MechMemDyn: Coupling Frustration Analysis with Membrane Dynamics to Target the TREM2-DAP12 Complex Interface.

Area of Science:

• Biochemistry and Structural Biology
• Computational Drug Discovery
• Neuroscience

Background:

• The TREM2-DAP12 complex is a key therapeutic target for Alzheimer's disease.
• Stabilizing this transmembrane protein-protein interaction is challenging due to its dynamic interface.
• Existing methods struggle to accurately predict the efficacy of potential drug compounds.

Purpose of the Study:

• To develop a novel computational framework for structure-based drug discovery of TREM2-DAP12 protein-protein interaction stabilizers.
• To address the limitations of conventional docking and AI-driven approaches in capturing interfacial energetics.
• To establish a physics-driven paradigm for targeting dynamic transmembrane protein interfaces.

Main Methods:

• Integration of protein frustration analysis with membrane-embedded molecular dynamics (MD) simulations.
• Mapping the local frustration landscape to identify critical contact networks at the protein interface.
• Developing a metric based on dampening distance fluctuations within key networks to predict ligand potency.

Main Results:

• MechMemDyn successfully identifies minimally frustrated contact networks essential for stabilizing the TREM2-DAP12 complex.
• The framework demonstrates a strong correlation between a ligand's ability to dampen interfacial fluctuations and its experimental potency.
• The proposed method significantly outperforms conventional static docking, AI-driven dynamic docking, and standard MD simulations in accuracy and reproducibility.

Conclusions:

• MechMemDyn provides a more accurate and reproducible method for cross-ligand comparison and drug design targeting the TREM2-DAP12 complex.
• This work establishes frustration analysis as a key tool for rationalizing protein-protein interaction stabilizer activity.
• The study presents a novel, physics-driven approach for designing stabilizers for dynamic transmembrane protein interfaces.