Aarhus University Seal / Aarhus Universitets segl

Drug Design

Computer Aided Drug Design

The Biomodelling group has access to a large variety of rational drug design methods. First, we concentrated on incorporating structure-based methods, especially molecular docking techniques. This has now been accomplished, and several students are working along these lines. In recent years we have successfully expanded the computational toolbox of the group to also include ligand based methods as pharmacophore modeling and QSAR techniques.

  • Structure Based Drug Design Studies - ATPases: The newly solved structure of an open E2-form of the calcium pump represents a long-sought template for proper modeling of the binding of clinically important inhibitors like omeprazol and ouabain. In collaboration with the PUMPKIN Centre we study the binding of these inhibitors to similarly open models of H+/K+-ATPase and Na+/K+-ATPase. Omeprazol, or Losec®, is a proton pump (H+/K+-ATPase) inhibitor used in the treatment of e.g. peptic ulcer and gastroesophageal reflux disease. Bufalin and ouabain belongs to the class of cardiac glycoside drugs used in the treatment of congestive heart failure and cardiac arrhythmia. The latter functions by binding to the Na+/K+-pump leading to an increased level of Ca2+-ions in the heart and thus better contraction of the muscle. The proposed studies are expected to give important new information regarding the actual drug-protein interaction and hereby also an improved understanding of the physiological consequences.
  • New Antibiotics: We are involved in rational design of new potential antibiotics. The drug target is bacterial elongation factor, EF-Tu, which has recently been characterized with two different inhibitors in the binding site. The structures revealed that the two molecules have critical regions, which must be occupied correctly in the binding pocket. The idea is to design hybrid compounds that take advantage of most of the interaction sites represented by the two molecules. If we succeed, this may have very large potential for applications due to the rising risk for drug resistance. So far, our studies have revealed that a cyclic hybrid can bind in the active site.
  • Endocrine Disrupting Chemicals: Pesticides, plastic softeners etc. are suspected to be involved in the recent observations of e.g. bisexual fish and the lowered fertility of female gardeners. It is known that these molecules, called endocrine disrupting chemicals (EDC), interact with the estrogen receptor (ER) thereby perturbing the normal hormonal balance by interfering in the signaling pathway of ER. We try to model the interaction of EDCs with ER. Initially, we use molecular docking to place the disruptors either in the active site or in other potential binding cavities, then we will simulate the effects of EDC, using the pesticide DDE, on the conformational behavior of ER, which is known to be important for proper function.
  • New Techniques for Molecular DockingWe have a long-term collaboration with the software company Molegro Aps with the overall aim of developing a computational protocol for improving molecular docking simulations. Our current projects focus on improving current methodology with respect to treatment of protein flexibility, developing better solvation models and scoring functions.
  • Ligand Based Methods: We have used QSAR and pharmacophor identification to quantify the binding of a series of analogs that bind to the human dopamine transporter, hDAT. We will continue to use these methods in the future for analogs of psychostimulants, ecstasy and amphetamine.