Planta Med 2014; 80 - WS17
DOI: 10.1055/s-0034-1394558

Linking ethnopharmacology with phenotypic and virtual screening for the identification of antiviral agents from natural sources

CE Mair 1, U Grienke 1, C Draschl 1, C Kramer 2, M Richter 3, S von Grafenstein 2, J Kirchmair 4, KR Liedl 2, M Schmidtke 3, JM Rollinger 1
  • 1Institute of Pharmacy/Pharmacognosy, Center for Molecular Biosciences Innsbruck, University of Innsbruck, Innrain 80 – 82, 6020 Innsbruck, Austria
  • 2Institute of General, Inorganic and Theoretical Chemistry, Center for Molecular Biosciences Innsbruck, University of Innsbruck, Innrain 80 – 82, 6020 Innsbruck, Austria
  • 3Department of Virology and Antiviral Therapy, Jena University Hospital, Hans-Knoell-Straße 2, 07745 Jena, Germany
  • 4Department of Chemistry and Applied Biosciences, Institute of Pharmaceutical Sciences, ETH Zurich, 8093 Zurich, Switzerland

In the last decades drug discovery was generally based on two main approaches: (a) phenotypic screening, where the identification of promising plant material is based on the possibility of constituents to trigger a desired biological response without knowledge of the underlying mode of action, and (b) the target-centric approach, where a particular hypothesis serves as a starting point for the identification of bio-actives interacting with a specific biomolecule [1]. As both approaches have their pros and cons, we applied a combination of these two strategies for the identification of antivirals from nature against acute respiratory infections (ARIs). Thereby, knowledge from folk medicine guided the selection of natural sources, followed by a phenotypic in vitro extract screening using three representative viruses supported by a target based virtual screening. After selection and extraction of natural starting material, 161 extracts were investigated in a cytopathic effect (CPE) inhibition assay against an influenza virus A (A/HK/68); 153 extracts were screened against human rhinovirus 2 (HRV2) and coxsackie virus B3 (CVB3). Cytotoxicity was determined for all extracts. Around 10% revealed IC50 < 30 µg/mL; further 10% showed an antiviral activity with IC50 of 30 to 100 µg/mL; about 80% had IC50 > 100 µg/mL and were considered as inactive. These phenotypic investigations were followed by an in silico screening of large molecular databases containing known constituents of the most promising extracts. The databases were screened in silico against selected antiviral targets to offer insights into the putative molecular mechanism and to guide phytochemical investigations. We exemplify this procedure on the antiviral extract and the constituents of Ruta graveolens L. and critically discuss the challenges, pitfalls and advantages of the ensemble approach.

Acknowledgements: Supported by the Austrian Science Fund (FWF P24587 & P23051) and the European Social Fund (ESF & TMWAT Project 2011 FGR 0137).

Keywords: Antiviral, virtual screening, phenotypic screening, influenza, ethnopharmacology, Ruta graveolens

References:

[1] Swinney, D.C., Anthony, J. (2011) Nat Rev Drug Discov 10, 507 – 519.