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

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.103.057836v1 308/2/679    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 Google Scholar Google Scholar Articles by Roman, D. L. Articles by Barker, E. L. Search for Related Content PubMed PubMed Citation Articles by Roman, D. L. Articles by Barker, E. L. Pubmed/NCBI databases Compound via MeSH Substance via MeSH Hazardous Substances DB TRITIUM

NEUROPHARMACOLOGY

Distinct Molecular Recognition of Psychostimulants by Human and Drosophila Serotonin Transporters

Department of Medicinal Chemistry and Molecular Pharmacology, School of Pharmacy and Pharmacal Sciences, Purdue University, West Lafayette, Indiana (D.L.R., S.N.S., D.E.N., E.L.B.); and Research Triangle Institute, Research Triangle Park, North Carolina (F.I.C.)

Received September 3, 2003; accepted October 16, 2003.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
In this study, human embryonic kidney (HEK)-293 cells stably expressing human, Drosophila, or a chimeric serotonin (5-hydroxytryptamine, 5-HT) transporter (hSERT, dSERT, and H^1–281D^282–476H^477–638, respectively) were used to explore the ability of two libraries of structurally distinct psychostimulants to inhibit 5-HT uptake. One library consisted of 3-phenyltropane analogs, whereas the second library consisted of several substituted amphetamines. hSERT exhibited a lower K_i value for all the compounds in both libraries compared with dSERT, whereas the chimeric SERT exhibited properties more closely resembling those of dSERT. This species selectivity was explored using computer-generated comparative molecular field analysis to model the interactions of the cocaine analogs and substituted amphetamines at hSERT, dSERT, and the cross-species chimera. Models for the 3-phenyltropane analogs indicate that a region exists around the aromatic ring where decreased electron density is favored, particularly for hSERT. This finding may indicate pi-pi stacking with an aromatic amino acid residue in SERT. Also, electronegative substituents in the 4'-position provide favorable interactions. This structural feature was demonstrated by increased potency of analogs with electronegative substituents on the aromatic ring that withdraw electron density. For the substituted amphetamines, key areas for interaction exist around the amine, an electrostatic component surrounding the 3-position on the aromatic ring, and a steric component surrounding the 4-position.

Cocaine and amphetamine remain two of the most abused drugs in present society. Widespread abuse of these substances carries grave physiological, psychological, and social consequences. Due to these problems, much research has focused on understanding how these two classes of abused drugs alter neurochemistry. Cocaine targets three major neurotransmitter systems in the brain: dopamine, serotonin, and norepinephrine (Eshleman et al., 1999; Uhl et al., 2002), in each case acting to inhibit the respective neurotransmitter transporter. Amphetamine and its neurotoxic derivatives are substrates for biogenic amine transporters, promoting nonvesicular release of neurotransmitter, with actions on dopaminergic, serotonergic, and noradrenergic systems (Amara and Sonders, 1998; Cozzi et al., 1999; Jones et al., 1999; Kantor et al., 1999; Reneman et al., 2002). Thus, these two structurally distinct drug classes both ultimately lead to increased neurotransmitter levels in the synapse.

Previous studies have shown that human and Drosophila SERTs exhibit different pharmacologies even though they possess approximately 51% sequence similarity (Hoffman et al., 1991; Usdin et al., 1991; Ramamoorthy et al., 1993; Blakely et al., 1994; Corey et al., 1994; Demchyshyn et al., 1994; Barker et al., 1998). The studies presented in this work demonstrate differences in the way the human and Drosophila SERT species variants, as well as a chimeric SERT, recognize cocaine and substituted amphetamines. In light of the structural diversity of these compounds, we hypothesize that distinct determinants of psychostimulant recognition by SERT exist and differences in these modes of recognition can be explored through molecular modeling and comparative molecular field analysis (CoMFA). To test this hypothesis, we have used a library of 3-phenyltropane analogs and substituted amphetamines to examine structural components of these compounds that are critical for recognition by hSERT, dSERT, and the H^1–281D^282–476H^477–638 chimera. The structural diversity of the compound classes provides a unique collection of probes for what may be multiple binding pockets or, at minimum, unique contact sites within the same pocket, for these two prototypical psychostimulants. In our studies, both libraries of compounds were screened using a [³H]5-HT uptake inhibition assay with HEK-293 cells stably expressing hSERT, dSERT, or the H^1–281D^282–476H^477–638 chimera. The K_i values for inhibition of [³H]5-HT transport were calculated from IC₅₀ values obtained from dose-response curves. The results of these assays were used as input for modeling using CoMFA. CoMFA has proven valuable in previous studies as a method to model protein-ligand interactions in systems that lack protein crystal structures or other definitive structural information (Carroll et al., 1994; Wilcox et al., 1998; Li et al., 1999; Newman et al., 1999). The six models presented here indicate differences in recognition on four levels. First, differences exist between the way hSERT and dSERT recognize 3-phenyltropane derivatives. Second, there are differences in the molecular recognition of the substituted amphetamines by the two species variants. Third, and perhaps most interesting, CoMFA maps reveal differences in the way SERTs, in general, recognize these two classes of psychostimulants. Fourth, the models suggest that the H^1–281D^282–476H^477–638chimera, which is an hSERT construct with transmembrane domains (TMDs) V–IX replaced with the dSERT sequence, exhibits distinct hybrid interactions with some components similar to hSERT and dSERT for the 3-phenyltropane analogs, whereas the determinants for recognizing the substituted amphetamines closely parallels dSERT. These maps and accompanying data indicate fundamental differences in the recognition of psychostimulants, suggesting unique molecular interactions with SERTs that promote high-affinity interactions with these drugs of abuse.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
[³H]5-HT uptake assays. The H^1–281D^282–476H^477–638 chimera was generated as described previously (Rodriguez et al., 2003). Briefly, it is a full-length 638 amino acid transporter, constructed from amino acids 1–281 and 477–638 with identity from hSERT and residues 282–476 with identity from dSERT.

Human embryonic kidney (HEK)-293 cells stably expressing either the human or Drosophila SERT were generous gifts from Dr. Randy D. Blakely (Vanderbilt University, Nashville, TN). All cells were maintained in Dulbecco's modified Eagle's medium with 10% dialyzed fetal bovine serum supplemented with 2 mM glutamine, 1% penicillin/streptomycin, and 600 mg/liter geneticin (G-418 sulfate) in a humidified 5% CO₂ incubator at 37°C.

Uptake assays were performed on cell suspensions in 96-well microtube racks. Confluent cells in 150-mm tissue culture dishes were washed with phosphate-buffered saline, scraped, and resuspended in Krebs-Ringer-HEPES (KRH) buffer supplemented with D-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). Total protein concentration was determined using the Bradford assay (Bradford, 1976). Cell suspensions were diluted in KRH/glucose to yield a concentration of 0.25 mg/ml total protein. Four hundred microliters (100 µg) of the cell suspension was added to each tube in a 96-well microtube rack. Amphetamine and 3-phenyltropane analogs were diluted serially in KRH buffer, and 50-µl aliquots of increasing concentrations of amphetamine or cocaine analogs were added to triplicate microtubes. When 3-phenyltropane analogs were assayed, the cells were preincubated with drug at 37°C for 10 min. This preincubation was not included when the substituted amphetamines were assayed, because these compounds have been shown to be substrates for SERT, and the intracellular accumulation of drug could affect 5-HT transport (Wall et al., 1995; Rodriguez et al., 2003). [³H]5-HT (110 Ci/mmol) was diluted to yield a final assay concentration of 20 nM in KRH buffer supplemented with 100 µM Lascorbic acid and 100 µM pargyline. Fifty microliters of the [³H]5-HT solution was added to each microtube, and uptake was allowed to proceed at 37°C for 10 min. Transport was terminated via vacuum filtration onto 96-well GF/B filter plates (presoaked with 0.3% polyethyleneimine) using a PerkinElmer Filtermate 96-well harvester (PerkinElmer Life Sciences, Boston, MA). Nonspecific uptake was determined using 10 µM fluoxetine. On average, specific 5-HT uptake was approximately 95% of total uptake. The filter plates were dried at ambient temperature overnight. The next day, scintillation cocktail was added before counting in a PerkinElmer TopCount NXT to determine accumulated [³H]5-HT. IC₅₀ values were determined using the nonlinear regression routine in GraphPad Prism version 3.0 (GraphPad Software Inc., San Diego, CA), and K_i values were calculated using the Cheng-Prusoff equation (Cheng and Prusoff, 1973). The K_m values for 5-HT uptake at each construct were 1.2 ± 0.2 µM (hSERT), 0.9 ± 0.2 µM (dSERT), and 0.8 ± 0.1 µM (H^1–281D^282–476H^477–638). Experiments were performed in triplicate and repeated in at least three separate assays.

CoMFA for 3-Phenyltropane Analogs. Computational molecular modeling studies were performed using the SYBYL version 6.8 software package (Tripos, St. Louis, MO) on a Silicon Graphics O2 workstation (Mountain View, CA) running IRIX version 6.5. The 3-phenyltropane analogs were sketched using library fragments in SYBYL. Atom types were automatically assigned by SYBYL and then manually checked. The compounds were subjected to a grid search (0–360°, 10° increment) to find the lowest energy conformation. This conformation was then minimized for all molecules using the conjugate gradient method with no initial simplex minimization, a gradient value of 0.05, and 10,000 maximum iterations. The compound set, as well as a consensus molecule containing the common tropane ring substructure, was inserted into the database. This consensus molecule containing the core structure of all the analogs was used as the template for SYBYL's database alignment tool. All of the structures were aligned using the tropane ring carbon skeleton as anchor points. The aligned molecules were then inserted into a molecular spreadsheet and pK_i values inserted as functional data. CoMFA data were generated using the parameters listed above. For the hSERT data, three components were determined to be optimal (q² = 0.63), whereas the dSERT (q² = 0.51) and H^1–281D^282–476H^477–638 chimera (q² = 0.54) models used two components that were determined to be optimal based on the partial least-squares (PLS) analysis.

CoMFA for Substituted Amphetamines. The amphetamine analogs were sketched using library fragments in SYBYL. Atom types were automatically assigned by SYBYL and then manually checked. The stereochemistry was assigned as S, based on the known S > R stereoselectivity for amphetamines at monoamine transporters. The compounds were subjected to a grid search (0–360°, 10° increment) to find the lowest energy conformation. This conformation was then minimized for all molecules using the conjugate gradient method with no initial simplex minimization, a gradient value of 0.05, and 10,000 maximum iterations. Alignment of the database was optimized using a template molecule based on the most rigid structure of (S)-5-methoxy-6-methyl-2-aminoindan, with the aromatic ring centroid and amine nitrogen of each compound held as the anchor points. The aligned molecules were then inserted into a molecular spreadsheet with pK_i values as functional data. CoMFA data were generated using the Tripos Standard CoMFA field class, with distance dielectric, no smoothing, and a steric and electrostatic cutoff of 30.0 kcal/mol. PLS analysis was validated using the leave-one-out option. Four components were initially selected for the PLS analysis, and then the number of optimal components was determined. For the hSERT, dSERT, and H^1–281D^282–476H^477–638chimera models three components were used with q² values of 0.65, 0.73, and 0.59, respectively.

Materials. Dulbecco's modified Eagle's medium, pargyline, and L-ascorbic acid were purchased from Sigma-Aldrich (St. Louis, MO). Dialyzed fetal bovine serum was purchased from Hyclone Laboratories (Logan, UT). Trypsin, L-glutamine, penicillin/streptomycin, and geneticin were from Invitrogen (Carlsbad, CA). Bradford protein assay dye concentrate was from Bio-Rad (Hercules, CA). Cell culture flasks and 150-mm dishes were from Falcon/B-D Labware (Mountain View, CA). Microtube racks (96-well) and tubes were purchased from Marsh Bio Products (Rochester, NY). [³H]5-HT trifluoroacetate (110 Ci/mmol) was purchased from Amersham Biosciences (Piscataway, NJ). Cocaine and CPT-D-tartrate were purchased from Sigma/RBI (Natick, MA). All other chemicals were purchased from Sigma-Aldrich or Fischer Scientific (Pittsburgh, PA) and were of the highest grade available. 3-Phenyltropane analogs RTI-142 and RTI-121 (Neumeyer et al., 1994), RTI-55 and RTI-112 (Carroll et al., 1994), -CFT (Neumeyer et al., 1996), RTI-32 (Holmquist et al., 1996), RTI-31 (Carroll et al.,1995), RTI-83 (Blough et al., 1996), and RTI-311 (Scheffel et al., 1997) were synthesized as described previously. Substituted amphetamines were synthesized using conventional methodologies and their structures confirmed with NMR, mass spectrometry, and elemental analysis, all of which matched expected criteria.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Comparison of 3-Phenyltropane Analogs and Substituted Amphetamines at hSERT, dSERT, and H^1–281D^282–476H^477–638 Chimera. Initial studies determined the sensitivities of hSERT, dSERT, and H^1–281D^282–476H^477–638 to cocaine and amphetamine derivatives. The potency of the 3-phenyltropane analogs and substituted amphetamines for inhibition of [³H]5-HT uptake was determined (Fig. 1). The K_i value for inhibition of 5-HT uptake in HEK-293 cells expressing hSERT (hSERT-HEK) was 10- to 100-fold lower, depending on the compound, compared with dSERT and H^1–281D^282–476H^477–638-expressing cells (dSERT-HEK and H^1–281D^282–476H^477–638-HEK, respectively) (Tables 1 and 2). Interestingly, potencies at H^1–281D^282–476H^477–638 generally paralleled dSERT for all compounds assayed. For the 3-phenyltropane analogs, the rank order of K_i values at hSERT was RTI-55 < [RTI-] 112 < -311 < -31 < -142 < -83 < -32 < -121 < cocaine < -CFT and < CPT-D-tartrate. At dSERT, the rank order of K_i values was RTI-55 < RTI-142 < -121 < -311 < -32 < -112 < cocaine, <-31, <-83 < -CFT and < CPT-D-tartrate. At the H^1–₂₈₁D^282–476H^477–638 chimera, the rank order of K_i values was RTI-55 = RTI-142 < cocaine < RTI-121 < -112 < -311 < -83 < -31 < -CFT < -32 and < CPT-D-tartrate.

View larger version (21K): [in this window] [in a new window]   Fig. 1. A, diagram of wild-type hSERT, dSERT, and the H1–281D282–476H477–638 chimera. The transporters were constructed as described under Materials and Methods, with hSERT sequence indicated in black and dSERT sequence indicated in white. B, inhibition of [3H]5-HT uptake at hSERT (), dSERT (), and H1–281D282–476H477–638 () using RTI-55 and 3,4-dichloroamphetamine. [3H]5-HT uptake assays were performed as described under Materials and Methods. Nonspecific uptake was determined with 10 µM fluoxetine. Data were plotted as a percentage of specific [3H]5-HT uptake. Dose-response curves depicted are representative of at least three experiments performed in triplicate.

For the substituted amphetamines, the order of increasing K_i values at hSERT was DCA < 4-MTA < 2-Me-MDA < MMAI < MDA <3-MTA < AMMI < DFA <6-Me-MDA <4-TFMA < AMMT. The order of increasing K_i values for the substituted amphetamines at dSERT was DCA < 4-MTA < 3-MTA < AMMT < MMAI < MDA <6-Me-MDA < 2-Me-MDA < DFA < 4-TFMA. The order of increasing K_i values for H^1–281D^282–476H^477–638 chimera was DCA < 4-MTA < 3-MTA = MDA = 6-Me-MDA < AMMT = AMMI < MMAI = 2-Me-MDA < 4-TFMA.

CoMFA of 3-Phenyltropane Analogs at hSERT, dSERT, and H^1–281D^282–476H^477–638 Chimera. A Silicon Graphics O2 workstation and the SYBYL software package were used to generate CoMFA maps for distinct models of 3-phenyltropane analog interaction with human (Fig. 2A), Drosophila (Fig. 2B), and the H^1–281D^282–476H^477–638 chimeric (Fig. 2C) SERTs, using RTI-55 as a model compound for reference. Overall, upon inspection, a high degree of similarity exists between the dSERT (Fig. 2B) and H^1–281 D^282–476H^477–638 (Fig. 2C) models. Also, the compounds assayed share very similar potencies for dSERT and the H^1–281D^282–476H^477–638 chimera. Four prominent features stand out in all three models. First, the region of disfavored steric bulk, near the bridgehead amine, depicted in yellow is reduced in size in the dSERT and H^1–281D^282–476H^477–638 models (Fig. 2, B and C), compared with hSERT (Fig. 2A). The pharmacological data indicate that addition of bulky groups to the amine nitrogen or to the C2 carboxylic acid carbon results in a loss of potency. For example, RTI-121 and RTI-311 are congeners of RTI-55 with bulky substitutions at position C2 or the amine nitrogen, respectively. Both of these compounds showed profound loss of activity at all three constructs compared with RTI-55 (Table 1). A second prominent feature is the blue field, denoting positive charge and hydrogen bond donors being favored. In hSERT (Fig. 2A), the blue field surrounds the aromatic ring, whereas in dSERT and H^1–281D^282–476H^477–638 (Fig. 2, B and C) the blue region is reduced in size and does not envelop the aromatic ring as completely as in the hSERT model. The third region of interest is the area denoting favored steric bulk, in green, near the 4'-substituted position of the aromatic ring. All three models share this green field. The effect of this region of steric bulk is evident when comparing -CFT, RTI-31, and RTI-55. The potency of the compound increases in going from F-, to Cl-, to I-, following increasing hydrophobicity of the substituent. The favorability of steric bulk is evident in that the 4'-ethyl substitution (RTI-83) was more potent than the 4'-methyl congener (RTI-32) of the 3-phenyltropane series (60 versus 110 nM at hSERT). The H^1–281D^282–476H^477–638 model shows this region of favored steric bulk that is larger than either parental hSERT or dSERT. In both hSERT and H^1–281D^282–476 H^477–638 lengthening the 4'-substituent from methyl to ethyl affords a modest increase in potency. By contrast, at dSERT, this changes lead to decreased affinity. The small yellow regions in this location in dSERT may indicate a reduced tolerance to steric bulk at this site, relative to hSERT and H^1–281D^282–476H^477–638. The fourth region of interest is the red field region of favored negative charge on the H^1–281 D^282–476H^477–638 model (Fig. 2C) also located near the 4'-position that is shared by the hSERT model (Fig. 2A) but not the dSERT model (Fig. 2B). The somewhat higher potency of RTI-31 (4'-Cl) over RTI-32 (4'-CH₃) at the hSERT and H^1–281 D^282–476H^477–638 chimera may reflect this favorability for negative substituents (such as chlorine or iodine) at the 4'-position.

View larger version (34K): [in this window] [in a new window]   Fig. 2. CoMFA maps of cocaine analog interactions with hSERT (A), dSERT (B), and H1–281D282–476H477–638 (C) chimera, arranged for cross-eyed viewing. RTI-55 is modeled within the field for directional orientation. Colored fields represent characteristics of the ligand that favorably or unfavorably affect high-affinity interactions. Steric contours are shown in green and yellow, whereas electrostatic contours are shown in blue and red. Green fields indicate molecular regions where steric bulk is favored for high-affinity interactions, whereas regions indicated by yellow are areas in which steric bulk is disfavored for high-affinity interactions. Blue fields indicate regions of the molecule where more positive charge and hydrogen bond donors are favored, and negative charge and hydrogen bond acceptors are disfavored for high-affinity interactions. Red fields indicate regions where negatively charged substituents and hydrogen bond acceptors are favored, and positive charge and hydrogen bond donors are disfavored.

CoMFA of Substituted Amphetamines at hSERT, dSERT, and H^1–281D^282–476H^477–638 Chimera. A Silicon Graphics O2 workstation and the SYBYL software package were used to generate CoMFA maps for distinct models of amphetamine derivative interaction with the 5-HT transporters. Figure 3 is the pharmacological data as a CoMFA model using 3,4-dichloroamphetamine for orientation for interactions with the human (Fig. 3A), Drosophila (Fig. 3B), and the H^1–281D^282–476H^477–638 chimeric (Fig. 3C) SERTs. The compounds in this training set all possess the aromatic core and ethylamine of the parent compound d-amphetamine, but vary by either single substitutions on the aromatic ring or the creation of a fused, sterically constrained bicyclic system with the aromatic ring and another saturated five- or six-membered ring. As with the 3-phenyltropane derivatives, inspection of Table 2 shows that the majority of compounds have similar potencies at dSERT and H^1–281 D^282–476H^477–638 that are, in general, lower than their potencies at hSERT. Inspection of the CoMFA fields shows a general similarity between those of dSERT and H^1–281D^282–476 H^477–638 consistent with the experimental data. The CoMFA fields for all three substituted amphetamine models show a region of favored steric bulk (green) in the vicinity of the 4-position of the aryl ring. This field changes size very slightly among the three models. In addition, a red region exists near the 3-position of the aryl ring, indicating a region where electronegative substituents would increase potency. This red region is considerably larger on the hSERT model (Fig. 3A) compared with the models in Fig. 3, B and C. There is also a sterically disfavored yellow region in the vicinity of the 5,6-positions of the aryl ring on all three models. The hSERT CoMFA (Fig. 3A) differs principally in that it has a much larger negative field at the 3-aryl position. In addition, the disfavored yellow region is displaced from its hSERT location in both the dSERT and H^1–281D^282–476H^477–638 models (Fig. 3, B and C). Each of the amphetamines tested in this training set contain the ethylamine moiety. The region around the amine differs in each of the three models. For hSERT (Fig. 3A), the region around the amine shows a small red field (negative electrostatics favored) as well as a slightly distal small blue (positive electrostatics favored) field. The dSERT and H^1–281D^282–476H^477–638 models show much larger red fields in this region. These results suggest a contact site with the amphetamine amine moiety exists on hSERT, but is different in dSERT and the H^1–281D^282–476 H^477–638 chimera.

View larger version (20K): [in this window] [in a new window]   Fig. 3. CoMFA maps of substituted amphetamine interactions with hSERT (A), dSERT (B), and H1–281D282–476H477–638 (C) chimera, arranged for cross-eyed viewing. 3,4-Dichloroamphetamine is modeled within the field for directional orientation. Colored fields represent characteristics of the ligand that favorably or unfavorably affect high-affinity interactions. Steric contours are shown in green and yellow, whereas electrostatic contours are shown in blue and red. Green fields indicate molecular regions where steric bulk is favored for high-affinity interactions, whereas regions indicated by yellow are areas in which steric bulk is disfavored for high-affinity interactions. Blue fields indicate regions of the molecule where more positive charge and hydrogen bond donors are favored, and negative charge and hydrogen bond acceptors are disfavored for high-affinity interactions. Red fields indicate regions where negatively charged substituents and hydrogen bond acceptors are favored, and positive charge and hydrogen bond donors are disfavored.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
The studies described in this article seek to examine the structural features of two different psychostimulants in the context of their common ability to interact with SERTs. Due to the striking differences in their molecular structure, yet common target, one might predict some overlapping features responsible for molecular recognition if shared contact sites exist.

Structure-Activity Relationships of 3-Phenyltropane Analogs at hSERT, dSERT, and H^1–281D^282–476 H^477–638 Chimera. The most potent 3-phenyltropane analog tested in this study at both hSERT and dSERT was RTI-55 (Carroll et al., 1994), whereas RTI-55, RTI-142, and cocaine are equipotent at the H^1–281D^282–476H^477–638 chimera. The 3-phenyltropane analog RTI-55 differs from cocaine in two ways. First, it lacks the ester linkage between the tropane substructure and the aromatic ring. Second, RTI-55 contains an iodine atom at the 4'-(para) position on the aromatic ring. These two changes to cocaine's structure combine to promote an increase in potency approximately 40-fold at hSERT (250 versus 7 nM) and greater than 3-fold at dSERT (190 versus 750 nM), but little difference at H^1–281D^282–476H^477–638, possibly indicating a unique mode of recognition by the chimera. Using RTI-55 and our CoMFA models, several possible conclusions were drawn about the high-potency interaction between 3-phenyltropane analogs and SERTs. Substitutions at the 4'-position on the aromatic ring seem to be important for potent interactions with SERT (as reflected by the red fields in Fig. 2, A and C). The 4'-fluoro is the most electron withdrawing substituent, followed by the 4'-chloro and 4'-iodo. In rank order of potency, the iodine-containing compound RTI-55 is most potent, followed by the chlorine-containing (RTI-31) and then the fluorine-containing analog (-CFT). These results are similar to those reported in previous studies focused on the cocaine binding site in purified rat brain striatum, a rich source of dopaminergic neurons (Carroll et al., 1994). This study, although performed on tissue containing DAT, supports the hypothesis that the importance of the substitution at the 4'-position may be general for high-affinity 3-phenyltropane analog interactions with biogenic amine neurotransmitter transporters (Carroll et al., 1994).

Our CoMFA models revealed similarities among hSERT, dSERT, and the H^1–281D^282–476H^477–638 chimera interaction with the aromatic ring of the 3-phenyltropane analogs. In the models, the electron withdrawing substituents seem to exert their effect through a delocalization of electrons in the aromatic ring. Interestingly, the dSERT and H^1–281D^282–476 H^477–638 chimera models (Fig. 2, B and C) show a much smaller area of favorable interactions around the aromatic ring faces. It has been proposed that pi-pi stacking interactions with residues in DAT may contribute to high-potency interactions with cocaine analogs (Blough et al., 2002). Aromatic amino acids present in hSERT, but absent in dSERT and H^1–281D^282–476H^477–638, could be important points of interaction or provide unique conformations. The H^1–281 D^282–476H^477–638 chimera may have contact points for cocaine in both its central dSERT (TMDs V–IX) and its hSERT (TMDs I-IV and X–XII) regions. The effect of alkyl substitutions on the aromatic ring have been shown previously to effect binding to both SERT and NET (Blough et al., 1996). The green region of favorable steric bulk in the vicinity of the 4-position is common to all three models (Fig. 2, A–C). The effect of this steric region is evident when comparing -CFT, RTI-32, and RTI-55. The potency of the compounds increases when going from F-, to Cl-, to I-, with increasing hydrophobicity of substituents. Steric bulk effects have also been observed for DAT binding sites in rat striatum, where a 4'-bromo substitution had similar potency as a 4'-methyl substitution (1.8 versus 1.7 nM) (Carroll et al., 1991). Similar van der Waals radii of the bromine and methyl substituent could explain the similarity in potency. This steric constraint has been observed with other cocaine derivatives at biogenic amine transporters (Neumeyer et al., 1996). Interestingly, our CoMFA models do not suggest a significant role for the charge associated with the amine nitrogen in contributing to potency. This observation is supported by the recent synthesis of non-nitrogen-containing phenyltropanes by Goulet et al. (2001) that effectively and potently inhibit 5-HT transport.

Structure-Activity Relationships of Substituted Amphetamines at hSERT, dSERT, and H^1–281D^282–476 H^477–638 Chimera. Amphetamine and its neurotoxic derivatives are known to interact with the biogenic amine transporters to cause an increase in synaptic neurotransmitters. Structure-activity relationships for substituted amphetamines have been studied at SERT, DAT, and NET. For instance, amphetamines with dual substitutions in the 3- and 4-positions become more selective for serotonergic effects (Nichols, 1994). Also, primary amines are the most potent for the biogenic amine neurotransmitter transporter family, but N-ethyl substitutions can differentially improve interactions with SERTs (Nichols, 1994). Our studies have focused on these two molecular areas to contribute to the understanding of substituted amphetamine structure-activity relationships.

In this study, we have generated CoMFA maps from pharmacological data for hSERT, dSERT, and the H^1–281D^282–476 H^477–638 chimera using a set of substituted amphetamines. Similarities and differences in their characteristics can be identified from these analyses. At all three SERT constructs the most potent substituted amphetamine assayed was 3,4,-dicholoroamphetamine, although the compound was 10-fold more potent at hSERT than dSERT and the H^1–281D^282–476 H^477–638 chimera (0.20 µM for hSERT versus 1.5 µM for dSERT and H^1–281D^282–476H^477–638). Structurally, the closest derivative in this study to 3,4-dicholoroamphetamine was 3,4-difluoroamphetamine, which differs only by the identity of the halide substitution. This fluorine-substituted compound, however, had markedly decreased potency at species variants compared with the chlorine-substituted variant (4.3 µM at hSERT, 56 µM at dSERT, and 12 µM at H^1–281 D^282–476H^477–638). The CoMFA models indicate a region of favored negative charge (Fig. 3, in red) in the area around the 3-position of the aromatic ring on each of the three models. However, this field is slightly displaced from the 3-position, indicating that this electrostatic effect may not be solely determined by the substitution at the 3-position. The green field near the 4-position indicates that the bulk of the substitution plays a major role in potency. The role of this position was examined in a previous study that revealed that bulk at the 4-position is critical in determining potency (Marona-Lewicka et al., 1995). The second most potent compound at hSERT, dSERT, and the H^1–281D^282–476H^477–638 chimera was 4-methylthioamphetamine (0.45, 4.6, and 3.8 µM, respectively). The position of the methylthiol group affects potency, with the 4-methylthioamphetamine exhibiting higher potency compared with 3-methylthioamphetamine, due to favorable steric interactions as indicated by the green region in Fig. 3.

In addition to distinctions surrounding the aromatic ring, differences in molecular fields are present near the amine. These species-specific interactions revealed similarities between dSERT and the H^1–281D^282–476H^477–638 chimera. The Drosophila SERT sequence in the H^1–281D^282–476H^477–638 chimera is localized to the TMD V–IX region, implicating these domains as possessing residues involved with interacting with the amine functionality. Previously published reports have identified TMD I as an important region of SERT involved with recognition of the amine of both 5-HT (Barker et al., 1999) and tryptamine derivatives (Adkins et al., 2001). Together, our data and these previous studies indicate that substrates likely interact with multiple TMDs, including TMD I and the TMD V-IX region, and, consequently, any model of SERT structure should consider multiple points of contact within the transporter to form a substrate translocation pathway. The models show two additional sites of similarity among the three SERTs. The green field (steric bulk favorability) could suggest that these bulky, hydrophobic substituents interact with a corresponding hydrophobic region on SERT. In addition, a sterically disfavored yellow region in the vicinity of the 5,6-positions of the aryl ring is common to all three models; however, its location is displaced in the dSERT and H^1–281D^282–476H^477–638 model compared with hSERT.

In conclusion, comparing and contrasting the two models of psychostimulant recognition by SERT provides a unique opportunity to reveal evidence of unique interaction sites for cocaine and substituted amphetamines at SERT. For the 3-phenyltropane analog models, differences exist in the way SERT species variants recognize these compounds on several levels: recognition of the phenyl ring, different substituents in the 4'-position of the aromatic ring, and the electrostatics of the phenyl ring. Similarity among these three models exists in the steric bulk components around the bridgehead amine and the 4'-position substitutions. For the substituted amphetamines favored negative charge around the 3- and 4-position is shared by all three SERTs. Also, differences in the way the three SERTs recognize the amine functionality exist. Because all known substrates for SERTs contain amine groups, the site involved with coordinating the amine may be critical for substrate recognition and translocation. The overall dissimilarity of the CoMFA fields for 3-phenyltropane and substituted amphetamine interactions is suggestive that the sites for interaction on SERT are very different. It is possible that the regions that bind to the 4'-aryl of the 3-phenyltropane analogs and the 4-aryl position of the substituted amphetamines could be the same, but there are no other structural features to suggest other similarities. We would argue, therefore, that the sites of interactions for these two families of psychostimulants are distinct, but have major determinants within the TMD V–IX region. Also, the uniqueness of the molecular interactions can likely be explained by differences in mechanism of action for these psychostimulants; the substituted amphetamines can act as substrates for the transporter, whereas the 3-phenyltropanes act as transport blockers (Wall et al., 1995). Future molecular studies should identify specific amino acids involved with recognition of psychostimulants and provide important new information about SERT function.

	Acknowledgements

 
We thank Fariborz Rakhshan and Vicki L. Croy for expert technical assistance and Kellie J. White for critical reading of the manuscript. We also thank Dr. Maximillian A. Silvestri for insightful discussion about molecular modeling and CoMFA.

	Footnotes

 
This work was supported by National Institute of Mental Health Grant MH60221 (to E.L.B.) and National Institute on Drug Abuse Grant DA05477 (to F.I.C.).

DOI: 10.1124/jpet.103.057836.

ABBREVIATIONS: 5-HT, 5-hydroxytryptamine, serotonin; SERT, serotonin transporter; CoMFA, comparative molecular field analysis; TMD, transmembrane domain; HEK, human embryonic kidney; KRH, Krebs-Ringer-HEPES; PLS, partial least-squares; DAT, dopamine transporter; NET, norepinephrine transporter, CPT-D-tartrate, (–)-2-carbomethoxy-3-phenyltropane tartrate; B-CFT, (–)-2--carbomethoxy-3--(4-fluorophenyl)-tropane 1,5-naphthalenedisulfonate; RTI-142, (–)-N-nor-3-(4-fluorophenyl)tropane-2-carboxylic acid methyl ester; RTI-121, 3-(4-iodophenyl)-tropane-2-carboxylic acid isopropyl ester hydrochloride; RTI-55, 3-(4-iodophenyl)tropane-2-carboxylic acid methyl ester tartrate; RTI-112, (–)-3-(3-methyl-4-chlorophenyl)tropane-2-carboxylic acid methyl ester tartrate; RTI-32, 3-(4-methylphenyl)tropane-2-carboxylic acid methyl ester tartrate; RTI-31, 3-(4-chorophenyl)tropane-2-carboxylic acid methyl ester tartrate; RTI-83, 3-(4-ethylphenyl)tropane-2-carboxylic acid methyl ester tartrate; RTI-311, N-allyl-N-nor-3-(4-iodophenyl)tropane-2-carboxylic acid methyl ester tartrate; DCA, 3,4-dichloroamphetamine; 4-MTA, 4-methylthioamphetamine; 2-Me-MDA, 2-methyl-3,4-methylenedioxyamphetamine; MMAI, 5-methoxy-6-methyl-2-aminoindan, MDA, 3,4-methylenedioxyamphetamine; 3-MTA, 3-methylthioamphetamine; AMMI, 1-aminomethyl-5-methoxyindan; DFA, 3,4-difluoroamphetamine; 6-Me-MDA, 6-methyl-3,4-methylenedioxyamphetamine; 4-TFMA, 4-trifluoromethylamphetamine; AMMT, 1-aminomethyl-6-metoxytetralin.

Address correspondence to: Dr. Eric L. Barker, Department of Medicinal Chemistry and Molecular Pharmacology, Purdue University School of Pharmacy, 575 Stadium Mall Dr., West Lafayette, IN 47907-2091. E-mail: ericb{at}pharmacy.purdue.edu

	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]

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, in Psychopharmacology: The Fourth Generation of Progress (Floyd E. Bloom and David J. Kupfer eds) pp 321–333, 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]

Blakely RD, De Felice LJ, and Hartzell HC (1994) Molecular physiology of norepinephrine and serotonin transporters. J Exp Biol 196: 263–281.[Abstract/Free Full Text]

Blough BE, Abraham P, Lewin AH, Kuhar MJ, Boja JW, and Carroll FI (1996) Synthesis and transporter binding properties of 3 beta-(4'-alkyl-, 4'-alkenyl- and 4'-alkynylphenyl)nortropane-2 beta-carboxylic acid methyl esters: serotonin transporter selective analogs. J Med Chem 39: 4027–4035.[CrossRef][Medline]

Blough BE, Keverline KI, Nie Z, Navarro H, Kuhar MJ, and Carroll FI (2002) Synthesis and transporter binding properties of 3beta-[4'-(phenylalkyl, -phenylalkenyl and -phenylalkynyl)phenyl]tropane-2beta-carboxylic acid methyl esters: evidence of a remote phenyl binding domain on the dopamine transporter. J Med Chem 45: 4029–4037.[CrossRef][Medline]

Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248–254.[CrossRef][Medline]

Carroll FI, Gao YG, Rahman MA, Abraham P, Parham K, Lewin AH, Boja JW, and Kuhar MJ (1991) Synthesis, ligand binding, QSAR and CoMFA study of 3 beta-(p-substituted phenyl)tropane-2 beta-carboxylic acid methyl esters. J Med Chem 34: 2719–2725.[CrossRef][Medline]

Carroll FI, Kotian P, Dehghani A, Gray JL, Kuzemko MA, Parham KA, Abraham P, Lewin AH, Boja JW, and Kuhar MJ (1995) Cocaine and 3 beta-(4'-substituted phenyl)tropane-2 beta-carboxylic acid ester and amide analogues. New high-affinity and selective compounds for the dopamine transporter. J Med Chem 38: 379–388.[CrossRef][Medline]

Carroll FI, Mascarella SW, Kuzemko MA, Gao Y, Abraham P, Lewin AH, Boja JW, and Kuhar MJ (1994) Synthesis, ligand binding, and QSAR (CoMFA and classical) study of 3 beta-(3'-substituted phenyl)-, 3 beta-(4'-substituted phenyl)- and 3 beta-(3', 4'-disubstituted phenyl)tropane-2 beta-carboxylic acid methyl esters. J Med Chem 37: 2865–2873.[CrossRef][Medline]

Cheng Y and Prusoff WH (1973) Relationship between the inhibition constant (K1) and the concentration of inhibitor which causes 50 per cent inhibition (I50) of an enzymatic reaction. Biochem Pharmacol 22: 3099–3108.[CrossRef][Medline]

Corey JL, Quick MW, Davidson N, Lester HA, and Guastella J (1994) A cocaine-sensitive Drosophila serotonin transporter: cloning, expression and electrophysiological characterization. Proc Natl Acad Sci USA 91: 1188–1192.[Abstract/Free Full Text]

Cozzi NV, Sievert MK, Shulgin AT, Jacob P, III and Ruoho AE (1999) Inhibition of plasma membrane monoamine transporters by beta-ketoamphetamines. Eur J Pharmacol 381: 63–69.[CrossRef][Medline]

Demchyshyn LL, Pristupa ZB, Sugamori KS, Barker EL, Blakely RD, Wolfgang WJ, Forte MA, and Niznik HB (1994) Cloning, expression and localization of a chloride-facilitated, cocaine-sensitive serotonin transporter from Drosophila melanogaster. Proc Natl Acad Sci USA 91: 5158–5162.[Abstract/Free Full Text]

Eshleman AJ, Carmolli M, Cumbay M, Martens CR, Neve KA, and Janowsky A (1999) Characteristics of drug interactions with recombinant biogenic amine transporters expressed in the same cell type. J Pharmacol Exp Ther 289: 877–885.[Abstract/Free Full Text]

Goulet M, Miller GM, Bendor J, Liu S, Meltzer PC, and Madras BK (2001) Non-amines, drugs without an amine nitrogen, potently block serotonin transport: novel antidepressant candidates? Synapse 42: 129–140.[CrossRef][Medline]

Hoffman BJ, Mezey E, and Brownstein MJ (1991) Cloning of a serotonin transporter affected by antidepressants. Science (Wash DC) 254: 579–580.[Abstract/Free Full Text]

Holmquist CR, Keverline-Frantz KI, Abraham P, Boja JW, Kuhar MJ, and Carroll FI (1996) 3 alpha-(4'-Substituted phenyl)tropane-2 beta-carboxylic acid methyl esters: novel ligands with high affinity and selectivity at the dopamine transporter. J Med Chem 39: 4139–4141.[CrossRef][Medline]

Jones SR, Joseph JD, Barak LS, Caron MG, and Wightman RM (1999) Dopamine neuronal transport kinetics and effects of amphetamine. J Neurochem 73: 2406–2414.[CrossRef][Medline]

Kantor L, Hewlett GH, and Gnegy ME (1999) Enhanced amphetamine- and K+-mediated dopamine release in rat striatum after repeated amphetamine: differential requirements for Ca2+- and calmodulin-dependent phosphorylation and synaptic vesicles. J Neurosci 19: 3801–3808.[Abstract/Free Full Text]

Li AH, Moro S, Forsyth N, Melman N, Ji XD, and Jacobson KA (1999) Synthesis, CoMFA analysis and receptor docking of 3,5-diacyl-2,4-dialkylpyridine derivatives as selective A3 adenosine receptor antagonists. J Med Chem 42: 706–721.[CrossRef][Medline]

Marona-Lewicka D, Rhee G-S, Sprague JE, and Nichols DE (1995) Psychostimulant-like effects of p-fluoroamphetamine in the rat. Eur J Pharmacol 287: 105–113.[CrossRef][Medline]

Neumeyer JL, Tamagnan G, Wang S, Gao Y, Milius RA, Kula NS, and Baldessarini RJ (1996) N-Substituted analogs of 2 beta-carbomethoxy-3 beta-(4'-iodophenyl) tropane (beta-CIT) with selective affinity to dopamine or serotonin transporters in rat forebrain. J Med Chem 39: 543–548.[CrossRef][Medline]

Neumeyer JL, Wang S, Gao Y, Milius RA, Kula NS, Campbell A, Baldessarini RJ, Zea-Ponce Y, Baldwin RM, and Innis RB (1994) N-omega-Fluoroalkyl analogs of (1R)-2 beta-carbomethoxy-3 beta-(4-iodophenyl)-tropane (beta-CIT): radiotracers for positron emission tomography and single photon emission computed tomography imaging of dopamine transporters. J Med Chem 37: 1558–1561.[CrossRef][Medline]

Newman AH, Izenwasser S, Robarge MJ, and Kline RH (1999) CoMFA study of novel phenyl ring-substituted 3alpha-(diphenylmethoxy)tropane analogues at the dopamine transporter. J Med Chem 42: 3502–3509.[CrossRef][Medline]

Nichols DE (1994) Medicinal chemistry and structure-activity relationships, in Amphetamine and Its Analogs (Cho AK and Segal DS eds) pp 3–41, Academic Press, San Diego.

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.[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]

Reneman L, Booij J, Lavalaye J, de Bruin K, Reitsma JB, Gunning B, den Heeten GJ, and van Den BW (2002) Use of amphetamine by recreational users of ecstasy (MDMA) is associated with reduced striatal dopamine transporter densities: a [123I]beta-CIT SPECT study–preliminary report. Psychopharmacology 159: 335–340.[CrossRef][Medline]

Rodriguez 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.[Abstract/Free Full Text]

Rudnick G and Clark J (1993) From synapse to vesicle: the reuptake and storage of biogenic amine neurotransmitters. Biochim Biophys Acta 1144: 249–263.[Medline]

Scheffel U, Lever JR, Abraham P, Parham KR, Mathews WB, Kopajtic T, Carroll FI, and Kuhar MJ (1997) N-substituted phenyltropanes as in vivo binding ligands for rapid imaging studies of the dopamine transporter. Synapse 25: 345–349.[CrossRef][Medline]

Schloss P and Williams DC (1998) The serotonin transporter: a primary target for antidepressant drugs. J Psychopharmacol 12: 115–121.

Usdin TB, Mezey E, Chen C, Brownstein MJ, and Hoffman BJ (1991) Cloning of the cocaine-sensitive bovine dopamine transporter. Proc Natl Acad Sci USA 88: 11168–11171.[Abstract/Free Full Text]

Uhl GR, Hall FS, and Sora I (2002) Cocaine, reward, movement and monoamine transporters. Mol Psychiatry 7: 21–26.[CrossRef][Medline]

Wall SC, Gu H, and Rudnick G (1995) Biogenic amine flux mediated by cloned transporters stably expressed in cultured cell lines: amphetamine specificity for inhibition and efflux. Mol Pharm 47: 544–550.[Abstract]

Wilcox RE, Tseng T, Brusniak MY, Ginsburg B, Pearlman RS, Teeter M, DuRand C, Starr S, and Neve KA (1998) CoMFA-based prediction of agonist affinities at recombinant D1 vs D2 dopamine receptors. J Med Chem 41: 4385–4399.[CrossRef][Medline]

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.103.057836v1 308/2/679    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 Google Scholar Google Scholar Articles by Roman, D. L. Articles by Barker, E. L. Search for Related Content PubMed PubMed Citation Articles by Roman, D. L. Articles by Barker, E. L. Pubmed/NCBI databases Compound via MeSH Substance via MeSH Hazardous Substances DB TRITIUM