Domain Motion Enhanced (DoME) Model for Efficient Conformational Sampling of Multidomain Proteins

Area of Science:

• Computational biology
• Biophysics
• Protein dynamics

Background:

• Simulating large conformational changes in multidomain proteins with all-atom molecular dynamics (MD) is computationally challenging due to slow timescales.
• Existing coarse-grained (CG) models often struggle to accurately capture the complex dynamics of these proteins.

Purpose of the Study:

• To develop a novel CG model that enables stable and efficient MD simulations of multidomain proteins.
• To improve the sampling of conformational transitions in proteins with significant domain motions.

Main Methods:

• Introduction of the "Motion Tree" concept to quantify domain motions and identify rigid units.
• Development of the domain motion enhanced (DoME) CG model, where interdomain interactions are inversely proportional to domain motion magnitude.
• Application of the DoME model combined with the Go model (DoME-Go) for simulating adenylate kinase (AdK).

Main Results:

• DoME-Go simulations of adenylate kinase (AdK) showed excellent agreement with a 10 μs all-atom MD simulation and experimental data.
• The DoME-Go model produced stable simulation trajectories resistant to temperature variations.
• Conformational transitions were sampled more efficiently compared to the conventional Go model, even with rigid domains.

Conclusions:

• The DoME model offers a significant advancement for simulating multidomain protein dynamics.
• Identifying protein domains and their interfaces is a valuable strategy for developing effective CG models.
• The DoME-Go approach provides a computationally efficient and accurate method for studying protein conformational changes.