Harnessing AlphaFold to reveal hERG channel conformational state secrets

Area of Science:

• Structural Biology
• Computational Biophysics
• Pharmacology

Background:

• Understanding transmembrane ion channel protein structure and function is crucial for designing effective therapies.
• The hERG channel (K_V11.1) is a key cardiac repolarizing channel and a critical drug safety target due to its association with arrhythmia risk.
• Predicting discrete conformational states of ion channels, particularly the hERG inactivated state, remains a significant challenge.

Purpose of the Study:

• To develop a targeted AlphaFold modeling approach using structural templates to predict distinct functional states of the hERG channel.
• To validate the predicted conformations against experimental data, including drug interactions and ion conduction properties.
• To enhance computational drug screening and uncover novel structure-function relationships for improved drug safety.

Main Methods:

• Utilized AlphaFold with carefully selected structural templates to guide protein conformation predictions.
• Validated predicted structures through molecular docking for drug interactions and molecular dynamics simulations for ion conduction.
• Analyzed protein residue interaction networks across closed, open, and inactivated states.

Main Results:

• AlphaFold successfully predicted hERG inactivation mechanisms and novel features explaining enhanced drug binding during inactivation.
• The targeted modeling approach improved agreement with experimental drug affinities compared to traditional single-state models.
• Identified critical residues driving state transitions, consistent with prior mutagenesis studies.

Conclusions:

• This methodology provides a validated approach for predicting discrete protein conformations using AlphaFold, advancing our understanding of hERG channel function and pharmacology.
• Leveraging state-dependent models significantly enhances computational screening for drug safety, reducing the risk of drug-induced arrhythmia.
• This integrated deep learning and experimental validation approach sets a new benchmark for drug discovery and therapeutic design.