Coarse-grained models for simulations of multiprotein complexes: application to ubiquitin binding

Area of Science:

• Computational Biology
• Biophysics
• Structural Biology

Background:

• Simulating multiprotein complexes with low binding affinity presents computational challenges.
• Accurate modeling requires capturing residue-level interactions, flexible linkers, and membrane interactions.

Purpose of the Study:

• To develop and validate a coarse-grained model for simulating thermodynamic and structural properties of multiprotein complexes.
• To apply the model to understand the binding of Vps27 to membrane-tethered ubiquitin.

Main Methods:

• Coarse-grained modeling with effective energy functions.
• Residue-level potentials for inter-domain interactions and Debye-Hückel electrostatics.
• Polymer models for flexible linkers and residue-dependent potentials for membrane interactions.
• Replica-exchange Monte Carlo simulations for parameterization and validation.

Main Results:

• The model accurately predicts binding affinities and structures for various protein complexes, with good agreement to experimental data.
• Simulated structures show high accuracy (RMSD < 5 Å in 70% of cases) and correct prediction of binding interfaces (90% of cases).
• Simulations of Vps27-ubiquitin binding reveal preferential binding to UIM domains, cooperative effects, and dynamic interactions.

Conclusions:

• The developed coarse-grained model is effective for simulating low-affinity multiprotein complexes.
• Cooperative and dynamic interactions between Vps27 and ubiquitin significantly enhance binding affinity.
• The findings provide insights into the multivesicular-body protein-sorting pathway.