Biophysical limits of protein-ligand binding

Area of Science:

• Biochemistry
• Structural Biology
• Computational Chemistry

Background:

• The concept of ligand efficiency (LE) was introduced by Kuntz et al. in 1999, initially focusing on drug-like molecules.
• Previous work indicated that metal binding interactions exhibit high ligand efficiencies.
• The Binding MOAD database provides a diverse dataset for examining binding phenomena across various small molecules.

Purpose of the Study:

• To investigate the physical limits of ligand efficiency across a broad spectrum of small molecules.
• To identify characteristics of complexes exhibiting the highest ligand efficiencies.
• To establish quantitative limits for ligand efficiency in protein-ligand interactions.

Main Methods:

• Analysis of protein-ligand complexes within the Binding MOAD database.
• Calculation of ligand efficiency based on binding energy and ligand atom count.
• Examination of binding site properties, including charge and buried surface area.
• Correlation of ligand and binding site characteristics with observed ligand efficiencies.

Main Results:

• The physical limit for ligand efficiency in the Binding MOAD database is -1.75 kcal/mol·atom.
• 95% of analyzed complexes fall below a "soft limit" of -0.83 kcal/mol·atom.
• A hard limit for ligand efficiency, based on buried molecular surface area, is -117 cal/mol·Å(2).
• Highly efficient complexes feature small, charged ligands within highly charged, buried binding sites.
• Optimal efficiency is driven by tight, complementary contacts between charged ligand groups and the binding pocket.

Conclusions:

• Ligand efficiency is fundamentally limited by physical constraints related to molecular complementarity and charge interactions.
• Small, charged ligands binding to complementary charged pockets represent the most efficient binding systems.
• Deviations in atomic fit (e.g., 0.24 Å misfit) significantly reduce achievable ligand efficiency.
• The established physical limits provide valuable insights for drug design and understanding molecular recognition.