Taking Protein-Protein Interactions to The Maximum Speed: Applying Coarse-Graining Simulations

Protein-Protein interactions (PPIs) play a key role in most biological processes and aberrant PPIs have been associated with many human diseases. Therefore, a detailed understanding of the structure and the dynamic interaction... [ view full abstract ]

Protein-Protein interactions (PPIs) play a key role in most biological processes and aberrant PPIs have been associated with many human diseases. Therefore, a detailed understanding of the structure and the dynamic interaction of PPIs is significant in drug design and discovery.
Current molecular dynamics (MD) all-atom (AA) models remain a popular tool for modeling such complex macro-molecular interactions; nevertheless, these models are restricted to small system sizes and short timescales (~100s of ns) because they are computationally very expansive.
For this reason, the Coarse-Graining (CG) MD approach was developed, which overcomes this restriction by clustering several atoms into one bead. In this way, the number of degrees of freedom in the system is significantly reduced and in return, the computational complexity is reduced. This allows simulating very large systems (~millions of atoms) for larger timescales (up to several μs), using CG-MD. However, the computational costs for large systems remains high, which is why Compute Canada was used to perform these simulations.
In this work, we employ state-of-the-art CG-MD to model the interactions of a HIV envelope glycoprotein (gp120) and a CD4 receptor. The CD4 receptor is exposed on the surface of T helper cells and induces, upon interaction with GP120, a cascade of conformational changes, leading to the fusion of the viral and the host cell membrane.
We assess the performance of our simulation by comparing the structure of the GP120-CD4 complex obtained from the CG-MD against those obtained from the docking software HADDOCK and the PDB (ID: 1GC1).