Negative Feedback Regulation of Nigrostriatal Dopamine Release: Mediation by Striatal D1 Receptors

There are four major dopamine (DA) systems in the brain: the nigrostriatal, mesocortical, mesolimbic, and tuberoinfundibular pathways (Missale et al., 1998; Cooper et al., 2003). The nigrostriatal DA system is essential for normal voluntary motor activity; its degeneration results in Parkinson's disease. The cell bodies of these neurons are localized in the substantia nigra and extend their axons to the dorsal striatum (Missale et al., 1998). The striatum has high densities of the two main subtypes of DA receptors: D1-like (D1 or D5), which are positively linked to adenylate cyclase; and D2-like (D2, D3, or D4), either not linked or negatively linked to adenylate cyclase (Kebabian and Calne, 1979). Studies have shown that D1- and D2-like receptors are generally segregated on separate neuronal populations of medium spiny GABA neurons in the striatum (Gerfen et al., 1990, 1995; Le Moine and Bloch, 1995). D1 receptors are localized on GABAergic neurons containing substance P and dynorphin that project to the substantia nigra (striatonigral neurons or the “direct pathway”; Gerfen et al., 1995). D2 receptors are found on enkephalin containing GABAergic neurons that project to the globus pallidus (“indirect pathway”). In addition, D2, but not D1, receptors are localized presynaptically on DA terminals (Hersch et al., 1995).

Physiological and neurochemical studies have shown that the activity of nigrostriatal neurons is regulated by short and long negative feedback loops (Westerink et al., 1994; Shi et al., 1997; for reviews, see Wolf et al., 1987 and Cooper et al., 2003). Evidence indicates that both D1 and D2 receptors participate in the autoregulation of DA release from the nigrostriatal pathway. Of these two receptor subtypes, D2-mediated inhibitory feedback regulation has been more firmly established. Multiple studies have demonstrated that released DA acts on presynaptic D2 autoreceptors on DA nerve terminals in the striatum to decrease DA synthesis and release (Imperato and Di Chiara, 1988; Westerink and de Vries, 1989). In addition, somatodendritically released DA acts on D2 receptors on DA cells in the substantia nigra to decrease DA neuronal activity and subsequent release of DA in the striatum (Santiago and Westerink, 1991). Activation of postsynaptic D2 receptors on striatopallidal neurons (the indirect pathway) engages a long loop negative feedback pathway (Shi et al., 1997). It is generally believed that D1 receptors on striatonigral projections (direct pathway) also mediate a long-loop inhibitory feedback system. However, the physiological evidence for this pathway remains equivocal. Although some reports have shown an inhibition of DA release after the administration of D1 agonists (Zetterstrom et al., 1986; Imperato and Di Chiara, 1988; Santiago et al., 1993; Rahman and McBride, 2001; Zackheim and Abercrombie, 2001), others have found no effect on release (Imperato et al., 1987) or basal DA cell firing (Carlson et al., 1987; Huang and Walters, 1992; Shi et al., 1997, 2000). Most of these studies used the benzazepine D1 agonists SKF 38393 [(±)-1-phenyl-2,3,4,5-tetrahydro-(1H)-3-benzazepine-7,8-diol] or SKF 82957 [R-(+)-6-chloro-7,8-dihydroxy-3-methyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine] and their D1-receptor specificity has been questioned recently (Zackheim and Abercrombie, 2001). However, studies with the D1 antagonist SCH 23390 [R-(+)-7-chloro-8-hydroxy-3-methyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine] have consistently found that blocking the D1 receptor increases DA release in the striatum (Imperato et al., 1987; Imperato and Di Chiara, 1988; Damsma et al., 1991).

To more fully elucidate the role of striatal D1 receptors in the negative feedback regulation of DA release, the present study used a more recently characterized full D1-like agonist, A-77636 [(-)-(1R,3S)-3-adamantyl-1-(aminomethyl)-3,4-dihydro-5,6-dihydroxy-1H-2-benzopyran)], that has a nonbenzazepine structure (Kebabian et al., 1992; Gulwadi et al., 2001; Chausmer and Katz, 2002). A-77636 has relatively high affinity for D1 receptors (K_i = 39.8 nM) and is functionally inactive at D2 receptors (EC₅₀ = 7–10 μM) (Kebabian et al., 1992). In vitro studies have shown that SCH 23390 competes with A-77636 at the D1 receptor because it blocks A-77636-induced receptor desensitization in SK-N-MC cells (Lin et al., 1996). The current work used in vivo microdialysis to determine the effects on extracellular DA from systemic administration of A-77636 given alone or in combination with SCH 23390. The localization of the relevant D1 receptors was tested by infusing SCH 23390 directly into the striatum through the microdialysis probe. It was hypothesized that systemic administration of A-77636 would dose dependently decrease extracellular DA concentrations in the striatum, and this would be blocked by concomitant infusions of SCH 23390 into the striatum.

Materials and Methods

Animals and Surgery. Male Sprague-Dawley rats (Zivic Miller, Hillson, PA or Harlan, Indianapolis, IN), weighing between 200 and 350 g at the time of surgery, were used. Rats were initially housed in pairs for 5 to 7 days before surgery. Rat housing was in a temperature-controlled room on a 12-h light/dark cycle. Food and water were available ad libitum. Before surgery, rats were anesthetized with a mixture of ketamine (70 mg/kg) and xylazine (6 mg/kg), injected i.m., and then mounted in a stereotaxic frame. After dura was removed, 21-gauge stainless steel guide cannula were implanted on the brain surface above the anterolateral striatum (AP, +1.1; ML, ±3.2; head angle, -3.3; coordinates from bregma according to Paxinos and Watson, 1998). The cannula and three skull screws were held in place with cranioplastic cement. Postsurgery, animals were housed individually for 3 to 5 days before microdialysis experiments were conducted.

Each rat was used once. After microdialysis experiments were finished, probe placements were verified histologically. Only animals whose probe placements were verified to be in the striatum were used in this study. All animal use procedures were in strict accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the local animal care committee.

Microdialysis. Microdialysis probes were of a concentric flow design (Pehek et al., 2001) and were designed to terminate in the striatum. The active dialyzing surface of the cellulose membrane (Spectra/Por Hollow; mol. wt. cutoff, 13,000; diameter, 216 μm) was 3.0 mm in length, extending from 3.0 to 6.0 mm ventral to the brain surface. Before the start of the experiments, rats were weighed and placed in clear Plexiglas chambers. Subsequently, microdialysis probes were carefully inserted through the guide cannulae in awake rats and secured in place with cyanoacrylate gel. Immediately afterwards, rats were tethered to counterbalance arms that allowed for relatively free movement around the cage. They were kept there with food and water available until the start of the experiments (18–24 h after probe implantation).

A microinfusion pump (PHD 2000; Harvard Apparatus Inc., Holliston, MA) and liquid swivels (Instech, Plymouth Meeting, PA) were used to perfuse the artificial cerebrospinal fluid (aCSF) through the probes at a rate of 1.0 μl/min. Perfusion began the morning of the experiment. Dialysate samples were collected every 30 min until basal DA concentrations were stable for at least three baseline samples. Drugs were then administered either systemically or into the brain with reverse dialysis. To perfuse the drug through the probe, tubing connections were switched from normal aCSF to aCSF containing the drug. This was performed so that flow rate and collection volumes were maintained. Sample collection continued for 3 to 4 h.

All experiments used a modified, commercially available artificial cerebrospinal fluid: Dulbecco's phosphate-buffered saline (137 mM NaCl, 2.7 mM KCl, 0.5 mM MgCl₂, 1.5 mM KH₂PO₄, 8.1 mM Na₂HPO₄, pH 7.4). CaCl₂ (1.2 mM) and glucose (10 mM) were added to this solution.

Drugs. All doses and concentrations refer to the salts. Doses of A-77636 (Acquas et al., 1997; Chausmer and Katz, 2002) and concentrations of SCH 23390 (Imperato et al., 1987; Zackheim and Abercrombie, 2001) were chosen based on amounts previously used in the literature. (R)-SCH 23390 hydrochloride (Sigma-Aldrich, St. Louis, MO) was administered intrastriatally via reverse dialysis. A 10 mM stock solution was made by dissolving the drug in 1 ml of deionized water. aCSF was then used to dilute this solution to the appropriate concentration (5.0–200 μM). The pH of the final aCSF solution was 7.4. A-77636 hydrochloride (Sigma-Aldrich) was dissolved in deionized water and administered s.c. in 1.0 ml/kg volumes. Doses were 0 (vehicle), 0.75, 1.5, and 3.0 mg/kg. For antagonist plus agonist experiments, rats were injected with A-77636 30 min after the beginning of the SCH 23390 infusion.

Chromatography. DA concentrations in dialysate samples were measured by reverse phase high-performance liquid chromatography coupled with electrochemical detection. Twenty-microliter samples were injected onto a 2-mm column (Ultracarb, 3-μm particle size, ODS 20; Phenomenex, Torrance, CA). The mobile phase consisted of 32 mM anhydrous citric acid, 54 mM sodium acetate trihydrate, 0.074 mM EDTA, 0.215 mM octylsulfonic acid, and 3% methanol (v/v), pH 4.2. To maintain separation of DA from its metabolites and 5-hydroxyindoleacetic acid, the pH of the mobile phase and the concentration of the octylsulfonic acid were adjusted as needed. A BAS LC-4C electrochemical detector with a glassy carbon electrode, maintained at a potential of +0.60 V relative to an Ag/AgCl reference electrode, was used.

Experimental Design: SCH 23390 Concentration Response. The effects of local administration of the D1 antagonist SCH 23390 were examined in this experiment. Drug was administered by reverse dialysis for 1 h directly into the striatum. Ten, 100, and 200 μM concentrations were infused in three separate groups of rats.

Experimental Design: A-77636 Dose Response. This study examined the dose-response characteristics of systemically administered A-77636 (s.c.) on DA release in the striatum. Four groups of rats were used: vehicle, 0.75, 1.5, and 3.0 mg/kg A-77636.

Experimental Design: SCH 23390 Blockade of 1.5 mg/kg A-77636. This experiment examined whether injections of A-77636 (1.5 mg/kg) would decrease dialysate DA and whether intrastriatal perfusion with SCH 23390 (5.0 μM) would attenuate these effects. There were four groups of rats: vehicle, A-77636 alone, SCH 23390 alone, and SCH 23390 + A-77636. All groups received either SCH 23390 or aCSF without drug followed by A-77636 or a vehicle injection.

Data Analysis. Data were expressed, analyzed, and graphed as the percentage of the last three baseline samples. Statistical analyses were performed using two-way repeated measures ANOVAs with time as the repeated measures factor and drug treatment as the independent factor. These analyses were followed by one-way repeated measures ANOVAs of each drug condition with time as the repeated measures factor. Degrees of freedom were adjusted for nonhomogeneity of variance by the Huynh-Feldt F-test. Thus, the reported p values are those after this adjusted test. Post hoc analyses comparing individual time points within a drug treatment were performed using Dunnett's test for comparing treatment means (postdrug time points) with a control (predrug baseline). Newman-Keuls post hoc tests were performed when comparing means between independent drug treatments. Significance level was set at p < 0.05.

Results

Microdialysis Probe Placements.Figure 1A shows the locations of representative probe placements in the anterior striatum. Note that, for clarity, not all placements are shown. Figure 1B illustrates a typical placement in a cresyl violet-stained brain section.

SCH 23390 Concentration Response.Figure 2 shows the comparison between the SCH 23390 groups. All three concentrations produced increases in dialysate DA with the maximal increases ranging from 206.07% of baseline after 10 μM to 326.97% for 100 μM and 443.89% for 200 μM. A two-way repeated measures ANOVA revealed significant main effects for drug concentration [F(2,17) = 4.36, p = 0.03], time [F(6,102) = 23.06, p < 0.0001], and a concentration by time interaction [F(12,102) = 3.54, p = 0.005]. One-way repeated measures ANOVAs followed by post hoc tests revealed that the 100 μM [F(6,30) = 18.22, p < 0.001] and 200 μM [F(6,30) = 11.25, p < 0.001] concentrations caused significant elevations in DA efflux at the 30- and 60-min time points. Basal levels of DA were 7.44 ± 1.35 pg for the 10 μM group (n = 8), 5.40 ± 0.58 pg for the 100 μM group (n = 6), and 6.99 ± 1.78 pg (n = 6) for the 200 μM group.

A-77636 Dose Response.Figure 3 shows the dose-response curve for A-77636 administration, using doses 0.75, 1.5, and 3.0 mg/kg s.c. A two-way ANOVA indicated a significant main effect for time [F(8,144) = 8.57, p < 0.0001] and a significant time by dose interaction [F(24,144) = 2.16, p = 0.01]. A one-way ANOVA for the vehicle data yielded no significant differences in basal DA efflux over time [F(8,24) = 0.91, ns; baseline concentrations, 6.76 ± 3.02 pg, n = 4]. Administration of the lowest dose of A-77636, 0.75 mg/kg, also had no significant effect on dialysate DA levels [F(8,24) = 1.39, ns; basal levels, 3.69 ± 1.21 pg/20 μl, n = 4]. The highest dose, 3.0 mg/kg, did significantly decrease DA release [F(8,56) = 7.92, p = 0.0014]. Post hoc tests revealed that these decreases were significantly different from basal levels at the 210- and 240-min time points in Fig. 3. A two-way ANOVA showed a significant time by drug interaction when 3.0 mg/kg A-77636 was compared with vehicle [F(8, 80) = 4.36, p < 0.001). Newman-Keuls post hoc analysis revealed that 3.0 mg/kg A-77636-induced dopamine release was significantly less than vehicle at the 150-, 180-, 210-, and 240-min time points (Fig. 3; p < 0.05). Baseline DA concentrations for this group were 6.86 ± 1.88 pg/20 μl, n = 8. Administration of 1.5 mg/kg A-77637 did not significantly decrease dialysate DA in comparison with predrug baseline values. However, a two-way ANOVA analysis of this dose versus vehicle showed a significant time by drug interaction between these groups [F(8,64) = 2.80, p = 0.01]. Newman-Keuls post hoc analysis revealed that dialysate DA in the 1.5 mg/kg A-77636 group was significantly lower than vehicle at the 210-min time point (Fig. 3). Baseline concentrations for the 1.5 mg/kg A-77636 group were 7.77 ± 2.08 pg/20 μl, n = 6. Interestingly, in the 1.5 mg/kg A-77636 group, there was an initial increase in basal DA levels that was significantly different from basal levels at 30- and 60-min postinjection [F(8,40) = 7.68, p = 0.0001]. There was also a trend toward an increase in dialysate DA in the other groups in this study.

SCH 23390 Blockade of 1.5 mg/kg A-77636.Figure 4 shows the comparison between the vehicle, SCH 23390 (5.0 μM) alone, A-77636 (1.5 mg/kg s.c.) alone, and the coadministration of both drugs. A two-way ANOVA yielded significant main effects for drug treatment [F(3,18) = 12.36, p < 0.001], time [F(8,144) = 4.71, p = 0.002] and a time by treatment interaction [F(24,144) = 3.77, p = 0.0002]. A one-way ANOVA followed by post hoc tests demonstrated that animals that received A-77636 alone showed a significant decrease in extracellular DA beginning 90 min postinjection (120-min time point on Fig. 4) and lasting to the end of the experiment [F(8,40) = 23.01, p < 0.0001]. The maximal decrease in dialysate DA levels was to 64.65% of baseline 120 min postinjection (baseline values, 7.06 ± 1.13 pg/20 μl, n = 6). For the SCH 23390 alone group, the infusion of a 5.0 μM concentration led to a borderline increase in extracellular DA [F(8,40) = 3.01, p = 0.0564; baseline DA levels, 6.86 ± 0.46 pg/20 μl, n = 6]. The maximal increase (168.68% of baseline concentration) occurred 90 min into the 4-h infusion. DA concentrations did not change over time in vehicle-treated animals [F(8,32) = 1.71, ns; baseline DA levels = 6.99 ± 1.13 pg/20 μl, n = 5]. Dialysate DA levels did change over time in the SCH + Abbott group [F(8,32) = 4.73, p < 0.001], but no significant decreases were observed, indicating that pretreatment with SCH 23390 attenuated the effects of A-77636 administration. A significant increase in DA efflux was observed at the 60-min time point. The SCH 23390 blockade of A-77636 was also revealed by a two-way ANOVA comparing the SCH + Abbott group with the Abbott alone group [significant time by drug interaction: F(8,72) = 2.82, p = 0.009]. Baseline concentrations for the SCH 23390 + A-77636 group were 6.31 ± 1.54 pg/20 μl, n = 5.

Discussion

The present results demonstrate that systemic administration of the isochroman D1-like agonist A-77636 dose dependently decreased DA release in the dorsal striatum. Intrastriatal perfusion with SCH 23390, a D1-like antagonist, increased DA efflux in a concentration-dependent manner. In addition, local infusion of SCH 23390 attenuated the effects of systemically administered A-77636. Overall, these results indicate that stimulation of D1-like receptors, localized within the striatum, decreases DA release from the nigrostriatal pathway. These results contribute to a body of evidence indicating that postsynaptic striatal D1 receptors participate in an inhibitory feedback pathway regulating the activity of midbrain DA neurons.

The present findings agree with many earlier reports using the systemic or intrastriatal administration of SCH 23390 and demonstrating increases in striatal DA (Imperato et al., 1987; Imperato and Di Chiara, 1988; Damsma et al., 1991). SCH 23390 is a highly selective ligand with much greater affinity for D1-like versus D2-like receptors (Hyttel, 1983). The only non-DA receptor that SCH 23390 has been reported to bind to is the 5-hydroxytryptamine_2C receptor (K_i = 9.3 nM), where it acts as an agonist (Millan et al., 2001). However, the present results are not likely to be due to 5-hydroxytryptamine_2C agonism because prior work indicates that this property would likely lead to a decrease, rather than an increase, in DA release (Gobert et al., 2000). Furthermore, SCH 23390 administration was able to block the effects of the D1 agonist A-77636, indicating an action on D1 receptors.

In contrast with prior work using SCH 23390, studies using D1 agonists have yielded conflicting results. D1 agonism fails to consistently decrease the electrical activity of DA cells (Carlson et al., 1987; Huang and Walters, 1992). Systemic administration of the benzazepine agonists SKF 38393 and SKF 82958 decreased striatal DA efflux in some studies but not others (Imperato et al., 1987; Imperato and Di Chiara, 1987, 1988; Zackheim and Abercrombie, 2001). Intrastriatal application of the full agonist SKF 82958 into the striatum decreased DA release, but this effect was not blocked by SCH 23390, suggesting a non-D1 receptor mechanism for benzazapine-class D1 agonists (Zackheim and Abercrombie, 2001). In fact, in vitro data demonstrate that these compounds may selectively depress slowly inactivating potassium currents (Nisenbaum et al., 1998). In contrast, local administration of the nonbenzazepine D1 agonist CY 208243 [(-)-4,6,6a,7,8,12b-hexahydro-7-methyl-indolo-(4,3-ab)phenanthoridine] reversed the augmentation of DA release by SCH 23390 (Imperato and Di Chiara, 1988). This agonist did not alter basal DA efflux, leading the authors to conclude that, under basal conditions, there is high dopaminergic tone at D1 receptors and, hence, further inhibition of DA release by exogenous compounds is not possible (but see Shi et al., 2000). In line with this, many studies have failed to find effects of D1 agonists on the basal firing of midbrain DA neurons (Shi et al., 1997). However, when D2 receptors are activated, D1 agonists decrease DA cell firing, suggesting that coactivation of these receptors is necessary for the expression of D1 feedback inhibition (Shi et al., 1997). Rahman and McBride (2001) reached similar conclusions in a recent study of the long-loop negative feedback pathway from the nucleus accumbens to the ventral tegmental area. Their data indicate that D1 and D2 receptors can independently modulate DA release in the nucleus accumbens but that a cooperative D1/D2 interaction is necessary to activate negative feedback to the ventral tegmental area.

In the present study, systemic administration of the structurally novel, nonbenzazepine agonist A-77636 significantly and dose dependently decreased basal DA release, in correspondence with a previous report (Acquas et al., 1997). In contrast to prior studies with other D1 agonists, the ability of A-77636 to alter basal DA may have resulted from its slow dissociation from and persistent binding to the D1 receptor (Lin et al., 1996). A-77636 has high selectivity for the D1 receptor (Kebabian et al., 1992; Chausmer and Katz, 2002; D'Aquila et al., 2002) and behaves as a D1 agonist in behavioral paradigms (Chausmer and Katz, 2002). Evidence that its present effects were D1-mediated is demonstrated by the ability of SCH 23390 to attenuate its effects. The dose dependence and magnitude of the effects of A-77636 differed between experiments 2.6 and 2.7 (Figs. 3 and 4). The effect was more robust and, compared with predrug baseline, was statistically significant at a lower dose (1.5 mg/kg) in the second study. This may have resulted from a tendency for dialysate DA levels to increase in all groups (including vehicle) in the A-77636 dose-response experiment. The only difference between the two studies was the rat supplier and exact room in the Animal Resource Center where housed. It is possible that the rats in the A-77636 dose-response experiment were more stressed by the s.c. injection procedure (note the person giving the injections was the same). Despite this variability, administration of A-77636 decreased DA efflux in both studies, and the decrease was clearly dose-dependent.

Infusions of SCH 23390, directly into the striatum, greatly attenuated the effects of A-77636 administration, indicating the agonist-induced decreases were due, at least in part, to the stimulation of D1 receptors localized in the striatum. The decreases were not abolished, particularly at the latter time points. This may have resulted from the increasing occupancy of D1 receptors over time by the slowly dissociating A-77636. Alternatively, agonist effects at sites other than the striatum may have contributed to the present effects. D1 receptors are dense in the pars reticulata of the substantia nigra, where they regulate GABA efflux and are in proximity to dendrites from pars compacta DA cell bodies (Trevitt et al., 2002). Thus, binding of A-77636 to these or D1 receptors in other brain areas may have contributed to the observed decreases in dialysate DA.

Evidence indicates that both D1 and D2 receptor subtypes are involved in the autoregulation (negative feedback) of DA neurons. These negative feedback systems have been termed “short-loop” and “long-loop” pathways (for review, see Wolf et al., 1987). A short-loop feedback circuit involving presynaptic D2 autoreceptors located on DA terminals in the striatum and cell bodies in the substantia nigra has been well established. Stimulation of these D2 receptors decreases the functioning of DA neurons by three mechanisms: decreasing impulse generation (somatodendritic D2 receptors), DA synthesis (nerve terminal D2 receptors), and DA release (nerve terminal D2 receptors). Long-loop feedback is believed to involve two parallel neuronal paths: a direct striatonigral pathway, stimulated by postsynaptic D1 receptors; and an indirect striatopallidal pathway, stimulated by postsynaptic D2 receptors (Le Moine et al., 1991; Gerfen et al., 1995; Shi et al., 2000). The cells of origin of both pathways are medium spiny GABAergic output neurons (Missale et al., 1998). The fact that D1 receptors are localized postsynaptically on striatonigral fibers (Caille et al., 1996) suggests the existence of this direct pathway regulating DA neuronal activity and subsequent transmitter release. However, the equivocal physiological and microdialysis studies with D1 agonists (see above) have not permitted a firm establishment of a physiological role for this circuit. The present results, in concert with other recent studies (Shi et al., 1997; Rahman and McBride, 2001), lend support to the existence and importance of a long negative feedback loop mediated by striatal D1 receptors.

One way this feedback pathway may operate is by D1-mediated activation of striatonigral GABAergic neurons. In the striatum, these neurons are innervated by DA terminals (Caille et al., 1996). They project to the substantia nigra where they make direct synaptic connections with DA neurons (Bolam and Smith, 1990; Caille et al., 1996). DA, released from nigrostriatal neurons, is presumed to bind to D1 receptors on striatonigral neurons, exciting these latter neurons and stimulating the efflux of GABA in the substantia nigra. GABA would then act to inhibit the activity of DA cells and, subsequently, the release of DA from these neurons. However, the details of this in vivo neuronal circuitry—the number of neurons involved—and the synaptic connectivity remain unclear at the present time. Because the current study did not examine DA or GABA efflux in the substantia nigra, circuitry other than striatonigral projections may have been involved in the present findings. For example, the feedback loop may actually involve local interactions within the striatum, such as interactions between DA terminals and GABAergic interneurons, glutamate afferents, or cholinergic interneurons (Rahman and McBride 2002). To examine this circuitry more fully, further neurochemical studies of the striatum and substantia nigra are necessary.

Previous studies have demonstrated cooperativity between D1 and D2 receptors (White, 1987; Wachtel et al., 1989; Keefe and Gerfen, 1995). At “normal”, i.e., basal, DA concentrations, D2 autoreceptors may mediate the inhibition of DA neurons. However, when DA efflux is increased, long-loop feedback inhibition may rely on the concurrent activation of both receptor subtypes (Shi et al., 2000; Rahman and McBride, 2001). Hence, sufficient endogenous tone at D2 receptors may be necessary for the expression of D1-mediated effects (Treseder et al., 2000; Smith et al., 2002). Thus, D1 stimulation-mediated inhibition of DA release may depend on factors such as the behavioral state of the animal (e.g., stressed versus calm) and the consequent basal dopaminergic tone.

In summary, these results indicate that activation of striatal D1 receptors contributes to the negative feedback regulation of DA release from nigrostriatal neurons. The relevant neuronal circuitry may involve long-loop striatonigral projections, although the precise delineation of the underlying circuitry awaits further studies. Increased knowledge of this circuitry may contribute to our understanding and treatment of diverse diseases such as Parkinson's disease and schizophrenia.