Virtual Screening Expands the Non-Natural Amino Acid Palette for Peptide Optimization

Area of Science:

• Medicinal Chemistry
• Computational Chemistry
• Drug Discovery

Background:

• Peptides are crucial in drug discovery, but optimization is limited by the small number of available amino acids.
• The chemical space for small molecule drugs is vast, yet peptide drug discovery has not fully exploited this potential.

Purpose of the Study:

• To develop a large, virtual library of synthesizable non-natural amino acids for peptide optimization.
• To demonstrate the utility of this library in virtual screening protocols for drug discovery.

Main Methods:

• Enumeration of nearly 380,000 non-natural amino acids to assess chemical diversity.
• Validation of a virtual screening workflow using a subset of 10,000 amino acids on the Keap1-Neh2 complex.
• In silico mutation and docking-based virtual screening to predict binding affinity.

Main Results:

• The virtual library exhibits significantly greater chemical diversity compared to natural and commercial amino acids.
• Virtual screening identified potential peptide candidates with substantially higher predicted affinity for the Keap1 protein.
• The developed protocol proved effective in exploring the extended non-natural amino acid chemical space.

Conclusions:

• The non-natural amino acid chemical space can be vastly expanded and effectively screened computationally.
• This approach offers a powerful strategy for accelerating peptide optimization and drug discovery.
• The developed virtual library and screening protocol represent a significant advancement in peptide-based therapeutics.