Adenosine Signaling through A1 Receptors Inhibits Chemosensitive Neurons in the Retrotrapezoid Nucleus.

A subset of neurons in the retrotrapezoid nucleus (RTN) function as respiratory chemoreceptors by regulating depth and frequency of breathing in response to changes in tissue CO2/H+. The activity of chemosensitive RTN neurons is also subject to modulation by CO2/H+-dependent purinergic signaling. However, mechanisms contributing to purinergic regulation of RTN chemoreceptors are not entirely clear. Recent evidence suggests adenosine inhibits RTN chemoreception in vivo by activation of A1 receptors. The goal of this study was to characterize effects of adenosine on chemosensitive RTN neurons and identify intrinsic and synaptic mechanisms underlying this response. Cell-attached recordings from RTN chemoreceptors in slices from rat or wild-type mouse pups (mixed sex) show that exposure to adenosine (1 µM) inhibits chemoreceptor activity by an A1 receptor-dependent mechanism. However, exposure to a selective A1 receptor antagonist (8-cyclopentyl-1,3-dipropylxanthine, DPCPX; 30 nM) alone did not potentiate CO2/H+-stimulated activity, suggesting activation of A1 receptors does not limit chemoreceptor activity under these reduced conditions. Whole-cell voltage-clamp from chemosensitive RTN neurons shows that exposure to adenosine activated an inward rectifying K+ conductance, and at the network level, adenosine preferentially decreased frequency of EPSCs but not IPSCs. These results show that adenosine activation of A1 receptors inhibits chemosensitive RTN neurons by direct activation of a G-protein-regulated inward-rectifier K+ (GIRK)-like conductance, and presynaptically, by suppression of excitatory synaptic input to chemoreceptors.

Purines and neuronal excitability: links to the ketogenic diet.

ATP and adenosine are purines that play dual roles in cell metabolism and neuronal signaling. Acting at the A(1) receptor (A(1)R) subtype, adenosine acts directly on neurons to inhibit excitability and is a powerful endogenous neuroprotective and anticonvulsant molecule. Previous research showed an increase in ATP and other cell energy parameters when an animal is administered a ketogenic diet, an established metabolic therapy to reduce epileptic seizures, but the relationship among purines, neuronal excitability and the ketogenic diet was unclear. Recent work in vivo and in vitro tested the specific hypothesis that adenosine acting at A(1)Rs is a key mechanism underlying the success of ketogenic diet therapy and yielded direct evidence linking A(1)Rs to the antiepileptic effects of a ketogenic diet. Specifically, an in vitro mimic of a ketogenic diet revealed an A(1)R-dependent metabolic autocrine hyperpolarization of hippocampal neurons. In parallel, applying the ketogenic diet in vivo to transgenic mouse models with spontaneous electrographic seizures revealed that intact A(1)Rs are necessary for the seizure-suppressing effects of the diet. This is the first direct in vivo evidence linking A(1)Rs to the antiepileptic effects of a ketogenic diet. Other predictions of the relationship between purines and the ketogenic diet are discussed. Taken together, recent research on the role of purines may offer new opportunities for metabolic therapy and insight into its underlying mechanisms.

Adenosine, ketogenic diet and epilepsy: the emerging therapeutic relationship between metabolism and brain activity.

For many years the neuromodulator adenosine has been recognized as an endogenous anticonvulsant molecule and termed a "retaliatory metabolite." As the core molecule of ATP, adenosine forms a unique link between cell energy and neuronal excitability. In parallel, a ketogenic (high-fat, low-carbohydrate) diet is a metabolic therapy that influences neuronal activity significantly, and ketogenic diets have been used successfully to treat medically-refractory epilepsy, particularly in children, for decades. To date the key neural mechanisms underlying the success of dietary therapy are unclear, hindering development of analogous pharmacological solutions. Similarly, adenosine receptor-based therapies for epilepsy and myriad other disorders remain elusive. In this review we explore the physiological regulation of adenosine as an anticonvulsant strategy and suggest a critical role for adenosine in the success of ketogenic diet therapy for epilepsy. While the current focus is on the regulation of adenosine, ketogenic metabolism and epilepsy, the therapeutic implications extend to acute and chronic neurological disorders as diverse as brain injury, inflammatory and neuropathic pain, autism and hyperdopaminergic disorders. Emerging evidence for broad clinical relevance of the metabolic regulation of adenosine will be discussed.

Correlated multisecond oscillations in firing rate in the basal ganglia: modulation by dopamine and the subthalamic nucleus.

Previous studies from this laboratory have shown that many neurons in the basal ganglia have multisecond (<0.5 Hz) periodicities in firing rate in awake rats. The frequency and regularity of these oscillations are significantly increased by systemically injected dopamine (DA) agonists. Because oscillatory activity should have greater functional impact if shared by many neurons, the level of correlation of multisecond oscillations was assessed by recording pairs of neurons in the globus pallidus and substantia nigra pars reticulata in the same hemisphere, or pairs of globus pallidus neurons in opposite hemispheres in awake, immobilized rats. Cross-correlation (90-180 s lags) and spectral analysis were used to characterize correlated oscillations. Thirty-eight percent of pairs recorded in baseline (n=50) demonstrated correlated multisecond oscillations. Phase relationships were near 0 or 180 degrees. DA agonist injection significantly increased the incidence of correlation (intra- and interhemispheric) to 94% (n=17). After DA agonist injection, phase relationships of globus pallidus/substantia nigra neuron pairs were exclusively concentrated near 180 degrees, and phases of interhemispheric pairs of globus pallidus neurons were concentrated near 0 degrees. After subthalamic nucleus lesion (n=8), the incidence of correlated multisecond oscillations (or of multisecond oscillations per se) was not changed, although the consistent phase relationship between the globus pallidus and substantia nigra pars reticulata was disrupted. Subthalamic lesion also blocked apomorphine-induced decreases in oscillatory period and increases in oscillation amplitude, and significantly attenuated apomorphine-induced changes in mean firing rate. The data demonstrate that multisecond oscillations in the basal ganglia can be correlated between nuclei, and that DA receptor activation increases the level of correlation and organizes internuclear phase relationships at these multisecond time scales. While the subthalamic nucleus is not necessary for generating or transmitting these slow oscillations, it is involved in DA agonist-induced modulation of mean firing rate, oscillatory period, and internuclear phase relationship. These data further support a role for DA in modulating coherent oscillatory activity in the basal ganglia, and for the subthalamic nucleus in shaping the effects of DA receptor stimulation on basal ganglia output.

Cocaine or selective block of dopamine transporters influences multisecond oscillations in firing rate in the globus pallidus.

Previous studies have shown that direct-acting dopamine agonists modulate the multisecond oscillations which are present in globus pallidus spike trains in vivo in awake rats. To investigate possible modulation by endogenous dopamine and by other monoamines, and by drugs with abuse potential, cocaine or selective monoamine uptake blockers were injected systemically during extracellular recording of single globus pallidus neurons and the results analyzed with spectral and wavelet methods. Both cocaine and the selective dopamine uptake blocker GBR-12909 significantly shortened the period of multisecond oscillations, as well as increasing overall firing rate. Cocaine effects were blocked by dopamine antagonist pretreatment, as well as by N-methyl-D-aspartate receptor antagonist (MK-801) pretreatment. Desipramine and fluoxetine (blockers of norepinephrine and serotonin uptake, respectively) had no significant effects on multisecond oscillations. The results suggest that dopamine has a primary role among monoamines in modulating multisecond oscillations in globus pallidus activity, and that tonic dopaminergic and glutamatergic transmission is necessary for normal slow oscillatory function.

Drugs used in the treatment of attention-deficit/hyperactivity disorder affect postsynaptic firing rate and oscillation without preferential dopamine autoreceptor action.

BACKGROUND: Current theories propose that low doses of catecholaminergic stimulants reduce symptoms in patients with attention-deficit/hyperactivity disorder by acting on autoreceptors to reduce catecholaminergic transmission; few data are available that directly address this hypothesis. METHODS: We investigated the autoreceptor and postsynaptic receptor actions of systemically administered stimulants on dopaminergic systems in rats with single-unit recording in the substantia nigra pars compacta and globus pallidus, respectively. RESULTS: Dose-response curves for rate indicated that the potencies of the indirect-acting agonists methylphenidate and D-amphetamine at dopaminergic autoreceptors were not greater than at postsynaptic receptors; in fact, D-amphetamine was more potent postsynaptically. In addition to effects on firing rate, spectral/wavelet analyses indicated that these drugs had prominent effects on postsynaptic multisecond oscillations. These oscillations were shifted by stimulants from baseline periods of approximately 30 sec to periods of 5-10 sec. Effects on pattern were found at doses as low as 1.0 mg/kg (methylphenidate) and 0.2 mg/kg (D-amphetamine). At this latter dose, D-amphetamine had little effect presynaptically. CONCLUSIONS: These and prior results demonstrate that there is no autoreceptor-preferring dose range of catecholaminergic stimulants; these drugs at low doses are unlikely to reduce motor activity by this mechanism. Nonetheless, they might affect attentive and cognitive processes by modulating multisecond temporal patterns of central activity.

Pre- and postsynaptic aspects of dopamine-mediated transmission.

Dopamine agonist administration induces changes in firing rate and pattern in basal ganglia nuclei that provide an insight into the role of dopamine in basal ganglia function. These changes support a more complex, integrated basal ganglia network than envisioned in early models. Functionally important effects on basal ganglia output involve alterations in burstiness, synchronization and oscillatory activity,as well as rate. Multisecond oscillations in basal ganglia firing rates are markedly affected by systemic administration of dopamine-receptor agonists. This suggests that coordinated changes in neuronal activity at time scales longer than commonly investigated play a role in the cognitive and motor processes that are modulated by dopamine.

Multisecond oscillations in firing rate in the globus pallidus: synergistic modulation by D1 and D2 dopamine receptors.

The firing rates of many basal ganglia neurons recorded in awake rats oscillate at seconds-to-minutes time scales, and the D1/D2 agonist apomorphine has been shown to robustly modulate these oscillations. The use of selective D1 and D2 antagonists suggested that both these receptor subfamilies are involved in apomorphine's effects. In the present study, spectral analysis revealed that baseline multisecond oscillations were significantly periodic in 71% of globus pallidus neurons. Baseline oscillations had a wide range of periods within the analyzed range, with a population mean of 32 +/- 2 s. Administration of the D1 agonist SKF 81297 (6-chloroPB) at 1.0 or 5.0 mg/kg significantly changed these oscillations, reducing means of spectral peak periods to 14 to 16 s (i.e., increasing oscillatory frequency). This effect was attenuated by D2 antagonist pretreatment. The D2 agonist quinpirole did not cause a significant population change in multisecond periodicities. The strongest effects on multisecond periodicities occurred after combined treatment with SKF 81297 and quinpirole. Low, ineffective doses of SKF 81297 and quinpirole, when combined, produced a significant increase in oscillatory frequency. Also, when quinpirole was administered after an already effective dose of SKF 81297, quinpirole shifted oscillations to an even faster range (typically to periods of <10 s). The dopaminergic control of multisecond periodicities in globus pallidus firing rate demonstrates D1/D2 receptor synergism, in that the effects of D1 agonists are potentiated by and partially dependent on D2 receptor activity. Modulation of multisecond oscillations in firing rate represents a novel means by which dopamine can influence globus pallidus physiology.

Dopamine agonist-mediated rotation in rats with unilateral nigrostriatal lesions is not dependent on net inhibitions of rate in basal ganglia output nuclei.

Current models of basal ganglia function predict that dopamine agonist-induced motor activation is mediated by decreases in basal ganglia output. This study examines the relationship between dopamine agonist effects on firing rate in basal ganglia output nuclei and rotational behavior in rats with nigrostriatal lesions. Extracellular single-unit activity ipsilateral to the lesion was recorded in awake, locally-anesthetized rats. Separate rats were used for behavioral experiments. Low i.v. doses of D1 agonists (SKF 38393, SKF 81297, SKF 82958) were effective in producing rotation, yet did not change average firing rate in the substantia nigra pars reticulata or entopeduncular nucleus. At these doses, firing rate effects differed from neuron to neuron, and included increases, decreases, and no change. Higher i.v. doses of D1 agonists were effective in causing both rotation and a net decrease in rate of substantia nigra pars reticulata neurons. A low s.c. dose of the D1/D2 agonist apomorphine (0.05 mg/kg) produced both rotation and a robust average decrease in firing rate in the substantia nigra pars reticulata, yet the onset of the net firing rate decrease (at 13-16 min) was greatly delayed compared to the onset of rotation (at 3 min). Immunostaining for the immediate-early gene Fos indicated that a low i.v. dose of SKF 38393 (that produced rotation but not a net decrease in firing rate in basal ganglia output nuclei) induced Fos-like immunoreactivity in the striatum and subthalamic nucleus, suggesting an activation of both inhibitory and excitatory afferents to the substantia nigra and entopeduncular nucleus. In addition, D1 agonist-induced Fos expression in the striatum and subthalamic nucleus was equivalent in freely-moving and awake, locally-anesthetized rats. The results show that decreases in firing rate in basal ganglia output nuclei are not necessary for dopamine agonist-induced motor activation. Motor-activating actions of dopamine agonists may be mediated by firing rate decreases in a small subpopulation of output nucleus neurons, or may be mediated by other features of firing activity besides rate in these nuclei such as oscillatory firing pattern or interneuronal firing synchrony. Also, the results suggest that dopamine receptors in both the striatum and at extrastriatal sites (especially the subthalamic nucleus) are likely to be involved in dopamine agonist influences on firing rates in the substantia nigra pars reticulata and entopeduncular nucleus.

Multisecond oscillations in firing rate in the basal ganglia: robust modulation by dopamine receptor activation and anesthesia.

Multisecond oscillations in firing rate in the basal ganglia: robust modulation by dopamine receptor activation and anesthesia. Studies of CNS electrophysiology have suggested an important role for oscillatory neuronal activity in sensory perception, sensorimotor integration, and movement timing. In extracellular single-unit recording studies in awake, immobilized rats, we have found that many tonically active neurons in the entopeduncular nucleus (n = 15), globus pallidus (n = 31), and substantia nigra pars reticulata (n = 31) have slow oscillations in firing rate in the seconds-to-minutes range. Basal oscillation amplitude ranged up to +/-50% of the mean firing rate. Spectral analysis was performed on spike trains to determine whether these multisecond oscillations were significantly periodic. Significant activity in power spectra (in the 2- to 60-s range of periods) from basal spike trains was found for 56% of neurons in these three nuclei. Spectral peaks corresponded to oscillations with mean periods of approximately 30 s in each nucleus. Multisecond baseline oscillations were also found in 21% of substantia nigra dopaminergic neurons. The dopamine agonist apomorphine (0.32 mg/kg iv, n = 10-15) profoundly affected multisecond oscillations, increasing oscillatory frequency (means of spectral peak periods were reduced to approximately 15 s) and increasing the regularity of the oscillations. Apomorphine effects on oscillations in firing rate were more consistent from unit to unit than were its effects on mean firing rates in the entopeduncular nucleus and substantia nigra. Apomorphine modulation of multisecond periodic oscillations was reversed by either D1 or D2 antagonists and was mimicked by the combination of selective D1 (SKF 81297) and D2 (quinpirole) agonists. Seventeen percent of neurons had additional baseline periodic activity in a faster range (0.4-2.0 s) related to ventilation. Multisecond periodicities were rarely found in neurons in anesthetized rats (n = 29), suggesting that this phenomenon is sensitive to overall reductions in central activity. The data demonstrate significant structure in basal ganglia neuron spiking activity at unexpectedly long time scales, as well as a novel effect of dopamine on firing pattern in this slow temporal domain. The modulation of multisecond periodicities in firing rate by dopaminergic agonists suggests the involvement of these patterns in behaviors and cognitive processes that are affected by dopamine. Periodic firing rate oscillations in basal ganglia output nuclei should strongly affect the firing patterns of target neurons and are likely involved in coordinating neural activity responsible for motor sequences. Modulation of slow, periodic oscillations in firing rate may be an important mechanism by which dopamine influences motor and cognitive processes in normal and dysfunctional states.

N-methyl-D-aspartate receptor blockade attenuates D1 dopamine receptor modulation of neuronal activity in rat substantia nigra.

It has been proposed that dopamine and glutamate affect basal ganglia output, in part, through interactions between D1 receptors and NMDA receptors. The present study examined whether N-methyl-D-aspartate (NMDA) receptor antagonists affect the neurophysiological responses of substantia nigra pars compacta (SNpc; dopaminergic) and pars reticulata (SNpr; non-dopaminergic) neurons to a systemically administered D1 dopamine agonist in two animals models of Parkinson's disease, reserpine treatment and nigrostriatal lesion. Previous studies using extracellular single unit recording techniques have shown that the D1 dopamine agonist SKF 38393 (10 mg/kg) exerts different effects on the firing rates of SNpr neurons after these two dopamine-depleting treatments, suggesting the involvement of multiple mechanisms. SKF 38393 consistently increased the firing rates of SNpr neurons in rats treated subchronically with reserpine, and markedly decreased SNpr firing rates in rats with nigrostriatal damage. Pretreatment with the non-competitive NMDA antagonist MK-801 (0.15 mg/kg i.v.) blocked, and the competitive NMDA antagonist (+/-)-CPP (30 mg/kg i.p.) attenuated, the rate effects of SKF 38393 in both dopamine-depleted preparations. SKF 38393 consistently inhibited the firing rate of SNpc dopamine neurons after acute reserpine treatment (10 mg/kg, 4-7 hours), an effect specifically mediated by D1 receptors. Pretreatment with MK-801 (0.1 mg/kg i.v.) or the competitive NMDA antagonist (+)-HA-966 (30 mg/kg i.v.) also effectively attenuated SKF 38393's inhibitory effect on SNpc dopamine neurons. Therefore, NMDA receptor blockade markedly reduces the ability of D1 receptor stimulation to modulate firing rates of both dopaminergic and non-dopaminergic cells in the substantia nigra. Although multiple mechanisms appear to underlie D1-mediated effects on substantia nigra firing rates in reserpine and 6-OHDA-treated rats, these results demonstrate a common dependence on glutamatergic transmission and a permissive role for NMDA receptor activation in the ability of D1 receptor stimulation to both enhance and reduce neuronal activity in the substantia nigra.

Effects of full D1 dopamine receptor agonists on firing rates in the globus pallidus and substantia nigra pars compacta in vivo: tests for D1 receptor selectivity and comparisons to the partial agonist SKF 38393.

Many studies have used the D1 agonist SKF 38393 to characterize D1 receptor influences on firing rates in basal ganglia nuclei in vivo. However, SKF 38393 is a partial agonist and so may not be ideal for delineating D1 receptor effects. This study characterizes the effects of four full D1 agonists, SKF 82958 (chloro-APB), SKF 81297 (6-chloro-PB), dihydrexidine and A-77636, on the firing rates of midbrain dopamine and globus pallidus neurons. Recordings were done in fully anesthetized or paralyzed, locally anesthetized rats, and drugs were given systemically intravenously. Dihydrexidine, SKF 81297 and A-77636 were free of rate effects on midbrain dopamine neurons (up to 10.2 mg/kg) and also did not antagonize the inhibitory effects of quinpirole. In contrast, SKF 82958 strongly inhibited dopamine cells through activation of D2 autoreceptors (ED50 = 0.70 mg/kg). Of these drugs, SKF 82958 also was the only one to increase pallidal unit firing rates when given alone (at 5.0 but not 1.0 mg/kg); the other compounds appeared to be selective for postsynaptic D1 receptors. The results suggest that SKF 82958 may be more properly classified as a mixed D1/D2 agonist. In addition, all four agonists strongly potentiated the pallidal response to quinpirole, demonstrating a D1 receptor potentiation of D2 receptor effects. The results support the role of D1 receptors in the midbrain and globus pallidus as previously characterized with SKF 38393. The similar actions of partial and full D1 agonists in these systems support evidence for a D1 receptor reserve and possibly an effector system other than adenylate cyclase.

Differing influences of dopamine agonists and antagonists on Fos expression in identified populations of globus pallidus neurons.

Dopamine agonists increase the activity of globus pallidus neurons, as shown electrophysiologically and with Fos expression. More recently it has been shown that decreased D2 receptor activity also causes pallidal Fos expression. Similar responses occur in the striatum, where both dopamine agonists and D2 blockade induce Fos, although in separate neuronal populations (i.e. striatonigral and -pallidal). The present experiments investigate the possible differential dopaminergic regulation of Fos within pallidal neuronal populations, which were classified as parvalbumin-positive or -negative (with parvalbumin immunostaining), or as projecting to various target nuclei (with retrograde transport of FluoroGold iontophoresed into these nuclei). Rats with prior nigrostriatal lesions received saline, D1 agonist, or D2 agonist. Rats with no lesions received saline, combined D1/D2 agonists, or the D2 antagonist eticlopride. Two hours after drug injection, rats were perfused and their brains processed for double-labeling: either Fos staining with parvalbumin staining, or Fos or parvalbumin staining in FluoroGold-labeled sections. Overall, dopamine drug treatments induced more Fos in parvalbumin-negative than -positive cells. However, unlike dopamine agonists, eticlopride induced significant Fos only in the parvalbumin-negative cells. Dopamine agonist-induced Fos was found in pallidal neurons projecting to each of the target nuclei examined, in both normal and nigrostriatal-lesioned rats. Eticlopride-induced Fos occurred only in pallidal neurons projecting to the striatum, providing functional evidence for pallidostriatal cells without axon collaterals to other nuclei. It was also found that pallidostriatal neurons were distinguished from other projection populations by a relative lack of parvalbumin immunoreactivity. Pallidal cells respond heterogeneously to dopaminergic treatments based on their projection target and expression of parvalbumin. The smaller Fos responses in parvalbumin-containing cells may be due largely to the calcium buffering by the parvalbumin itself. Also, the pattern of Fos expression in pallidostriatal neurons suggests that dopamine regulates activity in these cells differently than in other projection populations.

Cerebrocortical Fos expression following dopaminergic stimulation: D1/D2 synergism and its breakdown.

Using standard immunohistochemical techniques, we examined Fos expression in different areas and layers of cerebral cortex in rats following combined or separate stimulation of dopamine D1 and D2 receptors under normal conditions and following five days of reserpine (1 mg/kg/day), a treatment that causes a breakdown in requisite D1/D2 synergism. In normal animals, combined but not separate stimulation of D1 and D2 receptors elicited Fos expression in frontal and parietal, but not cingulate, cortex. Expression was highest in layer IV of primary somatosensory cortex; in frontal and secondary somatosensory cortex, Fos expression was lower and peaked in layer VI. Cortical Fos expression following amphetamine showed the same general pattern, and was blocked by either a selective D1 or D2 antagonist. Following reserpine treatment, stimulation of either D1 or D2 receptors gave rise to cortical Fos expression in patterns similar to each other and to combined D1/D2 stimulation in normal rats (except in frontal cortex in which separate D1 or D2 stimulation was unable to elicit Fos even following repeated reserpine treatment). The fact that cortical Fos expression was tightly associated with behavioral activation together with its laminar and areal distribution suggest that sensory input resulting from behavioral activation may be an important stimulus for this immediate-early gene response.

D1 dopamine receptors influence Fos immunoreactivity in the globus pallidus and subthalamic nucleus of intact and nigrostriatal-lesioned rats.

Studies of the globus pallidus (GP) and subthalamic nucleus (STN) have emphasized the role of D2 dopamine receptors, although effects of D1 receptor activation on GP firing rate and STN metabolism have been reported, especially in rats with nigrostriatal lesions. This study systematically investigated the effects of D1 and D2 receptor activation on the activity of the GP and STN in intact and 6-OHDA-lesioned rats using immunostaining for the immediate-early gene Fos. In intact rats, the D1 agonist SKF 38393 (20.0 mg/kg) produced a five-fold potentiation of the GP Fos expression due to the D2 agonist quinpirole produced significant Fos expression. In rats with prior nigrostriatal lesions, SKF 38393 (4.0 or 20.0 mg/kg) increased Fos immunostaining in both the GP and STN, while quinpirole increased it only in the GP. SKF 38393 effects in the GP and STN of nigrostriatal-lesioned rats were blocked completely by SCH 23390, and unaffected by eticlopride. These results are a novel demonstration of control of Fos expression by dopaminergic drugs in the STN and by D1 agonists in the GP.

Amphetamine- and cocaine-induced fos in the rat striatum depends on D2 dopamine receptor activation.

Amphetamine or cocaine injection causes expression of the immediate-early gene c-fos in the striatum. Previous studies have shown that dopamine D1 receptor activation is necessary for this effect, but have not established a consistent role for D2 receptors. We have investigated the involvement of D2 receptors in indirect dopamine agonist-induced striatal Fos-like immunoreactivity using the selective D2 antagonist eticlopride. Eticlopride treatment (0.5 mg/kg) caused Fos expression by itself, but also decreased Fos expression in the central striatum due to amphetamine (5.0 mg/kg) or cocaine (40 mg/kg) by 90% and 85%, respectively. In striatonigral neurons, identified by labeling with the retrograde tracer Fluorogold iontophoresed into the substantia nigra pars reticulata, the blockade of stimulant-induced Fos-like immunofluorescence by eticlopride was nearly complete, with decreases of 98% for amphetamine and 94% for cocaine. In striatonigral neurons, the D2 antagonist alone had minimal effect. We conclude that activation of both D1 and D2 receptor classes by dopamine agonists is necessary for induction of Fos in the striatonigral cells of normal rats. These results provide an important parallel to behavioral and electrophysiological work that also demonstrates D1/D2 interdependence in the control of normal basal ganglia functions.