A Transferable and Robust Computational Framework for Class A GPCR Activation Free Energies.

Area of Science:

• Biochemistry
• Computational Biology
• Pharmacology

Background:

• G-protein coupled receptors (GPCRs) mediate crucial cellular functions, but simulating their activation is complex due to required local and global motion capture.
• Previous methods for GPCR activation simulation, while successful for specific receptors like β1-adrenergic receptor (ADRB1), demanded extensive and error-prone collective variable (CV) refinement.

Purpose of the Study:

• To introduce a streamlined and user-friendly strategy for defining CVs in molecular simulations of GPCR activation.
• To reduce user intervention in the CV selection and refinement process while ensuring robust free energy convergence.
• To apply the new method to pharmacologically relevant apo-GPCRs, specifically ADRB1 and the μ-opioid receptor.

Main Methods:

• Developed an evolved strategy for defining collective variables (CVs) that minimizes user input.
• Applied the enhanced CV definition strategy to molecular simulations of ADRB1 and the μ-opioid receptor.
• Utilized a multiple replica enhanced sampling approach combined with the new CV strategy.

Main Results:

• The new method successfully reconstructed free energies for ADRB1 activation, consistent with previous tailored approaches.
• Novel biological insights into the activation mechanism of the μ-opioid receptor were obtained.
• The strategy demonstrated robustness in achieving free energy convergence for both studied receptors.

Conclusions:

• The proposed streamlined CV definition strategy is effective for simulating GPCR activation.
• This method significantly reduces the complexity and potential for error in preparing molecular simulations of GPCRs.
• The approach is readily applicable to other class A GPCRs, facilitating systematic studies of drug target activation mechanisms.