Exploring a Structural Data Mining Approach to Design Linkers for Head-to-Tail Peptide Cyclization.

Area of Science:

• Biochemistry
• Structural Biology
• Medicinal Chemistry

Background:

• Peptides are promising therapeutic agents but suffer from low bioavailability.
• Head-to-tail cyclization stabilizes peptide conformation, enhancing therapeutic properties like binding affinity and resistance to degradation.
• Rational design of cyclization linkers is crucial but lacks established frameworks.

Purpose of the Study:

• To develop a rational framework for designing head-to-tail cyclization linkers for peptides.
• To predict the conformation of cyclized peptides based on linker sequence and linear peptide structure.
• To demonstrate the application of this framework in improving peptide drug candidates.

Main Methods:

• Leveraging large-scale data mining of protein structures to identify suitable loop conformations for linkers.
• Developing a computational approach to predict cyclized peptide conformation from linear peptide structure and linker sequence.
• Applying the developed framework to design linkers for Nrf2-derived and urotensin II peptides.

Main Results:

• Accurate prediction of cyclized peptide conformation is achievable given linker sequence and linear peptide structure.
• A framework for rational linker design, constrained by length and amino acid composition, was established.
• Cyclization of an Nrf2-derived peptide resulted in a 26-fold increase in binding affinity.
• A novel bicyclic urotensin II analogue retained in vitro activity.

Conclusions:

• The developed approach provides the first framework for the rational design of head-to-tail cyclization linkers.
• This method shows significant potential for enhancing peptide-based drug design by improving stability and efficacy.
• The findings pave the way for developing more effective cyclic peptide therapeutics.