Lead optimization for new antimalarials and Successful lead identification for metalloproteinases: A Fragment-based approach Using Virtual Screening

Lead optimization for new antimalarials and Successful lead identification for metalloproteinases: A Fragment-based approach Using Virtual Screening Computer-aided drug design is an essential part of the modern medicinal chemistry, and has led to the acceleration of many projects. The herein d...

Full description

Saved in:
Bibliographic Details
Main Author: Silber, Katrin
Contributors: Klebe, Gerhard (Prof.) (Thesis advisor)
Format: Doctoral Thesis
Language:English
Published: Philipps-Universität Marburg 2007
Subjects:
Online Access:PDF Full Text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Lead optimization for new antimalarials and Successful lead identification for metalloproteinases: A Fragment-based approach Using Virtual Screening Computer-aided drug design is an essential part of the modern medicinal chemistry, and has led to the acceleration of many projects. The herein described thesis presents examples for its application in the field of lead optimization and lead identification for three metalloproteins. DOXP-reductoisomerase (DXR) is a key enzyme of the mevalonate independent isoprenoid biosynthesis. Structure-activity relationships for 43 DXR inhibitors are established, derived from protein-based docking, ligand-based 3D QSAR and a combination of both approaches as realized by AFMoC. As part of an effort to optimize the properties of the established inhibitor Fosmidomycin, analogues have been synthesized and tested to gain further insights into the primary determinants of structural affinity. Unfortunately, these structures still leave the active Fosmidomycin conformation and detailed reaction mechanism undetermined. This fact, together with the small inhibitor data set provides a major challenge for presently available docking programs and 3D QSAR tools. Using the recently developed protein tailored scoring protocol AFMoC precise prediction of binding affinities for related ligands as well as the capability to estimate the affinities of structurally distinct inhibitors has been achieved. Farnesyltransferase is a zinc-metallo enzyme that catalyzes the posttranslational modification of numerous proteins involved in intracellular signal transduction. The development of farnesyltransferase inhibitors is directed towards the so-called non-thiol inhibitors because of adverse drug effects connected to free thiols. A first step on the way to non-thiol farnesyltransferase inhibitors was the development of an CAAX-benzophenone peptidomimetic based on a pharmacophore model. On its basis bisubstrate analogues were developed as one class of non-thiol farnesyltransferase inhibitors. In further studies two aryl binding and two distinct specificity sites were postulated. Flexible docking of model compounds was applied to investigate the sub-pockets and design highly active non-thiol farnesyltransferase inhibitor. In addition to affinity, special attention was paid towards in vivo activity and species specificity. The second part of this thesis describes a possible strategy for computer-aided lead discovery. Assembling a complex ligand from simple fragments has recently been introduced as an alternative to traditional HTS. While frequently applied experimentally, only a few examples are known for computational fragment-based approaches. Mostly, computational tools are applied to compile the libraries and to finally assess the assembled ligands. Using the metalloproteinase thermolysin (TLN) as a model target, a computational fragment-based screening protocol has been established. Starting with a data set of commercially available chemical compounds, a fragment library has been compiled considering (1) fragment likeness and (2) similarity to known drugs. The library is screened for target specificity, resulting in 112 fragments to target the zinc binding area and 75 fragments targeting the hydrophobic specificity pocket of the enzyme. After analyzing the performance of multiple docking programs and scoring functions forand the most 14 candidates are selected for further analysis. Soaking experiments were performed for reference fragment to derive a general applicable crystallization protocol for TLN and subsequently for new protein-fragment complex structures. 3-Methylsaspirin could be determined to bind to TLN. Additional studies addressed a retrospective performance analysis of the applied scoring functions and modification on the screening hit. Curios about the differences of aspirin and 3-methylaspirin, 3-chloroaspirin has been synthesized and affinities could be determined to be 2.42 mM; 1.73 mM und 522 μM respectively. The results of the thesis show, that computer aided drug design approaches could successfully support projects in lead optimization and lead identification. fragments in general, the fragments derived from the screening are docked
Physical Description:249 Pages
DOI:10.17192/z2007.0753