Spike Protein Glycan Shield Evolution: Molecular Dynamics Simulation of Immune Evasion in Emerging Coronaviruses

Introduction

The spike glycoprotein of coronaviruses mediates host cell entry and is the primary target of neutralizing antibodies. A critical feature of this protein is its dense coating of N-linked glycans, collectively termed the glycan shield. This shield modulates antibody accessibility and can evolve under immune pressure, enabling viral escape [1, 2]. Understanding the dynamic behavior of the glycan shield at atomic resolution is essential for predicting immune evasion in emerging coronaviruses, including those of veterinary and zoonotic concern. Molecular dynamics (MD) simulations provide a powerful computational framework to model glycan conformational ensembles, quantify their interactions with antibodies, and map the evolutionary trajectories that alter shielding patterns [3, 2]. This article reviews the biophysical principles of glycan shield evolution, the computational methods used to simulate it, and the implications for vaccine design and therapeutic antibody development in veterinary contexts.

Glycan Shield Composition and Dynamics

Coronavirus spike proteins are heavily glycosylated, with each protomer harboring 20 to 30 N-glycosylation sequons depending on the species and variant [3, 4]. The glycan shield is not a static barrier; individual glycans sample multiple conformations on microsecond to millisecond timescales, creating a fluctuating landscape that can occlude or expose underlying protein epitopes [1, 2]. For example, the N343 glycan on the receptor-binding domain (RBD) of SARS-CoV-2 spike has been shown to modulate RBD conformational dynamics and co-receptor binding [3]. In the swine acute diarrhea syndrome coronavirus (SADS-CoV), cryo-electron microscopy (cryo-EM) structures revealed a unique glycan arrangement that differs from other coronaviruses, suggesting lineage-specific shielding strategies [4].

The composition of the glycan shield is determined by the host glycosylation machinery, which varies across species. This host-dependent glycosylation can influence cross-species transmission and immune recognition [4]. In emerging coronaviruses, mutations that introduce or remove glycosylation sites are frequently observed in variants that escape antibody neutralization [1, 5]. For instance, the Omicron lineage of SARS-CoV-2 acquired multiple glycan-related mutations that reorganized epitopes and reduced the efficacy of monoclonal antibodies such as sotrovimab [1]. Similarly, in the JN.1-derived subvariants LB.1, KP.2.3, KP.3, and KP.3.1.1, changes in glycan positioning were associated with altered spike stability and neutralization profiles [5].

Molecular Dynamics Simulation Approaches

MD simulations model the time-dependent behavior of atoms and molecules using classical force fields. For glycan shield studies, all-atom simulations with explicit solvent are typically employed, using packages such as GROMACS or AMBER [3, 2]. The CHARMM36 or GLYCAM force fields are commonly used for carbohydrate parameters. Simulations are initiated from cryo-EM or X-ray crystallography structures (e.g., PDB 6VXX, 6VYB for SARS-CoV-2 spike) and run for hundreds of nanoseconds to several microseconds to capture glycan conformational sampling [3, 2].

Free energy calculations, including molecular mechanics Poisson-Boltzmann surface area (MM-PBSA) and free energy perturbation (FEP), quantify the energetic contributions of individual glycans to antibody binding [1, 3]. These methods can predict how mutations at glycosylation sites alter the binding affinity of neutralizing antibodies. For example, Kumar et al. used MD simulations and free energy analysis to demonstrate that glycan shielding and epitope reorganization drive resistance to sotrovimab in Omicron variants [1]. Similarly, von Bülow et al. combined MD with antibody accessibility calculations to show that surface mutations in SARS-CoV-2 variants preferentially occur at positions with low glycan occlusion, consistent with immune pressure [2].

Rosetta-based modeling is also used to predict the effect of glycosylation on protein stability and to design glycan shield modifications [6]. Ishimaru et al. identified a conserved epitope in subdomain 1 of the spike that is partially shielded by glycans; MD simulations helped rationalize why this region remains vulnerable to broad neutralization [6].

The following Mermaid diagram summarizes a typical computational workflow for studying glycan shield evolution:

flowchart TD
    A[Sequence Surveillance Data: GISAID, NCBI], > B[Structural Modeling: Rosetta, AlphaFold2]
    B, > C[MD Simulation Setup: GROMACS, AMBER]
    C, > D[All-Atom MD Production Runs: 0.5-5 µs]
    D, > E[Trajectory Analysis: Glycan Conformational Sampling]
    D, > F[Free Energy Calculations: MM-PBSA, FEP]
    E, > G[Antibody Accessibility Mapping]
    F, > G
    G, > H[Identification of Immune Evasion Mutations]
    H, > I[Predictive Models for Vaccine Design]
    I, > J[Experimental Validation: Cryo-EM, Neutralization Assays]

Case Studies: SARS-CoV-2 Variants and Bat Coronavirus Precursors

SARS-CoV-2 Variants

The rapid emergence of SARS-CoV-2 variants has provided a natural experiment in glycan shield evolution. The Omicron lineage, in particular, accumulated multiple mutations near glycosylation sites. Kumar et al. demonstrated that the N440K and G446S mutations in the RBD, combined with altered glycan conformations at N343, reduced the binding of sotrovimab by over 100-fold [1]. MD simulations revealed that these mutations shifted the equilibrium of the N343 glycan toward a conformation that sterically blocked antibody access [1]. Li et al. extended these findings to JN.1-derived subvariants, showing that changes in the glycan shield contributed to reduced neutralization by vaccine-elicited sera [5]. The role of N343 glycosylation was further dissected by Ives et al., who used MD to show that this glycan stabilizes the RBD in an open conformation and modulates binding to the co-receptor ACE2 [3].

Bat Coronavirus Precursors

Bat coronaviruses are considered ancestral reservoirs for many zoonotic coronaviruses. Structural studies of bat coronavirus spike proteins, such as those from RaTG13 and related strains, reveal glycan shields that differ from SARS-CoV-2 in both density and positioning [4]. Guan et al. solved the cryo-EM structure of the SADS-CoV spike, which infects swine, and noted a unique glycan arrangement that may facilitate immune evasion in its porcine host [4]. MD simulations comparing bat and human coronavirus spikes can identify glycan sites that are conserved or variable, informing predictions of zoonotic spillover risk. For example, the N-linked glycan at position 343 is conserved across sarbecoviruses, but its conformational dynamics vary with sequence context [3]. Computational modeling of bat coronavirus spike-receptor interactions, as discussed in related articles on this portal (e.g., Molecular Dynamics Simulations of Bat Coronavirus Spike Protein-Receptor Interactions: Implications for Zoonotic Risk Assessment), relies on MD to assess how glycan shielding affects receptor binding and antibody escape.

Implications for Vaccine Design and Therapeutic Antibody Development

Understanding glycan shield evolution has direct implications for veterinary vaccine design. For coronaviruses affecting livestock and companion animals, such as porcine epidemic diarrhea virus (PEDV), transmissible gastroenteritis virus (TGEV), and feline coronavirus, the glycan shield can limit the efficacy of spike-based vaccines. Computational approaches that predict glycan shielding dynamics can guide the selection of epitopes that are less prone to occlusion [6, 2]. For instance, Ishimaru et al. identified a conserved epitope in subdomain 1 that remains accessible despite glycan shielding, providing a template for broadly neutralizing antibody design [6].

Therapeutic antibodies targeting the spike protein must contend with glycan-mediated resistance. MD simulations can pre-screen antibody candidates by modeling their binding in the presence of dynamic glycans [1, 2]. Free energy calculations can rank antibodies based on their predicted resilience to glycan shielding mutations. This approach is particularly valuable for developing pan-coronavirus therapeutics that target conserved, glycan-poor regions [6].

Sequence surveillance data from platforms such as GISAID and NCBI are essential for tracking glycan site mutations in real time. Integrating these data with structural modeling and MD simulations enables proactive identification of emerging variants with altered shielding properties [5, 2]. The computational framework described here is also applicable to other enveloped viruses, as discussed in related articles on Glycan Shield Engineering and the Computational Prediction of Immune Escape in Enveloped Viruses and Structural Bioinformatics of Viral Glycoprotein Glycan Shield Evasion.

Conclusion

The glycan shield of coronavirus spike proteins is a dynamic and evolvable barrier that plays a central role in immune evasion. Molecular dynamics simulations, combined with free energy calculations and structural bioinformatics, provide a detailed understanding of how glycan conformations and mutations alter antibody accessibility. Studies on SARS-CoV-2 variants and bat coronavirus precursors have revealed specific mechanisms of glycan-mediated resistance, including epitope reorganization and steric occlusion [1, 3, 2]. These insights are directly transferable to veterinary coronaviruses, where similar shielding strategies may operate. Continued integration of computational modeling with experimental structural biology and genomic surveillance will be critical for anticipating future immune escape and for designing robust vaccines and therapeutics.

References

[1] Kumar A, Yadav AJ, Tripathi T, et al. Glycan shielding and epitope reorganization drive sotrovimab resistance in SARS-CoV-2 Omicron variants. Arch Biochem Biophys. 2026. URL: https://pubmed.ncbi.nlm.nih.gov/42128042/

[2] von Bülow S, Sikora M, Blanc FEC, et al. Antibody accessibility determines location of spike surface mutations in SARS-CoV-2 variants. PLoS Comput Biol. 2023. URL: https://pubmed.ncbi.nlm.nih.gov/36693110/

[3] Ives CM, Nguyen L, Fogarty CA, et al. Role of N343 glycosylation on the SARS-CoV-2 S RBD structure and co-receptor binding across variants of concern. Elife. 2024. URL: https://pubmed.ncbi.nlm.nih.gov/38864493/

[4] Guan H, Wang Y, Perčulija V, et al. Cryo-electron Microscopy Structure of the Swine Acute Diarrhea Syndrome Coronavirus Spike Glycoprotein Provides Insights into Evolution of Unique Coronavirus Spike Proteins. J Virol. 2020. URL: https://pubmed.ncbi.nlm.nih.gov/32817223/ *** 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.

[5] Li P, Faraone JN, Hsu CC, et al. Neutralization and spike stability of JN.1-derived LB.1, KP.2.3, KP.3, and KP.3.1.1 subvariants. mBio. 2025. URL: https://pubmed.ncbi.nlm.nih.gov/40136024/

[6] Ishimaru H, Nishimura M, Shigematsu H, et al. Epitopes of an antibody that neutralizes a wide range of SARS-CoV-2 variants in a conserved subdomain 1 of the spike protein. J Virol. 2024. URL: https://pubmed.ncbi.nlm.nih.gov/38624232/