The distribution of neurons in the substantia nigra pars reticulata with input from the motor, premotor and prefrontal areas of the cerebral cortex in monkeys

We investigated the distribution of neurons in the substantia nigra pars reticulata (SNr) which received cortical input. The activities of single SNr neurons were studied extracellulary in awake monkeys. SNr neurons showed excitatory and/or inhibitory responses to cortical stimulation. These responses were considered to be mediated by the subthalamic nucleus and striatum, respectively. The neurons receiving inhibitory input from the motor, premotor and supplementary motor areas (Motor-related cortical areas) were located in the lateral part of the SNr, whereas those with input from the medial, dorsal and orbital areas of the prefrontal cortex (PFmdo) were frequently found in the rostro-medial part of this nucleus. SNr neurons with inhibitory input from the ventral periprincipal area (PSv) were mainly distributed in the intermedio-lateral portion, with some degree of overlap with input from other cortical areas. The distribution of the excitatory input was almost similar to that of inhibitory one, but the excitatory input from the PSv was much stronger than that from the PFmdo. Some SNr neurons receiving cortical input were proved to project to the thalamus. Our results support the existence of several parallel organization of the cortico-basal ganglia loop circuits [G.E. Alexander, M.R. DeLong, P.L. Strick, Parallel organization of functionally segregated circuits linking basal ganglia and cortex, Ann. Rev. Neurosci., 9, 1986, pp. 357-381.], but interaction between the loops can not be ignored.     

Evidence that the parafascicular projection to the subthalamic nucleus is glutamatergic

The parafascicular nucleus projection to the subthalamic neurones has an excitatory synaptic effect. We have examined the possible glutamatergic mediation of this pathway. The initial excitatory response elicited by electrical stimulation of the parafascicular neurones was inhibited by a microinjection of excitatory amino acid receptor antagonists into the subthalamic nucleus. The antagonists were the broad spectrum kynurenic acid, the NMDA selective antagonist d-AP-5 and the AMPA antagonist CNQX. Their effects were dose-dependent and reversible. The results suggest that the excitatory effect of the parafascicular neurones is mediated by AMPA and NMDA receptors.     

Role of the subthalamo-nigral input in the control of amygdala-kindled seizures in the rat

The substantia nigra pars reticulata (SNpr) is a critical site for the control of epileptic seizures. Potentiation of the inhibitory GABAergic input from the striatum to the SNpr suppresses primary or secondary generalized seizures in the rat. The purpose of this study was to examine the possible involvement of the excitatory glutamatergic input from the subthalamic nucleus to the SNpr in the control of both the electroencephalographic and the motor components of amygdala-kindled seizures in the rat. Microinjections of either an N-methyl-D-aspartate (NMDA) antagonist in the substantia nigra or a GABAA agonist in the subthalamic nucleus, significantly reduced motor seizures but did not modified the afterdischarges. These results provide evidence for the involvement of the subthalamo-nigral projection in the modulation and the propagation of the motor components of amygdala-kindled seizures.     

Electrophysiological and Fos immunohistochemical evidence for the excitatory nature of the parafascicular projection to the globus pallidus

Extracellular recordings and immunohistological detection of c-Fos-like immunoreactive proteins were used to determine the synaptic effect of the parafascicular projection to the globus pallidus. Electrical stimulation of the parafascicular neurons induced a single-spike excitatory response with a stable latency of 2.3 ms, suggesting a monosynaptically driven effect. Pharmacological stimulation of the parafascicular nucleus with carbachol increased tonically the pallidal discharge rate by 142%. The discharge rate of the pallidal neurons was described by 37% in parafascicular-lesioned rats. These results demonstrate the excitatory nature and the tonic action of the parafasciculopallidal projection. Carbachol activation of parafascicular neurons also induced the synthesis of c-Fos-like immunoreactive proteins in the pallidal neurons. Control experiments in subthalamic-lesioned rats showed that the parafascicular excitation of the pallidal neurons remained, but both electrophysiological and expression of c-Fos-like immunoreactive proteins were attenuated. This suggests that the direct parafascicular excitation of the pallidal neurons is indirectly reinforced by the previously described parafascicular excitatory input to the subthalamic nucleus. Conversely, the effect of this last input to the subthalamic nucleus is dramatically enhanced in rats with pallidal lesion. Our results demonstrate the complex role of the parafascicular nucleus in activating both the globus pallidus and the subthalamic nucleus, two closely related structures. These results illustrate the integrative capacities of the globus pallidus, whose activity is modulated by multiple afferents.     

High-performance liquid chromatographic analysis of the new antitumour drug N-benzoylstaurosporine (CGP 41 251) and four potential metabolites in micro-volumes of plasma

Molecular events underlying the mechanism by which brain injury elicits delayed transneuronal degeneration of neurons remote from the site of initial injury are not well understood. In rats, acute injury of the caudate nucleus (CN) and globus pallidus (GP) by local injection of excitotoxic ibotenic acid (IA) or by transient forebrain ischemia resulted in delayed cell death of neurons in the substantia nigra reticulata (SNr). To elucidate the involvement of glutamate receptor mediated hyperactivity of neurons produced by loss of inhibitory inputs in this delayed degeneration of SNr neurons, the region-specific expression of an immediate early gene, c-fos, and the effect of glutamate receptor antagonists on the c-fos expression were examined by using immunocytochemical and in situ hybridization analysis. Following unilateral IA-injection into the CN and GP, a robust expression of c-fos mRNA and Fos protein was induced specifically in neurons of both subthalamic nucleus (STN) and SNr deafferented by the IA-lesions 36 h after IA-injection. The delayed expression of Fos-protein in SNr neurons lasted for 48 h longer than that in STN neurons. Following unilateral IA-injection confined to the CN, an intense but short-term expression of Fos-protein was exhibited only in neurons of the deafferented SNr. c-fos mRNA and Fos protein were not expressed in neurons of the substantia nigra compacta at any time points examined. The induction of c-fos mRNA and Fos protein in neurons of the STN and SNr following IA-lesions of the CN and GP was reduced markedly by non-NMDA receptor antagonist (GYKI52466), but not by NMDA receptor antagonist (MK-801). The region-specific c-fos expression implies that deprivation of inhibitory afferents (disinhibition) due to destruction of presynaptic neurons can induce increased activity of postsynaptic neurons. The effect of GYKI52466 on the c-fos gene expression in neurons of the deafferented STN and SNr suggests that activation of non-NMDA receptors may be involved in a pathophysiological cascade for the transneuronal degeneration of SNr neurons.     

Relationships between the prefrontal cortex and the basal ganglia in the rat: physiology of the corticosubthalamic circuits

The prelimbic-medial orbital areas (PL/MO) of the prefrontal cortex are connected to the medial part of the subthalamic nucleus (STN) through a direct projection and an indirect circuit that involves the core of the nucleus accumbens (NAcc) and the ventral pallidum (VP). In the present study, the influence of the PL/MO on the discharge of STN cells has been characterized. The major pattern of the responses observed after stimulation of PL/MO consisted of two excitatory peaks often separated by a brief inhibitory period. The early excitation was most likely to be caused by the activation of direct cortical inputs because its latency matches the conduction time of the prefrontal STN projections. The late excitation resulted from the activation of the indirect PL/MO-STN pathway that operates through a disinhibitory process. Indeed, the late excitation was no longer observed after acute blockade of the glutamatergic corticostriatal transmission by CNQX application into the NAcc. A similar effect was obtained after the blockade of the GABAergic striatopallidal transmission by bicuculline application into the VP. Finally, the brief inhibition that followed the early excitation was likely to result from the activation of a feedback inhibitory loop through VP because this inhibition was no longer observed after the blockade of STN inputs by CNQX application into the VP. This study further indicates the implication of STN in prefrontal basal ganglia circuits and underlines that in addition to a direct excitatory input, medial STN receives an indirect excitatory influence from PL/MO through an NAcc-VP-STN disinhibitory circuit.     

A massively connected subthalamic nucleus leads to the generation of widespread pulses

A composite model of the subthalamic nucleus is developed from physiological and anatomical considerations. First, study of a geometric model of the anatomical arrangements of projection neurons within the nucleus indicates that they form a massively connected network. Second, given the excitatory nature of these neurons, their threshold and peak firing rates, a simple model of neuron responses reveals that large regions of this highly interconnected nucleus can respond to excitatory input in the form of a wide-spread uniform pulse. Such widespread pulses of activity may act as a braking signal that resets the major basal ganglia output nuclei.     

Position of the ventral pallidum in the rat prefrontal cortex-basal ganglia circuit

The ventral pallidum receives major inputs from the nucleus accumbens, a striatal region related to the prefrontal cortex. The ventral pallidum, through its projections to the mediodorsal nucleus of the thalamus, has been considered as the main output structure of the prefrontal-basal ganglia circuits. However, as shown recently, the ventral pallidum also sends efferents to the subthalamic nucleus and the substantia nigra, suggesting that it could participate in intrinsic basal ganglia circuits. The aim of the present investigation was to determine the position of the ventral pallidum in the prefrontal-basal ganglia circuit originating from the prelimbic and medial orbital areas. Following injections of biocytin (an anterograde tracer) into the region of the core of the nucleus accumbens receiving excitatory inputs from the prelimbic and medial orbital areas, axonal terminal fields were observed in a delineated dorsal region of the ventral pallidum. When the biocytin injections were made into this ventral pallidal region, anterogradely labelled fibres were observed in both the dorsomedial substantia nigra pars reticulata and the medial subthalamic nucleus, but not in the mediodorsal nucleus of the thalamus. Confirming these anatomical observations, electrical stimulation of the core of the nucleus accumbens induced an inhibition of the spontaneous activity (D=34.9+/-13.3 ms, L=9.2+/-3.3 ms) in 46.5% of the ventral pallidal cells. Among these responding cells, 43% were antidromically driven from the subthalamic nucleus, 30% from the substantia nigra pars reticulata and only 6% from the mediodorsal nucleus of the thalamus. These data demonstrate that the region of the ventral pallidum involved in the prefrontal cortex-basal ganglia circuit originating from the prelimbic and medial orbital areas represents essentially a ventral subcommissural extension of the external segment of the globus pallidus since it exhibits similar extrinsic connections and functional characteristics. In conclusion, in this prelimbic and medial orbital channel, the ventral pallidum cannot be considered as a major output structure but is essentially involved in intrinsic basal ganglia circuits.     

Tri-iodothyronine inhibits multilayer formation of the osteoblastic cell line, MC3T3-E1, by promoting apoptosis

Within the basal ganglia circuitry, recent conceptions of the subthalamic nucleus are that it fulfils integrative functions. We have previously shown that bilateral excitotoxic lesions of the subthalamic nucleus induce behavioural deficits in a five-choice serial reaction time task in the rat, consistent with attentional impairments and suggesting important roles of this basal ganglia structure in mechanisms of behavioural control. In the present study, we tested the effects of (i) blocking its excitatory inputs (originating mainly in the cerebral cortex and the parafascicular nucleus of the thalamus) via the NMDA receptors and (ii) stimulating its GABA receptors to mimick the influence of its inhibitory inputs (mainly from the globus pallidus). Bilateral microinfusions of APV (NMDA receptor antagonist) or muscimol (GABA-A receptor agonist) into the subthalamic nucleus were administered to rats trained in the same five-choice serial reaction time task. Both APV (0.125-0.5 microg) and muscimol (1-3 ng) reduced choice accuracy, slowed correct responses and increased omissions and perseverative responses. Premature responses tended to increase after APV but decrease after muscimol. Increased perseverations at the food magazine occurred only after muscimol infusions. These results reproduce many of the effects of lesions of the STN and are consistent with an integrative role for this structure in pallidal and thalamo-cortical processing.     

Complex pharmacology of natural cannabinoids: evidence for partial agonist activity of delta9-tetrahydrocannabinol and antagonist activity of cannabidiol on rat brain cannabinoid receptors

Delta9-tetrahydrocannabinol (delta9-THC), cannabinol and cannabidiol are three important natural cannabinoids from the Marijuana plant (Cannabis sativa). Using [35S]GTP-gamma-S binding on rat cerebellar homogenate as an index of cannabinoid receptor activation we show that: delta9-THC does not induce the maximal effect obtained by classical cannabinoid receptor agonists such as CP55940. Moreover at high concentration delta9-THC exhibits antagonist properties. Cannabinol is a weak agonist on rat cerebellar cannabinoid receptors and cannabidiol behaves as an antagonist acting in the micromolar range.     

Increase in meso-prefrontal dopaminergic activity after stimulation of CB1 receptors by cannabinoids

The intravenous administration of the psychoactive constituent of marijuana, delta9-tetrahydrocannabinol (delta9-THC) (62.5-1000 microg/kg), and the synthetic cannabinoid agonist WIN 55212,2 (WIN) (62.5-500 microg/kg), produced a dose-related increase in the firing rate and burst firing in the majority of antidromically identified meso-prefrontal dopaminergic neurons. In a restricted number of neurons (n=4), WIN administration did not increase firing rate but produced an increment of bursting activity. These effects of the cannabinoids were reversed by the intravenous administration of SR 141716 A, a selective cannabinoid antagonist (1 mg/kg), per se ineffective to modify the electrical activity of dopaminergic neurons. The results indicate that stimulation of cannabinoid CB1 receptors produces an activation of meso-prefrontal dopaminergic transmission. Considering that supranormal stimulation of D1 dopamine receptors in the prefrontal cortex has been shown to impair working memory, the present results suggest that the negative effects of cannabinoids on cognitive processes might be related to the activation of dopaminergic transmission in the prefrontal cortex.     

Metabotropic glutamate receptors in the rat nucleus accumbens

The effects of glutamate metabotropic receptors (mGluRs) on excitatory transmission in the nucleus accumbens were investigated using electrophysiological techniques in rat nucleus accumbens slices. The broad-spectrum mGluR agonist (1S,3R)-1-aminocyclopentyl-1,3-dicarboxylate, the mGluR group 2 selective agonists (S)-4-carboxy-3-hydroxyphenylglycine, (1S,3S)-ACPD) and (2S,1'S,2'S)-2-(2'-carboxycyclopropyl)glycine (L-CCG1), and the mGluR group 3 specific agonist L-2-amino-4-phosphonobutyrate (L-AP4) all reversibly inhibited evoked excitatory synaptic responses. The specific group 1 mGluR agonist (R,S)-3,5-dihydroxyphenylglycine [(R,S)-DHPG] did not depress transmission. Dose-response curves showed that the rank order of agonist potencies was: L-CCG1 > L-AP4 > (1S,3S)-ACPD. Group 2 and 3 mGluRs inhibited transmission via a presynaptic mechanism, as they increased paired-pulse facilitation, decreased the frequency of miniature excitatory postsynaptic currents and had no effect on their amplitude. The mGluRs did not inhibit transmitter release by reducing voltage-dependent Ca2+ currents through N- or P-type Ca2+ channels, as inhibition persisted in the presence of omega-conotoxin-GVIA or omega-Aga-IVA. The depression induced by mGluRs was not affected by specific antagonists of dopamine D1, GABA-B or adenosine A1 receptors, indicating direct effects. Finally, (R,S)-DHPG specifically blocked the postsynaptic afterhyperpolarization current (I(AHP)). Our results represent the first direct demonstration of functional mGluRs in the nucleus accumbens of the rat.     

Excitation of rat subthalamic nucleus neurones in vitro by activation of a group I metabotropic glutamate receptor

The subthalamic nucleus (SThN) provides a glutamate mediated excitatory drive to several other component nuclei of the basal ganglia, thereby significantly influencing locomotion and control of voluntary movement. We have characterised functionally the metabotropic glutamate (mGlu) receptors in the SThN using extracellular single unit recording from rat midbrain slices. SThN neurones fired action potentials spontaneously at a rate of 10 Hz which was increased by the group I/II mGlu receptor agonist (1S,3R)-1-aminocyclopentane-1,3-dicarboxylate (1S,3 R-ACPD; 1-30 microM) and the group I selective agonist (S, R)-dihydroxyphenylglycine (DHPG; 1-30 microM). However, both the group II selective agonist (1S,1'R,2'R,3'R)-2-(2,3-dicarboxycyclopropyl)glycine (DCG-IV; 1 microM) and the group III selective agonist (S)-2-amino-4-phosphonobutanoic acid (L-AP4; 10 microM) were without effect, indicating that the excitation was mediated by a group I mGlu receptor. The excitation caused by DHPG (3 microM) was reversed by co-application of the mGlu receptor antagonist (+)-alpha-methyl-4-carboxyphenylglycine (MCPG; 500 microM). Thus a group I mGlu receptor mediates excitation of SThN neurones, and suggests a use for group I mGlu receptor ligands for treatment of both hypo- and hyperkinetic disorders of basal ganglia origin, such as Parkinson's disease and Huntington's disease.     

Regulated expression of 5''-deleted mouse GLUT4 minigenes in transgenic mice: effects of exercise training and high-fat diet

Striatal lesions are known to cause the anterograde transneuronal degeneration of the substantia nigra pars reticulata (SNr) neurons in consequence to loss of GABAergic inhibitory striatonigral efferents. The present study was undertaken to examine whether long-term intraventricular administration of the GABA agonist muscimol could promote reformation of the striatonigral pathway arising from transplants by rescuing host SNr neurons from transneuronal death in rats with striatal ischemic lesions. Compared to nongrafted rats with striatal lesions, (i) a prominent axonal projection from the transplants to the ipsilateral substantia nigra, (ii) a significant increase in number of survived neurons in the ipsilateral SNr, and (iii) a significant reduction in number of apomorphine-induced turning behaviors were found in grafted animals with muscimol infusion, but not in those without muscimol administration. These findings suggest that preservation of the host target neurons for grafted cells may increase an efficacy of cerebral implants in establishment of the host-graft fiber connections, possibly, leading to functional restoration.     

Serum alpha L-fucosidase enzyme activity in ovarian and other female genital tract tumors

The effects of delta 9-tetrahydrocannabinol on single-unit activity in the subpostremal division of the nucleus tractus solitarii were investigated by extracellular recording in rat brain slices. The spontaneous firing rate of 54.8% of the recorded neurons was significantly changed after bath applications of delta 9-tetrahydrocannabinol. Putative nutrition-related neurons responding to a moderate increase in glucose concentration were selectively sensitive to delta 9-tetrahydrocannabinol. The delta 9-tetrahydrocannabinol-sensitive neurons were depressed by clonidine and are therefore likely to be adrenergic or noradrenergic. These observations suggest that some catecholaminergic, glucose-responsive neurons in the subpostremal nucleus tractus solitarii might mediate the influence of cannabinoids on feeding behaviour. Furthermore, most delta 9-tetrahydrocannabinol-sensitive neurons in the nucleus tractus solitarii showed opposite responses to delta 9-tetrahydrocannabinol and the 5-HT3 receptor agonist 1-phenylbiguanide, and might therefore be involved in the nausea-reducing effects of cannabinoids.     

Suppression of the humoral immune response by cannabinoids is partially mediated through inhibition of adenylate cyclase by a pertussis toxin-sensitive G-protein coupled mechanism

Cannabinoid compounds, including the major psychoactive component of marihuana, delta 9-tetrahydrocannabinol (delta 9-THC), have been widely established as being inhibitory on a broad array of humoral and cell-mediated immune responses. The presence of cannabinoid receptors has been identified recently on mouse spleen cells, which possess structural and functional characteristics similar to those of the G-protein coupled cannabinoid receptor originally identified in rat brain. These findings, together with those demonstrating that delta 9-THC inhibits adenylate cyclase in splenocytes, strongly suggest that certain aspects of immune inhibition by cannabinoids may be mediated through a cannabinoid receptor-associated mechanism. The objective of the present studies was to determine whether inhibition of adenylate cyclase is relevant to mouse spleen cell immune function and, if so, whether this inhibition is mediated through a Gi-protein coupled mechanism as previously described in neuronal tissue. Spleen cell activation by the phorbol ester phorbol-12-myristate-13-acetate (PMA), plus the calcium ionophore ionomycin, produced a rapid but transient increase in cytosolic cAMP, which was inhibited completely by immunosuppressive concentrations of delta 9-THC (22 microM) and the synthetic bicyclic cannabinoid CP-55940 (5.2 microM), which produced no effect on cell viability. Inhibition by cannabinoids of lymphocyte proliferative responses to PMA plus ionomycin and sheep erythrocyte (sRBC) IgM antibody-forming cell (AFC) response, was abrogated completely by low concentrations of dibutyryl-cAMP (10-100 microM). Inhibition of the sRBC AFC response by both delta 9-THC (22 microM) and CP-55940 (5.2 microM) was also abrogated by preincubation of splenocytes for 24 hr with pertussis toxin (0.1-100 ng/mL). Pertussis toxin pretreatment of spleen cells was also found to directly abrogate cannabinoid inhibition of adenylate cyclase, as measured by forskolin-stimulated accumulation of intracellular cAMP. These results indicate that inhibition of the sRBC AFC response by cannabinoids is mediated, at least in part, by inhibition of adenylate cyclase through a pertussis toxin-sensitive Gi-protein coupled cannabinoid receptor. Additionally, these studies further support the premise that cAMP is an important mediator of lymphocyte activation.     

Histopathological characterization of magnetic resonance imaging-detectable brain white matter lesions in a primate model of multiple sclerosis: a correlative study in the experimental autoimmune encephalomyelitis model in common marmosets (Callithrix jacchus)

N-methyl-D-aspartate (NMDA) and non-NMDA receptor-mediated manipulations of the cortical cholinergic input arising from the basal forebrain differentially affect cognitive function. We used [14C]-2-deoxyglucose autoradiography in conscious rats to map the effects of excitatory amino acid agonist infusions into the nucleus basalis magnocellularis (NBM) on cerebral functional activity, as reflected by local rates of glucose utilization. Acute stimulation of NBM neurones by local infusion of alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), 15 min before glucose use measurement, resulted in glucose use reductions in nine cortical regions innervated by NBM efferents including prefrontal, frontal, sensorimotor and cingulate cortices. NMDA infusions altered glucose use in two cortical areas. Both AMPA and NMDA markedly increased glucose use in the striatum and globus pallidus, with concomitant perturbations in striato-pallidal projection targets including the substantia nigra, entopeduncular nucleus, subthalamic nucleus and lateral habenular nucleus. In contrast, the GABAA agonist muscimol did not affect glucose use in the NBM or neocortical regions, but induced glucose use increases in several subcortical nuclei including the substantia nigra and entopeduncular nucleus. The delayed effects of excitotoxic lesions were assessed 3 weeks after basal forebrain infusions of AMPA, NMDA, ibotenate or quisqualate. Statistically significant glucose use changes only occurred in the hypothalamus after NMDA, and the NBM after ibotenate infusions, although reduced cortical metabolism was apparent following AMPA-induced lesions of the NBM. Results support a dissociation between the functional sequelae of NMDA and non-NMDA receptor-mediated events in the basal forebrain, and long-term compensatory functional adaptation following cortical denervation.     

Cannabidiol and (-)Delta9-tetrahydrocannabinol are neuroprotective antioxidants

The neuroprotective actions of cannabidiol and other cannabinoids were examined in rat cortical neuron cultures exposed to toxic levels of the excitatory neurotransmitter glutamate. Glutamate toxicity was reduced by both cannabidiol, a nonpsychoactive constituent of marijuana, and the psychotropic cannabinoid (-)Delta9-tetrahydrocannabinol (THC). Cannabinoids protected equally well against neurotoxicity mediated by N-methyl-D-aspartate receptors, 2-amino-3-(4-butyl-3-hydroxyisoxazol-5-yl)propionic acid receptors, or kainate receptors. N-methyl-D-aspartate receptor-induced toxicity has been shown to be calcium dependent; this study demonstrates that 2-amino-3-(4-butyl-3-hydroxyisoxazol-5-yl)propionic acid/kainate receptor-type neurotoxicity is also calcium-dependent, partly mediated by voltage sensitive calcium channels. The neuroprotection observed with cannabidiol and THC was unaffected by cannabinoid receptor antagonist, indicating it to be cannabinoid receptor independent. Previous studies have shown that glutamate toxicity may be prevented by antioxidants. Cannabidiol, THC and several synthetic cannabinoids all were demonstrated to be antioxidants by cyclic voltametry. Cannabidiol and THC also were shown to prevent hydroperoxide-induced oxidative damage as well as or better than other antioxidants in a chemical (Fenton reaction) system and neuronal cultures. Cannabidiol was more protective against glutamate neurotoxicity than either ascorbate or alpha-tocopherol, indicating it to be a potent antioxidant. These data also suggest that the naturally occurring, nonpsychotropic cannabinoid, cannabidiol, may be a potentially useful therapeutic agent for the treatment of oxidative neurological disorders such as cerebral ischemia.     

The subthalamic nucleus and the external pallidum: two tightly interconnected structures that control the output of the basal ganglia in the monkey

The aim of the present study was to elucidate the organization of the interconnections between the subthalamic nucleus and the two segments of the globus pallidus in squirrel monkeys. By making small deposits of tracers in the two segments of the globus pallidus, we demonstrate that interconnected neurons of the subthalamic nucleus and the external pallidum innervate, via axon collaterals, the same population of neurons in the internal pallidum. Furthermore, this organizational principle holds true for different functional regions of the pallidum and the subthalamic nucleus. Injections of biotinylated dextran amine were made in the dorsal (associative), ventrolateral (sensorimotor) and rostromedial (limbic) regions of the internal pallidum. Following these injections, there were rich clusters of labelled terminals in register with retrogradely labelled perikarya in related functional regions of the subthalamic nucleus and the external pallidum. At the electron microscopic level, the majority of labelled terminals in the external pallidum displayed the ultrastructural features of boutons from the subthalamic nucleus and were non-immunoreactive for GABA, whereas those in the subthalamic nucleus resembled terminals from the external pallidum and displayed GABA immunoreactivity. In both cases, the synaptic targets of the labelled terminals included labelled neurons. These observations suggest that the biotinylated dextran amine injected in the internal globus pallidus was transported retrogradely to perikarya in the external pallidum and the subthalamic nucleus and then anterogradely, via axon collaterals, to the subthalamic nucleus and the external pallidum respectively. This suggestion was supported by injections of biotinylated dextran amine or Phaseolus vulgaris-leucoagglutinin in regions of the external pallidum that corresponded to those containing retrogradely labelled cells following injections in the internal pallidum. The clusters of labelled cells and varicosities that resulted from these injections were found in regions of the subthalamic nucleus similar to those labelled following injections in the internal globus pallidus. Furthermore, terminals from the external pallidum and the subthalamic nucleus converged on the same regions in the internal globus pallidus. The results of the present tracing study define the basic network underlying the interconnections between the external segment of the globus pallidus and the subthalamic nucleus, and their connections with the output neurons of the basal ganglia in primates.