Coevolution combined with molecular dynamics simulations provides structural and mechanistic insights into the...

Area of Science:

• Structural Biology
• Computational Biology
• Biophysics

Background:

• Determining 3D structures of large protein complexes is challenging due to difficulties in protein expression and purification, despite advances in cryo-electron microscopy (cryo-EM).
• Computational methods like Direct Coupling Analysis (DCA) can predict protein-protein interactions but are limited by computational complexity and false positives, especially for large proteins.

Purpose of the Study:

• To develop and validate a novel computational approach combining modified Direct Coupling Analysis (DCA) with Molecular Dynamics (MD) simulations.
• To predict and elucidate the structural details and interfaces of large, multi-subunit protein complexes, particularly those lacking well-studied structures.

Main Methods:

• A modified Direct Coupling Analysis (DCA) approach was developed to predict likely protein-protein interaction interfaces.
• Molecular Dynamics (MD) simulations were integrated with the modified DCA to provide structural and mechanistic details of interacting peptides.
• The combined approach was applied to predict interactions within the Integrator complex (INTS9/INTS11), CPSF100/CPSF73 heterodimer, and an INTS4/INTS9/INTS11 heterotrimer.

Main Results:

• Predictions for the INTS9/INTS11 interaction interfaces showed high consistency with existing crystallographic data.
• The computational approach successfully predicted interactions for the previously less-studied CPSF100/CPSF73 and INTS4/INTS9/INTS11 complexes.
• Experimental data validated the predicted interactions for the CPSF and INTS complexes.

Conclusions:

• The combination of modified DCA and MD simulations offers a powerful computational strategy for determining the structure of large protein complexes.
• This integrated approach overcomes limitations of individual methods, providing valuable structural insights where experimental structure determination is difficult.