Table of Contents:
The protozoan disease malaria is one of the major inexorable infections worldwide, especially widespread throughout tropical regions. It is caused by parasites of the genus Plasmodium (P. falciparum, P. vivax, P. malariae and P. ovale) and transmitted by female Anopheles mosquitoes. The WHO estimates that approximately 350 to 500 million people are infected annually. Infections by P. falciparum, the most severe form of malaria, are responsible or at least contribute to more than a million deaths per year. Especially children below an age of five years are affected. The continuously increasing resistance of the vector towards insecticides as well as the emergence of multi-drug resistant parasites highly demands the development of new anti-infective substances favourably showing either a novel binding mode or a new mechanism of action, which is pivotal to combat this life-threatening disease. The aspartic proteases Plasmepsin I, II, IV and the closely related histoaspartic protease are believed to be attractive targets for the development of an antimalarial drug therapy. Most inhibitors synthesized so far to address the Plasmepsins are peptidomimetic transition state analogues addressing via a hydroxyl or hydroxyl-like moiety the two aspartates of the active site thereby replacing the native substrate. These peptidomimetic inhibitors are highly active but often suffer from low bioavailability and, additionally, are difficult to synthesize. Recently, substituted secondary amines, in particular substituted pyrrolidines, have proven to be micromolar inhibitors of other aspartic proteases such as HIV-protease. These results prompted us to investigate in greater detail the suitability of azacycles bearing a basic amino functionality as core element for the design and synthesis of Plm inhibitors. Novel non-peptidic Plm II and Plm IV inhibitors have been developed featuring a 2,3,4,7-tetrahydro-1H-azepine scaffold as core element. Equipped with suitable side-chains to address two of the enzyme’s specificity pockets, these inhibitors show activity up to the nanomolar range. Derived from a combined subpocket search and a combinatorial docking approach, first inhibitors with affinities in the low micromolar range could be identified. Structural modifications of the initial lead structure led to inhibitors exhibiting affinities in the submicromolar range. Overall, the experimentally determined Ki values of of the synthesized inhibitors were generally in good agreement with the design hypothesis. The found structure-activity relationships thus support the predicted binding mode for Plm II as well as Plm IV. In order to find new plasmepsin inhibitors with novel lead structures, a feature tree (Ftree) search was performed on one of the Plm inhibitors, bearing a tetrahydro-1H-azepine as core structural element. Ftree is a fast method to detect molecular similarity in terms of physicochemical properties between small organic compounds. In order to retrieve candidate molecules for testing with similar binding properties but exhibiting a molecular skeleton deviating from the query reference two different search strategies were applied, leading to new inhibitors with Ki values in the micromolar range. Novel scaffolds could be discovered. For the most promising one, based on a central thiophen moiety, a series of compounds was synthesized. By means of the Gewald reaction modifications of the P1, P1’ and P2’ substituents were easy accessible and within short time the inhibitor scaffold could be optimized to nanomolar range. Furthermore, a series of pyrrolidine derivatives, originally synthesized as HIV-1 protease inhibitors, was tested for Plm activity. Inhibitors in the nanomolar range were discovered for Plm II and IV. In order to explain structure-activity relationships detailed studies were carried out in finding putative binding modes. For pyrrolidine-diol-diester-derivatives and the substituted diamino pyrrolidine-diamide inhibitor resonable binding modes could be generated by using the Plm II conformer 1LF2. Putative binding modes for Plm II are in agreement with structure-activity relationships. Unfortunately, with the available crystallographic information it was not possible to predict a reasonable binding mode for the pyrrolidinedimethylene diamine derivatives. None of the three major binding site conformers seen in the presently available crystal structures allow accommodation. However, the structural studies indicate, that Plm II is a highly flexible protein. To tackle this problem in more detail, MD simulations of the uncomplexed state were carried out and transitions along the trajectory could be monitored. Supported by the MD simulations a putative binding mode for the inhibitors with a pyrrolidinedimethylene diamine scaffold could be suggested and provides further insights into binding pocket adaptions that possibly have implications for Plm II inhibitor design.