Exploring voltage-gated sodium channel conformations and protein-protein interactions using AlphaFold2

Area of Science:

• Structural biology
• Computational biophysics
• Pharmacology

Background:

• Voltage-gated sodium (NaV) channels are crucial for electrical signaling in cells.
• Developing subtype-selective NaV channel drugs is difficult due to conserved structures.
• Cryo-electron microscopy has advanced NaV channel structure determination but has limitations in capturing dynamic states.

Purpose of the Study:

• To evaluate AlphaFold2's ability to predict diverse NaV channel conformations.
• To assess AlphaFold Multimer's accuracy in modeling NaV channel complexes with accessory proteins.
• To explore how protein interactions influence NaV channel conformational dynamics.

Main Methods:

• Utilized AlphaFold2 for conformational sampling of NaV channels with enhanced techniques.
• Employed subsampled multiple sequence alignments and varied recycles for improved sampling.
• Applied correlation and clustering analyses to understand domain movements and state ensembles.
• Modeled NaV α-subunit interactions with β-subunits and calmodulin using AlphaFold Multimer.

Main Results:

• AlphaFold2 successfully modeled known, novel, and intermediate NaV channel conformations.
• Predicted structures revealed coordinated domain movements and recurring conformational states.
• AlphaFold Multimer accurately modeled NaV channel complexes with β-subunits and calmodulin.
• Protein partners significantly modulated the NaV α-subunit's conformational landscape and state coupling.

Conclusions:

• Deep learning methods show promise for understanding NaV channel structure, gating, and modulation.
• Predicted models offer valuable hypotheses but require experimental validation.
• Computational approaches can expand insights into the complex dynamics of ion channels.