In silico screening of natural product databases reveals new potential leads against neglected diseases
FC Herrmann, Thomas Jürgen Schmidt · Planta Medica · 2013
Infections with “protozoan” parasites, such as Malaria, Leishmaniasis or Trypanosomiasis cause tremendous suffering in wide parts of the world. Due to a lack of novel therapeutics, more research in this field is urgently required. Many natural products (NPs) have shown impressive activity while their mechanisms of action are mostly unknown. In this in silico study [1], structures of enzymes of major protozoan pathogens and NP databases have been studied to reveal new leads against these neglected diseases. Structures of NPs that had been tested against at least one major protozoan pathogen were extracted from literature (Pubmed [2]). The search was focused on phenolic plant constituents. Optimized 3D structures were collected in a database (1712 compounds; PheDB). A second database with 928 NPs published in [3] was created accordingly (NPDB). These two collections of potentially active substances were used for virtual screening. Structures of potential protein targets [3] were acquired from the Protein Databank [4]. Co-crystallized ligands were used to create pharmacophore queries which were employed for a virtual screening (VS) of the databases. The hits resulting from this VS were docked into the relevant enzymes' active centers. Docking-scores were compared to those of the co-crystallized ligands (CCL). About 40 enzymes were studied so far yielding promising results (better docking scores than CCL) for NPs of various classes (Table 1). Experimental validation, i.e. enzyme inhibition tests, have been initiated. Tab. 1: Results of the virtual screening of different protozoal proteins T. brucei T. cruzi Leishmania ssp. P. falciparum Number of investigated 10 9 13 13 Number of hits (PheDB) 71 235 133 162 Number of hits (NPDB) 85 101 403 123 Most promising target TbIGNH TcHPRT LmexGAPDH PfENR PDB-entry number 3FZ0 1TC1 1N1G 2OL4 PRS (%)* of the most promising ligand 154 174 222 146 *PRS (%)= Protein relevance score: Ratio in % between docking score for best docking ligand and co-crystalized ligand References: [1] For all calculations: Molecular Operating Environment 2011.10 (MOE), Chemical Computing Group, Montreal, Canada, http://www.chemcomp.com [2] http://www.ncbi.nlm.nih.gov/pubmed/ [3] Schmidt, T.J. et al. Curr Med Chem. 2012; 19: 2128 – 75 and 2176 – 228. [4] http://www.rcsb.org/pdb/home/home.do This work is part of the activities of ResNetNPND: http://www.uni-muenster.de/ResNetNPND/ Support of CCG, Montreal, is gratefully acknowledged