Download files
Abstract
Cyclic peptides are valuable lead compounds for the development of medicines targeting many different diseases, due to their high stability and their often strong affinity for biological targets. However, the synthesis of cyclic peptides is often inefficient, and their 3D shapes are not always optimal for target binding. This project aimed to provide a solution to both of these problems.
The cyclic peptide Sansalvamide A (San A) was selected as the scaffold for this project. San A is a pentapeptide that has potent activity against various cancer cell lines, but its medicinal development is hampered by synthetic difficulties. First, a series of analogues of San A were synthesised by incrementally increasing the flexibility of peptide backbone using glycine, beta-alanine or gamma-aminobutyric acid (GABA) in place of a leucine residue of San A. These replacements were shown to dramatically increase the efficiency of peptide cyclisation, which was rationalised by comparing the effective molarity (EM) of cyclisation of the corresponding linear precursor peptides.
Second, the cytotoxicity of the synthesised analogues was tested against human colon cancer cells. Overall, the analogues exhibited weaker activity than San A itself. However, the GABA-containing analogue had the most promising activity. Therefore this analogue was selected for further development.
Third, a series of conformationally-restricted GABA derivatives were targeted, in order to “rescue” the San A activity while retaining cyclisation efficiency. The library of GABA derivatives included two unusual backbone-fluorinated amino acids; these had previously been synthesised in the Hunter group, but the synthesis was inefficient. Accordingly, a new synthetic route was developed. The conformationally-restricted GABA derivatives were then elaborated into new San A analogues, and the cytotoxicity was measured.
Finally, to rationalise the observed trends in biological activity, preliminary efforts were made to determine the 3D geometries of all San A analogues synthesised by performing a series of NMR and molecular modelling experiments.
Overall, this project has contributed significantly to the medicinal development of cyclic peptides, by developing a new strategy for improving peptide cyclisation efficiency and by highlighting a new class of selectively fluorinated amino acids as novel tools for “fine-tuning” the conformations of cyclic peptides.