Molecular modeling and structure-activity relationship studies were performed to propose a model for binding of the neurotransmitter serotonin (5-HT) to the human serotonin transporter (hSERT). Homology models were constructed using the crystal structure of a bacterial homologue, the leucine transporter from Aquifex aeolicus, as the template and three slightly different sequence alignments. Induced fit docking of 5-HT into hSERT homology models resulted in two different binding modes. Both show a salt bridge between Asp98 and the charged primary amine of 5-HT, and both have the 5-HT C6 position of the indole ring pointing toward Ala173. The difference between the two orientations of 5-HT is an enantiofacial discrimination of the indole ring, resulting in the 5-hydroxyl group of 5-HT being vicinal to either Ser438/Thr439 or Ala169/Ile172/Ala173. To assess the binding experimentally, binding affinities for 5-HT and 17 analogues toward wild type and 13 single point mutants of hSERT were measured using an approach termed paired mutant-ligand analogue complementation (PaMLAC). The proposed ligand-protein interaction was systematically examined by disrupting it through site-directed mutagenesis and re-establishing another interaction via a ligand analogue matching the mutated residue, thereby minimizing the risk of identifying indirect effects. The interactions between Asp98 and the primary amine of 5-HT and the interaction between the C6-position of 5-HT and hSERT position 173 was confirmed using PaMLAC. The measured binding affinities of various mutants and 5-HT analogues allowed for a distinction between the two proposed binding modes of 5-HT and biochemically support the model for 5-HT binding in hSERT where the 5-hydroxyl group is in close proximity to Thr439.

The serotonergic system plays an important role in many psychiatric disorders. Its role in depression is well established (1). The majority of antidepressants, including TCAs, 6 cause increased synaptic serotonin (5-HT) levels via blockade of 5-HT reuptake into the presynaptic neuron (2-4) by competitive inhibition of hSERT. TCAs have been in clinical use since the 1950s, with imipramine being the first and most prominent compound (5). In severely depressed hospitalized patients, TCAs appear to be more efficacious than selective serotonin reuptake inhibitors (6). TCAs remain in widespread clinical use, especially for treatment-resistant depression (7).hSERT belongs to the neurotransmitter sodium symporter family (2, 8). These transporters utilize the electrochemical gradient of sodium and chloride ions to accumulate 5-HT against its own gradient (9 -11). No experimentally solved structures of the monoamine transporters exist, including hSERT and the dopamine and norepinephrine transporters. However, the structure of LeuT, a bacterial homolog of the neurotransmitter sodium symporters, in a substrate-occluded conformation, was reported in 2005 (12). Two sodium ions (12) and a chloride ion bind near the central substrate site (13-14) structurally and functionally coupling sodium and chloride binding to substrate binding. Recently, different transport mechanisms have been suggested (15,16).Subsequently, a low affinity noncompetitive binding site for TCAs in the extracellular vestibule of the LeuT 11 Å above the central binding site was identified (17,18). The relevance of the LeuT vestibular site for TCA binding to the physiologically relevant target, hSERT, is a matter of debate. This study identifies the central binding site, not the putative vestibular site, as relevant for TCA binding to hSERT and furthermore describes and validates the orientation of TCAs within this site.In this paper, we present induced fit docking studies of imipramine and selected analogs in the previously described homology models of hSERT (19). We present binding affinity studies of 10 imipramine analogs (Fig. 1 Copenhagen Ø, Denmark. 4 To whom correspondence may be addressed. E-mail: birgit@chem.au.dk. 5 To whom correspondence may be addressed. E-mail: owiborg@post.tele.dk. 6 The abbreviations used are: TCA, tricyclic antidepressant; 5-HT, serotonin; hSERT, human SERT; WT, wild type; PaMLAC, paired mutation ligand analog complementation; MD, molecular dynamics; r.m.s., root mean square.

Translocation through the extracellular vestibule and binding of leucine in the leucine transporter (LeuT) have been studied with molecular dynamics simulations. More than 0.1 mus of all-atom molecular dynamics simulations have been performed on different combinations of LeuT, bound substrate, and bound structural Na(+) ions to describe molecular events involved in substrate binding and in the formation of the occluded state and to investigate the dynamics of this state. Three structural features are found to be directly involved in the initial steps of leucine transport: a Na(+) ion directly coordinated to leucine (Na-1), two aromatic residues closing the binding site toward the extracellular vestibule (Tyr-108 and Phe-253), and a salt bridge in the extracellular vestibule (Arg-30 and Asp-404). These features account for observed differences between simulations of LeuT with and without bound substrate and for a possible pathway for leucine binding and thereby formation of the occluded LeuT binding site.

The two enantiomers of the antidepressant citalopram inhibit the human serotonin transporter substantially differently. Previous studies revealed Tyr95 and Ile172 as important for citalopram binding, however, the overall orientation of the ligands in the binding site and the protein-ligand interaction points remain unknown. The binding of S- and R-citalopram to a human serotonin transporter homology model are herein examined via docking simulations including induced fit effects. For a better description of the formal charges of the ligand when bound inside the protein, polarization effects of the protein were included by additional quantum-polarized ligand docking calculations, where ligand charges are evaluated using QM/MM calculations. By this approach a much clearer picture emerged of the positions of the functional groups of citalopram. The two enantiomers are predicted to bind in the substrate binding pocket with opposite orientations of their aromatic groups. The predicted binding modes are experimentally validated using human wild type and 15 serotonin transporter mutants and 13 optically pure citalopram analogues. Important protein-ligand interaction points were identified validating one binding model for each enantiomer. In the validated model of the high affinity enantiomer, S-citalopram, the fluorine atom is located near Ala173 and Thr439 and the cyano group is in close proximity of Phe341; these contacts are found to be reversed for the R-enantiomer.

We present 158 ns of unrestrained all-atom molecular dynamics (MD) simulations of the human estrogen receptor α ligand binding domain (ERα LBD) sampling the conformational changes upon binding of estradiol. The pivotal role of His524 in maintaining the protein structure in the biologically active agonist conformation is elucidated. With His524 modeled as the ε-tautomer, a conserved hydrogen bond to the ligand is found in the active complex. Helices 3 and 11 are held together by a hydrogen-bonding network from His524 to Glu339 via Glu419 and Lys531, arresting the ligand in the binding pocket and creating the “mouse trap” binding site for helix 12 (H12). The simulations reveal how His524 serves as a communication point between the two. When estradiol is bound, His524 is positioned correctly for the hydrogen bond network to be established. H12 is then positioned for interaction with the co-activator protein, leading to the biologically active complex. The conformational dynamics of ERα LBD is further investigated from simulations of antagonist and apo conformations of the protein. These simulations suggest a likely sequence of events for the transition from the inactive apo structure to the transcriptionally active conformation of ERα LBD. Stable conformations are identified where H12 is placed neither in the “mouse trap” nor in the co-activator binding groove, as is the case for antagonist structures of ERα LBD. Finally, the influence of such conformations on the biological function of ERα is discussed in relationship to the interaction with selective estrogen receptor modulators and endocrine-disrupting compounds.

Bovine chymosin is an aspartic protease that selectively cleaves the milk protein kappa-casein. The enzyme is widely used to promote milk clotting in cheese manufacturing. We have developed models of residues 97-112 of bovine kappa-casein complexed with bovine chymosin, using ligand docking, conformational search algorithms, and molecular dynamics simulations. In agreement with limited experimental evidence, the model suggests that the substrate binds in an extended conformation with charged residues on either side of the scissile bond playing an important role in stabilizing the binding pose. Lys111 and Lys112 are observed to bind to the N-terminal domain of chymosin displacing a conserved water molecule. A cluster of histidine and proline residues (His98-Pro99-His100-Pro101-His102) in kappa-casein binds to the C-terminal domain of the protein, where a neighboring conserved arginine residue (Arg97) is found to be important for stabilizing the binding pose. The catalytic site (including the catalytic water molecule) is stable in the starting conformation of the previously proposed general acid/base catalytic mechanism for 18 ns of molecular dynamics simulations.

Endocrine-disrupting compounds (EDCs) accumulating in nature are known to interact with nuclear receptors. Especially important is the human estrogen receptor R (hERR), and several EDCs are either known or suspected to influence the activity of the ligand-binding domain (LBD). We here present a comparative docking study of both well-known hERR ligands and small organic compounds, including selected polychlorinated biphenyls (PCBs), plasticizers, and pesticides, that are all potentially endocrinedisrupting, into different conformations of the hERR LBD. Three newly found quasi-stable structures of the hERR LBD are examined along with three crystallographic conformations of the protein, either the apo structure or using a protein structure with a bound agonist or antagonist ligand. The possible interactions between the protein and the potentially EDCs are described. It is found that most suspected EDCs can bind in the steroid binding cavity, interacting with at least one of the two hydrophilic ends of the steroid binding site. DDE, DDT, and HPTE are predicted to bind most strongly to the hERR LBD. It is predicted that these compounds can interact with the three conformations of hERR LBD with comparable affinities. The metabolic hydroxylation of aromatic compounds is found to lead to an increase in the binding affinity of PCBs as well as DDT. Docking into the quasi-stable conformations of the hERR LBD leads to computed binding affinities similar to or better than those calculated for the three X-ray structures, revealing that the new structures may be of importance for assessing the function of the influence of EDCs on nuclear receptors.

Biological function of proteins is frequently associated with the formation of complexes with small-molecule ligands. Experimental structure determination of such complexes at atomic resolution, however, can be time-consuming and costly. Computational methods for structure prediction of protein/ligand complexes, particularly docking, are as yet restricted by their limited consideration of receptor flexibility, rendering them not applicable for predicting protein/ligand complexes if large conformational changes of the receptor upon ligand binding are involved. Accurate receptor models in the ligand-bound state (holo structures), however, are a prerequisite for successful structure-based drug design. Hence, if only an unbound (apo) structure is available distinct from the ligandbound conformation, structure-based drug design is severely limited. We present a method to predict the structure of protein/ligand complexes based solely on the apo structure, the ligand and the radius of gyration of the holo structure. The method is applied to ten cases in which proteins undergo structural rearrangements of up to 7.1 A backbone RMSD upon ligand binding. In all cases, receptor models within 1.5 A backbone RMSD to the target were predicted and close-to-native ligand binding poses were obtained for eight of ten cases in the top-ranked complex models. The developed protocol is expected to enable structure modeling of protein/ligand complexes and structure-based drug design for cases where crystal structures of ligand-bound conformations are not available.