Address correspondence and reprint requests to Dr Joshua M. Gulley, Department of Pharmacology, University of Colorado Health Sciences Center, Campus Box C-236, 4200 E Ninth Avenue, Denver, CO 80262, USA. E-mail: email@example.com
In heterologous expression systems, dopamine transporter (DAT) cell-surface localization is reduced after relatively prolonged exposure to d-amphetamine (AMPH) or dopamine (DA), suggesting a role for substrate-mediated regulation of transporter function. Here, we investigated whether brief, repeated periods of substrate exposure modulated transporter function, first, in an in vitro model system and, second, in intact rat brain. In human DAT-expressing Xenopus laevis oocytes, repeated exposure to low micromolar concentrations of DA, AMPH or tyramine markedly reduced transport-mediated currents. This functional down-regulation was attenuated by inclusion of a protein kinase C (PKC) inhibitor and probably reflects DAT redistribution, as cell-surface [3H]WIN 35 428 binding was significantly lower following DA exposure. High-speed chronoamperometry was used to measure clearance of exogenously applied DA in dorsal striatum (STR) and nucleus accumbens (NAc) of anesthetized rats. In STR, frequent (every 2 min) applications of DA altered DA clearance parameters in a manner consistent with profound down-regulation of DAT function. Similar changes were not observed in NAc or after repeated vehicle (ascorbic acid) application. Together, our results suggest that brief, repeated periods of substrate exposure lead to rapid down-regulation of DAT activity and that this type of regulation can occur in vivo in STR, but not NAc.
Regulation of DAT function has also been shown to result from exposure to transporter substrates, most notably AMPH, methamphetamine (METH), and DA. In human embryonic kidney (HEK) 293 cells expressing fluorescent epitope-tagged (FLAG) hDAT, prolonged exposure to AMPH (2 µm, 60 min) reduces [3H]DA uptake and transport-associated currents (Saunders et al. 2000). In these studies, confocal microscopy indicated this functional inhibition was due to trafficking of DAT away from the cell surface. Preincubation of rat striatal synaptosomes with a wide range of METH concentrations, including 10 µm METH for as little as 5 min, reduces DAT Vmax, and this effect is not due to residual METH and can be blocked by the PKC inhibitor NPC15437 (Kim et al. 2000; Sandoval et al. 2001). Acute systemic administration of METH also results in reduced uptake of [3H]DA into striatal synaptosomes, which appears to be unrelated to METH-induced hyperthermia or neurotoxicity (Metzger et al. 2000). However, reports of DA-mediated regulation of DAT function have been less consistent. Similar to AMPH, in HEK 293 cells expressing FLAG-tagged hDAT, 10 µm DA (60-min exposure) reduces subsequent uptake of [3H]DA, and 100 µm DA increases localization of DAT in the cytosol versus the cell membrane (Saunders et al. 2000). In contrast, however, exposure to 10 µm DA for 40 min failed to produce DAT internalization in Madin–Darby canine kidney (MDCK) cells expressing green fluorescent protein (GFP)-tagged hDAT (Daniels and Amara 1999). Thus, it appears that AMPH, METH, and perhaps DA can down-regulate DAT function. However, with the exception of METH, these effects have been demonstrated only in vitro. Furthermore, the extent to which functional changes occur likely depends on the system investigated.
Previous studies from our laboratory have indicated that, in urethane-anesthetized rats, local applications of DA (200 µm; 25–150 nL) into STR or NAc at 5-min intervals for up to 60 min produce electrochemical signals with relatively rapid decays that are highly reproducible and sensitive to DAT inhibitors such as cocaine and nomifensine (Cass et al. 1993b; Zahniser et al. 1999). This observation, along with those in the literature discussed above, made us question whether DA-induced down-regulation of DAT occurs in response to relatively brief periods of substrate exposure and whether this type of regulation occurs in vivo. Thus, our first aim was to investigate if brief, repeated exposure to a variety of substrates would produce functional alterations in a model system. If so, we wanted to determine if this down-regulation was PKC-dependent and correlated with a reduction in cell-surface expression of DAT. Our second aim was to determine the extent to which DA-induced changes in DAT function could be observed in vivo and if substrate-induced regulation was similar in different DA neuronal projection areas. We addressed this aim by using high-speed chronoamperometry to measure DA clearance following local application of DA at several different time intervals and in two different brain regions.
Materials and methods
For in vitro electrophysiological experiments, we used oocytes surgically removed from female Xenopus laevis frogs that were purchased from Xenopus I (Ann Arbor, MI, USA). Oocytes were defolliculated as previously described (Doolen and Zahniser 2001), with stage V and VI oocytes selected for injection with hDAT cRNA (see below). For in vivo electrochemical experiments, 45 male Sprague–Dawley rats (225–465 g), obtained from Charles Rivers Laboratory (Sasco, Omaha, NE, USA), were tested. Rats were housed up to three per cage with a 12 h/12 h light–dark cycle and ad libitum food and water. All animal use procedures were in strict accordance with the NIH Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee at the University of Colorado Health Sciences Center.
hDAT cRNA preparation and oocyte expression
Capped cRNA was transcribed from a linear oocyte expression vector pOTV containing the 1.9-kb hDAT cDNA insert (Sonders et al. 1997) using mMessage mMachine with T7 polymerase (Ambion, Austin, TX, USA). Oocytes were injected with water-diluted cRNA (∼10 ng) and maintained at room temperature for 3–5 days in frog Ringer's buffer (FRB; 96 mm NaCl, 2 mm KCl, 1.8 mm CaCl2, 1 mm MgCl2 and 5 mm HEPES, pH 7.5) supplemented with 2.5 mm sodium pyruvate, 0.5 mm theophylline, 100 U/mL penicillin, 100 µg/mL streptomycin and 50 µg/mL gentamycin.
Two-electrode voltage clamp recording
Currents were measured with two-electrode voltage clamp using glass microelectrodes filled with 3 m KCl (Sonders et al. 1997). Analog signals were passed through a Warner OC-725B amplifier (Warner Instruments, Hamden, CT, USA) and a DigiData 1200 interface, where currents were low-pass filtered (100 Hz) and digitized (at 2048 Hz). Stimulation parameter control and the acquisition and analysis of data were provided by pClamp6 (Axon Instruments, Foster City, CA, USA) and MacLab using a MacLab/2e interface (AD Instruments, Castle Hill, Australia).
Individual oocytes were superfused with FRB at room temperature in a 0.5-mL recording chamber at 2 mL/min. Oocytes were initially clamped at −60 mV and subsequently given a series of 10 mV, 400 ms jumps in membrane potential that ranged from −120 to + 40 mV. However, here we present only the data from voltage jumps between −120 and −20 mV because currents recorded at these potentials reflect primarily DAT-mediated transport, whereas currents between −10 and +40 mV reflect primarily constitutive leak currents (Sonders et al. 1997). Currents were recorded during the last 20 s of each 1-min superfusion with FRB containing one of three DAT substrates: 3 or 10 µm DA, 10 µm tyramine, or 2 µm AMPH. These concentrations were chosen in order to compare the results to previous studies where transport-associated currents have been characterized (Sonders et al. 1997; Doolen and Zahniser 2001). Steady-state currents were measured and averaged during the last 100 ms of each voltage jump. Substrate-induced inward currents were measured by performing off-line subtraction of currents in the absence of drug from currents measured with substrate present (Idrug–Ibuffer). While oocytes were exposed to substrates for 1 min every 5 min, currents were recorded every 10 min. In experiments where oocytes were exposed to substrate and the PKC inhibitor bisindolylmaleimide I (BIM), 1 min superfusion with DA or AMPH was followed by 1 min superfusion with BIM (1 µm). This BIM concentration has been shown to reverse phorbol ester-induced decreases in [3H]DA uptake in hDAT-expressing oocytes (Zhu et al. 1997). Control experiments using an inactive form of this PKC inhibitor, bisindolylmaleimide V (BIM V; 1 µm), were performed in the same manner.
Radioligand binding to hDAT
Binding to hDAT was quantified in intact oocytes at room temperature in the 0.5 mL chamber by 20-min superfusion (2 mL/min) with FRB that contained an approximate half-maximal concentration of 2β-carbomethoxy-3β-(4-fluorophenyl)[3H]tropane ([3H]WIN 35 428; specific activity 84.5 Ci/mmol; 4 nm). Non-specific binding was determined in the presence of 1 mm cocaine. Following superfusion, oocytes were quickly washed three times in ice-cold FRB and dissolved in 0.25 mL of 2% sodium dodecyl sulfate. Radioactivity was measured by liquid scintillation spectroscopy.
In vivo electrochemistry
Recording electrodes (30-µm diameter carbon fiber) were coated with 5% Nafion solution to reduce attraction of interferent ions such as ascorbic acid (AA) and 3,4-dihydroxyphenylacetic acid (Gerhardt et al. 1984; Gerhardt and Hoffman 2001) and calibrated in vitro for their response to AA (250 µm) and increasing concentrations of DA (2–8 µm or 2–12 µm, in 2 µm increments; Zahniser et al. 1998). Only those exhibiting a high selectivity of DA over AA (≥ 1000 : 1) and a linear DA response (r2 ≥ 0.997) were used. After calibration, sticky wax was used to attach electrodes to a single- or double-barrelled pipette (tip opening: 10–20 µm) such that tips were parallel and separated by 150–300 µm. Single-barrel pipettes and one barrel of double-barrelled pipettes were filled with 200 µm DA and 100 µm AA (in saline or 0.1 m phosphate buffered saline, pH 7.4). The second barrel of the double-barrel pipettes was filled with 100 µm AA.
Rats were anesthetized (1.5 g/kg urethane, intraperitoneally) and placed in a stereotaxic frame, with body temperature maintained at 37°C via a heating pad coupled to a rectal thermometer. Holes were drilled in the left and right side of the skull overlying STR and NAc (0.7–1.5 mm anterior and 1.5–2.2 mm lateral to bregma; Paxinos and Watson 1998), with an additional hole drilled just anterior to the interaural line for insertion and attachment (via dental cement) of a Ag/AgCl reference electrode. An opening was made in the dura for stereotaxic insertion of the recording assembly; the skull and exposed brain surface were bathed in saline for the duration of the experiment.
High-speed chronoamperometric measurements were made using an IVEC-10/FAST-12 system (Quanteon, LLC, Lexington, KY, USA), which applied square-wave pulses of 0.00–0.55 V (with respect to reference) at a frequency of 5 Hz. Resulting oxidation and reduction currents were digitally integrated during the last 80 ms of each 100-ms pulse, and signal changes were converted to DA concentrations based on the in vitro calibration (Gerhardt et al. 1984; Zahniser et al. 1998). Electrode/pipette assemblies were lowered into STR or NAc (4.0–5.0 mm and 6.5–8.0 mm ventral to the skull, respectively; Paxinos and Watson 1998) and a stable background current was established and set to zero prior to pressure-ejecting DA (5–20 psi for 0.1–5 s) at calibrated volumes (10–200 nL). For a given assembly and recording location, an ejection volume was chosen that resulted in signal amplitudes of 0.5–2.75 µm in STR and 1.0–6.75 µm in NAc. Volumes were monitored for consistency during each ejection (Friedemann and Gerhardt 1992). Ejections were made at 5-min intervals, and signals used to determine ‘baseline’, were established when signal amplitude and clearance time varied by ≤ 15% for three consecutive applications. Before baselines were established, the mean (± SEM) number of ejections and volume per ejection at each recording location was 2.59 ± 0.21 and 66.5 ± 6.40 nL, respectively.
Subsequent DA ejections were made for 60 or 90 min at constant intervals (CI; every 5 min throughout the entire experiment; n = 12 recordings) or variable intervals (VI; every 2, 3, 5 or 10 min for 30-min periods; n = 21 recordings). In separate experiments using double-barrel pipettes (n = 3 recordings), a sequence of ejections of AA at 2- and 4-min time points and DA at 6-min time points was delivered repeatedly within a 30-min period, resulting in a total of 10 AA and 5 DA applications. The volume, pressure and delivery time of AA ejections was identical to those of the corresponding DA ejections. To test whether chronoamperometric voltage pulses enhanced the formation of DA oxidation products, CI (5 min) and VI (2, 5 or 10 min) ejections of DA (n = 6 recordings) were given in the absence of applied voltage pulses except during periods before (90 s) and after (60 s) signal measurement. In the CI group, signals were measured after every 5-min interval DA ejection. In the VI group, they were measured after every DA ejection at 5- and 10-min intervals, but only after every third ejection at 2-min intervals.
In experiments in which data were obtained from the same animal (n = 16 recordings from eight rats), recording locations were separated by ≥ 500 µm or were made on the opposite side of the brain. After experiments were completed, a small current was passed through the recording electrode to produce a marking lesion. The brain was then removed and stored in buffered formalin (4% w/v) for at least 1 week. Subsequently, coronal sections (40 µm) of STR and NAc were made using a vibratome, mounted to glass slides, and stained with cresyl violet to localize recording sites.
Chemicals were purchased from Sigma (St. Louis, MO, USA) or Fisher (Pittsburgh, PA, USA), with the following exceptions: BIM and BIM V were purchased from Calbiochem (La Jolla, CA, USA), [3H]WIN 35 428 was purchased from NEN Life Science Products (Boston, MA, USA), and Nafion (5% solution) was purchased from Aldrich (Milwaukee, WI, USA).
Data and statistical analysis
Data are presented as mean ± SEM. In the oocyte experiments, changes in currents measured at −120 and −20 mV were evaluated with repeated measures anova followed by Tukey post-hoc tests comparing 0, 30 and 60 min time points. These two voltages were chosen because they represented the most hyperpolarized and depolarized potentials, respectively, of the voltage-jump range that measured transport-associated currents (Sonders et al. 1997). Group differences in specific [3H]WIN 35 428 binding between untreated and substrate-exposed oocytes were analyzed with unpaired Student's t-test.
For in vivo electrochemistry, two signal parameters were analyzed from DA oxidation currents: maximal amplitude (Amax) and signal time course (T80). Amax reflects the maximal extracellular DA concentration detected, whereas T80 is the time for the signal to rise to Amax and decay by 80%. Both parameters are affected by DAT inhibitors (e.g. Zahniser et al. 1999) and reflect changes in DA clearance (Schmitz et al. 2001; May et al. 1988; Wightman et al. 1988; Ng et al. 1992; Cass et al. 1993b; Nicholson 1995; Suaud-Chagny et al. 1995). Data were normalized within each recording by obtaining a mean value for parameters during the three-point baseline period, setting this value as 100%, and expressing all data (including the three points composing the baseline) as a percentage of baseline. The statistical significance between CI and VI groups was determined using unpaired Student's t-test to compare data from the last VI time point with that from the CI group at the same time point.
In vitro DAT function
First, we tested the extent to which brief, repeated exposure to different DAT substrates could alter transporter function in hDAT-expressing oocytes. After 1-min exposures to 3 or 10 µm DA, 10 µm tyramine, or 2 µm AMPH once every 5 min, we observed a gradual, but consistent, decline over time in the magnitude of transport-associated currents. These currents were recorded at 10-min intervals, but only the 30-min (seven exposures) and 60-min (13 exposures) results are summarized in Fig. 1. Furthermore, since substrates do not induce inward currents in control (water-injected) oocytes, substrate-induced regulation was tested only in hDAT-expressing oocytes.
With DA, the inhibition of DAT function appeared to be partially related to concentration because the changes produced by 3 µm DA (Fig. 1a) were not evident as quickly as those produced by 10 µm DA (Fig. 1b). At 30 min, there was a trend for 10 µm DA to decrease currents by 25 and 21%, compared with baseline at both −120 and −20 mV, respectively. These voltages are the extremes of those used here to measure DAT-mediated transport currents. At 60 min, this concentration of DA significantly reduced currents at −120 and −20 mV by 65 and 58% compared with baseline, respectively. At this same time point, 3 µm DA produced a larger, but considerably more variable, effect on currents, reducing them by 72 and 95% compared with baseline at −120 and −20 mV, respectively. DA-induced reductions in current were not reversed upon cessation of exposure to either of the two concentrations or after extensive washing (data not shown). In addition, a single exposure to 10 µm DA followed by a 60-min wash period did not significantly alter currents compared with baseline, demonstrating that DA-induced currents are stable over a 60-min period in the absence of repeated DA exposure (data not shown).
Reductions in current magnitude were also seen after exposure to tyramine and AMPH. Of the three substrates tested, AMPH produced a significant decrease after the fewest repeated exposures. When oocytes were exposed to 10 µm tyramine for 1 min once every 5 min for 60 min, transport-associated currents were significantly reduced by 39% compared with baseline at −120 mV (Fig. 1c). Similar to 3 µm DA, currents recorded at 30 min after repeated tyramine exposure were not different from baseline. In contrast, seven exposures to 2 µm AMPH significantly reduced currents by 31 and 45% compared with baseline at −120 and −20 mV, respectively (Fig. 1d). After 13 exposures to AMPH, currents measured at −120 mV were further decreased by 70% compared with baseline, whereas those measured at −20 mV were reduced by 75% compared with baseline.
To determine if substrate-induced inhibition of DAT function involved a PKC-dependent mechanism, successive DA- or AMPH-induced transport currents were measured in hDAT-expressing oocytes superfused with the PKC inhibitor BIM (1 µm; Fig. 2). When BIM was present, 10 µm DA exposure reduced current magnitude to a lesser extent than DA exposure alone (Fig. 2a). This effect was most pronounced at −120 mV; current reduction measured at −20 mV after repeated DA exposure was similar whether in the absence or presence of BIM. However, similar voltage-dependency was not seen for the effect of BIM on repeated 2 µm AMPH exposure, and the BIM inhibition of AMPH-induced regulation was more complete (Fig. 2b). Thus, after 13 exposures to AMPH along with BIM pretreatment, transport-associated currents did not differ from baseline. Pretreatment with BIM V (1 µm), which does not block PKC, had no effect on the functional down-regulation produced by 2 µm AMPH exposure (data not shown).
Radioligand binding assays were performed to determine the extent to which changes in cell-surface expression of DAT may underlie the observed substrate-mediated reduction in transport-associated currents. Specific [3H]WIN 35 428 binding is dependent on the presence of relatively high Na+ concentrations (Reith and Coffey 1993; Doolen and Zahniser, unpublished observations), and the Na+ concentration inside oocytes is low (∼6 mm; Barish 1983) compared with the extracellular superfusion buffer (96 mm). Thus, specific [3H]WIN 35 428 binding to intact oocytes measures hDATs present on the cell surface. As shown in Fig. 3, specific binding was reduced significantly by 37% in intact hDAT-expressing oocytes after 40 min of repeated exposure to DA (10 µm), compared with those treated only with FRB. The reduction in binding was similar to the reduction in DAT currents (∼ 34 and 37% at −120 and −20 mV, respectively) produced by this length of exposure to 10 µm DA.
In vivo DAT function
In the second series of experiments, we tested whether brief, repeated exposure to DA could alter in vivo DA clearance, a measure of DAT activity in brain. We obtained a total of 18 electrochemical recordings from the STR of 16 rats and 15 recordings from the NAc of 13 rats. All electrode positions were verified by histological reconstruction. These reconstructions indicated that striatal recordings were made primarily in the dorsomedial quadrant of the STR and accumbal recordings were made mostly in the core, rather than shell, region.
The mean values over the 15-min baseline period for Amax, T80 and DA ejection volume for the various groups are compared in Table 1. By chance, mean regional baseline Amax and T80 values were generally lower for the CI groups, but non-parametric analysis (i.e. Mann–Whitney test) indicated that these differences were not statistically significant. However, the nearly three- and twofold higher Amax and T80 values, respectively, between baseline STR and NAc recordings were significantly different. These regional differences are consistent with other reports (Cass et al. 1992; Jones et al. 1995; David et al. 1998) and likely reflect the higher number of DATs in STR, compared with NAc.
Table 1. Baseline electrochemical signal parameters in rats with recording electrodes in either STR or NAc subsequently used in CI or VI DA clearance experiments
Brain region Ejection interval
p < 0.05 compared with respective CI and VI groups from STR.
We have shown previously (Cass et al. 1993b; Zahniser et al. 1999) that applications of identical amounts of DA pressure-ejected into the STR at 5-min intervals result in electrochemical signals with Amax and T80 values that are relatively stable over time. Here, we observed similar results (Fig. 4a), with Amax and T80 values decreasing ∼25% and ∼3% below baseline, respectively, by the end of the 105-min recording period (Fig. 5a). In contrast, when the time interval between ejections was reduced to 2 min, there was a steady rise in both Amax and T80 values, consistent with a reduction in DA clearance (Fig. 4b). The changes in signal parameters in the STR reached an asymptote after 24 min of ejections (t = 39 min). Amax and T80 values increased by 156 and 44% compared with baseline, respectively, at the final 2-min interval ejection (t = 45 min; Fig. 5a). These values were significantly different from those measured at the same time point in the 5-min CI recordings, where Amax and T80 were reduced by 12 and 4% compared with baseline, respectively. The consistency of this result is evident by the fact that in all cases (n = 6), Amax and T80 were increased by a period of 2-min interval DA ejections. Additionally, in all cases each electrochemical signal decayed to its initial level before the initiation of a subsequent DA ejection. The increases in Amax and T80, which are indicative of reduced DAT function (Wightman et al. 1988; Ng et al. 1992; Cass et al. 1993b; Suaud-Chagny et al. 1995), were reversed during subsequent periods of DA ejections at 5- and 10-min intervals (Fig. 5a). When the order of the 2-, 5- and 10-min interval ejections was changed such that 30 min of 5-min interval ejections followed baseline recording, DA signal parameters measured in the STR remained stable initially and increased significantly only after the between-ejection interval was decreased to 2 min (Fig. 5b). Thus, the results showing increased DA signal parameters with 2-min, but not 5-min, ejection intervals did not depend upon the order of testing.
In contrast to STR, varying the time interval between DA ejections in NAc failed to significantly increase Amax or T80 values (Fig. 6). After 30 min of DA ejections every 2 min, Amax and T80 values were increased by 27 and 18% compared with baseline, respectively, for the VI group and by 5 and 2% compared with baseline, respectively, for the CI group. By the end of the recording period (t = 105 min), Amax in the VI group had decreased by 34% while Amax in the CI group had increased by 5%, compared with baseline. This trend in the VI group towards decreases in signal parameters was also observed in T80 values, but in neither case were these reductions statistically significant. As already noted in Table 1, high baseline Amax values (up to 7.00 µm) were achieved in NAc when DA ejection volumes similar to those used in STR recordings were used. When Amax values in NAc were matched to those in STR (< 2.75 µm) by using lower ejection volumes (10–25 nL), we still found no significant effect of 2-min interval DA ejections on either Amax or T80 in NAc (data not shown).
We next performed several experiments to confirm the specificity of the effect of the repeated 2-min DA ejections in STR. In the first of these, 100 µm AA (DA vehicle) was ejected at two of every three 2-min intervals and DA was ejected at the third interval. Thus, the resulting electrochemical signal was measured every 6 min This was done to control for potential changes in striatal tissue at the recording site due to the more frequent fluid applications and potential influences due to the more frequent applications of AA. As shown in Fig. 7(a), repeated AA had no effect on DA signal Amax or T80 values. Furthermore, AA produced < 5% change in the background electrochemical signal (data not shown). This observation confirmed the in vivo integrity of the Nafion coating previously applied to the carbon fiber electrode. Subsequent DA applications every 2 min led to increases in Amax and T80 that were 143 and 47% above baseline, respectively, after the last 2-min interval ejection (Fig. 7a). These changes were significantly different from those observed for the CI group at the same time point. Signal parameters again returned toward baseline during subsequent DA ejections at 5-min intervals.
Next, we delivered VI DA ejections in the 2-, 5- and 10-min order as before, but with chronoamperometric voltage pulses turned off except during the period encompassing electrochemical signal measurement. We did this to address the role in the observed DA-induced regulation of quinone formation, which is potentially enhanced by our more frequent electrochemical measurements. These DA oxidation products have been shown to inhibit DAT function (Berman et al. 1996; Whitehead et al. 2001). When DA ejections were given at a CI of 5 min for the duration of the experiment, signal parameters remained consistent, similar to when chronoamperometric voltage pulses were delivered without interruption at 5 Hz. However, applications of DA every 2 min increased Amax and T80 (Fig. 7b). After the final 2-min DA ejection, these values were increased by 267 and 66% compared with baseline, respectively. These values were significantly different from those observed in the CI group at the same time point, where Amax was increased by 7% and T80 was reduced by 4%, compared with baseline.
Because we observed an effect of repeated DA ejections in the STR when delivery intervals were 2 min, but not 5 min, we investigated the effect of applying DA at an intermediate-time interval. After the baseline-recording period, DA was ejected at 3-min intervals for 30 min, followed by 5-min intervals for 30 min (Fig. 8). When data were summarized in CI and VI groups, there was little overall effect of DA ejected at 3-min intervals on either Amax or T80 (Fig. 8a). However, inspection of individual recordings demonstrated a bimodal distribution. After the last 3-min ejection, Amax and T80 were significantly increased by 138 and 23% compared with baseline, respectively, in three of nine recordings (Fig. 8b). In the remaining six recordings, Amax and T80 were not different from baseline at this same time point (Fig. 8b). Thus, DA application at 3-min intervals appeared to represent the threshold for DA to induce down-regulation of DAT function in STR.
Our results provide evidence for down-regulation of DAT function following brief, repeated periods of substrate exposure both in vitro and in vivo. In hDAT-expressing oocytes repeated, but short-term, exposure to three different DAT substrates led to marked down-regulation of DAT function. These changes were dependent on the number of substrate exposures and were not seen in oocytes recorded over the same time period but not exposed repeatedly to substrates. Furthermore, the functional down-regulation produced by the repeated exposure to either DA or AMPH in oocytes was dependent on PKC activation and appeared to be due at least partially to reduction in cell-surface expression of DAT. In rat STR following repeated local applications of DA at 2-min, but not 5-min, intervals, we observed robust changes in in vivo DA clearance also consistent with DAT inhibition or down-regulation. Interestingly, this DA-induced DAT regulation was observed in STR, but not NAc.
Substrate-mediated inhibitions of transport-associated currents, which were most consistent after 60-min of repeated AMPH exposure, had several distinct characteristics. For example, they were common to the three different DAT substrates tested – DA, tyramine and AMPH – at moderate to high concentrations. The higher concentration of DA (10 µm) and the concentration of AMPH (2 µm) used here have previously been shown to reduce [3H]DA uptake by 20–40%, reduce AMPH-induced currents, and decrease cell-surface expression in HEK 293 cells stably transfected with FLAG-hDAT (Saunders et al. 2000). However, exposure to DA or AMPH was continuous for 60 min. Furthermore, in MDCK cells expressing GFP-hDAT, 40 min exposure to 10 µm DA failed to produce changes in DAT function or cellular localization, whereas phorbol ester treatment altered both measures (Daniels and Amara 1999). The discrepancy between the results of Saunders et al. (2000) and our experiments and those of Daniels and Amara (1999) may be due in part to the cellular context within which DAT was studied, different levels of DAT cell surface expression, and/or the particular hDAT construct employed.
Another characteristic of substrate-induced inhibition of hDAT function in our study was its dependence on PKC activation. This was demonstrated for DA and AMPH, as coexposure to the PKC inhibitor BIM, but not its inactive form BIM V, attenuated substrate-induced changes in transport-mediated currents. Likewise, METH-induced reductions in rat synaptosomal DAT activity are attenuated by the PKC inhibitor NPC15437 (Sandoval et al. 2001). A number of studies have confirmed that PKC mediates the inhibition of DAT function produced by phorbol esters (for review see Zahniser and Doolen 2001). Furthermore, it has become increasingly clear that DAT down-regulation resulting from PKC activation is due to a reduction in the cell-surface expression of DAT (Pristupa et al. 1998; Daniels and Amara 1999; Melikian and Buckley 1999). In the present experiments, the ability of substrates to inhibit hDAT function in oocytes also appears to be due to altered trafficking of DAT, as we observed a decrease in the cell surface binding of [3H]WIN 35 428 after repeated exposure to DA. Concluding there is an absolute role for PKC activation in substrate-mediated DAT down-regulation should be done with caution, however, because BIM did not completely block the DA-induced loss of hDAT function. Furthermore, it is likely that PKC regulates multiple pathways that may in turn affect DAT function and trafficking.
While providing a less direct measure of DAT function than in vitro analyses, electrochemical detection of exogenous DA clearance in the intact rat brain offers a means to determine if the substrate-mediated changes in transporter function observed in model systems also occur in the native, in vivo situation. Delivering DA at 2-min intervals, we found that repeated DA exposure markedly increased Amax and T80 in the STR of anesthetized rats. Amax, which is indicative of local DA concentrations (Gerhardt et al. 1984; Zahniser et al. 1998), was increased up to 266% above baseline by the end of a 30-min period of DA ejections delivered every 2 min. T80 was increased up to 66% above baseline. Previously, similar changes in DA signal parameters have been shown to result from systemic injection or local infusion of DA uptake inhibitors such as mazindol, nomifensine, and cocaine (Stamford et al. 1984; Ng et al. 1992; Cass et al. 1993a; Hoffman and Gerhardt 1998; Zahniser et al. 1999). Detailed kinetic analyses have suggested that signal amplitude and decay time primarily reflect transporter activity (Stamford et al. 1984; May et al. 1988; Wightman et al. 1988; Ng et al. 1992; Cass et al. 1993b; Nicholson 1995; Suaud-Chagny et al. 1995; Wu et al. 2001). Diffusion of applied DA contributes to certain aspects of the signal parameters (Rice and Nicholson 1989; Wightman and Zimmerman 1990). However, its contribution, especially in STR where a large number of DATs is present, appears to be relatively minimal (Garris and Wightman 1995; Giros et al. 1996; Schmitz et al. 2001). Also, it is important to note that the contribution from diffusion would not be expected to change over the course of repeated DA applications. Thus, the effects on signal parameters we observed here likely reflect an inhibition of DAT function produced by repeated substrate exposure.
It is tempting to speculate that the functional down-regulation we observed in STR results from substrate-induced reduction in surface expression of DAT, such as we saw in our oocyte experiments and others have reported in different expression systems (Saunders et al. 2000) and synaptosomes prepared from rat striatum (Kim et al. 2000; Metzger et al. 2000). We did not test this possibility directly, and to our knowledge, a direct measure of in vivo, substrate-mediated transporter internalization has not been described. However, we performed several control experiments to confirm the specificity of repeated DA in producing the functional change. Specifically, the inability of repeated AA application to alter DAT activity suggests that neither the vehicle nor repeated pressure ejection of exogenous fluid influenced the amplitude or decay time of clearance signals. Furthermore, the observation that the effect of repeated DA exposure persists when chronoamperometric voltage pulses were utilized only briefly, suggests that the effect was not due to an enhanced buildup of reactive oxygen species formed by the oxidation of DA (i.e. quinones). This possibility is also not likely given the high extracellular concentration of endogenous AA in STR and the presence of AA in the DA solution we applied. Furthermore, the return of the signal parameters to baseline after the time interval between DA ejections was increased to 5 or 10 min would not be expected if repeated DA exposure was producing quinone-induced toxicity.
Although DA delivered in 2-min intervals led to robust changes in signal parameters measured in the STR, this same manipulation failed to alter those recorded in the NAc. In fact, 2-min recordings in NAc were quite stable, similar to those obtained in both NAc and STR when DA was delivered every 5 min (i.e. CI group). Using fast-scan cyclic voltammetry coupled with DA iontophoresis in awake, unrestrained rats, Kiyatkin et al. (2000) also found DA applications into NAc, when repeated every 1.5–2.5 min, had no effect on the stability of DA signal parameters over a 30-min test period. Furthermore, METH-induced decreases in [3H]DA uptake are observed in synaptosomes prepared from STR, but not NAc (Kokoshka et al. 1998). The regional specificity we observed for the effects of DA may be due in part to differences in DAT function and localization in the NAc versus the STR. For example, the present results and previous reports (May and Wightman 1989; Cass et al. 1992; McElvain and Schenk 1992; Garris and Wightman 1994; Wu et al. 2001) indicate the rate of uptake is slower in the NAc, which likely reflects a lower density of transporters compared with the STR (Marshall et al. 1990; Richfield 1991; Mennicken et al. 1992; Cass et al. 1993b). It may be the case that, because of the reduced capacity in NAc, it is important for DAT to be resistant to substrate-induced regulation. It remains to be established, however, why substrate-mediated down-regulation of function would be less likely to occur in NAc.
The importance of the 2-min substrate exposure time in our in vivo experiments is not known and may simply represent an arbitrary point at which we could easily observe functional changes. However, we did find a variable effect on signal parameters when DA was ejected at 3-min intervals. It therefore appears that 3-min DA ejection intervals are at or near a threshold for the regulatory effects to emerge in STR. It may be the case that DA ejections at 2 min, but not 3 or 5 min, consistently disrupt endogenous, DAT recycling processes. It is also noteworthy that exposure to DA every 5 min, which was without effect in STR and NAc, was optimal for functional down-regulation in hDAT-expressing oocytes. It is unclear whether this discrepancy relates specifically to kinetic differences between these effects or simply reflects differences in native versus model systems.
In summary, we have found evidence for substrate-induced changes in the function of DAT both in vitro and in vivo. In oocytes, the functional changes are associated with a decrease in plasma membrane localization of DAT. While technical challenges make it difficult to test for this mechanism in vivo, the effects of repeated DA exposure we observed in the STR of urethane-anesthetized rats are consistent with this hypothesis. A number of issues remain to be resolved. For example, it is not clear from our experiments whether substrate-induced down-regulation of DAT requires the translocation of substrate across the membrane or can be produced solely by binding to the transporter. Furthermore, the extent to which DA receptor mechanisms might contribute to changes in DAT function, especially in vivo, is not well understood. Electrochemical studies have suggested that activation of DA D2 receptors up-regulates DAT function (Meiergerd et al. 1993; Cass and Gerhardt 1994; Dickinson et al. 1999), an effect opposite to that of substrates. However, recent studies of DAT function in mesencephalic cultures have not supported this conclusion (Prasad and Amara 2001).
The physiological significance of substrate-induced down-regulation of function, which may seem to be a paradoxical response to persistent elevations in neurotransmitter levels, is unknown. Because the primary function of transporter proteins appears to be neurotransmitter clearance, it might be expected that repeated substrate treatment would lead to an up-regulation of function, such as has been seen with the GABA transporter GAT1 (Bernstein and Quick 1999). On the other hand, DA can be cytotoxic (Rabinovic et al. 2000), and down-regulation of DAT induced by relatively high extracellular concentrations of DA may represent a protective mechanism for DA neurons. Furthermore, exposure to the DAT inhibitor cocaine has been shown recently to produce the opposite effect: functional up-regulation and increased cell surface expression of DAT (Daws et al. 2002; Little et al. 2002).
This work was supported by the National Institutes of Health grants DA 04216, DA 14204, DA 15050 and AA 07464. We thank Dr Michael J. Zigmond for helpful discussions of this work.