| Summary: | <p>Proton dependent oligopeptide transporters (POTs) play a central role in nitrogen homeostasis by coupling the uptake of dipeptides and tripeptides to the proton electrochemical gradient across the plasma membrane. In human, members of this transporter family, PepT1 and PepT2, are critical modulators of drug pharmacokinetics as they facilitate the uptake and retention of numerous orally administered drugs including the β-lactam antibiotics. Rationally designing drugs to target these transporters is therefore an attractive approach to improving bioavailability. To this end, the binding of peptides to a bacterial homolog, PepT<sub>St</sub>, was modelled based on recently determined crystal structures. A range of computational methods to predict the free energy of binding were evaluated and a hybrid approach, where the end-point methods were used to classify peptides into strong and poor binders and a theoretically exact method for refinement, was able to accurately predict ligand affinities. This approach was utilised to investigate the substrate preference of PepT<sub>St</sub> and the results were validated using <em>in vitro</em> transport assays. To extend this study to the human peptide transporters, homology models of PepT1 and PepT2 were built using the crystal structures of PepT<sub>So</sub> and mouse and rat extracellular domains (ECDs) as the templates. Essential residues as proposed by various mutational studies align well with the binding cavity, suggesting that the models are structurally sound. Applying the free energy methods to predict the affinities of peptides and drugs to the homology model of PepT1, however, resulted in discrepancies with experimental data, highlighting the importance of a high-resolution crystal structure in binding affinity predictions. Based on the results for PepT<sub>St</sub>, a binding model for peptide prodrugs and β-lactam antibiotics to the human PepT1 was proposed. Overall, this thesis provides a framework for future computational studies using free energy methods to understand drug interactions with pharmaceutically relevant transporters.</p>
|
|---|