Anmerkungen:
Diese Datenquelle enthält auch Bestandsnachweise, die nicht zu einem Volltext führen.
Beschreibung:
Protein Tyrosine Phosphatase 1B (PTP1B) is a validated drug target for the treatment of diabetes type 2 and obesity. Until now, development of suitable modulators has been hampered by the polarity of the binding site and related bioavailability issues of the molecules. The design of selective inhibitors of PTP1B against the closely related T-Cell protein tyrosine phosphatase (TC-PTP), which was associated to severe side effects in animal studies, proved even more challenging. Over the years progress wasmade, but known PTP1B inhibitors only achieved atmaximummoderate selectivity over TC-PTP. This study aims to break the traditional boundaries of PTP1B selectivity by deliberately exploiting structural differences of both proteins. Due to their high similarity this requires thorough analysis of their static structures as well as their flexible behavior. The goal was therefore pursued with two different approaches: In Part I of the study a detailed analysis of protein complexes with selective ligands was performed including their flexible behavior as determined frommolecular dynamics simulations. Since common analysismethods were not able to explain the selectivity of the investigated ligands, a newmethod was developed which is able to assess parameters of ligand affinity that are not covered by currently available methods: steric complementarity of the ligand to the protein together with ligand strain. The developed tool allows to assess those properties on high numbers of molecular dynamics frames to calculate ligand shape fit in a flexible context. It further enables to trace back the ligand atoms or parts responsible for good or bad shape fit. In Part II of this study the flexible behavior of the apoproteins was studied. Since surface properties are highly similar in both proteins, the analysis focused on binding site shapes. For this, a novel approach was chosen that translates binding site shapes from molecular dynamics simulations into point maps and subsequently uses clustering and difference calculations to ...