Benchmarking Alchemical Relative Binding Free Energy Calculations for Nucleotide Binding to Multimeric ATPases.

Area of Science:

• Biochemistry and Biophysics
• Computational Chemistry
• Structural Biology

Background:

• Multimeric ATPases bind nucleotides at interfaces, complicated by cooperative interactions.
• Experimental measurement of binding affinities in these systems is challenging.

Purpose of the Study:

• To benchmark alchemical relative binding free energy (RBFE) calculations for nucleotide binding in six classes of multimeric ATPases.
• To assess the impact of substrates (DNA/RNA) and identify limitations of RBFE in complex systems.

Main Methods:

• Large-scale RBFE calculations using fixed-charge force fields across 55 interfacial sites in F1-ATPase, MalK, MCM, Rho, FtsK, and gp16.
• Simulations conducted with and without substrates, requiring extensive sampling (>20 ns) due to ligand properties.
• Structural analysis of protein flexibility and deviations during simulations.

Main Results:

• RBFE reproduced experimental binding preferences for 91% of sites in F1-ATPase, MalK, and MCM, which showed high structural stability.
• Agreement dropped to 60% for Rho, FtsK, and gp16, exhibiting greater structural variability.
• AlphaFold3 modeling of gp16 suggested distinct functional states compared to cryo-EM structures.

Conclusions:

• RBFE methods show significant predictive potential for nucleotide binding in multimeric ATPases.
• Key error sources include structural fidelity, protein flexibility, ligand pose, and alchemical transformation artifacts.
• Computational modeling, including AlphaFold3, can aid in understanding functional states of ATPases.