Computational design of a protein crystal

Area of Science:

• Protein engineering
• Structural biology
• Materials science

Background:

• Protein crystals are crucial for catalysis, materials science, structural biology, and drug development.
• Designing specific protein crystal structures is challenging due to protein complexity and noncovalent interactions.

Purpose of the Study:

• To present a computational method for designing proteins that self-assemble into macroscopic crystals.
• To engineer a de novo protein with a specific three-dimensional crystal structure and symmetry.

Main Methods:

• Computational design of a three-helix coiled-coil protein for self-assembly into a P6 space group crystal.
• Ensemble creation of crystalline structures with targeted symmetry.
• Sequence design for identified "designable" structures based on energy landscape minima.
• Experimental characterization of designed proteins, including X-ray crystallography.

Main Results:

• A de novo designed protein self-assembled into a polar, layered, 3D crystal with a "honeycomb-like" structure and hexameric channels.
• The crystal structure of the designed protein, determined at 2.1 Å resolution, showed sub-Å agreement with the computational model (backbone root mean square deviation).

Conclusions:

• The computational approach successfully enabled the de novo design of proteins forming predictable, macroscopic crystals.
• This method has significant potential for creating novel nanostructured materials and improving X-ray crystallographic analysis of proteins.