QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.103.054593v1 307/1/34    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ravna, A. W. Articles by Dahl, S. G. Search for Related Content PubMed PubMed Citation Articles by Ravna, A. W. Articles by Dahl, S. G.

CELLULAR AND MOLECULAR

Molecular Mechanism of Citalopram and Cocaine Interactions with Neurotransmitter Transporters

Department of Pharmacology, Institute of Medical Biology, University of Tromsø, Tromsø, Norway

Received May 20, 2003; accepted June 20, 2003.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
The selective serotonin reuptake inhibitors (SSRIs) and cocaine bind to the neural serotonin (5-HT) transporter (SERT) and thus inhibit presynaptic reuptake of 5-HT and elevate its concentration in the synaptic cleft. Cocaine also binds to the dopamine transporter (DAT) and to the noradrenaline transporter (NET) and inhibits presynaptic reuptake of dopamine and noradrenaline. SERT, DAT, and NET belong to the sodium/neurotransmitter symporter family, which is predicted to have a molecular structure with 12 transmembrane -helices (TMHs) and intracellular amino- and carboxy terminals. We used an electron density projection map of the Escherichia coli Na⁺/H⁺ anti-porter, and site-directed mutagenesis data on DAT and SERT to construct 3-dimensional molecular models of SERT, DAT and NET. These models were used to simulate the molecular interaction mechanisms of the SSRI, S-citalopram, its less potent enantiomer, R-citalopram and of cocaine with the transporters. In the SERT model, a single amino acid (Tyr95) in TMH1 determined the transporter selectivity of S-citalopram for SERT over DAT and NET. A dipole-dipole interaction was formed between the hydroxy group of Tyr95 in SERT and the nitril group of S-citalopram, but could not be formed by S-citalopram in DAT and NET where the corresponding amino acid is a phenylalanine. The lower binding affinity of R-citalopram may be due to sterical hindrance at the binding site. The tropane ring of cocaine interacted with Tyr95 in SERT and with the corresponding phenylalanines in NET and DAT. This may explain why cocaine, but not S-citalopram, has high binding affinity to all three transporters.

Binding studies of cocaine, imipramine, and the SSRIs paroxetine and citalopram indicate the existence of more than one binding site on the SERT. Cocaine, citalopram, and imipramine act as competitive inhibitors of 5-HT reuptake (Barker and Blakely, 1995; Sur et al., 1998). Paroxetine does not bind at the 5-HT recognition site and seems to inhibit reuptake by allosteric modulation of the transporter (Akunne et al., 1992).

Cocaine is nonselective and inhibits reuptake of 5-HT, dopamine, and noradrenaline (Barker and Blakely, 1995). Imipramine also inhibits noradrenaline reuptake by the noradrenaline transporter (NET) but has slightly lower affinity for NET than for SERT. SSRIs have from 300 to 3,500 times higher binding affinity to SERT than to NET (Tatsumi et al., 1997). S-Citalopram is a highly SERT-selective compound and about 30-fold more potent as a 5-HT reuptake inhibitor than its enantiomer R-citalopram (Hyttel, 1994; Owens et al., 2001).

SERT, DAT, and NET belong to the sodium/neurotransmitter symporter family (SNF) of proteins that act as co-transporters of sodium ions, chloride ions, and transmitter molecules (Barker and Blakely, 1995). A sodium gradient across the cell membrane provides energy for an inward movement of the transmitter molecule as part of the regulatory mechanisms for the neurotransmitter. The SNF transporter family belongs to the larger group of secondary transporters, which is predicted to have a molecular structure with 12 transmembrane -helices (TMHs) and the amino- and carboxy terminals localized intracellularly (Reizer et al., 1994; Rothman et al., 1996; Kaback and Wu, 1997; Chen et al., 1998). The sequence identity within the SNF family is 40 to 90% (Reizer et al., 1994).

The human SERT is a membrane protein with 630 amino acids (Ramamoorthy et al., 1993). The region including the N-terminal and TMH1 is more conserved within the SNF transporter family than the subsequent segments. Hydropathy plots of SNF protein sequences indicate that this region has a hydrophilic character and appears to be an amphipatic -helix with several conserved polar residues localized at one side of the helical axis (Reizer et al., 1994). This region was predicted by the Swiss-Prot database prediction service (Bairoch and Apweiler, 1999) to be TMH1. Functionally important residues for ligand interactions of SERT and DAT are situated in this region. These include Asp98 (Barker et al., 1999) and Tyr95 (Barker et al., 1998) in SERT and Asp79 (Kitayama et al., 1992) and Phe76 (Lin et al., 1999) in DAT.

No detailed three-dimensional molecular structure of any SNF protein has been reported. The secondary transporter proteins include the Na⁺/H⁺ antiporter (NhaA) and the lactose permease symporter (lac permease). An electron density projection map of the Escherichia coli NhaA (Williams, 2000) has confirmed a topology with 12 membrane-spanning domains. Site-directed mutagenesis studies and biophysical studies on lac permease also indicate 12 TMHs (Kaback and Wu, 1997). Based on the hypothesis that the E. coli NhaA, the E. coli lac permease, and the mammalian SNF proteins all belong to a common protein folding class, the electron density projection map of NhaA (Williams, 2000) provides an experimental basis for construction of three-dimensional molecular models of SERT, NET, and other secondary transporters.

We have previously proposed a complete model of the DAT based on the NhaA projection map (Ravna et al., 2003). In the present study, a molecular model of the entire SERT molecule and a 12-TMH model of NET were constructed in a similar way. S-Citalopram and cocaine were docked into putative binding sites in the SERT, NET, and DAT models to examine the molecular mechanisms behind the differences in their binding affinities. R-Citalopram was docked into the putative SERT binding site to explain the difference in binding affinity between the two enantiomers of citalopram.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Computational Procedures. The MIDAS plus software (Ferrin et al., 1988) was used for interactive molecular graphics, and the AMBER 5.0 program package (University of California, San Francisco, CA) (Cornell et al., 1995) was used for empirical force field calculations of molecular structures. Energy minimizations were performed by 500 cycles of steepest descent minimization followed by 2,000 steps of conjugate gradient minimization or until convergence with an energy gradient difference between successive steps of 0.0001 kcal/mol for the ligands and 0.02 kcal/mol for SERT and the transporter-ligand complexes. A distance dependent dielectric function ( = 4r_ij), where r_ij is the distance between atoms i and j, and a 15-Å cutoff radius was used in molecular mechanics calculations of three-dimensional structures and in molecular dynamics (MD) simulations. Molecular dynamics are thermally driven molecular movements, which are important for the activity and mechanisms of biologically active molecules (Karplus and McCammon, 2002). The present MD simulations were done as a means of structural refinement since the addition of thermal energy allows the structure to move over potential energy barriers during the simulations. To increase the strength of the hydrogen bonds in the 12 TMHs during MD refinement of the SERT model and of the SERT-ligand complexes, the backbone O-N intramolecular helical interactions, excluding proline residues, were restrained with a constant of 5.0 kcal/mol during the energy calculations. Similar strategies for constraining transmembrane helices, together with a dielectric constant of 4 and a distance dependent dielectric function ( = 4r_ij), has also been used in previous MD studies of membrane proteins (Sylte et al., 2001). The SERT model was subjected to 30 ps of MD with "gradual heating" (0–300 K) followed by 150 ps of MD simulation at 300 K.

Ligand Molecules. The chemical structures of citalopram and cocaine are shown in Fig. 1. Atomic coordinates, point charges, and empirical force field parameters for the molecular mechanics calculations were taken from previous molecular modeling studies (Ravna and Edvardsen, 2001; Ravna et al., 2003).

View larger version (14K): [in this window] [in a new window]   Fig. 1. Schematic representation of docking positions: S-citalopram (positions 1 and 2) and cocaine (positions 3 and 4) in the putative ligand binding area in SERT.

Construction and Refinement of the SERT Model. The SERT model was constructed from a model of NhaA (Ravna et al., 2001) and the coordinates of a recent model of DAT (Ravna et al., 2003). The NhaA model, which was based on the electron density projection map of E. coli NhaA (Williams, 2000), was used to generate the transmembrane part of the DAT model (Ravna et al., 2003). The loops and terminals of DAT were constructed from multiple amino sequence alignments of SNF proteins, secondary structural predictions, loop searches in the protein data bank (PDB) database (Berman et al., 2000), and experimental ligand binding studies of SERT and DAT. The SERT model was constructed from the DAT model (Ravna et al., 2003) by substituting the amino acids of DAT into the corresponding amino acids of SERT, according to a previously published amino acid sequence alignment (Ravna et al., 2003).

Deletions in the SERT sequence aligned to the DAT sequence in the extracellular loop between TMH3 and TMH4 (five residues), and seven residues in the carboxy terminal were removed from the DAT coordinate file. Insertions in the SERT sequence aligned to the DAT sequence, three single amino acids and one triplet of amino acids, were added and connected to the existing loop conformations in the PDB file. An initial three-dimensional model of the first 16 amino acids in the amino terminal of SERT was obtained from the PDB database (Berman et al., 2000).

The side chains of the entire model were energy minimized, followed by 50 ps of MD simulation at 300 K. Refinement of the loops was performed by energy minimization with constrained TMHs, followed by 30 ps of MD with gradual heating from 0 to 300 K during the simulations. The entire model was then energy minimized and subjected to 30 ps of MD with gradual heating (0–300 K) followed by 150 ps of MD at 300 K, giving a total of 180 ps simulation. Backbone O-N intramolecular helical interactions were restrained during calculations with the entire model. The average structure from coordinates sampled each picosecond during the last 50 ps of the MD refinement was calculated using the CARNAL module of the AMBER program package.

Construction of the 12 TMH NET Model. A model of the 12 TMHs of NET was constructed using the SERT model as a template by substituting the amino acids the transmembrane segments of SERT into the corresponding amino acids of NET, according to the amino acid sequence alignment (Ravna et al., 2003), and removing the loop- and terminal segments from the coordinate file. The 12 TMH NET model was energy minimized.

Docking and Refinement of SERT-Ligand Interactions. Several putative binding positions for S-citalopram and cocaine in the SERT model were considered. Nevertheless, the two positions for each ligand that were most in agreement with results from previous site-directed mutagenesis studies (Table 1) were considered in detail, both using Asp98 of TMH1 as an anchoring point. The two S-citalopram positions in SERT differed regarding the orientation of the fluorobenzyl group, which was oriented toward Phe263 and Met260 (TMH4) in one position (position 1; Fig. 1) and toward Ile172 (TMH3) in the other position (position 2; Fig. 1). The two initial positions of cocaine in SERT differed regarding the orientation of the two ester groups relative to Tyr267 (TMH4) and Tyr289 (TMH5) (Fig. 1). In both positions, the ester groups formed hydrogen bonds with these tyrosine residues. The benzoate carboxyl ester interacted with Tyr267 in one position (position 3) and with Tyr289 the other position (position 4) and vice versa for the other ester group. The four ligand-transporter complexes were energy minimized until convergence followed by MD refinement with 30 ps of gradual heating from 0 to 300 K. R-Citalopram was docked into the SERT model in positions analogous to docking positions 1 and 2 for S-citalopram.

DAT-Ligand Docking. To examine the molecular interactions responsible for the differences in binding affinities of cocaine and S-citalopram to SERT and DAT, these ligands were also docked into the DAT model (Ravna et al., 2003) using Asp79 as an anchoring point and similar positions as used for SERT. The docking of cocaine into the DAT model was performed as described in a previous DAT modeling study (Ravna et al., 2003).

NET-Ligand Docking. S-Citalopram and cocaine were docked into the putative binding area of the NET model in positions analogous to the docking positions for SERT using Asp75 as an anchoring point to examine why S-citalopram binds selectively to SERT while cocaine is nonselective.

	Results

Top Abstract Materials and Methods Results Discussion References

 
The average SERT structure calculated from all coordinate sets observed between 130 and 180 ps of MD simulation is shown in Fig. 2. In the model, a pore is formed between TMHs 1, 3, 4, 5, 7, and 11 and possibly TMHs 2, 8, and 12 (Fig. 3). This pore might represent a 5-HT translocation pathway and ligand binding area.

View larger version (44K): [in this window] [in a new window]   Fig. 2. Backbone structure of energy refined SERT model viewed in the membrane plane.

View larger version (15K): [in this window] [in a new window]   Fig. 3. Schematic representation of the 12 TMH arrangements of SERT viewed from the extracellular side.

During 150 ps of MD refinement, TMHs 1, 4, and 5 moved 0.5–1 Å closer to the pore center, making the pore too narrow for ligand docking. TMH12 tilted about 5 degrees toward the extracellular end of TMH4, and TMHs 9 and 10 both moved about 1.5 Å toward TMH11 during MD simulation. The remaining helices were positioned similarly to their positions before MD simulation.

S-Citalopram-SERT and Cocaine-SERT Interactions. S-Citalopram-SERT (position 1) and cocaine-SERT (position 3) complexes after 30 ps of MD with gradual heating are shown in Fig. 4, A and B. Both ligands interacted with Asp98, Tyr95, Tyr267, and Tyr289. The nitril group of S-citalopram interacted with Tyr95 and Tyr267. The positively charged nitrogen atoms of the ligands interacted with Asp98, forming a salt bridge. Other residues interacting with both ligands were Ile172, Tyr176 (TMH3), Met260, Phe263 (TMH4), Ala285, Leu292 (TMH5), and Phe551 (TMH11). Ser372, Ser375 (TMH7), and Leu 547 (TMH11) lined this ligand binding area and may also interact with the ligands.

View larger version (38K): [in this window] [in a new window]   Fig. 4. S-Citalopram (A), cocaine (B), and R-citalopram (C) in their putative binding sites on SERT. Color coding of atoms: N, blue; O, red; C, gray; H, dark gray; F, green.

R-Citalopram-SERT Interactions. Figure 4C shows R-citalopram docked at the putative binding area in SERT with the fluorobenzyl group oriented toward Phe263 and Met260 in TMH4, corresponding to position 1 for S-citalopram. Both R- and S-citalopram interacted with Asp98 and Tyr95 of TMH1, but they differed regarding the interaction with TMH4. S-Citalopram interacted with Phe263 and Met260, while the fluorobenzyl group of R-citalopram overlapped with the aromatic side chain of Phe263 and thus hindered the ligand sterically to fit into the binding site. R-Citalopram did not interact with Met260. When the enantiomeres where positioned with the phenyl group facing TMH3 (position 2), the fluorobenzyl group of S-citalopram interacted with Ile172 (TMH3), while the fluorobenzyl group of R-citalopram interacted with Gly94 (TMH1).

S-Citalopram-DAT and Cocaine-DAT Interactions. The nitril group of S-citalopram was close to Phe76, which corresponds to Tyr95 in SERT. Cocaine also interacted with Phe76, forming a hydrophobic interaction between the benzene ring of this amino acid and the tropane ring of cocaine. Both ligands interacted with Asp79 (TMH1), Tyr252 (TMH4), and Tyr274 (TMH5) of DAT.

S-Citalopram-NET and Cocaine-NET Interactions. Docking of S-citalopram and cocaine into the putative binding area of NET in positions analogous to 1 for SERT (Fig. 1) indicated that both ligands formed salt bridges with Asp75, which corresponds to Asp98 in the SERT structure (Fig. 5). In NET, the amino acid residue corresponding to Tyr95 in SERT is Phe72. As observed in the docking of S-citalopram in DAT, the nitril group of S-citalopram and the tropane ring of cocaine interacted with Phe72. Both ligands interacted with Tyr249 (TMH4) and Tyr271 (TMH5).

View larger version (35K): [in this window] [in a new window]   Fig. 5. S-Citalopram (A) and cocaine (B) in their putative binding sites on NET. Color coding of atoms as in Fig. 4.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
The present SERT model was based on an assumed structural resemblance between the SNF transporters and NhaA (Ravna et al., 2001). These transporters share a common functional mechanism, using an ion gradient as energy source for translocation of molecules against a concentration gradient, which supports the assumption of a common folding motif. This has not yet been confirmed, however, and it should be noted that the sequence identity within the transmembrane regions between NhaA and the SNF proteins SERT, DAT, and NET is rather low (8–11%). Nevertheless, the sequence identity within the transmembrane regions between G-protein coupled receptors (GPCR) and bacteriorhodopsin is 6–11% (Hibert et al., 1991), and in a similar way, an electron density projection map of bacteriorhodopsin was successfully used as a template in early molecular models of GPCRs (Hibert et al., 1991) before the rhodopsin structure was available. The X-ray structure of the GPCR rhodopsin (Palczewski et al., 2000) confirmed that these proteins share a similar overall three-dimensional folding pattern with seven TMHs and showed that using bacteriorhodopsin as a template was a reasonable approach for the initial models of GPCRs.

No X-ray structures of structurally related proteins are available as a template for modeling of secondary transporter proteins. The recently published X-ray structures of the E. coli MsbA, which belongs to the multidrug resistance ATP binding cassette transporters (Chang and Roth, 2001), and the bacterial multidrug efflux transporter AcrB (Murakami et al., 2002) were not considered as suitable templates due to large differences in their functional mechanisms compared with the SNF transporter proteins. It therefore appears that NhaA currently provides the best available structural basis for molecular modeling of SNF transporter proteins.

A previous 12 TMH SERT model was based only on studies of cocaine and citalopram binding to mutated SERT and DAT proteins (Ravna and Edvardsen, 2001), with focus on the binding site of cocaine since no template structure then was available. The previous SERT model proposed two aqueous pores, one as a possible 5-HT translocation and cocaine binding area. The electron density projection map of the Escherichia coli NhaA (Williams, 2000), which later was published, provides a better structural basis for modeling of 12 TMH secondary transporters and should allow construction of more precise structural models of SERT and other SNF proteins, provided that the hypothesis of their common folding pattern is correct. Both the present and the previous SERT models have aqueous pores consisting of TMHs 1, 2, 3, 4, 5, 7, and 11. In addition, the present SERT model indicate that TMHs 8 and 12 in may be involved 5-HT translocation and cocaine binding. The cocaine binding area in both models include residues Tyr95, Asp98 (TMH1), Ile172, Tyr176 (TMH3), Tyr267 (TMH4), Tyr289 (TMH5), and Phe551 (TMH11) (Table 1). The loops and terminals were included in the present SERT and DAT models to get a more correct distribution of masses and electrostatic fields in the simulations than in a model of only the TMHs.

Docking of S-citalopram and cocaine into the putative binding area of the SERT model indicated that both ligands participate in hydrogen bonds with Tyr176 (TMH3), Tyr267 (TMH4), and Tyr289 (TMH5). Similarly, a hydrogen bond interaction between the benzoate carboxyl of cocaine and the hydroxyl group of a tyrosine residue has been demonstrated in the crystal structure of a bacterial cocaine esterase complex (Larsen et al., 2002). In the present SERT model, the benzoate carboxyl of cocaine interacted with Tyr267 in docking position 3 and Tyr289 in docking position 4 (Fig. 1). The positively charged nitrogen atoms of both ligands interacted with Asp98 and created a salt bridge, and a dipole-dipole interaction of the nitril group of S-citalopram and Tyr95 was observed for both docking positions. There was a hydrophobic interaction between the benzene ring of Tyr95 and the tropane ring of cocaine. It has been demonstrated, using the high-affinity fluorescent cocaine-analog RTI-233, that the cocaine interaction area in SERT is highly hydrophobic (Rasmussen et al., 2001). Hydrophobic residues in the binding area of the SERT model are Ile172, Met180 (TMH3), Met260, Phe263 (TMH4), Ala285, Leu292 (TMH5), and Phe551 (TMH11). Met180 has recently been proposed to be part of a citalopram binding pocket in a cross-species analysis comparing human and bovine SERT (Mortensen et al., 2001). In the present SERT model, Met180 was located in the binding area and directly involved in binding of S-citalopram.

The importance of S-citalopram binding to Tyr95 in SERT has been experimentally demonstrated with a Tyr95Phe mutant (Barker et al., 1998). The present models of S-citalopram interacting with DAT and NET were in agreement with this hypothesis. The amino acid residue corresponding to Tyr95 in SERT is Phe76 in DAT and Phe72 in NET (Table 2). Phe76 in the DAT model and Phe72 in the NET model did not participate in dipole-dipole interactions with the nitril group of S-citalopram and do not have the possibility of forming hydrogen bonds. Therefore, the interaction between the nitril group of S-citalopram and a corresponding phenylalanine residue in DAT and NET would be weaker than between the nitril group of S-citalopram and Tyr95 in SERT. The interaction between the nitril group of S-citalopram and Tyr95 in SERT might explain why S-citalopram is a selective 5-HT reuptake inhibitor. The tropane ring of cocaine interacted with the benzene rings of Tyr95 in SERT, Phe76 of DAT, and Phe72 of NET, which might explain why cocaine has almost the same affinity for all three transporters.

The role of Tyr271 (TMH5) as a determinant for cocaine binding to NET has been confirmed by site-directed mutagenesis studies (Paczkowski and Bryan-Lluka, 2001). The roles of Tyr249 and Phe531 in cocaine binding to NET is unclear, however. The corresponding amino acids in human DAT, Tyr251, and Phe534 (Tyr533 in rDAT) are involved in cocaine binding as shown by site-directed mutagenesis studies (Kitayama et al., 1996; Mitsuhata et al., 1998), while involvement of Tyr249 and Phe531 in cocaine binding to NET not has been confirmed (Paczkowski and Bryan-Lluka, 2001).

When comparing the docking of the two citalopram enantiomers into SERT, S-citalopram fitted well into the binding area in positions 1 and 2. In position 1, R-citalopram was hindered sterically by overlap between the phenyl group of R-citalopram and Phe263 in TMH4 (Fig. 4C). In position 2, the R-enantiomer did not have contact with Ile172 in TMH3, which the S-enantiomer had. S-Citalopram also had van der Waals contact with Phe263 and Met260 in TMH4 in position 1 and with Ile172 in TMH3 in position 2. This may explain why R-citalopram is about 30-fold less potent than S-citalopram as a 5-HT reuptake inhibitor (Hyttel, 1994; Owens et al., 2001).

The ligand binding area in SERT was modeled from experimental results from studies of cocaine and citalopram binding to mutated SERT and DAT proteins (Table 1). The interpretation of such site-directed mutagenesis data are not unambiguous, however, since the results do not show whether the observed effects are due to direct disruption of side chain-ligand interactions or caused by overall structural changes of the binding region induced by the mutations.

According to studies on lac permease, the 12 TMHs of secondary transporters may be loosely packed in an active state associated with transport of substrate molecules. Water molecules are likely to be present in the cavities, and widespread cooperative conformational changes including sliding and tilting motions of the TMHs may occur during ion and substrate transport (Kaback and Wu, 1997). Therefore, the SERT transport mechanism may involve several conformational changes in the transporter, both in TMHs and in loop segments. Several experimental studies provide support for the involvement of TMH1 (Adkins et al., 2001), TMH3 (Chen et al., 1997), and TMH7 (Penado et al., 1998; Kamdar et al., 2001) in SERT and DAT substrate recognition and transport.

Explicit water molecules or membrane phospholipids were not included in the calculations since the present transporter models have a relatively low level of accuracy. The omission of solvation and membrane molecules may influence the structure and dynamics of the transporter models. To compensate for this, calculations were performed with a distance dependent dielectric function ( = 4r) and hydrogen bonding restraints on TMHs.

Structural information about SERT and its molecular interactions with cocaine and citalopram is important for understanding the molecular mechanisms of action of these drugs and for development of drugs with improved potency and selectivity. The present model may be used as a tool for designing of further experimental structural studies of SERT, DAT, NET, and their ligand interactions.

In conclusion, the interaction of the nitril group of S-citalopram with Tyr95 of SERT might explain why S-citalopram has higher affinity for SERT than for DAT and NET. Sterical hindrance between the phenyl group of R-citalopram and Phe263 may explain why R-citalopram is 30-fold less potent than S-citalopram as a 5-HT reuptake inhibitor. The tropane ring of cocaine interacted with the benzene rings of Tyr95 in SERT and also with the corresponding phenylalanines in NET and DAT, which may explain why cocaine has relatively high binding affinities to all three transporters. Coordinates of the SERT, NET, and DAT models are available from the authors upon request.

	Footnotes

 
The present study was supported by computer time on the HP RISC super-computer at the University of Tromsø (Tromsø, Norway).

DOI: 10.1124/jpet.103.054593.

ABBREVIATIONS: 5-HT, serotonin; SSRIs, serotonin reuptake inhibitors; SERT, serotonin transporter; NET, noradrenaline transporter; DAT, dopamine transporter; SNF, sodium/neurotransmitter symporter family; TMH, transmembrane -helix; NhaA, Na⁺/H⁺ antiporter; lac permease, lactose permease symporter; MD, molecular dynamics; PDB, protein data bank; GPCR, G-protein coupled receptors; RTI-233, N-(N-methyl-N-(4-nitrobenzo-2-oxa-1,3-diazol-7-yl)ethanolamine ester hydrochloride.

Address correspondence to: Dr. Svein G. Dahl, Department of Pharmacology, I.M.B., University of Tromsø, N-9037 Tromsø, Norway. E-mail: sgd{at}fagmed.uit.no

	References

Top Abstract Materials and Methods Results Discussion References

 

Adkins EM, Barker EL, and Blakely RD (2001) Interactions of tryptamine derivatives with serotonin transporter species variants implicate transmembrane domain I in substrate recognition. Mol Pharmacol 59: 514–523.[Abstract/Free Full Text]

Akunne HC, de Costa BR, Jacobson AE, Rice KC, and Rothman RB (1992) [3H] cocaine labels a binding site associated with the serotonin transporter in guinea pig brain: allosteric modulation by paroxetine. Neurochem Res 17: 1275–1283.[CrossRef][Medline]

Bairoch A and Apweiler R (1999) The SWISS-PROT protein sequence data bank and its supplement TrEMBL in 1999. Nucleic Acids Res 27: 49–54.[Abstract/Free Full Text]

Barker EL and Blakely RD (1995). Norepinephrine and serotonin transporters: molecular targets of antidepressant drugs, in Psychopharmacology: The Fourth Generation of Progress (Bloom FE and Kupfer DJ eds) pp 321–334, Raven Press, New York.

Barker EL, Moore KR, Rakhshan F, and Blakely RD (1999) Transmembrane domain I contributes to the permeation pathway for serotonin and ions in the serotonin transporter. J Neurosci 19: 4705–4717.[Abstract/Free Full Text]

Barker EL, Perlman MA, Adkins EM, Houlihan WJ, Pristupa ZB, Niznik HB, and Blakely RD (1998) High affinity recognition of serotonin transporter antagonists defined by species-scanning mutagenesis. An aromatic residue in transmembrane domain I dictates species-selective recognition of citalopram and mazindol. J Biol Chem 273: 19459–19468.[Abstract/Free Full Text]

Berman HM, Westbrook J, Feng Z, Gilliland G, Bhat TN, Weissig H, Shindyalov IN, and Bourne PE (2000) The Protein Data Bank. Nucleic Acids Res 28: 235–242.[Abstract/Free Full Text]

Chang G and Roth CB (2001) Structure of MsbA from E. coli: a homolog of the multidrug resistance ATP binding cassette (ABC) transporters. Science (Wash DC) 293: 1793–1800.[Abstract/Free Full Text]

Chen JG, Liu-Chen S, and Rudnick G (1998) Determination of external loop topology in the serotonin transporter by site-directed chemical labeling. J Biol Chem 273: 12675–12681.[Abstract/Free Full Text]

Chen JG, Sachpatzidis A, and Rudnick G (1997) The third transmembrane domain of the serotonin transporter contains residues associated with substrate and cocaine binding. J Biol Chem 272: 28321–28327.[Abstract/Free Full Text]

Cornell WD, Cieplak P, Bayly CI, Gould IR, Merz KM, Ferguson DM, Spellmeyer DC, Fox T, Caldwell JW, and Kollman PA (1995) A 2nd generation force-field for the simulation of proteins, nucleic-acids and organic-molecules. J Am Chem Soc 117: 5179–5197.[CrossRef]

Ferrin TE, Huang CC, Jarvis LE, and Langridge R (1988) The MIDAS display system. J Mol Graphics 6: 13.

Hibert MF, Trumpp-Kallmeyer S, Bruinvels A, and Hoflack J (1991) Three-dimensional models of neurotransmitter G-binding protein-coupled receptors. Mol Pharmacol 40: 8–15.[Abstract]

Hyttel J (1994) Pharmacological characterization of selective serotonin reuptake inhibitors (SSRIs). Int Clin Psychopharmacol 9 (Suppl 1): 19–26.

Kaback HR and Wu J (1997) From membrane to molecule to the third amino acid from the left with a membrane transport protein. Q Rev Biophys 30: 333–364.[CrossRef][Medline]

Kamdar G, Penado KM, Rudnick G, and Stephan MM (2001) Functional role of critical stripe residues in transmembrane span 7 of the serotonin transporter. Effects of Na⁺, Li⁺ and methanethiosulfonate reagents. J Biol Chem 276: 4038–4045.[Abstract/Free Full Text]

Karplus M and McCammon JA (2002) Molecular dynamics simulations of biomolecules. Nat Struct Biol 9: 646–652.[CrossRef][Medline]

Kitayama S, Morita K, Dohi T, Wang JB, Davis SC, and Uhl GR (1996) Dissection of dopamine and cocaine binding sites on the rat dopamine transporter expressed in COS cells. Ann N Y Acad Sci 801: 388–393.[CrossRef][Medline]

Kitayama S, Shimada S, Xu H, Markham L, Donovan DM, and Uhl GR (1992) Dopamine transporter site-directed mutations differentially alter substrate transport and cocaine binding. Proc Natl Acad Sci USA 89: 7782–7785.[Abstract/Free Full Text]

Larsen NA, Turner JM, Stevens J, Rosser SJ, Basran A, Lerner RA, Bruce NC, and Wilson IA (2002) Crystal structure of a bacterial cocaine esterase. Nat Struct Biol 9: 17–21.[CrossRef][Medline]

Lee SH, Chang MY, Lee KH, Park BS, Lee YS, and Chin HR (2000) Importance of valine at position 152 for the substrate transport and 2-carbomethoxy-3-(4-fluorophenyl)tropane binding of dopamine transporter. Mol Pharmacol 57: 883–889.[Abstract/Free Full Text]

Lin Z, Wang W, Kopajtic T, Revay RS, and Uhl GR (1999) Dopamine transporter: transmembrane phenylalanine mutations can selectively influence dopamine uptake and cocaine analog recognition. Mol Pharmacol 56: 434–447.[Abstract/Free Full Text]

Mitsuhata C, Kitayama S, Morita K, Vandenbergh D, Uhl GR, and Dohi T (1998) Tyrosine-533 of rat dopamine transporter: involvement in interactions with 1-methyl-4-phenylpyridinium and cocaine. Brain Res Mol Brain Res 56: 84–88.[Medline]

Mortensen OV, Kristensen AS, and Wiborg O (2001) Species-scanning mutagenesis of the serotonin transporter reveals residues essential in selective, high-affinity recognition of antidepressants. J Neurochem 79: 237–247.[CrossRef][Medline]

Murakami S, Nakashima R, Yamashita E, and Yamaguchi A (2002) Crystal structure of bacterial multidrug efflux transporter AcrB. Nature (Lond) 419: 587–593.[CrossRef][Medline]

Owens MJ, Knight DL, and Nemeroff CB (2001) Second-generation SSRIs: human monoamine transporter binding profile of escitalopram and R-fluoxetine. Biol Psychiatry 50: 345–350.[CrossRef][Medline]

Paczkowski FA and Bryan-Lluka LJ (2001) Tyrosine residue 271 of the norepinephrine transporter is an important determinant of its pharmacology. Brain Res Mol Brain Res 97: 32–42.[Medline]

Palczewski K, Kumasaka T, Hori T, Behnke CA, Motoshima H, Fox BA, Le Trong I, Teller DC, Okada T, Stenkamp RE, Yamamoto M, and Miyano M (2000) Crystal structure of rhodopsin: a G protein-coupled receptor. Science (Wash DC) 289: 739–745.[Abstract/Free Full Text]

Penado KM, Rudnick G, and Stephan MM (1998) Critical amino acid residues in transmembrane span 7 of the serotonin transporter identified by random mutagenesis. J Biol Chem 273: 28098–28106.[Abstract/Free Full Text]

Ramamoorthy S, Bauman AL, Moore KR, Han H, Yang-Feng T, Chang AS, Ganapathy V, and Blakely RD (1993) Antidepressant- and cocaine-sensitive human serotonin transporter: molecular cloning, expression and chromosomal localization. Proc Natl Acad Sci USA 90: 2542–2546.[Abstract/Free Full Text]

Rasmussen SG, Carroll FI, Maresch MJ, Jensen AD, Tate CG, and Gether U (2001) Biophysical characterization of the cocaine binding pocket in the serotonin transporter using a fluorescent cocaine analogue as a molecular reporter. J Biol Chem 276: 4717–4723.[Abstract/Free Full Text]

Ravna AW and Edvardsen O (2001) A putative three-dimensional arrangement of the human serotonin transporter transmembrane helices: a tool to aid experimental studies. J Mol Graph Model 20: 133–144.[CrossRef][Medline]

Ravna AW, Sylte I, and Dahl SG (2001) Molecular model of the Echerichia coli Na⁺/H⁺ antiporter NhaA. Receptors and Channels 7: 319–328.[Medline]

Ravna AW, Sylte I, and Dahl SG (2003) Molecular model of the neural dopamine transporter. J Comput Aided Mol Des 7: 319–328.

Reizer J, Reizer A, and Saier MH Jr (1994) A functional superfamily of sodium/solute symporters. Biochim Biophys Acta 1197: 133–166.[Medline]

Rothman A, Padan E, and Schuldiner S (1996) Topological analysis of NhaA, a Na⁺/H⁺ antiporter from Escherichia coli. J Biol Chem 271: 32288–32292.[Abstract/Free Full Text]

Sur C, Betz H, and Schloss P (1998) Distinct effects of imipramine on 5-hydroxy-tryptamine uptake mediated by the recombinant rat serotonin transporter SERT1. J Neurochem 70: 2545–2553.[Medline]

Sylte I, Bronowska A, and Dahl SG (2001) Ligand induced conformational states of the 5-HT_1A receptor. Eur J Pharmacol 416: 33–41.[CrossRef][Medline]

Tatsumi M, Groshan K, Blakely RD, and Richelson E (1997) Pharmacological profile of antidepressants and related compounds at human monoamine transporters. Eur J Pharmacol 340: 249–258.[CrossRef][Medline]

Williams KA (2000) Three-dimensional structure of the ion-coupled transport protein NhaA. Nature (Lond) 403: 112–115.[CrossRef][Medline]

This article has been cited by other articles:

				L. Celik, B. Schiott, and E. Tajkhorshid Substrate Binding and Formation of an Occluded State in the Leucine Transporter Biophys. J., March 1, 2008; 94(5): 1600 - 1612. [Abstract] [Full Text] [PDF]

				F. A. Paczkowski, I. A. Sharpe, S. Dutertre, and R. J. Lewis {chi}-Conotoxin and Tricyclic Antidepressant Interactions at the Norepinephrine Transporter Define a New Transporter Model J. Biol. Chem., June 15, 2007; 282(24): 17837 - 17844. [Abstract] [Full Text] [PDF]

				S. G. Dahl and I. Sylte From genomics to drug targets. J Psychopharmacol, July 1, 2006; 20(4 Suppl): 95 - 99. [Abstract] [PDF]

				L. K. Henry, J. R. Field, E. M. Adkins, M. L. Parnas, R. A. Vaughan, M.-F. Zou, A. H. Newman, and R. D. Blakely Tyr-95 and Ile-172 in Transmembrane Segments 1 and 3 of Human Serotonin Transporters Interact to Establish High Affinity Recognition of Antidepressants J. Biol. Chem., January 27, 2006; 281(4): 2012 - 2023. [Abstract] [Full Text] [PDF]

				M. B. Larsen, B. Elfving, and O. Wiborg The Chicken Serotonin Transporter Discriminates between Serotonin-selective Reuptake Inhibitors: A SPECIES-SCANNING MUTAGENESIS STUDY J. Biol. Chem., October 1, 2004; 279(40): 42147 - 42156. [Abstract] [Full Text] [PDF]

				S. G. Dahl, I. Sylte, and A. W. Ravna Structures and Models of Transporter Proteins J. Pharmacol. Exp. Ther., June 1, 2004; 309(3): 853 - 860. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.103.054593v1 307/1/34    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ravna, A. W. Articles by Dahl, S. G. Search for Related Content PubMed PubMed Citation Articles by Ravna, A. W. Articles by Dahl, S. G.