This study aims to investigate the potential mechanisms by which the fusion proteins EFK8-mPhl p1-GS-mBet v1 and EFK8-mDer p1-GS-mDer p2 reduce IgE binding affinity and cytotoxicity through computational structural biology methods. Preliminary experiments indicated that after self-assembly of the fusion peptide EFK8 (FEFEFKFK), the IgE binding affinity and cytotoxicity of allergenic proteins were significantly reduced. We hypothesized that EFK8-mediated self-assembly may form oligomers that mask or alter the spatial conformation of antigenic epitopes, thereby diminishing their immunogenicity. This study employed Alphafold3 and HADDOCK 2.4 for structural prediction and interaction analysis of the fusion proteins and their oligomers to validate this hypothesis.

1. Structure Prediction and Modeling Strategy

• Monomeric Fusion Protein Modeling:

Using AlphaFold3 (https://alphafoldserver.com/) to predict the three-dimensional structures of the monomeric allergen proteins, complete EFK8-mPhl p1-GS-mBet v1 and EFK8-mDer p1-GS-mDer p2 monomers, and their dimers. Focus on analyzing the influence of the Linker region (GSGSGSGS) on the spatial conformation of monomeric hypoallergenic proteins and the potential oligomerization tendency of the EFK8 peptide segment.

• Dimer Interaction Analysis:

Flexible docking analysis was performed using HADDOCK 2.4 (HADDOCK Web Server) for the homodimers constructed from EFK8-mPhl p1-GS-mBet v1 and EFK8-mDer p1-GS-mDer p2. The cluster with the lowest HADDOCK score was selected for energy decomposition and interface analysis, including van der Waals forces, electrostatic interactions, desolvation energy, and buried surface area (BSA).

2. Key Parameters and Validation Metrics

•HADDOCK Scoring System:

Integrates van der Waals energy, electrostatic potential energy, and constraint violation energy to assess structural significance via Z-Score evaluation.

• Spatial Hindrance Assessment:

Evaluates binding site accessibility by measuring atomic distances between key sites and changes in solvent-accessible surface area (SASA).

1. EFK8-mPhl p1-GS-mBet v1

Molecular docking analysis of this complex revealed a low HADDOCK score (-100.0 ± 5.6) and a significant electrostatic potential contribution (-191.4 ± 22.8), indicating that the assembly process is highly dependent on charge complementarity. This aligns with the Alphafold3 prediction concluding that “electrostatic complementarity dominates assembly.” Additionally, the large buried surface area (2176.1 ± 26.7 Ų) and van der Waals energy (-57.2 ± 2.5) further indicate a tightly bound and stable dimer interface. More importantly, this assembly directly leads to spatial shielding of the N-terminal β-sheet (CFEIKCT) in mPhl p1, reducing exposed hydrophilic residues and mechanistically explaining the decreased IgE affinity. Furthermore, the high constraint violation energy (126.3 ± 28.0) suggests potential conformational tension in the linker region, providing direction for subsequent optimizations in linker design. Integrating the diagrams with Alphafold3 predictions reveals that EFK8-guided oligomerization (particularly during higher-order polymer formation) significantly reduces IgE binding capacity through steric hindrance and epitope masking, while electrostatic complementarity serves as the key energy driver for this assembly process. This discovery provides crucial structural biology insights for designing hypoallergenic allergen vaccines. (Figure 1).

Figure 1 (A) Protein structure, (B) dock surface and (C) HADDOCK analysis result of EFK8-mPhl p1-GS-mBet v1 dimer.

Alphafold3 structural prediction indicates that the introduction of the (GS)₄ linker causes the N-terminal α-helix of Der p1 (CWAFSGVAA) to come into excessive spatial proximity with the central β-sheet of Der p2 (KIEIKASI), forming a narrow region of only 4.9 Å. This may hinder antibody binding to key epitopes through steric hindrance effects. Further molecular docking analysis of the EFK8-mDer p1-GS-mDer p2 system via HADDOCK reveals a low HADDOCK score (-110.9 ± 2.2) with stable interfacial binding (van der Waals energy: -74.2 ± 4.3; electrostatic potential: -65.4 ± 10.5), and a buried surface area of 2163.1 Ų, indicating extensive coverage of surface residues (Figure 2). Should this system form higher-order multimers, the steric hindrance between adjacent subunits is expected to further obscure the IgE epitopes on mDer p1 and mDer p2, providing a structural explanation for the reduced immunogenicity of the fusion protein.

Figure 2 (A) Protein structure, (B) dock surface and (C) HADDOCK analysis result of EFK8-mDer p1-GS-mDer p2 dimer.

Through computational simulations, we revealed that EFK8-mediated fusion protein oligomerization may physically shield sensitizing epitopes, thereby reducing IgE binding capacity. However, the current model still has the following limitations and areas for improvement:

1. Multiscale Validation and Extended Analysis

Oligomer Quantification and Morphology Validation: Determine actual oligomer size and distribution via techniques like Dynamic Light Scattering (DLS) to provide more accurate oligomeric state constraints for modeling.

2. Linker and Self-Assembly Module Optimization

Flexible Linker Replacement: Test more flexible linkers (e.g., (GGGGS)n) or rigid linkers (e.g., α-helix-forming peptides) to balance domain independence and interactions.

EFK8-mediated self-assembly reduces the IgE-binding capacity and cytotoxicity of fusion allergens through steric hindrance effects. Computational analysis integrating Alphafold3 and HADDOCK provides the structural basis for this mechanism. Future work will focus on experimental validation of the polymeric state and immunogenicity characterization to assess its efficacy and controllability in allergen immunotherapy.