![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
NEUROPHARMACOLOGY
Department of Medicinal Chemistry and Molecular Pharmacology, Purdue University School of Pharmacy and Pharmaceutical Sciences, West Lafayette, Indiana (C.C.W., D.E.N., E.L.B.); and Research Triangle Institute, Research Triangle Park, North Carolina (F.I.C.)
Received for publication
January 4, 2008
Accepted
March 18, 2008.
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
). The transport of a molecule of serotonin (5-hydroxytryptamine; 5-HT) into the presynaptic neuron is thought to be electrogenically coupled to the inward transport of one molecule each of Na+ and Cl- and the countertransport of one K+ molecule (Gu et al., 1994). This uptake process regulates the temporal and spatial actions of 5-HT on postsynaptic receptors and subsequent physiological processes such as mood, sleep, memory, appetite, and libido (Amara and Kuhar, 1993; Barker and Blakely, 1995).
In addition to regulating the downstream actions of 5-HT, SERT is the target of several structurally dissimilar compounds such as cocaine, amphetamines, tricyclic antidepressants (TCAs), and the selective serotonin reuptake inhibitors (Owens et al., 1997; Amara and Sonders, 1998). Substrate translocation for the SERT and the related dopamine and norepinephrine transporters is thought to occur by an alternating access mechanism that exposes the permeation pathway successively to the extracellular and intracellular sections of the transporter (Zhang and Rudnick, 2006).
The third transmembrane helix (TMH) of SERT previously has been shown to contain residues that are associated with serotonin and cocaine binding (Chen et al., 1997b). Chen and colleagues (1997b) used cysteine-scanning mutagenesis to demonstrate that two mutants, I172C and I179C, were sensitive to inactivation by [2-(trimethylammonium)ethyl] methanethiosulfonate (MTSET) but were insensitive to (2-sulfonatoethyl)methanethiosulfonate. In addition, a cysteine substitution at residue Tyr176 completely blocked transport. Furthermore, mutants I172C and Y176C had reduced cocaine affinity, and Y176C had reduced 5-HT affinity in the presence of MTSET. The authors concluded that Ile172 and Tyr176 were not only on a helical face accessible to the aqueous environment but were also involved in 5-HT and cocaine binding (Chen et al., 1997b).
Recent studies by Henry et al. (2006) probed the contributions of TMH I and TMH III to antagonist recognition and determined that the human serotonin transporter mutant I172C displayed reductions in antagonist potencies, whereas substrate [5-HT, 3,4-methylenedioxy-N-methyl-amphetamine (MDMA)] recognition was not affected. Furthermore, the authors used engineered cysteine mutants to determine that an N101C/I179C double mutant was able to coordinate cadmium, providing evidence that some residues in TMH I and TMH III may be in proximity to each other. Studies by our group using engineered zinc binding sites also have confirmed the proximity of hSERT TMH I and TMH III (White et al., 2006). Dose-response assays revealed that the double hSERT mutants V102C/M180C and V102C/I179C are more sensitive to zinc inhibition compared with the hSERT V102C single mutant (White et al., 2006). The increased sensitivity to zinc inhibition suggests that TMH I and TMH III are oriented in close proximity within the plasma membrane. A SERT homology model (D. E. Nichols, C. C. Walline, and E. L. Barker, unpublished data) generated from the coordinates of the Aquifex aeolicus leucine transporter structure (Yamashita et al., 2005) supports the proximity of TMH I and TMH III. Our model places residues Asn101 and Ile179 approximately 12 Å from each other. In addition, the carbonyl oxygen of Asn101 is predicted to be within 3 Å of a sodium ion and, therefore, may be involved in sodium binding or 5-HT translocation. Furthermore, measured distances between V102C and I179C or M180C were less than 9 Å and were in general agreement with the experimental data (White et al., 2006).
We now extend our research on TMH III by using species-scanning mutagenesis to probe the structure-activity relationships and molecular recognition of three distinct compound groups, the amphetamines, 3-phenyltropane analogs, and antidepressants. Comparative molecular field analysis (CoMFA) was used with selectivity fields that use the difference in pharmacological activities between the wild-type transporter and a point mutant as a dependent variable for structure-activity analysis (Baskin et al., 2003). This analysis allows visualization of the impact of mutations on ligand potency, providing new understanding for ligand recognition by SERT. To the best of our knowledge, this represents the first time selectivity field analysis has been used to discriminate the specific contribution of point mutations on ligand recognition. In addition, selectivity field analysis may be useful as a unique tool for determining whether changes in drug potency at a mutant are caused by specific protein-ligand interactions or disruption of binding sites caused by conformational changes.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
5-HT Transport Assay. HeLa cells were grown in Dulbecco's modified Eagle's medium with 5% bovine calf serum and 5% fetal clone I and supplemented with 2 mM glutamine, and 1% penicillin/streptomycin. Cells were maintained in a humidified 5% CO2 incubator at 37°C. Cells were transfected using the vaccinia virus T7 method using a 24- or 96-well format (Fuerst et al., 1986; Blakely et al., 1991; Roman et al., 2003). In brief, HeLa cells were plated at a density of 1 x 105 cells/well in 24-well culture plates (PerkinElmer Life and Analytical Sciences, Waltham, MA) or 2 x 104 cells/well in 96-well plates 15 to 18 h before transfection. Cells were infected with 10 plaque-forming units of vaccinia virus per cell and transfected with 300 (24 well) or 100 (96 well) ng cDNA/well using the Lipofectin (Invitrogen, Carlsbad, CA) transfection method according to the manufacturer's protocol. Protein expression was allowed to increase during a 6 to 8-h incubation. For [3H]5-HT uptake assays, cells were washed once with 37°C Krebs-Ringer-HEPES (KRH) buffer supplemented with calcium chloride and glucose (120 mM NaCl, 4.7 mM KCl, 2.2 mM CaCl2, 10 mM HEPES, 1.2 mM KH2PO4, 1.2 mM MgSO4, 1.8 g/l D-glucose, pH 7.4). Cells were preincubated for 10 min at 37°C with increasing inhibitor concentrations. Transport assays with 20 nM [3H]5-HT, 100 µM L-ascorbic acid and 100 µM pargyline were performed for 10 min at 37°C. Uptake was terminated by washing three times with ice-cold KRH. Nonspecific uptake was determined using a vector transfection. Microscint 20 scintillation cocktail (PerkinElmer Life and Analytical Sciences) was added to each well, and accumulated [3H]5-HT was determined on a PerkinElmer TopCount NXT. IC50 values were determined with a nonlinear regression analysis in GraphPad Prism version 3.0 (GraphPad Software Inc., San Diego, CA).
Kinetics Assay. Saturation transport assays were performed using HeLa cells (1 x 105 cells/well) in 24-well culture plates. At the time of assay, cells were washed once with KRH buffer. Cells were incubated with increasing concentrations of [3H]5-HT (final concentration ranged from 10 µM to 625 nM) with a specific activity diluted to 
0.1 Ci/mmol with unlabeled compound. Uptake was terminated exactly at 10 min by washing cells with ice-cold KRH buffer. Non-specific uptake was determined for each concentration using a vector transfection.
Amphetamine-Induced Efflux Assay. HeLa cells were plated in 24-well plates and transfected as described above. Cells were incubated at 37°C with [3H]5-HT (20 nM final concentration) for 10 min (hSERT) or 30 min [Drosophila SERT (dSERT) and hSERT A169D mutant] to provide for equivalent loading. Cells were treated with increasing concentrations of dichloroamphetamine for 10 min, and efflux was terminated by washing cells with ice-cold KRH buffer. The amount of [3H]5-HT remaining in the cells was determined on a PerkinElmer TopCount NXT.
Basal Efflux Assay. HeLa cells were plated in 24-well plates and transfected as described above. Cells were washed once with 37°C KRH. Next, [3H]5-HT was added to each well at a final concentration of 40 nM. Radiolabeled substrate and cells were incubated for 30 min at 37°C. Cells were washed two times with 37°C, and the second wash was left on for 2 min. Then, 250 µl of 37°C KRH was added to each well (except those used to define total accumulation) and aspirated after desired time. At the completion of the assay, all wells were washed one time with 800 µl of ice-cold KRH.
Sodium Dependence Assay. HeLa cells were plated in 24-well plates and transfected as described above. At the time of assay, cells were washed once with sodium-free phosphate-buffered saline supplemented with calcium and magnesium (PBS-CM) (137 mM LiCl, 4.3 mM Na2HPO4, 1.4 mM KH2PO4, 2.7 mM KCl, 0.1 mM CaCl2, and 1 mM MgCl2 at pH 7.4). Cells were incubated in sodium-free PBS-CM isotonically supplemented with increasing concentrations of NaCl and [3H]5-HT (40 nM final concentration) at 37°C for 10 min. Uptake was terminated by washing cells in ice-cold sodium-free PBS-CM. Nonspecific uptake was determined using a vector-transfected control.
CoMFA Selectivity Fields. Computational molecular modeling studies were performed using SYBYL version 7.1 software package (Tripos, St. Louis, MO) on a Silicon Graphics O2 workstation (SGI, Mountain View, CA) running IRIX version 6.5. The antidepressants, cocaine analogs, and amphetamines were sketched in SYBYL using library fragments. Atom types were manually checked and the active enantiomers of MDA, MDMA, fluoxetine, and citalopram were defined as S. S-Fluoxetine was chosen because the potency is nearly identical to the R enantiomer and allowed for easy alignment with the other molecules. 1S,4S-sertraline was used and included in the molecular modeling. The more active stereoisomer of tomoxetine was defined as R, which is referred to in the text as the more common name atomoxetine. Each compound was subsequently minimized using the conjugate gradient minimization method with no initial simplex minimization, a gradient value of 0.05 kcal/mol, and 10,000 maximal iterations using MMFF94s force-field parameters. A common substructure was drawn for the antidepressants using the two phenyl rings to align all of the minimized antidepressants. Three antidepressants, the selective serotonin reuptake inhibitor fluvoxamine, the serotonin/norepinephrine reuptake inhibitor venlafaxine, and the serotonin transport blocker zimelidine, were manually aligned to a phenyl ring and selected centroid atoms in the remaining ring or chain because they lack two phenyl rings. In addition, the side-chain torsions for clomipramine were defined based on the orientation observed bound in the crystal structure of the leucine transporter (Singh et al., 2007). The remaining antidepressant side chains were manually oriented to align to clomipramine so that the nitrogen atoms overlapped. Likewise, the tropane ring of cocaine was used to align cocaine and the RTI molecules (Research Triangle Institute, Research Triangle Park, NC). In addition, the amphetamines were aligned to a phenethylamine template. One drug, dichlorophenylpiperazine, was excluded from the CoMFA because the piperazine ring did not align well with the ethylamine side chain. A database for each drug class was created that contained the compound set as well as the substructure. Then, the pKi values for the antidepressants, cocaine analogs, or amphetamines for each mutant were converted into selectivity field data using a modification of the equation as defined by Baskin et al. (2003): (log Ki(hSERT mutant)) - (log Ki(wild-type hSERT)). The selectivity field data were inserted into an explicit data column in a molecular spreadsheet. In addition, a CoMFA column was created using the Tripos Standard CoMFA field class, with steric and electrostatic cutoffs at 30.0 kcal/mol. Molecular probe regions were manually defined with a minimum of 2-Å boundaries outside the molecule in the X, Y, and Z directions. CoMFA data were generated using one to three components based on the partial least-squares analysis for each molecule. Standard CoMFA contour specifications were used (80 for favored and 20 for disfavored) for steric and electrostatic contributions.
Data Analysis. The Km and Vmax in saturation experiments and IC50 values in transport assays were determined using nonlinear curve-fitting analysis (Prism 3.0; GraphPad Software Inc.). IC50 values were converted to Ki values using the Cheng-Prusoff equation (Cheng and Prusoff, 1973). Statistical analyses of Ki values for [3H]5-HT uptake inhibition assays were performed with Prism using a one-way analysis of variance followed by a Dunnett's post hoc test.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
|
Recognition of Amphetamines by hSERT Mutants. The potencies of the amphetamines were evaluated at the hSERT mutants I167L, A169D, F170I, I172M, and S174M (Table 2). For the hSERT F170I mutant, there were no statistically significant changes in amphetamine potency. For the hSERT I167L mutant, only 3,4-methylenedioxy-2-N-N-dipropylindan (N-dipropyl-indan) and AMMT had significantly increased potency at the hSERT I167L mutant. However, the remaining mutants exhibited differences in potencies for several of the amphetamines.
|
Selectivity field CoMFAs were performed on the hSERT A169D mutant (Baskin et al., 2003). Selectivity fields allow the differences in biological activity between a wild-type transporter and a mutant transporter to be visualized as colored CoMFA fields. The fields generated from explicit data were expressed as the difference between activity of the wild-type and mutant transporters using the following equation: (log Ki(hSERT mutant)) - (log Ki(wild-type hSERT)). The hSERT A169D mutant showed a significant increase in potency for 10 amphetamines [3,4-difluoroamphetamine, MDMA, DiFMDA, 2-Me-MDA, 4-trifluoromethylamphetamine, AMMT, 1-aminomethyl-5-methoxyindan, 3,4-dichlorophenylpiperazine, (R)-3-methoxy-4-methyl-N-methyl-amphetamine, and N-dipropyl-indan; Table 2]. However, the selectivity field CoMFA for the A169D mutant showed no predominating fields, indicating that the increase in potency of the amphetamine analogs is probably not a direct result of hydrogen bonding or steric interactions but rather may result from a global structural change in the SERT protein (Fig. 1).
|
|
|
). Of the five hSERT mutants constructed with a dSERT residue, only the A169D mutant gained a net negative charge. We hypothesized that introduction of the charged aspartate may disrupt the translocation of positively charged amphetamines, leading to decreased amphetamine-induced efflux in the case of dSERT. Substrates such as p-chloroamphetamine and MDMA promote a reversed transport mechanism in hSERT that can be used to explore the mechanism of the bidirectional 5-HT translocation process (Sitte et al., 2000). Carrier-mediated efflux for hSERT, dSERT, and the hSERT A169D mutant was assessed using dichloroamphetamine (DCA) to determine the potency of these drugs at inducing efflux. DCA produced a modest efflux of 5-HT from hSERT (Fig. 4A). However, concentrations of DCA up to 10 µM were not able to induce an appreciable efflux from the hSERT A169D mutant or dSERT (Fig. 4B). Indeed, for A169D, efflux of preloaded [3H]5-HT occurred in the absence of drug, indicating a possible alteration in the 5-HT translocation cycle. Therefore, the basal efflux rate for hSERT, the A169D mutant, and dSERT was determined. The initial rate of loss of radiolabeled 5-HT from preloaded cells was determined from remaining accumulated 5-HT at 0, 5, 10, and 15 s (Fig. 4C). When total transport is normalized to 100%, dSERT and the A169D mutant efflux a higher percentage of preloaded 5-HT over time (Fig. 4C). It is interesting to note that the A169D mutant and dSERT both retained only 75% of transported 5-HT after 15 s compared with 95% for hSERT (Fig. 4C). The actual rates of basal [3H]5-HT efflux in the presence of KRH buffer for hSERT, the A169D mutant, and dSERT were 25 ± 5 x 10-21, 0.8 ± 0.1 x 10-21, and 1.5 ± 0.7 x 10-21 mol/cell/min, respectively. The rate of spontaneous efflux for WT hSERT is statistically significantly greater than for either the A169D mutant or dSERT (p < 0.05, one-way analysis of variance, Bonferroni's post hoc test), and the rates for the A169D mutant and dSERT do not differ from each other. However, this greater hSERT efflux rate is influenced by the increased overall number of 5-HT molecules transported by hSERT compared with dSERT and the hSERT A169D mutant. These results indicate that by substituting the aspartate present in dSERT for the corresponding alanine at position 169 in hSERT, the serotonin efflux process is altered, and basal efflux resembles that of dSERT. This finding may indicate that A169D is involved in a substrate exchange mechanism because our SERT homology model places residue 169 near the Na+ binding site and near the cytoplasmic interface for substrate translocation.
|
Sodium Dependence Assay and Transport Competition Assay with MDMA. We constructed an hSERT homology model from the LeuTAa crystal structure, a bacterial transporter and ortholog to the monoamine transporters (Yamashita et al., 2005). The model was based on a manual sequence alignment of the leucine transporter and SERT (D. E. Nichols, C. C. Walline, and E. L. Barker, unpublished data). Our homology model revealed that the Ala169 residue is in the same plane of the transporter as the two hypothesized sodium ions necessary for leucine transport (Yamashita et al., 2005). We hypothesized that the gain in potency for many of the amphetamine analogs in the A169D mutant could be due to an alteration in sodium requirements. We determined that the sodium dependence for the A169D mutant was not saturable over the concentration of sodium used in our experiments (Fig. 5A). In contrast, hSERT displayed one-site binding kinetics and had a Km value of 40 mM (Fig. 5B). Our results indicate that the hSERT A169D mutant may have reduced affinity for sodium or a more complex interaction with the cation similar to the sodium requirements for the dopamine and norepinephrine transporters (Gu et al., 1994).
|
Recognition of Cocaine (3-Phenyltropane) Analogs by hSERT Mutants. 3-Phenyltropane analogs were used to evaluate antagonist inhibition potency for the TMH III mutants. Only one construct, hSERT I172M, showed a reduction in antagonist potency for the cocaine analogs (see Table 3). All analogs except RTI-55 and RTI-121 exhibited reduced potency compared with the wild-type SERT. Both RTI-55 and RTI-121 contain an iodine substitution in the 4-position on the phenyl ring. The presence of the electronegative iodine atom will make the phenyl ring slightly more positive relative to an unsubstituted phenyl ring. The selectivity field CoMFA indicated that positive charge around the phenyl ring increases the potency of the cocaine analogs at the I172M mutant (Fig. 6). Indeed, our data for RTI-55 and RTI-121 support this model. Therefore, areas of positive charge and/or hydrogen bond donation may contribute to a high potency interaction of the cocaine analogs when Met substitutes for Ile at position 172. The predominant red selectivity field suggests this interaction could be a direct contact site for the cocaine analogs. In our homology model based on the leucine transporter, Ile172 lies directly below the Glu493-Arg104 "gate" and lies along the likely permeation pathway. Although Ile172 can interact with local aromatic residues through van der Waals interactions, Met172 also can interact with aromatic rings through a sulfur-pi complex (Tatko and Waters, 2004). Furthermore, Ile172 is a rather compact residue compared with Met, whereas methionine projects far out into the putative permeation pathway. Thus, we hypothesize that the nonhalogenated cocaine analogs do not bind as well in the I172M mutant because of the steric projection of Met172 into the permeation pathway, impeding access to an area in the transporter that may be involved in cocaine binding. In contrast, the iodinated cocaine analogs have increased affinity to the Met172 mutation because they have the ability to form a complex between their more positive aromatic ring and the polarizable sulfur atom of the methionine that may compensate for the larger steric size of the methionine residue. This reasoning is compatible with our results suggesting that Ile172 may be a contact site for cocaine binding.
|
|
Recognition of Antidepressants by hSERT Mutants. Twenty antidepressants were screened for uptake inhibition potency at the TMH III mutants (Table 4). No statistically significant changes of antagonist potency were seen for the I167L, A169D, F170I, or S174M mutants. In contrast, statistically significant decreases in antagonist potency at the I172M mutant were seen for all of the antidepressants except for amoxapine, maprotiline, and 3
-bis-(4-fluorophenyl)-methoxytropane, a potent dopamine transporter inhibitor (Newman et al., 1994). The selectivity field CoMFA of the antidepressant data showed no one predominating colored field, indicating that the loss of potencies for the antidepressant set may be the result of a global conformational change and not due to disruption of a direct binding site (Fig. 7).
|
|
Alternatively, CoMFAs assume that all compounds in the test set are binding in the same location within a biological target and share a common core structure. The broad structural diversity associated with the core chemical structures of the antidepressant compound set may not be ideal for CoMFA selectivity field analysis because the antidepressants may be binding to multiple sites within SERT, thus confounding the analysis and resulting in the presence of multiple fields.
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
) determined that the GABA transporter (GAT) residue Tyr140, a residue conserved in the third TMH throughout the sodium/chloride-dependent transporter family, is critical for GABA recognition and transport. Furthermore, residue Tyr289 of the neuronal glycine transporter GLYT2 has an important role in glycine binding and transport, ion dependence, and sodium selectivity (Ponce et al., 2000). GAT Tyr140 and GLYT2 Tyr289 are both homologous to Tyr176 in the hSERT. hSERT Tyr176 as well as I172C and I179C were shown to be sensitive to inactivation by MTSET (Chen et al., 1997b). Residues Ile172, Tyr176, and Ile179 were suggested to face toward the inside of the pore, and residue Ile172 was proposed to reside close to the substrate binding site (Chen et al., 1997a,b). An hSERT homology model constructed from the crystal structure of a prokaryotic amino acid transporter from A. aeolicus (LeuTAa) supports these experimental data and places residues Ile172, Tyr176, and Ile179 facing the putative substrate binding site (Beuming et al., 2006). Therefore, the third transmembrane domains of GAT, GLYT2, and SERT each contain residues critical for substrate recognition, sodium binding, and antagonist binding, supporting the hypothesis that THM III is part of the permeation pathway and is critical for substrate translocation and antagonist recognition.
Our group used species-scanning mutagenesis to generate five hSERT TMH III mutants where the amino acid identity in dSERT was substituted into the equivalent position in hSERT. The potencies of the amphetamines, 3-phenyltropane analogs, and antidepressants were evaluated at the hSERT mutants. Subsequently, CoMFAs were performed using a modification of the approach developed by Baskin and colleagues (2003), where the difference in pharmacological activity between the WT and mutant transporter is expressed as a dependent variable in the CoMFA spreadsheet. For our purposes, the pharmacological activity of the wild-type hSERT was subtracted from the pharmacological activity of the mutant to determine the role of the point mutation with respect to antagonist or substrate potency (Baskin et al., 2003). Therefore, the colored contour fields are a visual representation of the effect of the mutation on amphetamine, cocaine, or antidepressant interaction with the SERT protein. The resulting fields directly illustrate the steric or electrostatic interactions induced by the mutation that alter drug potency. Traditional CoMFAs require visual inspection of two separate contour maps to determine differences in pharmacological activity between two proteins, with comparisons performed in a qualitative manner. The use of CoMFA selectivity fields semiquantitatively represents the change in drug potency between two proteins. The size of the selectivity field can be directly observed and eliminates the need to visually compare differences in field size for two separate analyses.
Our results from the uptake competition assays revealed significant decreases in transport inhibition potencies for nearly all the amphetamines at the S174M mutant. In the LeuTAa, Ser174 corresponds to Ala105. Ala105, Val103, Ile106, and Met403 participate in a hydrophobic cluster that would stabilize the helical packing forces between TMHs I and X. In contrast, in the SERT, Ser174 seems to participate in a hydrogen-bonded network in a homologous region that includes Tyr171 and Lys490. Again, this network would stabilize packing between helices III and X. Mutating residue Ser174 would disrupt this interaction, loosening the association between helices III and X. Helix X is directly adjacent to the critical Arg104, and any change in the relative conformation of TMH X is likely to affect function, thus potentially disrupting the ability of the amphetamines to inhibit transport. The selectivity field CoMFA indicates that alterations in hydrogen bonding or charges in various positions around the amphetamine structure could be responsible for changes in drug potency. It is possible that by replacing the one-carbon hydroxy-substituted serine with a three-carbon thioether containing methionine that critical hydrogen bonding or electrostatic interactions could have been disrupted. Alternatively, a narrow green field is shown wrapping around the amphetamine structure, which suggests that introduction of the methionine residue leads to a transporter that disfavors bulky amphetamines. The presence of multiple, small fields suggests that changes in drug potency probably result from a global change in protein structure and are not representative of a direct contact site for the amphetamines at residue Ser174.
Six of the amphetamine analogs tested (DiFMDA, MDA, 2-Me-MDA, 6-methyl-3,4-methylenedioxyamphetamine, -ethylparachloroamphetamine, and 3-methoxy-4-methyl--ethylphenethylamine) showed significant decreases in potency for transport inhibition at the I172M mutant. The selectivity field CoMFA shows several colored fields, indicating that this mutation causes a disruption in antagonist recognition and may indicate a global conformational change of the SERT protein.
Additional selectivity field CoMFAs were performed for the cocaine analogs and the antidepressants at the I172M mutant. The cocaine analog analysis indicated that positive charge around the phenyl ring increases the relative potency of the drug for the I172M mutant. The presence of one large field for the cocaine analog analysis indicates that this position may represent a direct contact site for the cocaine analogs (Fig. 6). This result supports previous research suggesting that Ile172 and Tyr176 are in proximity to the 5-HT and cocaine binding site (Chen et al., 1997b).
However, analysis of the antidepressants provided several selectivity fields, which suggest that the I172M mutation may cause a global disruption in antidepressant recognition by influencing the conformation associated with nearby residues that may be involved in directly interacting with the antidepressants (Fig. 7). Furthermore, the fact that the 3-phenyltropane analogs revealed evidence consistent with a direct interaction suggests distinct but possibly overlapping sites for the different classes of drugs. Alternatively, the differences found in the core chemical structure of our antidepressant compound set did not allow for a molecular alignment in the CoMFA that was as good as that for the 3-phenyltropanes or the amphetamines. This poor alignment may have masked a direct interaction with one or more antidepressants. Earlier studies implicated residues Ile172 and Tyr95 as determinants of citalopram potency (Henry et al., 2006). The residue homologous to Ile172 in the LeuTAa structure projects into a hydrophobic binding pocket to accommodate the hydrophobic methyl groups of the substrate (Yamashita et al., 2005). In addition, cocaine analogs are thought to bind in a highly hydrophobic microenvironment (Rasmussen et al., 2001). Altogether, our results support the involvement of Ile172 with cocaine recognition but not in the binding of any of the antidepressants.
Two very recent reports have described a TCA binding site within the LeuTAa structure (Singh et al., 2007; Zhou et al., 2007). Both groups identified residues extracellular to the gating residues Arg104 and Glu493 as contact sites for the TCAs. Based on our hSERT homology model, these putative TCA binding site residues range from 10 to 22 Å from the TMH III residue Ile172. The selectivity field analysis indicates that hSERT mutant I172M is probably not a direct contact site for the antidepressant compound set.
Our group has previously shown that substituted amphetamines are not effectively transported by dSERT and therefore have decreased efficacy for inducing 5-HT efflux (Rodríguez et al., 2003). In addition, unlabeled 5-HT was less capable of inducing substrate efflux. It is possible that the aspartate in the A169D mutant and wild-type dSERT may be disrupting the release of substrate, resulting in a loss of amphetamine-induced efflux (Fig. 4B). A lack of substrate-induced efflux is characteristic of channels and may indicate that the A169D mutant and dSERT are operating in a channel-like mode (Rodríguez et al., 2003). This finding may suggest that the aspartate in position 169 is involved with gating SERT channel-like properties.
Sodium dependence assays indicated that the hSERT A169D mutant was not saturable over the range of sodium concentrations used in our studies. We believe that this altered sodium dependence is probably the result of an increase in the Km value for sodium at the A169D mutant. This reduced affinity for sodium could disrupt 5-HT transport, and this alteration in uptake capacity may be overcome in the presence of higher sodium concentrations (Fig. 5). It is interesting to note that the potency of MDMA did not change with differing sodium concentrations. Therefore, the sodium requirements for the hSERT A169D mutant may perturb 5-HT transport without affecting the potency of other substrates.
In conclusion, our results indicate that hSERT transmembrane helix III probably contributes to the permeation pathway and that residues Ile172 and Ala169 could represent important contact sites for antagonist recognition and substrate transport. The selectivity field CoMFAs used the difference in biological activities between hSERT and selected point mutants to visualize the impact of changes in activity. This analysis is a useful tool to distinguish between general conformational changes and potential direct protein-ligand interactions. In addition, the hSERT mutant A169D had altered sodium requirements, suggesting that this residue is near the sodium-coordinating site. This result is consistent with our homology model, which places the aspartate less than 10 Å from a Na+ binding site in LeuTAa. Thus, our study has used a novel computational method to integrate mutagenesis and structure-activity relationship data to explore hSERT TMH III as an important domain for substrate selectivity and antagonist binding.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
ABBREVIATIONS: SERT, serotonin transporter; 5-HT, 5-hydroxytryptamine, serotonin; TCA, tricyclic antidepressant; TMH, transmembrane helix; MTSET, [2-(trimethylammonium)ethyl]methanethiosulfonate; MDMA, 3,4-methylenedioxy-N-methyl-amphetamine; hSERT, human SERT; CoMFA, comparative molecular field analysis; KRH, Krebs-Ringer-HEPES; dSERT, Drosophila SERT; PBS-CM, phosphate-buffered saline supplemented with calcium and magnesium; MDA, 3,4-methylenedioxyamphetamine; N-dipropyl-indan, 3,4-methylenedioxy-2-N-N-dipropylindan; AMMT, 1-aminomethyl-6-methoxytetralin; DiFMDA, 1-(2,2-difluorobenzo[d][1,3]dioxol-5-yl)propan-2-amine; 2-Me-MDA, 2-methyl-3,4-methylenedioxyamphetamine; DCA, 3,4-dichloroamphetamine; RTI-31, 3β-(4-chorophenyl)tropane-2β-carboxylic acid methyl ester tartrate; RTI-32, 3β-(4-methylphenyl)tropane-2β-carboxylic acid methyl ester tartrate; RTI-55, 3β-(4-iodophenyl)tropane-2β-carboxylic acid methyl ester tartrate; RTI-83, 3β-(4-ethylphenyl)tropane-2β-carboxylic acid methyl ester tartrate; RTI-112,(-)-3β-(3-methyl-4-chlorophenyl)tropane-2β-carboxylic acid methyl ester tartrate; RTI-121, 3β-(4-iodophenyl)-tropane-2β-carboxylic acid isopropyl ester hydrochloride; RTI-142, (-)-N-nor-3β-(4-fluorophenyl)tropane-2β-carboxylic acid methyl ester; RTI-147, 3β-(4-chlorophenyl)tropane-2β-pyrrolidinecarboxamide hydrochloride; RTI-311, N-allyl-N-nor-3β-(4-iodophenyl)tropane-2β-carboxylic acid methyl ester tartrate.
Address correspondence to: Dr. Eric L. Barker, Department of Medicinal Chemistry and Molecular Pharmacology, Purdue University, 575 Stadium Mall Drive, West Lafayette, IN 47907. E-mail: ericb{at}pharmacy.purdue.edu
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Amara SG and Kuhar MJ (1993) Neurotransmitter transporters: recent progress. Annu Rev Neurosci 16: 73-93.[Medline]
Amara SG and Sonders MS (1998) Neurotransmitter transporters as molecular targets for addictive drugs. Drug Alcohol Depend 51: 87-96.[CrossRef][Medline]
Barker EL and Blakely RD (1995) Norepinephrine and serotonin transporters: molecular targets of antidepressants drugs, in Pharmacology: The Fourth Generation of Progress (Bloom FE and Kupfer DJ eds) pp 321-333, Raven Press, New York.
Baskin II, Tikhonova IG, Palyulin VA, and Zefirov NS (2003) Selectivity fields: comparative molecular field analysis (CoMFA) of the glycine/NMDA and AMPA receptors. J Med Chem 46: 4063-4069.[CrossRef][Medline]
Beuming T, Shi L, Javitch JA, and Weinstein H (2006) A comprehensive structure-based alignment of prokaryotic and eukaryotic neurotransmitter/Na+ symporters (NSS) aids in the use of the LeuT structure to probe NSS structure and function. Mol Pharmacol 70: 1630-1642.
Bismuth Y, Kavanaugh MP, and Kanner BI (1997) Tyrosine 140 of the gamma-aminobutyric acid transporter GAT-1 plays a critical role in neurotransmitter recognition. J Biol Chem 272: 16096-16102.
Blakely RD, Clark JA, Rudnick G, and Amara SG (1991) Vaccinia-T7 RNA polymerase expression system: evaluation for the expression cloning of plasma membrane transporters. Anal Biochem 194: 302-308.[CrossRef][Medline]
Chen JG, Liu-Chen S, and Rudnick G (1997a) External cysteine residues in the serotonin transporter. Biochemistry 36: 1479-1486.[CrossRef][Medline]
Chen JG, Sachpatzidis A, and Rudnick G (1997b) The third transmembrane domain of the serotonin transporter contains residues associated with substrate and cocaine binding. J Biol Chem 272: 28321-28327.
Cheng Y and Prusoff WH (1973) Relationship between the inhibition constant (K1) and the concentration of inhibitor which causes 50% inhibition (I50) of an enzymatic reaction. Biochem Pharmacol 22: 3099-3108.[CrossRef][Medline]
Fuerst TR, Niles EG, Studier FW, and Moss B (1986) Eukaryotic transient-expression system based on recombinant vaccinia virus that synthesizes bacterio-phage T7 RNA polymerase. Proc Natl Acad Sci U S A 83: 8122-8126.
Gu H, Wall SC, and Rudnick G (1994) Stable expression of biogenic amine transporters reveals differences in inhibitor sensitivity, kinetics, and ion dependence. J Biol Chem 269: 7124-7130.
Henry LK, Field JR, Adkins EM, Parnas ML, Vaughan RA, Zou MF, Newman AH, and Blakely RD (2006) 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 281: 2012-2023.
Newman AH, Allen AC, Izenwasser S, and Katz JL (1994) Novel 3 alpha-(diphenylmethoxy)tropane analogs: potent dopamine uptake inhibitors without cocaine-like behavioral profiles. J Med Chem 37: 2258-2261.[CrossRef][Medline]
Owens MJ, Morgan WN, Plott SJ, and Nemeroff CB (1997) Neurotransmitter receptor and transporter binding profile of antidepressants and their metabolites. J Pharmacol Exp Ther 283: 1305-1322.
Ponce J, Biton B, Benavides J, Avenet P, and Aragon C (2000) Transmembrane domain III plays an important role in ion binding and permeation in the glycine transporter GLYT2. J Biol Chem 275: 13856-13862.
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.
Rodríguez GJ, Roman DL, White KJ, Nichols DE, and Barker EL (2003) Distinct recognition of substrates by the human and Drosophila serotonin transporters. J Pharmacol Exp Ther 306: 338-346.
Roman DL, Walline CC, Rodriguez GJ, and Barker EL (2003) Interactions of antidepressants with the serotonin transporter: a contemporary molecular analysis. Eur J Pharmacol 479: 53-63.[CrossRef][Medline]
Singh SK, Yamashita A, and Gouaux E (2007) Antidepressant binding site in a bacterial homologue of neurotransmitter transporters. Nature 448: 952-956.[CrossRef][Medline]
Sitte HH, Scholze P, Schloss P, Pifl C, and Singer EA (2000) Characterization of carrier-mediated efflux in human embryonic kidney 293 cells stably expressing the rat serotonin transporter: a superfusion study. J Neurochem 74: 1317-1324.[Medline]
Tatko CD and Waters ML (2004) Investigation of the nature of the methionine-
interaction in β-hairpin model systems. Protein Sci 13: 2515-2522.[CrossRef][Medline]
White KJ, Kiser PD, Nichols DE, and Barker EL (2006) Engineered zinc-binding sites confirm proximity and orientation of transmembrane helices I and III in the human serotonin transporter. Protein Sci 15: 2411-2422.[CrossRef][Medline]
Yamashita A, Singh SK, Kawate T, Jin Y, and Gouaux E (2005) Crystal structure of a bacterial homologue of Na+/Cl-dependent neurotransmitter transporters. Nature 437: 215-223.[CrossRef][Medline]
Zhang YW and Rudnick G (2006) The cytoplasmic substrate permeation pathway of serotonin transporter. J Biol Chem 281: 36213-36220.
Zhou Z, Zhen J, Karpowich NK, Goetz RM, Law CJ, Reith ME, and Wang DN (2007) LeuT-desipramine structure reveals how antidepressants block neurotransmitter reuptake. Science 317: 1390-1393.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
