Abstract

The serotonin transporter (SERT), among the transporters for dopamine (DAT) and norepinephrine (NET) member of the SLC6 gene family, is an integrated membrane protein facilitating the rapid reuptake of previously released serotonin from the synaptic cleft and thus the main mechanism of terminating the neurochemical signal using an electrochemical sodium gradient. Involved in various diseases of the central nervous system, such as depression, it is with NET the main target for antidepressant therapy and together with DAT targeted by psychostimulants[1]. Up to now, no crystal structure of a human neurotransmitter:sodium symporter (NSS) is available. High resolution crystal structures of a bacterial homologue leucine transporter (LeuT) were resolved[2]. Similar in structure and function, it shares a sequence identity of ∼45% with the NSS in the binding site[3]. The detailed molecular mechanisms of binding and inhibition of the transporters is still a matter of discussion. To get insights into the molecular mechanisms of protein ligand interactions, in-vitro and in-silico methods are combined to answer questions about interactions of tricyclic antidepressants (TCA) with SERT at atomic level within the framework of the research project SFB35 (Spezialforschungsbereich 35, www.sfb35.at). Starting with homology modelling[4] of the human transporters based on the ‘open-to-out’ conformation of the template LeuT, subsequent docking experiments[5] with selected inhibitors have been performed. A protocol was elaborated to avoid bias from scoring functions in an early step, that uses primarily hierarchical clustering and protein–ligand–interactions combined with experimental data. The selection of the docking solutions was finally driven by the results of the pharmacological assays and ended up in a newly proposed binding mode for imipramine in SERT[6]. To further assess the docking complexes in explicit water, molecular dynamics simulations[7] were performed on the finally selected docking poses. In total, ≈ 60 simulations were carried out on the Vienna Scientific Cluster (www.vsc.ac.at). The trajectories were analysed with respect to the stability of the ligand, specific protein-ligand distances and hydrogen bond networks. As final part, pharmacophore modelling[8] and virtual screening was applied. A database of 300000 purchasable drug–like compounds (www.lifechemicals.com) was used for screening against the validated model. The finally selected compounds have been experimentally tested at the Institute of Pharmacology, Medical University of Vienna. The experiments revealed active compounds with a binding affinity in low μM range. Bibliography [1] A. S. Kristensen, J. Andersen, T. N. Jørgensen, L. Sørensen, J. Eriksen, C. J. Løland, K. Strømgaard, U. Gether, Pharmacol Rev 2011, 63, 585–640. [2] S. K. Singh, C. L. Piscitelli, A. Yamashita, E. Gouaux, Science 2008, 322, 1655–1661. [3] T. Beuming, L. Shi, J. A. Javitch, H. Weinstein, Mol Pharmacol 2006, 70, 1630–1642. [4] A. Sali, T. L. Blundell, J Mol Biol 1993, 234, 779–815. [5] Chemical Computing Group, MOE: Molecular Operating Environment, 2009. www.chemcomp.com. [6] S. Sarker, R. Weissensteiner, I. Steiner, H. H. Sitte, G. F. Ecker, M. Freissmuth, S. Sucic, Mol Pharmacol 2010, 78, 1026–1035. [7] B. Hess, C. Kutzner, D. van der Spoel, E. Lindahl, Journal of Chemical Theory and Computation 2008, 4, 435–447. [8] G. Wolber, T. Langer, J Chem Inf Model 2005, 45, 160–169.