Abbreviations used : aCSF, artificial CSF ; DA, dopamine ; DAT, dopamine transporter ; DOPAC, 3,4-dihydroxyphenylacetic acid ; HVA, homovanillic acid ; T80, time required for the dopamine signal to reach its maximum and to decay by 80% ; [3H] WIN 35,428, (-)2β-[3H]carbomethoxy-3β-(4-fluorophenyl)tropane or [3H]CFT.
Address correspondence and reprint requests to Dr. S. D. Dickinson at Department of Pharmacology, C236, University of Colorado Health Sciences Center, 4200 East Ninth Avenue, Denver, CO 80262, U.S.A.
Abstract : Presynaptic D2 dopamine (DA) autoreceptors, which are well known to modulate DA release, have recently been shown to regulate DA transporter (DAT) activity. To examine the effects of D2 DA receptor deficiency on DA release and DAT activity in dorsal striatum, we used mice genetically engineered to have two (D2+/+), one (D2+/-), or no (D2-/-) functional copies of the gene coding for the D2 DA receptor. In vivo microdialysis studies demonstrated that basal and K+-evoked extracellular DA concentrations were similar in all three genotypes. However, using in vivo electrochemistry, the D2-/- mice were found to have decreased DAT function, i.e., clearance of locally applied DA was decreased by 50% relative to that in D2+/+ mice. In D2+/+ mice, but not D2-/- mice, local application of the D2-like receptor antagonist raclopride increased DA signal amplitude, indicating decreased DA clearance. Binding assays with the cocaine analogue [3H]WIN 35,428 showed no genotypic differences in either density or affinity of DAT binding sites in striatum or substantia nigra, indicating that the differences seen in DAT activity were not a result of decreased DAT expression. These results further strengthen the idea that the D2 DA receptor subtype modulates activity of the striatal DAT.
Dopaminergic neurotransmission in the CNS plays a key role in the control of motor, cognitive, and reward processes. Dopaminergic neurotransmission is a complex and tightly controlled process that involves synthesis, storage, release, receptor binding, subsequent activation of signal transduction systems, and termination of the action of the neurotransmitter. Extracellular levels of dopamine (DA) within the striatum are thought to largely depend on a balance between vesicular release of DA and reuptake of the released DA through the DA transporter (DAT). Multiple mechanisms have been identified for the short- and long-term regulation of DA release, whereas relatively little is known about mechanisms by which DAT may be acutely regulated.
One well-characterized mechanism by which DA release can be modulated is via DA autoreceptors. DA receptors are divided into two general classes, the D1-like family, composed of the D1 and D5 receptor subtypes, and the D2-like family, composed of the D2, D3, and D4 subtypes (see Sokoloff and Schwartz, 1995). Release-modulating DA autoreceptors belong to the D2-like family (see Langer, 1997). Studies using in vitro slice techniques, synaptosomes, and in vivo microdialysis have provided strong evidence that activation of these terminal autoreceptors with D2-like receptor agonists inhibits DA release within the rodent striatum (Dwoskin and Zahniser, 1986 ; Altar et al., 1987 ; Westerink and de Vries, 1989). D2 DA receptors have been localized on dopaminergic axon terminals (Sesack et al., 1994), and at least a subpopulation of these receptors is likely the release-modulating DA autoreceptors. Recently, however, both in vivo and in vitro studies using microdialysis or electrochemical methods have found that D2-like receptor ligands can also modulate rat striatal DAT activity (Meiergerd et al., 1993 ; Parsons et al., 1993 ; Cass and Gerhardt, 1994). DAT activity is increased by the D2-like receptor agonist quinpirole and decreased by the D2-like receptor antagonists pimozide, sulpiride, and raclopride. It is important that Meiergerd et al. (1993) observed these effects in striatal synaptosomes, suggesting that autoreceptors are involved. Much of the evidence implicating the D2 DA autoreceptor in the regulation of DA release and DAT activity is pharmacological in nature and thus relies on the specificity of the agents used for a given subtype of DA receptor. However, because no compounds are absolutely selective for any one particular DA receptor subtype, this approach is limited.
The availability of genetically altered mice deficient in the D2 DA receptor allows further investigation into the role of the D2 receptor in modulating DA release and DAT function. For our studies we used D2 DA receptor mutant mice generated and characterized by Kelly et al. (1997, 1998). In these mice, D2 DA receptor density in striatum of the heterozygous (D2+/-) mice is ~50% of that in wild-type (D2+/+) mice, whereas no specific D2 receptor binding is detectable in striata of knockout (D2-/-) mice. In the present study in vivo microdialysis, in vivo electrochemistry, and quantitative autoradiography were used to examine the consequences of D2 receptor deficiency on DA release, DAT function, and DAT expression.
MATERIALS AND METHODS
Male and female D2 DA receptor null mice (weighing 20-35 g) were generated at the Vollum Institute (Oregon Health Sciences University, Portland, OR, U.S.A.). The methods used to produce these mice are reported in detail elsewhere (Kelly et al., 1997). In brief, the D2 DA receptor gene was mutated by homologous recombination in embryonic stem cells with a targeting vector to delete all of exon 7 and the 5′ half of exon 8, the region encoding the majority of the putative third intracellular loop, the last two transmembrane domains, and all of the carboxy terminus. Blastocyst injection was used to generate chimeric mice of the mixed 129/Sv × C57BL/6J background. F1 heterozygous mice sired by the chimeras were interbred to produce F2 mice of all three possible genotypes : D2+/+, D2+/-, and D2-/-. Backcrossing F2 mice to the C57BL/6J strain for five generations resulted in the incipient congenic N5 mice of all three genotypes used here. All genotypes were confirmed by Southern blotting.
Mice were housed in groups of two to five under a 12-h light/dark cycle with food and water freely available. All animal use procedures were in accordance with the NIH Guide for the Care and Use of Laboratory Animals and were approved by the Animal Care and Use Committee at the University of Colorado Health Sciences Center.
In vivo microdialysis
Mice were anesthetized with urethane [12.5% (wt/vol), 1.25-1.5 g/kg i.p.) and placed into a stereotaxic apparatus (Kopf or Stoelting Instruments) fitted with a mouse adaptor. Core body temperature was maintained at 37°C with a heating pad. Microdialysis probes were lowered stereotaxically [0.6 mm anterior to bregma, 1.7 mm lateral from midline, 4.0 mm below dura (Franklin and Paxinos, 1997)] into dorsal striatum.
Microdialysis experiments were performed using methods previously described (Hebert and Gerhardt, 1997 ; Hoffman et al., 1997). Concentric-design microdialysis probes were constructed using 0.3 mm o.d. hollow cellulose fibers (ENKAAG, Germany ; molecular weight cutoff = 12,500) and had an exposed membrane length of 2 mm. Probes were perfused at a rate of 1 μ1/min using a computerized multisyringe pump (World Precision Instruments, Sarasota, FL, U.S.A.). Solutions were loaded into 1-ml-capacity gas-tight Hamilton syringes that were connected to the inlet of the probe via a short length of polyethylene tubing. All solutions were gassed with 95% O2/5% CO2 and adjusted to pH 7.4 before use. Dialyzing solutions consisted either of a normal artificial CSF (aCSF ; 123 mM Nacl, 3 mM KCl, 1 mM CaCl2, 1 mM MgCl2, 25 mM NaHCO3, 1 mM NaH2PO4, 5.9 mM glucose, and 0.5 mM ascorbic acid) or a modified aCSF containing high potassium (23 mM NaCl and 100 mM KCl with all other reagents the same as normal aCSF). In vitro probe recoveries were determined before each experiment by collecting a 20-μl sample from probes immersed in a standard solution of 0.1 mM DA, 0.1 mM 3,4-dihydrophenylacetic acid (DOPAC), and 1 mM homovanillic acid (HVA) in aCSF, maintained at 37°C. Probe recoveries ranged from 7 to 16%, with a mean ± SEM of 10 ± 1% (n = 25) ; all data were corrected for probe recovery.
In vivo samples were collected every 20 min ; data from the first 20-min sample were discarded as they reflected tissue damage caused by probe insertion. After three stable 20-min fractions were collected for baseline measurement, normal aCSF perfusate was switched to high-K+ aCSF, and a 20-min evoked-release fraction (S1) was collected. After a 60-min recovery period, a second period of stimulation (S2) was induced with high-K+ aCSF. In a subset of animals, the 60-min recovery period was followed by a 20-min perfusion with 100 μM raclopride in aCSF. S2 was then induced by high-K+ aCSF that also contained 100 μM raclopride. Basal extracellular levels of DOPAC and HVA were used as indicators of correct probe placement. At the conclusion of each experiment, probe placement and integrity were verified by visual examination of the probe tract during postmortem coronal tissue sectioning. Data from two mice were excluded as a result of incorrect probe placement. The average of three 20-min fractions before K+ stimulation was taken to represent basal outflow. Effects of genotype on spontaneous and K+-evoked release of DA, DOPAC, and HVA were analyzed using one-way ANOVAs.
Dialysis samples (20 μl) were analyzed via HPLC coupled with electrochemical detection as previously described (Hebert and Gerhardt, 1997). In brief, the mobile phase consisted of a citrate/acetate buffer solution (pH 4.0-4.1) containing 0.3 mM 1-octane sulfonic acid, 0.67 mM citric acid monohydrate, 0.1 M sodium acetate, and 0.13 mM EDTA with 7% methanol delivered at a flow rate of 2 ml/min. Samples were injected into a Keystone ODS-Hypersil column (C18 ; particle size, 3 μm ; 100 × 4.6 mm), and detection was performed using an analytical cell (ESA model 5011) with dual coulometric detectors. An oxidation potential of 0.4 V was used to quantitate HVA levels, whereas signals from the second reducing electrode, set at —0.25 V, were used to detect DA and DOPAC. DA, DOPAC, and HVA were quantified based on peak heights and retention times relative to known standard concentrations of each compound.
In vivo electrochemistry
Mice were prepared for surgery as discussed above. A 2-cm incision was made in the scalp, and the skull overlying anterior cortex was removed using a hand-held drill (Dremel) fitted with a small engraver's bit. The dura was removed, and an electrochemical electrode/micropipette assembly was lowered slowly under stereotaxic control [1-1.5 mm anterior to bregma, 1.2-1.5 mm lateral from midline, 2-2.7 mm below dura (Franklin and Paxinos, 1997)] into dorsal striatum. Electrode/micropipette assemblies were constructed by attaching Nafion-coated carbon fiber electrodes to either a single-barrel or a double-barrel micropipette. The carbon fiber electrodes each contained a single carbon fiber (30 μm in diameter) sealed in a glass capillary with a 100-150-μm-length exposed. Electrodes were coated with Nafion to give a selectivity for DA over ascorbic acid of at least 500 : 1 and were calibrated in vitro before implantation as previously described (Gratton et al., 1989 ; Hebert and Gerhardt, 1997). All electrodes responded in <1 s to addition of an aliquot of DA during calibration. The tips of the micropipettes were 10-15 μm in diameter and were positioned 270-300 μm from the tip of the electrode. The micropipettes contained solutions of 400 μM DA and 100 μM ascorbic acid in saline or 200 μM raclopride in saline (pH 7.4). A Ag/AgCl reference electrode was made fresh each day and placed in a small hole drilled in the skull over posterior cerebral cortex. A small amount of petroleum jelly was used to cover the hole and to provide a waterproof seal over the reference electrode.
High-speed chronoamperometry was performed as previously described (Friedemann and Gerhardt, 1992 ; Cass and Gerhardt, 1994). The oxidation (0.55 V) and resting (0.0 V) potentials were applied sequentially for 100 ms each. Finite volumes (25-200 nl) of 400 μM DA were pressure-ejected at 5-min intervals from the micropipette. Once stable signals were obtained, i.e., the same volume of DA produced three signals with amplitudes that did not differ from each other by > 10%, the same volume of DA was locally applied every 5 min for at least 35 min to establish basal DAT function. In a subset of animals, once a stable baseline was achieved, 200 μM raclopride was ejected from the second barrel of a two-barrel micropipette at 30-60 s before the next DA application. Raclopride was applied at twice the volume of DA so that the picomoles of DA and raclopride were equal. Signals were measured at 5-min intervals for at least 35 min after raclopride ejection.
The electrochemical signal parameters used for analysis are shown in Fig. 1. The maximal height of the DA oxidation signal is the amplitude. The time course (T80) of a signal is the time required for the DA signal to reach its maximum and to decay by 80%. Clearance rate is calculated as the slope of the pseudolinear decay portion of the DA oxidation signal (between the time for 20% and 60% decay from the maximum).
It is well known that striatal tissue exhibits marked heterogeneity of DA-related parameters (see, e.g., Boyson et al., 1986 ; Graybiel and Moratalla, 1989 ; Friedemann and Gerhardt, 1992), and it has been shown that the amplitude and time course (T80) of exogenous DA signals can vary within rat dorsal striatum (Cline et al., 1995). To assess effects of genotype on DAT function and eliminate potential confounds resulting from the heterogeneity of striatal tissue and the ensuing variations in signal parameters, we only used electrode placements that resulted in signals with an amplitude between 1.5 and 3.5 μM DA and a T80 of <120 s for determination of baseline measurements of DAT function.
One-way ANOVAs were used to analyze effects of genotype on baseline parameters. Tukey's test was used for post hoc analysis. The effects of raclopride were assessed as a percentage of baseline for each genotype ; an unpaired t test was used to determine significance.
Mice of each genotype were killed by cervical dislocation and decapitated. Brains were removed, immediately frozen in powdered dry ice, and stored at -70°C. Coronal sections (10 μm) were cut with a cryostat at the levels of the striatum/nucleus accumbens and substantia nigra/ventral tegmental area. Indirect saturation curves were generated by incubating the slide-mounted brain sections with 4 nM (-)2β-[3H]carbomethoxy-3β-(4-fluorophenyl)tropane ([3H]WIN 35,428 or [3H]CFT) and with no drug (total binding), one of eight concentrations of unlabeled WIN 35,428 (0.3 nM-1 μM), or benztropine (30 μM, to define nonspecific binding) for 90 min in ice-cold 30 mM sodium phosphate buffer (pH 7.4) containing 0.32 M sucrose (Reith and Coffey, 1993). Sections were then washed twice for 1 min in ice-cold buffer without sucrose, dipped in ice-cold water, and immediately dried. Slides and standards were apposed to film for 2-6 weeks. Films were analyzed with an MCID M2 image analysis system (Imaging Research, St. Catharines, Ontario, Canada). InPlot software (GraphPad, San Diego, CA, U.S.A.) was used to fit the indirect saturation curves. Data were analyzed with one-way ANOVAs to assess effects of genotype on binding parameters.
DA HCl, DOPAC, HVA, and benztropine mesylate were obtained from Sigma Chemical Co. (St. Louis, MO, U.S.A.). Raclopride tartrate was a gift from Astra Alab AB (Södertälje, Sweden). [3H]WIN 35,428 was obtained from Du Pont-New England Nuclear (Boston, MA, U.S.A.). Unlabeled WIN 35,428 was provided by the National Institute on Drug Abuse (Research Triangle Park, NC, U.S.A.).
Extracellular concentrations of DA, DOPAC, and HVA
Basal and K+-evoked extracellular DA concentrations were measured in dorsal striata of urethane-anesthetized D2+/+, D2+/-, and D2-/- mice (Table 1). Basal concentrations trations of extracellular DA ranged from 8 to 11 nM and were not significantly different among the three genotypes. Compared with basal extracellular concentrations of DA, concentrations of the DA metabolites DOPAC and HVA were ~150- and 400-fold higher, respectively, in all three genotypes. Similar to DA, no genotypic differences were observed.
Table 1. Basal and K+-stimulated extracellular concentrations of DA and its metabolites in dorsal striata of urethane-anesthetized D2 DA receptor mutant mice Dialysate was collected every 20 min. Basal levels reflect the mean ± SEM value of the last three 20-min fractions collected before K+ stimulation. KCl (100 mM) was substituted for 100 mM NaCl in the aCSF perfusate for 20 min to evoke DA release. Data are mean ± SEM values.
Inclusion of 100 mM KCl in the aCSF perfusate for 20 min (S1) stimulated a marked, 100-fold increase in extracellular DA concentrations in all three genotypes (Table 1). However, there were no significant differences in DA, DOPAC, or HVA concentrations among the three genotypes in response to K+ stimulation. DA concentrations gradually returned to basal values during the 60-min washout period (data not shown). A second K+ stimulation (S2) 60 min later caused a similar rise in DA concentrations, resulting in an average S2/S1 ratio of 0.93 ± 0.07 (n = 8). Inclusion of the D2-like receptor antagonist raclopride (100 μM) in the aCSF perfusate before and during S2 produced no significant effect on extracellular DA concentrations in striata of the D2+/+, D2+/-, or D2-/- mice (data not shown).
Basal DAT function
Exogenous DA clearance, which reflects DAT activity, was measured using in vivo voltammetric recording in dorsal striatum of urethane-anesthetized D2+/+, D2+/-, and D2-/- mice. Basal DA signal parameters in striatum of the three genotypes are shown in Fig. 2. Baseline signals were determined on the basis of amplitude so that volumes (in picomoles) of DA ejected were adjusted to give amplitudes between 1.5 and 3.5 μM in all mice. Therefore, no effect of genotype was observed on this parameter (Fig. 2A). However, significantly higher volumes of DA were required to obtain signals within this range of amplitudes in D2+/+, as compared with D2-/-, mice (Fig. 2B). On average, 95 nl of DA was ejected in D2+/+ mice, whereas 35 nl was ejected in D2-/- mice. This result suggests that striatal DA was cleared more quickly in D2+/+ mice. This observation is supported by the DA signal time course data : D2+/+ mice had T80 values that were significantly lower by 50% than D2-/- mice (Fig. 2C). The DA clearance rate in dorsal striata of D2+/+ mice was 0.2 μM DA/s, and the clearance rate was again significantly reduced by 50% in D2-/- mice (Fig. 2D). Taken together, these data indicate that basal DAT function in dorsal striatum is decreased in D2 receptor null mutant mice and are consistent with the suggestion that D2 DA receptors modulate DAT velocity.
One-way ANOVAs revealed a significant effect of genotype on baseline volume, T80, and clearance rate [F(2,25) = 4.58, 5.17, and 3.74, respectively ; p values < 0.05]. Tukey's post hoc analysis indicated that D2-/- mice required less DA, had a longer T80, and showed decreased clearance of DA than did D2-/- mice (all p values < 0.05). D2+/- mice did not differ significantly from either D2+/+ or D2-/- mice in any measure (p values > 0.05 by Tukey's test).
Inhibition of DAT function by raclopride
In a subset of the animals used for baseline measurements, the regulation of striatal DAT activity by D2 DA receptors was further examined by testing the effect of locally coapplied raclopride, a D2-like receptor antagonist, on DA clearance [equal amounts (in picomoles) of raclopride and DA applied ; Table 2]. Signal amplitude was increased by 23% above baseline by raclopride in D2+/+ mice, indicating that DAT activity was significantly attenuated by D2-like receptor blockade (p < 0.05 by unpaired t test ; Fig. 3). In contrast, raclopride produced no significant effect on DA signal amplitude in either D2-/- or D2+/- mice (Table 2). Furthermore, no significant changes were induced by raclopride in either T80 values or clearance rates in any of the three genotypes (Table 2).
Table 2. DA signal parameters in dorsal striata of D2 DA receptor mutant mice before and after application of the D2-like DA receptor antagonist raclopride Data are mean ± SEM values. Baseline values are the mean ± SEM of two or three samples immediately before raclopride application. Equal amounts (in picomoles) of the 200 μM raclopride solution were applied 30-60 s before DA. Raclopride values are the mean ± SEM results of the seven signals after raclopride application (5-35 min). Percent baseline was calculated for each mouse [(raclopride mean/baseline mean) × 100] within each genotype.
Clearance rate (μM/s)
aSignificantly different from 100%, p < 0.05, shown in Fig. 3.
In vitro radioligand binding assays and quantitative autoradiographic analysis were used to determine the affinity and number of DAT binding sites associated with nigrostriatal and mesolimbic DA neurons in the D2+/+, D2+/-, and D2-/- mice. Indirect [3H]WIN 35,428/WIN 35,428 saturation curves were generated in regions containing the neuronal terminals (dorsal striatum and nucleus accumbens) as well as cell bodies (substantia nigra and ventral tegmental area). In all four regions, the curves were monotonic, indicating a single binding site. The curves generated in the dorsal striata of the three genotypes are shown in Fig. 4. The affinities (Ki values) for [3H]WIN 35,428 ranged from 25 to 53 nM and were not different among the four brain regions (Table 3). The number of [3H]WIN 35,428 binding sites (Bmax) showed the expected regional rank order : dorsal striatum (2.5 pmol/mg of protein) > nucleus accumbens (1.4 pmol/mg) > ventral tegmental area (0.50 pmol/mg) ≥ substantia nigra (0.33 pmol/mg ; Table 3). However, within any one brain region there were no significant differences in either Ki or Bmax values for DAT binding sites among the D2+/+, D2+/-, and D2-/- mice (F values < 2.5, p values > 0.05 by one-way ANOVAs ; Table 3).
Table 3. Affinity and number of DAT binding sites in brain regions of D2 DA receptor mutant mice Data are mean ± SEM values.
The present experiments demonstrated that DAT activity, but not expression, was diminished in D2-/- null mutant mice. Compared with D2+/+ mice, only one-third the amount of DA was required to produce baseline signals with mean amplitudes of 2.5 μM in dorsal striata of D2-/- mice, suggesting that DA was cleared more slowly in D2-/- mice and was therefore available for electrochemical detection. This difference could have resulted from decreased DAT activity and/or expression in the D2-/- mouse striatum. However, the effect of genotype on DAT function did not appear to result from altered transporter expression because no differences in the density of striatal [3H]WIN 35,428 binding sites between the D2-/- and D2+/+ mice were observed. Diminished DAT activity in D2-/- mice was also indicated by significantly longer signal time courses (T80 values) and decreased DA clearance rates. Blockade of D2-like DA receptors with raclopride increased DA signal amplitudes, indicative of antagonism of DAT activity, in D2+/+, but not in D2+/- or D2-/-, mice. These results are also consistent with the conclusion that DAT velocity is normally accelerated in response to activation of D2 DA receptors and that this modulatory effect is lost in the D2-/- mice. Thus, the present findings support and extend previous reports of DAT modulation by D2-like receptor ligands (Meiergerd et al., 1993 ; Parsons et al., 1993 ; Cass and Gerhardt, 1994). The previous studies used D2-like receptor agonists or antagonists to alter DAT activity, thus leaving open the possibility that either D3 and/or D4 receptors were involved or that these drugs were acting directly on the DAT. The present report that specific, targeted removal of D2 receptors by a genetic strategy decreased DAT activity suggests that the effects of pharmacological manipulation on DAT function reported previously (Meiergerd et al., 1993 ; Parsons et al., 1993 ; Cass and Gerhardt, 1994) were indeed a result of antagonism of D2 receptors.
It is interesting that DA clearance signal parameters in D2+/- mice, although not significantly different from those in wild-type mice, strongly resembled those in D2-/- mice. This finding indicates that a 50% loss of D2 receptors is sufficient to impact DAT function, suggesting that there is no population of spare receptors for this effect. A similar conclusion was reached by Kelly et al. (1998) in regard to striatal postsynaptic D2 receptors involved in locomotor behaviors. Work by Meller and colleagues supports the idea that a spare receptor population may exist for one effect of a D2-like receptor, whereas another effect of that receptor may be mediated by a population with no spare receptors. No receptor reserve was seen in striatal D2-like receptors regulating either agonist inhibition of tyrosine hydroxylase (Bohmaker et al., 1989) or acetylcholine release in striatum (Meller et al., 1988). In contrast, this group did find a spare receptor population of striatal D2-like release-modulating autoreceptors (Yokoo et al., 1988).
A possible mechanism for the observed modulation of DAT by presynaptic D2 receptors comes from recent reports that the DAT is voltage-dependent (Sonders et al., 1997 ; Zahniser et al., 1998). Specifically, in Xenopus oocytes expressing human DAT, transporter velocity is increased by hyperpolarization of the membrane and decreased by depolarization (Sonders et al., 1997). Activation of D2 DA receptors opens inwardly rectifying potassium channels, resulting in transient hyperpolarization of the membrane (Lacey et al., 1987). This hyperpolarization could, in turn, increase DAT velocity and DA uptake. Alternatively, activation of D2 receptors could alter DAT activity via signal transduction cascades and changes in DAT phosphorylation state. In the electrochemical studies described here using D2+/+ mice, it is likely that the locally applied DA binds to presynaptic D2 receptors and thereby increases DAT activity. With either pharmacological blockade of D2 receptors or where D2 receptors are absent as in the D2-/- mice, no such modulation can occur after application of DA, with the end result being diminished DAT function. The present findings that basal DAT function is attenuated in D2-/- mice compared with D2+/+ mice and that application of raclopride decreases DAT activity in D2+/+ mice are consistent with this hypothesis.
Differential DA release among the three genotypes cannot explain the observed differences in DAT function. Although it might be expected that mice lacking the inhibitory effects of presynaptic D2 autoreceptors would show greater DA release, D2-/- and D2+/+ mice exhibited similar extracellular concentrations of basal and K+-evoked DA in striatum. It has been suggested that dialysate DOPAC levels are more a reflection of DA synthesis than of release (Zetterström et al., 1988). Thus, our finding that extracellular levels of DOPAC and HVA were not significantly altered among genotypes suggests similar rates of DA synthesis in the D2-/- mutant and D2+/+ mice. This conclusion is consistent with the findings in another D2 receptor null mutant mouse in which no changes were observed in the expression of mRNA or protein for tyrosine hydroxylase, the enzyme responsible for DA synthesis (Baik et al., 1995 ; Saiardi et al., 1998). Likewise, striatal tissue levels of DA and its metabolites are not different among the three genotypes used here (Kelly et al., 1998). These results suggest that compensatory mechanisms have developed in D2-/- mice to maintain DA synthesis and release. It is interesting that the lack of effect of the receptor deletion on DA levels parallels the subtle behavioral effects seen in these mice (as reported by Kelly et al., 1998). It is possible that there could be small changes in DA release parameters that were undetectable under the conditions used ; however, significant changes in DA levels, such as those seen in the DAT knockout mouse (Giros et al., 1996), would have been detected. Future studies using alternative microdialysis methods, e.g., no-net flux, and/or longer probe implantation times should be performed to investigate more fully basal extracellular DA levels in these mice.
The D2-/- mice have provided strong evidence that DA autoreceptors localized on both the somatodendritic and terminal regions of nigrostriatal neurons are of the D2 subtype. Although the baseline electrophysiological properties of nigral DA neurons do not differ between D2+/+ and D2-/- mice, exposure to the D2-like receptor agonist quinpirole induces hyperpolarization and inhibition of spontaneous firing only in D2+/+ mice (Mercuri et al., 1997). Likewise, quinpirole dose-dependently reduces stimulation-evoked release of [3H]DA from striatal synaptosomes prepared from D2+/+, but not D2-/-, mice (L'hirondel et al., 1998). Our microdialysis experiments with K+ stimulation showed no consistent augmentation of DA release with the D2-like receptor antagonist raclopride in any of the three genotypes. Most likely, this result reflects the fact that extracellular DA concentrations were markedly elevated in response to K+, and modulatory effects of autoreceptors are more easily detected with low levels of stimulated release (Dwoskin and Zahniser, 1986). Future studies should address the effects of low-level stimulation on DA release in the null mutant mice.
It is possible that in the D2-/- mutant mice, D3 DA receptors serve as inhibitory autoreceptors, thereby resulting in similar striatal DA release in D2-/- and D2+/+ mice. However, the results discussed above from L'hirondel et al. (1998) and Mercuri et al. (1997) in D2-/- mice strongly argue against this. Furthermore, whether D3 DA receptors even exist as autoreceptors has been controversial (see discussion in Koeltzow et al., 1998). However, it has been shown that D3 receptors can inhibit DA release in a clonal dopaminergic cell line (Tang et al., 1994). Also, Tepper et al. (1997) used in vivo D2 or D3 receptor antisense knockdown in rat substantia nigra and found that the inhibitory effects of the DA agonist apomorphine on cell firing were attenuated after treatment with the D3- as well as the D2- selective antisense oligonucleotides. A recent article by Koeltzow et al. (1998) investigating autoreceptor function in D3-/- mice supports the idea that D3 receptors can modulate DA release in vivo ; however, this effect appears to be mediated via postsynaptic D3 receptors involved in short feedback loops, rather than via presynaptic D3 autoreceptors. In any case, our results clearly demonstrated that D3 receptors do not compensate for the loss of D2 receptors in modulating DAT activity.
Meiergerd et al. (1993), using in vitro electrochemical recording and striatal synaptosomes, provided definitive evidence that presynaptic D2-like autoreceptors can modulate DAT activity. Thus, we have assumed that it is the loss of presynaptic D2 autoreceptors that has impacted DAT function in the D2-/- mice. However, the D2 null mutant mice also lack postsynaptic D2 DA receptors. As such, it is possible that the loss of these postsynaptic D2 receptors alters short- and/or long-loop feedback systems, which ultimately affect DAT activity. Functionally, a lack of postsynaptic D2 receptor activity often results in increased dopaminergic neuronal activity and DA release (Chiodo and Bunney, 1983 ; Imperato and Di Chiara, 1985 ; Westerink and de Vries, 1989). Another compensatory mechanism might be to slow DA uptake velocity, i.e., DAT function, to maintain extracellular levels of DA. Our studies cannot rule out this possibility.
In summary, the lack of D2 DA receptors altered striatal DAT activity but, surprisingly, not DA release. These results provide further support for the conclusions of previous studies (Meiergerd et al., 1993 ; Parsons et al., 1993 ; Cass and Gerhardt, 1994) that D2 DA receptors modulate DAT activity. The molecular mechanism(s) underlying this receptor-mediated modulation will be the focus of future experiments. However, in the D2-/- mice the altered DAT activity does not appear to involve altered transporter expression. Recently, it has been recognized that both DAT activity and expression are regulated. Diverse mechanisms, such as changes in membrane potential, altered posttranslational modifications via protein kinases and phosphatases, nitric oxide, eicosainoids, and oxygen radicals, and changes in gene expression, have all been implicated (see Pogun et al., 1994 ; Zahniser et al., 1995 ; Fleckenstein et al., 1997 ; Reith et al., 1997 ; Sonders et al., 1997 ; Zhu et al., 1997). Because DAT is crucial for normal DA neurotransmission and motor behaviors (Giros et al., 1996), the regulation of its activity and/or expression represents yet another example of neuroplasticity that may contribute to rapid alterations in synaptic transmission as well as to long-term changes in response to chronic drug administration and neurodegeneration.