A potential path to the development of small-molecule inhibitors is to identify small-molecules that mimic the interactions of short peptides with proteins. The present study uses Perl scripts and a MySQL database to build a unique dataset of 258 protein-peptide interactions (ProPep) from structures contained in the Protein Data Bank. The physiochemical and structural nature of protein-peptide interfaces were analysed in part using a novel amino acid pictogram analysis alongside accessible surface area, residue pairing and amino acid composition analysis. The results indicate that, for the peptide, proline residues and tyrosine residues play specific roles in protein-peptide interfaces. Furthermore it was observed that the peptide residues are significantly more buried than the residues of the cognate protein surface. The virtual screening program LIDAEUS was used to mine chemical databases to identify novel peptidomimetic compounds that have evincible binding to protein targets of therapeutic interest. Target-based and fragment-based virtual screening identified a series of potential compounds targeting the interaction between p21 and PCNA. Whilst the docking results were promising, results from testing in biological assays were inconclusive. A target-based virtual screening approach to identify small-molecule mimics of the interaction between the GnRH peptide and GnRH receptor yielded two promising compounds that demonstrated weak binding in biological assays. A third study to identify small-molecules binding to the SH3 domain of PSD-95 produced some promising hit compounds that as yet have not been tested in binding assays.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:652095 |
| Date | January 2008 |
| Creators | Harding, Simon D. |
| Publisher | University of Edinburgh |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://hdl.handle.net/1842/10935 |
Page generated in 0.0036 seconds