Analysis of the selectivity and promiscuity of ligand binding for fragment-based design

• Applicant: Professor Dr. Oliver Koch
• Subject Area: Pharmacy
• Term: since 2021
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 462329126

A huge amount of bioactivity data and protein structures are publicly available and allow different data mining and knowledge discovery approaches. The basic insight behind this work is the identification of a ligand fragment that binds in a specific binding subpocket that also occurs in unrelated proteins with different functions. In all these proteins this fragment was also part of known active ligands that occupied this binding subpocket. One identified pair also showed a high similarity in ligand binding with the potential to develop a dual inhibitor targeting both proteins. The final goal will be to identify other common fragments and their specific binding environment. If such a specific binding environment is identified in an interesting protein target, the knowledge of the specific binding fragment can be used for rational fragment-based design approaches. These fragments will be identified by an in-depth analysis of available bioactivity data and protein structures using scaffold-based analysis and binding pocket comparisons. Furthermore, this analysis will provide further insights into the selectivity and promiscuity of ligand binding for the rational development of polypharmacology-based ligands. The latter means the development of ligands with desired synergistic effects that modulate more than one protein target. The underlying analysis comprises three different workflows: 1) The scaffold-based analysis of bioactivity data to identify promiscuous scaffolds or fragments shared by ligands that bind to different and unrelated protein targets. 2) A comprehensive binding site comparison of available protein structures to identify similar subpockets that bind similar ligand-fragments. 3) The development of an automated workflow for the generation of target-specific scoring functions that can be used for the identification of polypharmacology-based ligands based on identified common fragments. For this purpose, a specially designed virtual library will be created containing virtual ligands designed around these fragments. Furthermore, identified fragments are of particular interest as potential starting points for fragment-based design if it is possible to crystallise such fragments with a target protein. Therefore, crystallisation experiments with identified fragments will be performed and compared with available crystallographically based fragment screens.

• DFG Programme: Research Grants