Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

The dopamine transporter (DAT) has been implicated as a primary binding site for several drugs of abuse, including cocaine and methamphetamine (Amara and Kuhar, 1993) as well as for therapeutic drugs such as methylphenidate, mazindol, and bupropion, which are used to treat attention deficit disorder, as an anorexiant and as an antidepressant, respectively. Cocaine inhibits the inward transport of DA by the DAT, and methamphetamine, a substrate for the DAT, competes with DA at the DAT and releases DA from intracellular stores. In human subjects, the magnitude of the self-reported high following cocaine administration is correlated with the degree of DAT occupancy (Volkow et al., 1997). The DAT is a member of the sodium- and chloride-dependent biogenic amine transporter family predicted to contain 12-membrane spanning regions as determined by hydrophobicity analysis, internal amino- and carboxyl-terminals, three to five glycosylation sites, and several internal consensus phosphorylation sites.

The development of therapeutic agents that interfere with the binding of abused drugs while permitting the functioning of the DAT is one possible strategy for treating cocaine and methamphetamine abuse (Rothman, 1990; Johnson and Vocci, 1993; Vocci et al., 1995; Kuhar et al., 1999). Multiple lines of evidence support the possibility of identifying a drug that can block binding of an abused substance to the transporter but not affect basal uptake. These include differential effects on substrate uptake and cocaine-analog binding following single nucleotide substitutions in the DAT (for review, see Uhl et al., 1998) and protection of different fragments of the DAT from proteolytic digestion by structurally dissimilar DAT uptake inhibitors (Vaughan, 1995). Also, differing domains are important in determining the affinity of substrates and inhibitors, and maximal transport rates, as revealed by chimeras constructed of portions of the DAT and another member of this family of transporters, the norepinephrine transporter (Giros et al., 1994; Buck and Amara, 1995). These results suggest that structurally dissimilar inhibitors and substrates bind to overlapping, but nonidentical sites, on the DAT.

The use of cocaine in humans generally either does not affect or decreases the transcription levels of DAT mRNA (Little et al., 1998; Chen et al., 1999). In contrast, chronic cocaine exposures increase DAT protein density in both rat and human brains (Staley et al., 1994b; Little et al., 1998). In rats, withdrawal from cocaine causes a decrease in mRNA levels (Cerruti et al., 1994) and a decrease in DAT protein in the frontal cortex (Hitri and Wyatt, 1993).

Several model systems have been used to characterize drug interactions with the DAT, including rat striatal preparations and cell lines that express the recombinant human or rat DAT. In addition, membranes from cryopreserved human caudate have been used to measure cocaine analog binding (Little et al., 1993; Staley et al., 1994a). Studies with rat striatum revealed that post-mortem degradation and cryopreservation cause a modest decrease in the density of the protein as determined using radioligand binding (Janowsky et al., 1987) and decrease the affinity and maximal uptake rate of DAT (Haberland and Hetey, 1987). Good correlations have been reported between drug potency at inhibition of binding of cocaine analogs or of structurally dissimilar ligands at the human and rat DATs (Rothman et al., 1994). There has been very little reported on the pharmacology of drug effects on DA transport in human tissue due to the difficulties in measuring robust transport rates after autolysis and freezing. In the present study, we have characterized the transport rate of [³H]DA, and the potency of drugs at inhibiting [³H]DA uptake, in synaptosomes prepared from cryopreserved human caudate, and the density of binding sites for the cocaine analog [¹²⁵I]RTI-55. The kinetic characteristics determined in human tissue are compared with those obtained with freshly prepared and cryopreserved rat striatum. The inhibitory potencies of drugs for uptake and binding are compared among human, rat, and recombinant heterologously expressed DAT.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials

Human caudate tissue from non-neurological control subjects was provided by the Brain Endowment Bank (DA 06227; AGO5128). All cases were evaluated for common drugs of abuse, alcohol, and prescription medications and positive urine screens were confirmed by quantitative analysis of blood. Cases with a positive screen for compounds that might affect the dopaminergic system, such as prescription drugs, drugs administered during medical intervention, or drugs of abuse, were classified as "polydrug" and were excluded from this study. Caudates were sampled from drug-free and age-matched control subjects selected from accidental deaths with no cocaine or metabolites detected in toxicology screens of blood or brain tissue. Caudate tissues were blocked from the anterior sectors of the striatum and equilibrated in ice-cold sucrose solution. All donors were male, with an average age of 49 ± 2 years and had an average brain weight of 1.47 ± 0.05 kg. Autolysis time ranged from 7.5 to 25 h with an average of 14.9 ± 2.2 h. Caudate samples from six donors were used for the binding assays, and samples from 16 donors were used for the uptake assays.

Male Sprague-Dawley rats were obtained from Harlan (Indianapolis, IN) and group housed two to three per cage at the Portland Veteran's Administration Medical Center Veterinary Medical Unit. Rats were maintained on a 12-h light/dark cycle with continuous access to food (Purina rat chow) and water. Twenty-four rats were used in the uptake and binding assays.

RTI-55 was a generous gift from Dr. F. Ivy Carroll (Research Triangle Institute, Research Triangle Park, NC). All other drugs were purchased from Research Biochemicals International (Natick, MA) or Sigma (St. Louis, MO). [¹²⁵I]RTI-55 and [³H]DA (specific activity 53 Ci/mmol) were purchased from NEN (Boston, MA).

Cryopreservation of Human or Rat Striatal Tissue

Sucrose (0.32 M) was prepared in Sorenson's buffer and placed in bags, which were then preweighed. The volume of sucrose was chosen to fully cover tissue samples. Fresh tissue was placed in the bag, which was heat-sealed to exclude air and to ensure the tissue remained submerged, and the bag was reweighed. The samples were packed in a thick (1.5-inch) Styrofoam cooler and packed with cotton wool balls to fill the volume of the cooler. The top was closed tightly and the entire cooler placed in a 72°C freezer. The insulating effects of the cotton and the thick-walled cooler, combined with the antifreeze effects of sucrose, caused a very slow freezing process without artifact. Long-term storage was in a 80°C freezer.

For cryopreserved rat striatal synaptosome preparations, striata were removed and placed in a bag with sucrose (0.32 M). The bag was sealed and placed in a Styrofoam container in a 80°C freezer.

[¹²⁵I]RTI-55 Binding to Human Caudate Membranes

A section of tissue (~0.5 mg) was cut from the cryopreserved sample tissue over dry ice, placed in ice-cold sucrose (0.32 M), and allowed to thaw slowly. The tissue was homogenized with a PowerGen 700 Polytron (5 s, setting 1) in sucrose (0.32 M) and the homogenate was centrifuged at 800g at 4°C for 10 min. The resulting supernatant was centrifuged at 28,000g at 4°C for 20 min. The supernatant was discarded and the pellet was resuspended in sucrose (0.32 M, 100 v/w of original tissue) immediately before use.

The assay included 50 µl of membrane preparation, 25 µl of [¹²⁵I]RTI-55 (40-60 pM), 25 µl of drug, and Krebs-HEPES buffer (pH 7.4; 25 mM HEPES, 122 mM NaCl, 2.5 mM CaCl₂, 1.2 mM MgSO₄, 10 µM pargyline, 100 µM tropolone, 0.2% glucose, and 0.02% ascorbic acid), for a final assay volume of 250 µl. Membranes were preincubated with drug for 10 min before the addition of the [¹²⁵I]RTI-55. Following a 90-min incubation at 25°C in the dark, binding was terminated by filtration over Whatman GF/C filters using a Tomtec 96-well cell harvester. Filters were washed for six seconds with ice-cold saline. Specific binding was defined as the difference in binding observed in the presence and absence of mazindol (5 µM). Three or more independent experiments with each drug were conducted with duplicate determinations.

For association experiments, buffer, membrane preparation, and mazindol (5 µM) were placed in wells. [¹²⁵I]RTI-55 was added to wells at the times indicated and the assay was terminated as described above. For saturation binding experiments, the specific activity of [¹²⁵I]RTI-55 was reduced by addition of increasing concentrations of RTI-55 (0.04-20 nM). For experiments designed to detect the role of serotonin transporters in binding, equilibrium saturation assays were conducted in parallel in the presence of buffer or 100 nM fluoxetine in a final volume of 0.25 ml with identical tissue preparations and RTI-55 dilutions. Proteins were determined by a modification of the method of Lowry et al. (1951).

For radioligand binding to rat brain preparations, the dissected areas were homogenized with a Polytron in ice-cold 0.32 M sucrose, and centrifuged at 1000g for 10 min. The resulting supernatant was centrifuged at 30,000g, and the pellet was suspended in 150 volumes of 0.32 M sucrose.

Filtration Assay for Inhibition of [³H]DA Uptake in Synaptosomes

Synaptosome Preparation. A block of cryopreserved human caudate tissue was placed into 10 to 15 ml of room temperature sucrose (0.32 M) until thawed, and then removed from sucrose and weighed. Sucrose (0.32 M, 10 volumes per original wet weight of tissue, 4°C) was added and tissue was homogenized by hand with 10 to 20 strokes of a glass-Teflon homogenizer. The homogenate was centrifuged at 1000g for 10 min at 4°C, and the supernatant was decanted and centrifuged at 20,000g for 20 min at 4°C. The pellet was resuspended in 12 to 13 volumes of glucose (0.32 M) per original wet weight of tissue at 4°C. For rat synaptosomes, fresh or cryopreserved striata were prepared as described above and the final pellet was resuspended in 125 volumes of glucose (0.32 M) per original wet weight.

Uptake Inhibition Assay. Krebs-HEPES buffer (350 µl) and drugs (50 µl) were added to 1-ml tubes (Marsh Biomedical, Rochester, NY) and placed in a 25°C water bath. Synaptosomes (50 µl) were added and preincubated with the drugs for 10 min. The assay was initiated by the addition of [³H]DA (50 µl, 20 nM final concentration). Specific uptake was defined as the difference in uptake observed in the absence and presence of mazindol (5 µM). Uptake was terminated after 5 min by filtration through Whatman GF/C filters presoaked in 0.05% polyethylenimine. Scintillation fluid was added to each filtered spot and radioactivity remaining on the filters was determined using a Wallac -plate reader. Each experiment was conducted with duplicate determinations, and three or more independent experiments for each drug competition curve were conducted.

[³H]Dopamine Uptake in Rat Brain Preparations. For rat striatal postnuclear preparations, striata were dissected and homogenized with a glass-Teflon homogenizer in ice-cold 0.32 M sucrose (4 ml). The suspension was brought up to 150 volumes of 0.32 M sucrose per original wet weight of tissue and centrifuged at 1000g for 10 min. Aliquots of the resulting supernatant (50 µl) were added to the assay, as described above for human brain tissue. All other procedures were as described above for the uptake assay, except the assay was conducted with triplicate determinations.

Data Analysis

Competition and saturation binding and uptake data were analyzed by nonlinear regression using GraphPad Prism 3.0, with IC₅₀ values converted to K_i values using the Cheng-Prusoff correction: K_I = IC₅₀/(1 + [L*]/K_d*), where L* and K_d* are the concentration and the affinity constant, respectively, of the radioligand (Cheng and Prusoff, 1973). The K_d values used in the Cheng-Prusoff equation for the binding of [¹²⁵I]RTI-55 to human caudate, rat caudate and HEK-hDAT cells were 1.43, 3.37, and 1.18 nM, respectively. Correlations (r) were calculated using the nonparametric method of Spearman, with significance set at p < 0.05.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Radioligand Binding. [¹²⁵I]RTI-55 binding to human caudate preparations increased linearly with protein, ranging from 2 to 16 µg/assay (resuspension volumes of 800 to 100 volumes per original wet weight), with deviation from linearity at 33 and 64 µg of protein (25 and 50 v/w). Using ~16 µg of protein per assay, binding of [¹²⁵I]RTI-55 to human caudate approached equilibrium at 60 min. The t_1/2 value for association was 16.5 ± 1.1 min and the apparent association rate (k_obs) was 0.042 ± 0.003 min¹ (n = 3, Fig. 1A). A single exponential association rate was adequate to describe the data in all cases. Using an equilibrium saturation binding assay, the K_d values for [¹²⁵I]RTI-55 binding to human caudate and rat striatal membranes were 1.43 ± 0.33 nM (n = 4) and 3.37 ± 0.45 nM (n = 7), respectively (Fig. 1B). For all experiments, the saturation data were best fit to a single site (Fig. 1B, Scatchard inset). The difference in K_d values between the rat and human preparations was significant (p < 0.05). The B_max values in the human and rat preparations were 1.38 ± 0.42 and 4.07 ± 0.81 pmol/mg of protein, respectively. Thus, there was only a 3-fold lower density of DAT in human membrane preparations compared with rat membrane preparations, even after the processes of autolysis during autopsy and cryopreservation.

View larger version (21K): [in this window] [in a new window]   Fig. 1.   [125I]RTI-55 binding to human caudate and rat striatal membrane preparations. A, time course of association of [125I]RTI-55 with human caudate membrane preparations. Data shown are the mean cpm ± S.E.M. of three independent experiments, each conducted with duplicate determinations. The protein concentration was ~15 µg/assay and nonspecific binding was determined in the presence of 5 µM mazindol. B, equilibrium saturation binding of [125I]RTI-55 to human caudate and rat striatal membrane preparations. Data shown are the mean ± S.E.M. of four independent experiments each conducted with duplicate determinations. Inset, representative Scatchard replot of an individual experiment for membrane preparations from human and rat. Bound = pmol/mg of protein, B/F = pmol/(mg of protein · nM).

Uptake. In preliminary experiments, we optimized assay conditions for [³H]DA uptake by human caudate synaptosomes. Specific [³H]DA uptake by synaptosomes prepared from cryopreserved human caudate was linear for up to 20 min (data not shown) and all assays were conducted using a 5-min incubation at 22°C. Uptake of [³H]DA increased linearly from 25 to 130 µg of protein per assay. For all subsequent assays, 128 ± 4 µg of protein (from an original wet weight of approximately 7 mg resuspended in 12-13 volumes of 0.32 M glucose) were used for all assays.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Effect of butaclamol on the concentration-response curve of DA in competition with [3H]DA uptake by synaptosomes prepared from human caudate. The dotted line indicates nonspecific uptake as defined by 5 µM mazindol. Addition of butaclamol prevented binding of DA and [3H]DA to D1-like and D2-like DA receptors. Data shown are the mean ± S.E.M. of four (butaclamol) or five (vehicle) experiments, each conducted with duplicate determinations. The uptake data in the presence of butaclamol were transformed to picomoles per milligram of protein as shown in the lower curve in Fig. 3. Inset, concentration-response curve for inhibition of [3H]DA uptake by butaclamol (n = 3).

View larger version (19K): [in this window] [in a new window]   Fig. 3.   [3H]DA uptake by synaptosomes prepared from nonfrozen and cryopreserved rat striatum and from cryopreserved human caudate. Data shown are the mean ± S.E.M. of four (cryopreserved rat), five (cryopreserved human), or eight (fresh rat) experiments, each conducted in duplicate. Note the break in the y-axis to allow presentation of data from rat and human preparations in the same graph. The Vmax and Km values are given in Table 1.

Drug Interactions with the DAT. Twenty-two compounds, including therapeutic and abused drugs, were tested for their potency at inhibiting [¹²⁵I]RTI-55 binding to membranes from cryopreserved human caudate and fresh rat striatum. There was a high degree of correlation for the effects of the drugs in both tissues (Table 2; Fig. 4A). There was also a high degree of correlation when K_i values in human caudate were compared with K_i values obtained using HEK-hDAT cell membranes (Eshleman et al., 1999) (Fig. 4B).

View larger version (39K): [in this window] [in a new window]   Fig. 4.   Correlations of drug potency for inhibition of [125I]RTI-55 binding between human caudate and rat striatum (A) and between human caudate and HEK-hDAT cells (B). The Ki values for inhibitory potency of drugs interacting with HEK-hDAT cell membranes were taken from Eshleman et al. (1999). Unity is indicated by the dotted line. A, human caudate and rat striatum. Analysis with Spearman's nonparametric correlation, which makes no assumption about the distribution of the values and bases calculations on rank, yielded a correlation coefficient r = 0.97, p < 0.0001. B, human caudate and HEK-hDAT cells. Analysis with Spearman's nonparametric correlation yielded a correlation coefficient r = 0.91, p < 0.0001.

View larger version (14K): [in this window] [in a new window]   Fig. 5.   Representative experiment for drug inhibition of [3H]DA uptake by human caudate synaptosomes. For this experiment, 250,000 cpm of [3H]DA were added to each well, [3H]DA uptake in the absence of drugs was 27,000 cpm, and nonspecific uptake was 2,150 cpm. Specific uptake ranged from 2,000 to 24,000 cpm (0.04-0.5 pmol) over the course of the uptake experiments, varying with the tissue samples.

View larger version (39K): [in this window] [in a new window]   Fig. 6.   Correlations of drug potency for inhibition of [3H]DA uptake between human caudate and rat striatum (A) and between human caudate and HEK-hDAT cells (B). The IC50 values for inhibitory potency of drugs interacting with HEK-hDAT cells were taken from Eshleman et al. (1999). Unity is indicated by the dotted line. A, human caudate and rat striatum. Analysis of the data with Spearman's nonparametric correlation yielded a correlation coefficient of r = 0.97, p < 0.0001. B, human caudate and HEK-hDAT cells. The experiments with HEK-hDAT cells were identical to those conducted with human caudate except that the incubation time was 10 min and experiments were conducted in triplicate. Analysis with Spearman's nonparametric correlation yielded a correlation coefficient r = 0.95, p < 0.0001.

View larger version (28K): [in this window] [in a new window]   Fig. 7.   Correlation of the potencies of drugs for inhibition of [3H]DA uptake and [125I]RTI-55 binding in human caudate preparations. Results are plotted as the logarithm of the IC50 value (Table 2) for inhibition of [3H]DA uptake versus the logarithm of the Ki value (Table 2) for inhibition of [125I]RTI-55 binding. Unity is indicated by the dotted line. Analysis of the data with Spearman's nonparametric correlation yielded a correlation coefficient of r = 0.64, p = 0.001.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

We have characterized the transport rate of [³H]DA, the potency of drugs at inhibiting [³H]DA uptake in human and rat synaptosomes, and the potency of drugs at, and the density of, binding sites for the cocaine analog [¹²⁵I]RTI-55. The affinity of [¹²⁵I]RTI-55 for native human and rat transporters was similar, but the density of binding sites was 3-fold less in autopsied, cryopreserved human caudate compared with the density of binding sites in fresh rat striatum. The affinity of DA was similar for the DAT in human caudate and rat striatum, while the maximal rate of uptake in preparations from frozen human brain was 0.5 to 7% of the rate of uptake in rat striatal preparations. The potencies of a wide range of compounds at inhibiting uptake by the rat and human brain preparations and the recombinant hDAT were strikingly similar. However, the neurotransmitters, DA, norepinephrine, and serotonin were at least 5- to 10-fold less potent at inhibition of [³H]DA uptake in HEK-hDAT cells compared with both human and rat preparations. Other substrates did not show this differential potency.

The use of cryopreserved human caudate for biochemical uptake and binding assays allows the verification of drug interactions with model systems, such as heterologously expressed hDAT and rat striatal preparations. However, the effects of post-mortem delays until freezing and cryopreservation may introduce artifacts into the system. Using rat striatal preparations, the changes in K_m and V_max values for DA transport that are induced by post-mortem autolysis at 22°C have been modeled by Haberland and Hetey (1987) as an exponential degradation with time delays from euthanasia until assay of 12 to 24 h. At 24 h post mortem, they observed a significant decrease in V_max with little change in the affinity for DA. Similarly, we observed that cryopreservation of rat striatum caused a 50% decrease in maximal uptake rate (Table 1).

Although similar K_m values were measured for DA uptake in cryopreserved human caudate synaptosomes and fresh or cryopreserved rat striatal preparations, there was a 15- to 100-fold difference in V_max values. In contrast, there was only a 3-fold lower density of DAT in human caudate compared with rat striatal preparations. If the density of DAT in human caudate is one-third of the density in rat striatum in vivo, then either the turnover rate of the DAT in the human tissue is lower or the lower maximal transport rate in the human preparation is due to the process of autolysis during autopsy. These data, which reveal a much lower rate of uptake in the human preparation than would be predicted by the ratio of uptake/binding in rat striatal preparations, suggest that many transporters in cryopreserved human tissue are inactivated and that membrane integrity is decreased so that fewer synaptosomes are formed, or that synaptosomes are "leaky" and do not retain transported DA. However, the remaining active transporters in human tissue function similarly to those in freshly prepared rat tissue, as suggested by the comparable K_m values for DA and inhibitory potencies of a structurally diverse set of compounds (Table 2).

Addition of (+)-butaclamol in the kinetic analysis of DA uptake minimized binding of DA to both DA D₁-like and D₂-like receptors (Neve and Neve, 1997). When the receptors are in their high-affinity states, the K_d for DA is in the low nanomolar concentration range (Neve and Neve, 1997). In human caudate, there is uneven distribution of D₁, D₂, and D₃ receptors and DAT. D₁ receptor density displays a rostral-to-caudal declining gradient in the putamen, D₂ receptor density increases from rostral to caudal in the caudate and putamen, and D₃ receptors are concentrated in the ventral striatum. DA uptake sites increase in a rostral-to-caudal gradient in the caudate (Joyce et al., 1991; Piggott et al., 1999). There is also a differential distribution of these proteins in regard to striosome-matrix compartmentalization. Since the samples of caudate used for uptake assays may be from different subregions of the caudate, the rate of [³H]DA uptake and the inhibitory effect of (+)-butaclamol on [³H]DA binding to receptors are variable.

The current results, which indicated a single binding site for [¹²⁵I]RTI-55 in human caudate preparations, are not in agreement with other reports that indicated two affinity states for [¹²⁵I]RTI-55 binding (K_d values = 0.1 and 1.8 nM, Staley et al., 1994a; K_d values = 0.066 and 1.52 nM, Little et al., 1993; and K_d values = 0.21 and 0.76 nM, Rothman et al., 1994). In rat striatal preparations, both one-site (Wall et al., 1993) and two-site (Boja et al., 1992) binding models have been observed for [¹²⁵I]RTI-55. The affinity of [¹²⁵I]RTI-55 binding for human caudate was similar to the lower affinity site in each of these studies. The different buffering systems used by Staley et al. (1994a, sucrose phosphate), Rothman et al. (1994, sodium phosphate), and Little et al. (1993, Tris/NaCl) compared with the buffer described above (Krebs-HEPES) may account for the differences. Staley et al. (1995), using the cocaine analog RTI-121, observed a decrease in the density of a lower affinity binding site when using a high-salt buffer compared with sucrose phosphate buffer. Buffer-dependent changes may thus be due in part to an ion (Na⁺) binding motif that could allosterically regulate the residues important for ligand binding (McElvain and Schenk, 1992). Our results differ from those of Staley et al. (1995) in that the density of the high-affinity, subnanomolar site is so low as to be undetectable under the current assay conditions. It is also possible that one of the binding sites for [¹²⁵I]RTI-55 measured by others is the serotonin transporter. In our preparation, a concentration of fluoxetine that saturated the serotonin transporter had no effect on the measured density of binding sites, indicating that most of the binding sites were the DAT. However, many of the Hill slopes for the binding competition data in Table 2 are less than 1; these shallower displacement curves may be due to ligand binding to the serotonin transporter. Alternatively, the lower Hill slopes may indicate binding to multiple sites on the DAT.

Although there were differences in apparent affinity states for RTI-55 in these preparations, there was excellent correlation between the potencies of seven drugs (Staley et al., 1994a; Spearman's nonparametric test, r = 0.96, p = 0.003) or eight drugs (Little et al., 1993; Spearman's nonparametric r = 0.90; p = 0.005) at inhibiting [¹²⁵I]RTI-55 binding compared with the data presented here. Furthermore, the excellent correlation between inhibition of [¹²⁵I]RTI-55 binding to rat and human membrane preparations is consistent with the results reported by others who used [³H]2-carbomethoxy-3-(4-fluorophenyl)tropane as radioligand (Kuhar et al., 1999).

The three biogenic amine neurotransmitters were at least an order of magnitude more potent at inhibition of [³H]DA uptake by the native hDAT, and 5-fold more potent at inhibition of [³H]DA uptake by the rat DAT (Table 1; Fig. 6) than they were at inhibiting uptake by the HEK-hDAT cells (Eshleman et al., 1999). The absence of the vesicular monoamine transporter in the HEK cells, which may lead to increases in the intracellular concentration of free neurotransmitter following uptake, may be one of the reasons for this difference in potency. Glycosylation differences between native and recombinant transporters may also play a role in the neurotransmitter potency differences (Patel et al., 1993; Vaughan et al., 1996). Other substrates did not show this differential potency. Possible differences in amino acid sequence and carbohydrate moieties may have less impact on uptake compared with effect on release (Johnson et al., 1998), suggesting that these measures of transporter function are dependent on different sites.

If the criteria for a possible therapeutic for cocaine abuse includes the ability to inhibit cocaine analog binding while permitting function of the DAT, then none of the compounds tested with the native human transporter qualify as pharmacotherapies (Table 2). Only three drugs were more potent at blocking binding than they were at blocking uptake: desipramine (3.2-fold), GBR-12935 (1.9-fold), and nortriptyline (2.2-fold). GBR-12909 and other analogs of GBR-12935 do not maintain self-administration under some conditions (Glowa et al., 1996; Tella et al., 1996), photoaffinity analogs of GBR 12935 and cocaine bind to different peptide fragments of DAT (Vaughan, 1995), and GBR-12909 does not show stimulant effects in clinical trials (Sogaard et al., 1990). Although the roles of the DAT and of cocaine-induced increases in extracellular dopamine concentrations following blockade of the DAT in addiction continue to be debated (Rothman and Glowa, 1995; Rocha et al., 1998; Spanagel and Weiss, 1999), the use of cryopreserved human tissue or heterologously expressed DAT can help elucidate the initial events of drug interactions. The high degree of correlation between drug IC₅₀ values in cryopreserved human tissue and HEK-hDAT cells further validates the use of HEK 293 cells expressing the recombinant hDAT as a model for studying the interactions of drugs with the transporter. Transport and binding site characteristics suggest that the rat DAT and recombinant human DAT are good models for studying drug interactions with the DAT in human brain.