyzed using MD simulations [53] .Steered molecular dynamics (SMD) is an in silico mechanobiological methodology for investigating the mechanical response of proteins during unfolding as well as the unbinding procedure of ligands from them [54] [55] [56] . In the current study, we report molecular dynamics simulations and steered molecular dynamics simulations of human ACE2 in the proximity of both SARS-CoV and SARS-CoV-2 spike (S) protein receptor-binding domain (RBD) to determine their pairwise non-bonded interactions and effect of these interactions on the mechanical response of ACE2 respectively. We also utilize SMD to explore the binding forces of coronavirus spike RBDs to ACE2. Since ACE2 is the primary cellular receptor for the SARS-CoV and the SARS-CoV-2, any changes in the mechanisms of attachment of ACE2 with SARS-CoV and SARS-CoV-2 spike (S) protein receptor-binding domains (RBD) is relevant to the understanding of the host cell invasion and for developing interventions to prevent attachment.

The initial three-dimensional structures of SARS-CoV spike RBD with ACE2 and SARS-CoV-2 spike RBD with ACE2 have been obtained from RCSB Protein Data Bank. The SARS-CoV model has been developed using X-ray diffraction data [57] while the model for SARS-CoV-2 was constructed using cryo-Electron Microscopy data [58] . Both of these models were experimentally validated before submitting to Protein Data Bank. The corresponding PDB ID of SARS-CoV spike RBD-ACE2 complex and SARS-CoV-2 spike RBD-ACE2 complex are 2AJF and 6M17, respectively [57, 58] . The models were chosen due to their availability and similarity i.e., both models utilized ACE2 homodimer. The SARS-CoV complex model (2AJF) contains two spike-RBD chains (E and F) bound with two ACE2 protein chains (A and B) (Fig. 1) .

The SARS-CoV-2 model comprises of two spike-RBD chains (E and F) with ACE2 dimer (chain B and D) along with the neutral amino acid transporter B 0 AT1 (chain A and C). As the primary objective of this study is to investigate the interactions between coronavirus spike RBD and ACE2, we have removed the amino acid transporter B 0 AT1 from model 6M17 (Fig. 2) . 

Molecular dynamics (MD) and steered molecular dynamics (SMD) simulations were performed using NAMD 2.12, a parallel molecular dynamics code [59] . NAMD was developed by the Theoretical and Computational Biophysics Group at the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. All the parameters were obtained from CHARMm (Chemistry at HARvard Macromolecular Mechanics) force field [60] . It consists of functions and constants to define energy expression. CHARMm uses both bonded and non-bonded interaction terms. In this study, we utilize non-bonded interactions. At first, both models are minimized at 0 K temperature and 0 bar pressure using conjugate-gradient method [61] .

Further, both models are brought to 310 K temperature and 1.01325 bar pressure to mimic the human physiological condition. The models are run for five ns using a timestep of 0.5 fs until they reach the equilibrium condition. Thermodynamic and conformational equilibration of structures are characterized by total energy, and root mean squared deviation (RMSD), respectively. These equilibrated models are further utilized for nonbonded energy calculations and steered molecular dynamics (SMD) simulations.

Constant velocity SMD of ACE2 is done to assess its mechanical behavior by pulling its one terminal while keeping the other terminal fixed. In case of SARS-CoV model, the Nterminal was pulled and the C-terminal was kept fixed with and without the presence of spike RBD (Fig. 3a) . In the SARS-CoV-2 model, the boundary atom of peptidase domain (residue id 615) remained fixed while pulling its N-terminal (Fig. 4a) . 

The full-length ectodomain structure of human ACE2 is characterized by the claw-like Nterminal peptidase domain and C-terminal collectrin domain [62] . In this case, the Nterminal peptidase domain of ACE2 serves as the cellular receptor of concave surfaced SARS-CoV spike receptor-binding domain (RBD) [57] . The SARS-CoV spike RBD is 174 residues long, with the terminal residues being cysteine (CYS) and glutamic acid (GLU) [57] . Each unit of ACE2 homodimer consists of 597 residues with serine (SER) and

aspartic acid (ASP) as the terminal residues. The spike RBD attaches to each ACE2 protomer resulting in the complex formation with two spike protein chains (E and F) with Table 5 . These specific residues have been chosen based on the suggestion of structure resolving study [57] and they been shown to contribute 29.2% (-48.5 kcal/mol) of the total interaction energy.

In terms of the secondary structure, turn and coil components of ACE2 contribute to the majority (62%) of the non-bonded interactions (-103 kcal/mol). The rest of the interactions originate primarily from the alpha-helices of ACE2 (Supplementary Table 2 ). In terms of the tertiary structure, more than 99% of the interactions arise from the polar residues of ACE2. Also, the non-bonded interactions between the two chains of ACE2 (A and B) are calculated as -285 kcal/mol in the absence of SARS-CoV spike RBD. This interaction energy is reduced to -107 kcal/mol in the ACE2-spike RBD complex. In both cases, the interactions are predominantly electrostatic.

The constant velocity steered molecular dynamics (SMD) [59, 63] method has been used to investigate the mechanical response of ACE2 through its modeling of unfolding to external loading. As shown in Figure 6a , Chain B of ACE2 is stretched at constant velocity The binding force of SARS-CoV spike RBD and the human ACE2 has also been explored utilizing the constant velocity SMD simulations. The C-terminal of SARS-CoV spike RBD chain F is pulled away from ACE2 chain B to determine the amount of force required to separate the spike RBD from the spike RBD-ACE2 complex (Fig. 6(c) ). The detachment of the spike RBD from ACE2 is seen as two important unlatching events at about 20 Å and 121 Å displacement, which are indicated by two sharp peaks in the forcedisplacement curve. The first peak is observed at 8,030 pN force at 20 Å (Fig. 6(d) ). The 

The spike (S) protein of SARS-CoV-2 attaches to the peptidase domain (PD) of human ACE2 [58] . Due to the homodimerization of ACE2, two spike (S) protein receptor-binding domains (RBDs) attach to ACE2 dimer, where each PD binds with one RBD. Chain B and D of ACE2 are attached to spike RBD chain E and F, respectively (Fig.2) . The SARS-CoV-2 spike RBD consists of 183 residues with cysteine (CYS) at the N-terminal and leucine (LEU) at the C-terminal. The ACE2 protomer is a full-length model having 748

residues, which N-terminal is isoleucine (ILE), and C-terminal is arginine (ARG). [58] The peptidase domain consists of about 80% of the total residues of ACE2 (residues 21 to 615). The non-bonded interactions between every residue of chain B of ACE2 and chain E of SARS-CoV-2 spike RBD have been computed to probe the interactions within the SARS-CoV-2 spike RBD-ACE2 complex and are shown in Fig. 7(a) and Supplementary   Table 3 . The total non-bonded interactions between them (chain B of ACE2 and SARS-CoV-2 spike RBD chain E) are observed to be -356 kcal/mol with electrostatic and VDW interactions of -289 kcal/mol and -67 kcal/mol respectively. Among the 20 different residues of ACE2, two residues contribute to almost 86% of the total non-bonded attractive interactions between ACE2 and SARS-CoV-2 RBD (Fig. 7(b) ). The non-bonded interactions of specific residues of ACE2 with SARS-CoV-2 spike RBD are provided in Supplementary Table 5 . These specific residues have been chosen based on the suggestion of structure resolving study [58] and they been shown to contribute 43.2% (-153.8 kcal/mol) of the total interactions. Table 4 ). Polar residues of ACE2 participate in more than 99% of these interactions. Also, the total non-bonded interactions within the ACE2 dimer (between chain B and D) are observed to be -648 kcal/mol and -808 kcal/mol in the absence and presence of SARS-CoV-2 spike RBD respectively. In both cases, the electrostatic interactions produce more than 80% of the total interactions.

The (Fig. 8(b) ). The peak of 3,803 pN at a displacement of 112 Å marks the complete unfolding of α1 helix. The unfolding of α2 helix is accomplished The C-terminal of SARS-CoV-2 spike RBD chain E is pulled away from ACE2 chain B to evaluate the binding force of spike RBD within the SARS-CoV-2 spike RBD-ACE2 complex (Fig. 8c) . The detachment of the spike RBD from ACE2 is characterized by two unlatching events at about 46.2 Å and 129 Å, which are indicated by two sharp peaks in the force-displacement curve. The first peak has a maximum force of 6713.18 pN at 46.2 Å displacement (Fig. 8d) . The highest peak of 7759.95 pN at a displacement of 129 Å signifies the complete separation of spike RBD chain E from the SARS-CoV-2 spike RBD-ACE2 complex. The remainder of the plot is the mechanical response of spike RBD chain E alone.

The simulations indicate that electrostatic interactions dominate the non-bonded interactions between coronavirus spike RBD and ACE2 for both the SARS-CoV-2 and SARS-CoV. One of the major differences between the two coronaviruses spike-RBD interaction with ACE2 is that in the case of SARS-CoV-2, the majority of attractive interaction energies are primarily mediated by just two residues of ACE2; GLU and ASP, whereas, for SARS-CoV, the attractive interaction energies are spread out over four residues GLU, ASP, LYS and GLN (Fig. 5a & 5b and Fig. 7a & 7b) . The two residue Certain residues (K31, M82, R357) were found to be oppositely interacting (attraction/repulsion) between SARS-CoV and SARS-CoV-2 spike RBD.

The mechanical response of ACE2 is obtained from the force-displacement plot of the complex using constant velocity pulling ( Fig. 6b and Fig. 8b) . The sharp peaks in the plot result from the unfolding of helices or coils in ACE2. The ACE2 force-displacement curve shifts upward, i.e., increased force is needed to cause the same displacement when the coronaviruses are attached to the ACE2 for both the coronaviruses. This force increment is the result of spike RBD binding interactions with the ACE2 peptidase domain (PD) that results in changes to the ACE2 unfolding behavior making the response stiffer. The force needed to pull the RBD by itself is found to be around 1000 pN, and this magnitude of the force is subtracted from the net force displacement of the RBD-ACE2 complex for both the coronaviruses. The plots shown in Fig. 6b and The binding force of spike RBD towards the ACE2 cell receptor is evaluated from the force-displacement behavior obtained by pulling the spike RBD at a constant velocity. We observe two peaks in the force-displacement plots for both coronaviruses resulting from two specific unlatching events that lead to the detachment of the spike RBD from ACE2.

Both the spike RBDs from the two coronaviruses exhibit a two-step unlatching, leading to detachment, as shown in Supplementary Figures 1 and 2 ACE2 occurs at a lower force than the SARS-CoV. The mechanical pull-off of the coronavirus RBD from ACE2 enabled using SMD is a different phenomenon than measurement of dissociation constant using surface plasmon resonance [64, 65] . On pulling the spike RBD from the attachment to ACE2, conformational changes begin in the RBD, which progressively reduce interaction energies at the ACE2-spike RBD interface.

These changes are influenced by factors like ACE2-RBD interface area, two-residue interactions of SARS-CoV-2 spike RBD as opposed to multi-residue interaction of SARS-CoV spike RBD with ACE2, RBD structure, unfolding rate of RBD, ACE2 conformation etc. As seen in Figures 6d and 8d , the pulled terminal of SARS-CoV-2 spike RBD needed to be displaced by a larger distance (580 Å) than SARS-CoV (500 Å) for the force magnitude during pulling to become zero, indicating the influence of factors described above on the deformation. It appears that these factors likely cause faster reduction of non-bonded interactions in spike-ACE2 complex during the pulling of SARS-CoV-2 RBD compared to SARS-CoV RBD, resulting in lower unlatching force of SARS-CoV-2 spike RBD.

Since SMD simulations presented here mimic single protein pulling experiments with AFM, it is suggested that future AFM experiments that evaluate binding mechanisms of these complexes consider the effect of the unfolding of spike RBD. SMD simulations are thus a useful methodology to observe an accurate sequence of events in the pulling away of the spike RBD from ACE2. Understanding the important biological consequence of the formation of the ACE2-coronavirus spike RBD complex is aided by this additional viewpoint of stiffening of ACE2 on attachment to coronavirus, and the pulling force of the spike RBD from the ACE2-coronavirus spike RBD complex.

This study utilizes computational techniques to explain the initial host cell response due to coronavirus infections. It has been performed by capturing the molecular interactions and changes in the mechanical response of coronavirus cellular receptor angiotensinconverting enzyme 2 (ACE2) in the presence of SARS-CoV and SARS-CoV-2 spike (S) protein receptor-binding domain (RBD). Molecular dynamics (MD) simulation has been employed to determine the non-bonded interactions, while steered molecular dynamics (SMD) was used to describe the mechanical response of ACE2. The binding force of coronavirus spike (S) RBDs from the ACE2 has also been investigated by using SMD.

The major findings of our study are summarized below:

 Of the attractive non-bonded interactions of SARS-CoV-2 RBD with ACE2, 86% result from just two ACE2 residues; GLU and ASP. On the other hand, 89% of the SARS-CoV spike RBD interaction energy is spread over four ACE2 residues, including LYS and GLN, besides GLU and ASP. These observations suggest potential sites of intervention to inhibit attachment of spike RBD to ACE2.

 The non-bonded interaction energies between SARS-CoV-2 spike RBD and ACE2 are more than twice the interaction energies between SARS-CoV spike RBD and ACE2.

 The pull-off force of the spike RBD from the ACE2 is higher in magnitude for the SARS-CoV. On pulling the spike RBD from the attachment of ACE2, continuous conformational changes begin in the RBD, which progressively reduce interaction energies at the ACE2-spike RBD interface and hence influence total pull-off force.

 The attachment of spike RBD with ACE2 results in the stiffening of ACE2 for both SARS-CoV-2 and SARS-CoV. The relative change in stiffness due to the attachment of spike RBD is higher for SARS-CoV-2 (54%) compared to the SARS-CoV (9%). The significantly larger relative stiffness of ACE2 on the SARS-CoV-2 attachment as compared to the SARS-CoV attachment points towards differences in the biological response of ACE2.

Since the host entry modes of the two coronaviruses compared here are similar, it is interesting to note the differences in the mechanisms of interactions, two ACE2 residues for SARS-CoV-2 versus multiple attachment residues for SARS-CoV. The stronger nonbonded interaction energies between SARS-CoV-2 and ACE2 result in a much stiffer ACE2 on attachment to the coronavirus spike RBD than for the SARS-CoV. Overall, the evaluation of these mechanisms of attachment and the resulting binding forces are critical to the development of therapies beyond vaccines that prevent the attachment and subsequent entry into host cells. Further mechanobiological studies that relate mechanical changes to the severity of the ARDS would provide a definitive answer to this important health concern.

Coronaviridae: The viruses and their replication, Fundamental virology

Detection of Novel SARS-like and Other Coronaviruses in Bats from Kenya

Discovery of Seven Novel Mammalian and Avian Coronaviruses in the Genus Deltacoronavirus Supports Bat Coronaviruses as the Gene Source of Alphacoronavirus and Betacoronavirus and Avian Coronaviruses as the Gene Source of Gammacoronavirus and Deltacoronavirus

DBatVir: the database of bat-associated viruses

Coronaviruses: An Overview of Their Replication and Pathogenesis

A pneumonia outbreak associated with a new coronavirus of probable bat origin

A dynamic compartmental model for the Middle East respiratory syndrome outbreak in the Republic of Korea: A retrospective analysis on control interventions and superspreading events

A novel coronavirus genome identified in a cluster of pneumonia cases-Wuhan

China Novel, A Novel Coronavirus from Patients with Pneumonia in China

Coronavirus disease 2019 (COVID-19): situation report

Discovery of a rich gene pool of bat SARS-related coronaviruses provides new insights into the origin of SARS coronavirus

Genomic characterisation and epidemiology of 2019 novel coronavirus: implications for virus origins and receptor binding

Angiotensin-converting enzyme 2 is a functional receptor for the SARS coronavirus

SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor

Structure of the SARS-CoV-2 spike receptor-binding domain bound to the ACE2 receptor

The coronavirus spike protein is a class I virus fusion protein: Structural and functional characterization of the fusion core complex

Pre-fusion structure of a human coronavirus spike protein

Cryoelectron microscopy structure of a coronavirus spike glycoprotein trimer

Architecture of the SARS coronavirus prefusion spike

Receptor Usage and Cell Entry of Porcine Epidemic Diarrhea Coronavirus

Porcine Aminopeptidase N is a Functional Receptor for the PEDV Coronavirus

A 193-amino acid fragment of the SARS coronavirus S protein efficiently binds angiotensin-converting enzyme 2

Coronavirus spike proteins in viral entry and pathogenesis

Angiotensin-Converting Enzyme 2 Metabolizes and Partially Inactivates Pyr-Apelin-13 and Apelin-17 Physiological Effects in the Cardiovascular System

A novel angiotensin converting enzyme related carboxypeptidase (ACE2) converts angiotensin I to angiotensin 1-9

Tissue distribution of ACE2 protein, the functional receptor for SARS coronavirus. A first step in understanding SARS pathogenesis

A human homolog of angiotensin-converting enzyme -Cloning and functional expression as a captopril-insensitive carboxypeptidase

Single-cell RNA expression profiling of ACE2, the receptor of SARS-CoV-2

Angiotensinconverting enzyme 2 protects from severe acute lung failure

Pulmonary surfactant

Role of angiotensinconverting enzyme 2 (ACE2) in COVID-19

A novel coronavirus associated with severe acute respiratory syndrome

Clinical course and outcomes of critically ill patients with SARS-CoV-2 pneumonia in Wuhan, China: a single-centered, retrospective, observational study, The Lancet

Cellular entry of the SARS coronavirus

Theory of protein folding: The energy landscape perspective

Mechanisms of protein folding

Simulations of protein folding and unfolding

Pathways to a protein folding intermediate observed in a 1-microsecond simulation in aqueous solution

Folding-unfolding transitions in single titin molecules characterized with laser tweezers

Reversible unfolding of individual titin immunoglobulin domains by AFM

Elasticity and unfolding of single molecules of the giant muscle protein titin

Recombinant human ACE2: acing out angiotensin II in ARDS therapy

Mechanical Stress and the Induction of Lung Fibrosis via the Midkine Signaling Pathway

An insight into quartz mineral interactions with kerogen in Green River oil shale

Modeling the Behavior of Organic Kerogen in the Proximity of Calcite Mineral by Molecular Dynamics Simulations

Modeling the Nanoscale Kerogen Inclusions in Green River Oil Shale, Poromechanics VI

Molecular mechanics of the swelling clay tactoid under compression, tension and shear

Molecular Origin of Compressibility and Shear Strength of Swelling Clays, International Conference of the International Association for Computer Methods and Advances in Geomechanics

Dissociation Mechanisms of G-actin Subunits Govern Deformation Response of Actin Filament

Differences in Interactions Within Viral Replication Complexes of SARS-CoV-2 (COVID-19) and SARS-CoV Coronaviruses Control RNA Replication Ability

Steered molecular dynamics investigations of protein function

Atomic force microscope

Biological applications of optical forces

Structure of SARS coronavirus spike receptor-binding domain complexed with receptor

Structural basis for the recognition of SARS-CoV-2 by full-length human ACE2

Scalable molecular dynamics with NAMD

CHARMM General Force Field: A Force Field for Drug-Like Molecules Compatible with the CHARMM All-Atom Additive Biological Force Fields

Iterative minimization techniques for abinitio total-energy calculations -molecular-dynamics and conjugate gradients

ACE2 X-ray structures reveal a large hinge-bending motion important for inhibitor binding and catalysis

Computational Molecular Dynamics: Challenges, Methods, Ideas

Cryo-EM structure of the 2019-nCoV spike in the prefusion conformation

Stabilized coronavirus spikes are resistant to conformational changes induced by receptor recognition or proteolysis

Pathophysiological characteristics and therapeutic approaches for pulmonary injury and cardiovascular complications of coronavirus disease

The authors gratefully acknowledge the use of computational resources at the North Dakota State University Center for Computationally Assisted Science and Technology 2) Tables containing the magnitudes of interaction energies between the spike-RBDs and ACE2.