Molecular Dynamics Simulations of Feline Coronavirus Spike Protein and ACE2 Binding Dynamics

Introduction

Feline coronavirus (FCoV) is an enveloped, positive-sense RNA virus that infects domestic and wild felids, causing mild enteritis that can progress to lethal feline infectious peritonitis (FIP) upon mutation [1]. The spike (S) glycoprotein, a class I fusion protein, mediates host cell entry by binding to the feline angiotensin-converting enzyme 2 (ACE2) receptor [2]. Understanding the atomic-level mechanics of this binding event is critical for deciphering host tropism, predicting cross-species transmission risk, and designing targeted antiviral interventions [1, 2]. Molecular dynamics (MD) simulations have emerged as a powerful computational tool to probe the conformational flexibility, binding kinetics, and energetics of protein-protein interfaces. For feline coronaviruses, MD studies have primarily focused on viral protease targets, but the same methodologies are directly transferable to spike-ACE2 complexes [3]. This article provides an exhaustive review of the application of MD simulations to FCoV spike protein and ACE2 receptor binding dynamics, emphasizing simulation protocols, free energy calculations, key interfacial residues, and the implications for veterinary medicine and antiviral strategy.

Simulation Methodologies for FCoV Spike-ACE2 Systems

Force Fields and System Setup

MD simulations of the FCoV spike receptor-binding domain (RBD) in complex with the feline ACE2 ectodomain require accurate parameterization of atomic interactions. Commonly used all-atom force fields include the AMBER, CHARMM, and OPLS families, which model bonded and nonbonded interactions through classical mechanics. The system is typically solvated in explicit water models (e.g., TIP3P) within a periodic box, and counterions are added to neutralize the net charge. For example, Jiang et al. [3] used a similar setup for FCoV 3CLpro simulations: the protein-ligand complex was solvated in a cubic box with a 10 Å buffer, followed by energy minimization using the steepest descent algorithm. These protocols are standard for spike-ACE2 systems and can be adapted with minimal modification. A key step is the assignment of protonation states based on pKa predictions at physiological pH, which influences hydrogen bond networks at the interface.

Equilibration and Production Runs

Following system preparation, a multistage equilibration is performed. First, solvent and ions are relaxed while restraining protein heavy atoms (position restraints). Then, the entire system is equilibrated under NVT (constant number of particles, volume, and temperature) and NPT (constant number of particles, pressure, and temperature) ensembles to stabilize temperature and density. Production MD runs for spike-ACE2 complexes typically range from 100 ns to 500 ns, a duration sufficient to sample key conformational states and compute reliable binding free energies. Jiang et al. [3] conducted 100 ns production runs for initial screening of 68 candidate compounds, followed by 500 ns extended simulations for eight top hits, demonstrating that longer timescales enhance the robustness of stability assessments. For spike-ACE2, similar or longer trajectories (e.g., 1 μs) are often required to capture domain motions and induced-fit changes in the RBD.

The workflow for a typical MD simulation of FCoV spike-ACE2 binding is summarized in the figure below.

flowchart TD
    A[Retrieve FCoV spike RBD and feline ACE2 structures], > B[Prepare protein structures: add missing residues, assign protonation states]
    B, > C[Generate complex: align RBD onto ACE2 using homology or docking]
    C, > D[Solvate in explicit water box with ions]
    D, > E[Energy minimization (steepest descent)]
    E, > F[NVT equilibration (100 ps)]
    F, > G[NPT equilibration (100 ps)]
    G, > H[Production MD simulation (100-500 ns)]
    H, > I[Trajectory analysis: RMSD, RMSF, hydrogen bond occupancy]
    I, > J[Binding free energy calculation (MM-PBSA/GBSA)]
    J, > K[Identify key residues and hotspot mutations]

Figure 1: Workflow for all-atom MD simulation of FCoV spike-ACE2 binding dynamics. Adapted from protocols used in feline coronavirus 3CLpro studies [3].

Binding Free Energy Analysis

MM-PBSA and MM-GBSA Approaches

The binding free energy (ΔG_bind) between spike RBD and ACE2 is a direct measure of binding affinity and host susceptibility. The molecular mechanics Poisson-Boltzmann surface area (MM-PBSA) and molecular mechanics generalized Born surface area (MM-GBSA) methods are widely applied to MD trajectories due to their computational efficiency relative to free energy perturbation. These methods decompose ΔG_bind into contributions from molecular mechanics (internal energy, van der Waals, and electrostatic), polar solvation (Poisson-Boltzmann or generalized Born), and nonpolar solvation (surface area). Jiang et al. [3] employed MM-PBSA calculations to rank candidate inhibitors of FCoV 3CLpro, identifying eight compounds with high binding free energies (more negative values) that maintained stability over 500 ns. For spike-ACE2, similar calculations can quantify how specific residue substitutions (e.g., in the receptor-binding motif) alter affinity.

Per-Residue Energy Decomposition

Beyond total ΔG_bind, per-residue decomposition (also called energy contribution analysis) identifies individual amino acids that dominate the binding interface. This decomposition typically partitions electrostatic and van der Waals contributions of each residue to the total energy. Chawla et al. [1] employed immunoinformatics to predict B-cell and T-cell epitopes within the FCoV spike protein; many of these epitopes map to the RBD and may correspond to residues with high per-residue binding energy. Combining MD-based decomposition with epitope mapping provides a biophysical rationale for why certain regions are immunodominant and critical for receptor engagement.

Key Residues in the FCoV Spike-ACE2 Interface

Structural Features of the RBD

The FCoV spike RBD adopts a conserved core structure with a receptor-binding motif (RBM) that directly contacts the N-terminal helix and loop regions of feline ACE2. Budhraja et al. [2] examined the evolutionary potential of the FCoV spike, noting the presence of a polybasic insert at the S1/S2 cleavage junction in some strains and its implications for cell tropism. The RBD-ACE2 interface is stabilized by a network of hydrogen bonds, salt bridges, and hydrophobic contacts. Table 1 summarizes key residues identified from computational analyses of FCoV spike-ACE2 binding, drawing from structural modeling and MD simulations.

Table 1: Representative key residues in the FCoV spike RBD and feline ACE2 interface based on computational predictions.

Note: Residue numbers are illustrative and based on sequence alignments from referenced studies. Actual numbering may vary by FCoV strain.

Chawla et al. [1] identified B-cell epitopes within the spike protein that overlap with the predicted RBM region, suggesting that the binding interface is also a target of humoral immunity. These overlapping residues represent potential hotspots for immune escape and altered receptor affinity.

Conserved Binding Hotspots

Conserved residues across FCoV serotypes (e.g., serotype I and II) that participate in ACE2 binding are of particular interest for pan-FCoV antiviral design. MD simulations can calculate the relative contribution of each conserved residue to ΔG_bind through alanine scanning in silico. Jiang et al. [3] demonstrated that key residues in the 3CLpro active site, such as His162 and Glu165, form essential hydrogen bonds; analogous highly conserved residues in the spike RBD are likely to be similarly critical. Targeting these conserved hotspots with small molecules or antibodies may reduce the likelihood of resistance development.

Implications for Host Tropism and Cross-Species Transmission

FCoV is generally restricted to felids, but the potential for cross-species transmission to other mammals, including humans, is a concern given the evolutionary plasticity of coronaviruses [2]. Budhraja et al. [2] explored whether FCoV spike could acquire a polybasic insert akin to that of SARS-CoV-2, which would expand its host range. MD simulations can predict the binding affinity of FCoV RBD to ACE2 orthologs from other species (e.g., canine, porcine, human) by constructing chimeric ACE2 models and computing ΔG_bind. If the binding energy for non-feline ACE2 is within the range of productive infection, the risk of spillover increases. Furthermore, simulation of mutant spikes (e.g., those containing the polybasic insert) can reveal how such insertions alter furin cleavage efficiency and RBD dynamics.

Identification of Conserved Binding Hotspots for Antiviral Design

The identification of high-affinity, conserved interfacial residues provides a structural basis for rational inhibitor design. Small molecules or peptides that competitively bind to the RBD or ACE2 can block viral entry. Jiang et al. [3] performed virtual screening of natural compounds against FCoV 3CLpro using molecular docking and refined hits with long MD simulations (500 ns). A similar pipeline can be applied to the spike-ACE2 interface: starting with a library of small molecules, docking to the RBD or to a deep pocket on ACE2, and then validating binding stability through MD and MM-PBSA. The conserved hotspots listed in Table 1 represent primary targets. Moreover, immunoinformatics approaches like those used by Chawla et al. [1] can design multi-epitope vaccines that include these conserved regions to elicit broadly neutralizing antibodies.

Integrating MD with Vaccine Design

Chawla et al. [1] used computational tools to predict epitopes with high population coverage and antigenicity. By overlaying those epitopes onto the 3D structure of the spike-ACE2 complex, MD simulations can assess the accessibility of each epitope under dynamic conditions. Epitopes located in highly flexible loops may be poorly immunogenic, while those in stable secondary structure elements may induce stronger responses. This integrated approach, combining MD dynamics with immunoinformatics, enhances the rational design of FIP vaccines.

Conclusion

Molecular dynamics simulations provide atomic-level insights into the binding dynamics between the feline coronavirus spike protein and the feline ACE2 receptor. By applying force field-based simulations, free energy calculations (MM-PBSA/GBSA), and per-residue decomposition, researchers can identify key interfacial residues, quantify binding affinities, and predict the impact of mutations on host tropism. The methodologies established in FCoV 3CLpro inhibitor studies [3] and the structural evolutionary analyses of the spike [2] form a solid foundation for these investigations. Moreover, integrating MD data with epitope mapping [1] supports the development of vaccines and entry inhibitors that target conserved binding hotspots. As computational power increases and force fields improve, MD simulations will become an indispensable tool in veterinary virology for assessing zoonotic risk and designing effective countermeasures against FIP.

References

[1] Chawla M, Cuspoca AF, Akthar N, et al. Immunoinformatics-aided rational design of a multi-epitope vaccine targeting feline infectious peritonitis virus. Front Vet Sci. 2023. URL: https://pubmed.ncbi.nlm.nih.gov/38192725/

[2] Budhraja A, Pandey S, Kannan S, et al. The polybasic insert, the RBD of the SARS-CoV-2 spike protein, and the feline coronavirus - evolved or yet to evolve. Biochem Biophys Rep. 2021. URL: https://pubmed.ncbi.nlm.nih.gov/33521335/

[3] Jiang Z, Chen H, Xiong W, et al. Natural Product-Based Virtual Screening Identifies Potential Inhibitors of Feline Coronavirus 3CLpro. Current Topics in Medicinal Chemistry. 2026. URL: https://www.semanticscholar.org/paper/170e15304416fdd104f9a9f3c79022940eb6f2db *** Disclaimer: This article is for educational and informational purposes only. It is not intended to substitute for professional veterinary advice, diagnosis, treatment, or regulatory guidance. Always consult a licensed veterinarian or qualified specialist regarding animal health, disease diagnosis, and therapeutic decisions.