Three-dimensional organization of the striosomal compartment and patchy distribution of striatonigral projections in the matrix of the cat caudate nucleus

The striatal compartments, striosome and matrix, are embedded in largely distinct resting state functional networks

The striatum is divided into two interdigitated tissue compartments, the striosome and matrix. These compartments exhibit distinct anatomical, neurochemical, and pharmacological characteristics and have separable roles in motor and mood functions. Little is known about the functions of these compartments in humans. While compartment-specific roles in neuropsychiatric diseases have been hypothesized, they have yet to be directly tested. Investigating compartment-specific functions is crucial for understanding the symptoms produced by striatal injury, and to elucidating the roles of each compartment in healthy human skills and behaviors. We mapped the functional networks of striosome and matrix in humans in vivo. We utilized a diverse cohort of 674 healthy adults, derived from the Human Connectome Project, including all subjects with complete diffusion and functional MRI data and excluding subjects with substance use disorders. We identified striatal voxels with striosome-like and matrix-like structural connectivity using probabilistic diffusion tractography. We then investigated resting state functional connectivity (rsFC) using these compartment-like voxels as seeds. We found widespread differences in rsFC between striosome-like and matrix-like seeds (p < 0.05, FWE corrected for multiple comparisons), suggesting that striosome and matrix occupy distinct functional networks. Slightly shifting seed voxel locations (<4 mm) eliminated these rsFC differences, underscoring the anatomic precision of these networks. Striosome-seeded networks exhibited ipsilateral dominance; matrix-seeded networks had contralateral dominance. Next, we assessed compartment-specific engagement with the triple-network model (default mode, salience, and frontoparietal networks). Striosome-like voxels dominated rsFC with the default mode network bilaterally. The anterior insula (a primary node in the salience network) had higher rsFC with striosome-like voxels. The inferior and middle frontal cortices (primary nodes, frontoparietal network) had stronger rsFC with matrix-like voxels on the left, and striosome-like voxels on the right. Since striosome-like and matrix-like voxels occupy highly segregated rsFC networks, striosome-selective injury may produce different motor, cognitive, and behavioral symptoms than matrix-selective injury. Moreover, compartment-specific rsFC abnormalities may be identifiable before disease-related structural injuries are evident. Localizing rsFC differences provides an anatomic substrate for understanding how the tissue-level organization of the striatum underpins complex brain networks, and how compartment-specific injury may contribute to the symptoms of specific neuropsychiatric disorders.

Ten-m3 plays a role in the formation of thalamostriatal projections

The importance of the thalamostriatal pathway for a myriad of brain functions is becoming increasingly apparent. Little is known about the formation of this pathway in mice. Further, while Ten-m3, a member of the Ten-m/teneurin/Odz family, is implicated in the proper wiring of mature thalamostriatal projections, its developmental time course is unknown. Here, we describe the normal development of thalamostriatal projections arising from the parafascicular nucleus (PFN) and show a role for Ten-m3 in its formation. Ten-m3 is expressed in both the PFN and the striatum by embryonic day 17 (E17). By postnatal day 3 (P3), it had a patchy appearance in the striatum, overlaid on a high dorsal-low ventral expression gradient in both structures. In wild-type mice, axons from the PFN begin to innervate the striatum by E17. By P3, terminals had ramified but were not confined to any striatal subregion. By P7, the axons had begun to avoid striosomes. The first indication of clustering of thalamic terminals within the striatal matrix was also seen at this time point. The compartmental targeting and clustering of PFN projections became more apparent by P10. Analysis of Ten-m3 knockout mice showed that while the early developmental progression of the thalamostriatal pathway is conserved, by P10 differences emerged, with a loss of topographic precision and the absence of terminal clustering. No evidence of the involvement of EphA7 downstream of Ten-m3 was found. Overall, our results suggest that Ten-m3 plays a role in the consolidation and refinement of thalamic axons to a specific subregion of the striatal matrix.

Transient compartmentalization and accelerated volume growth coincide with the expected development of cortical afferents in the human neostriatum

The neostriatum plays a central role in cortico-subcortical circuitry underlying goal-directed behavior. The adult mammalian neostriatum shows chemical and cytoarchitectonic compartmentalization in line with the connectivity. However, it is poorly understood how and when fetal compartmentalization (AChE-rich islands, nonreactive matrix) switches to adult (AChE-poor striosomes, reactive matrix) and how this relates to the ingrowth of corticostriatal afferents. Here, we analyze neostriatal compartments on postmortem human brains from 9 postconceptional week (PCW) to 18 postnatal months (PM), using Nissl staining, histochemical techniques (AChE, PAS-Alcian), immunohistochemistry, stereology, and comparing data with volume-growth of in vivo and in vitro MRI. We find that compartmentalization (C) follows a two-compartment (2-C) pattern around 10PCW and is transformed into a midgestational labyrinth-like 3-C pattern (patches, AChE-nonreactive perimeters, matrix), peaking between 22 and 28PCW during accelerated volume-growth. Finally, compartmentalization resolves perinatally, by the decrease in transient "AChE-clumping," disappearance of AChE-nonreactive, ECM-rich perimeters, and an increase in matrix reactivity. The initial "mature" pattern appears around 9 PM. Therefore, transient, a 3-C pattern and accelerated neostriatal growth coincide with the expected timing of the nonhomogeneous distribution of corticostriatal afferents. The decrease in growth-related AChE activity and transfiguration of corticostriatal terminals are putative mechanisms underlying fetal compartments reorganization. Our findings serve as normative for studying neurodevelopmental disorders.

An MRI method for parcellating the human striatum into matrix and striosome compartments in vivo

The mammalian striatum is comprised of intermingled tissue compartments, matrix and striosome. Though indistinguishable by routine histological techniques, matrix and striosome have distinct embryologic origins, afferent/efferent connections, surface protein expression, intra-striatal location, susceptibilities to injury, and functional roles in a range of animal behaviors. Distinguishing the compartments previously required post-mortem tissue and/or genetic manipulation; we aimed to identify matrix/striosome non-invasively in living humans. We used diffusion MRI (probabilistic tractography) to identify human striatal voxels with connectivity biased towards matrix-favoring or striosome-favoring regions (determined by prior animal tract-tracing studies). Segmented striatal compartments replicated the topological segregation and somatotopic organization identified in animal matrix/striosome studies. Of brain regions mapped in prior studies, our human brain data confirmed 93% of the compartment-selective structural connectivity demonstrated in animals. Test-retest assessment on repeat scans found a voxel classification error rate of 0.14%. Fractional anisotropy was significantly higher in matrix-like voxels, while mean diffusivity did not differ between the compartments. As mapped by the Talairach human brain atlas, 460 regions were significantly biased towards either matrix or striosome. Our method allows the study of striatal compartments in human health and disease, in vivo, for the first time.

Peripherally administered persistent organic pollutants distribute to the brain of developing chicken embryo in concentrations relevant for human exposure

Persistent organic pollutants (POPs) can reach the fetal brain and contribute to developmental neurotoxicity. To explore the distribution of POPs to the fetal brain, we exposed chicken embryos to a POP mixture, containing 29 different compounds with concentrations based on blood levels measured in the Scandinavian human population. The mixture was injected into the allantois at embryonic day 13 (E13), aiming at a theoretical concentration of 10 times human blood levels. POPs concentrations in the brain were measured at 0.5, 1, 2, 4, 6, 24, 48, and 72 h after administration. Twenty-seven of the individual compounds were detected during at least one of the time-points analyzed. Generally, the concentrations of most of the measured compounds were within the order of magnitude of those reported in human brain samples. Differences in the speed of distribution to the brain were observed between the per- and polyfluoroalkyl substances (PFASs), which have protein binding potential, and the lipophilic polychlorinated biphenyls (PCBs), organochlorine pesticides (OCPs) and brominated flame retardants (BFRs). Based on pharmacokinetic modeling, PFASs were best described by a one compartment model. PFASs displayed relatively slow elimination (Kel) and persisted at high levels in the brain. Lipophilic OCPs and PCBs could be fitted to a 2-compartment model. These showed high levels in the brain relative to the dose administrated as calculated by area under the curve (AUC)/Dose. Altogether, our study showed that chicken is a suitable model to explore the distribution of POPs into the developing brain at concentrations which are relevant for humans.

Application of computer-assisted navigation in mandibular angle osteotomy

Objective

To compare the effectiveness, accuracy, and surgical safety of a navigation technique with those of a traditional technique for intraoperative mandibular angle osteotomy.

Methods

Forty-three postsurgical patients with mandibular angle hypertrophy who were admitted to our Department from June 2014 to June 2017 were retrospectively reviewed. Of these patients, 23 underwent mandibular angle osteotomy using computer-assisted navigation (navigation group), and 20 underwent osteotomy using a traditional technique (traditional group). Postoperative computed tomography images were analyzed by three-dimensional software. Each patient’s facial proportion indices were measured using Mimics 19.0 software, and statistical comparisons and analyses were performed preoperatively and postoperatively.

Results

The postoperative facial contour morphology and facial proportion were improved in both groups; the navigation group showed greater improvement. The difference between the predicted and postoperative values was smaller in the navigation group than traditional group. The postoperative shape of the mandibular angle sample was similar to the preoperative predicted shape in the navigation group. No complications occurred in the navigation group, but paresthesia occurred in 17% of patients in the traditional group.

Conclusions

Mandibular angle osteotomy aided with computer-assisted navigation is more effective, accurate, and safe than the traditional technique and represents a promising clinical approach.

Distinct migratory behaviors of striosome and matrix cells underlying the mosaic formation in the developing striatum

The striatum, the largest nucleus of the basal ganglia controlling motor and cognitive functions, can be characterized by a labyrinthine mosaic organization of striosome/matrix compartments. It is unclear how striosome/matrix mosaic formation is spatially and temporally controlled at the cellular level during striatal development. Here, by combining in vivo electroporation and brain slice cultures, we set up a prospective experimental system in which we differentially labeled striosome and matrix cells from the time of birth and followed their distributions and migratory behaviors. Our results showed that, at an initial stage of striosome/matrix mosaic formation, striosome cells were mostly stationary, whereas matrix cells actively migrated in multiple directions regardless of the presence of striosome cells. The mostly stationary striosome cells were still able to associate to form patchy clusters via attractive interactions. Our results suggest that the restricted migratory capability of striosome cells may allow them to cluster together only when they happen to be located in close proximity to each other and are not separated by actively migrating matrix cells. The way in which the mutidirectionally migrating matrix cells intermingle with the mostly stationary striosome cells may therefore determine the topographic features of striosomes. At later stages, the actively migrating matrix cells began to repulse the patchy clusters of striosomes, presumably enhancing the striosome cluster formation and the segregation and eventual formation of dichotomous homogeneous striosome/matrix compartments. Overall, our study reveals temporally distinct migratory behaviors of striosome/matrix cells, which may underlie the sequential steps of mosaic formation in the developing striatum. J. Comp. Neurol. 525:794-817, 2017. © 2016 Wiley Periodicals, Inc.

Dynamic ordering of early generated striatal cells destined to form the striosomal compartment of the striatum

The mature striatum is divided into a labyrinthine system of striosomes embedded in a surrounding matrix compartment. We pulse-labeled striosomal cells (S cells) and matrix cells (M cells) in cats with (3) H-thymidine and followed their distributions during fetal and postnatal development. We identified three maturational phases in S-cell distributions. The early phase (sampled at embryonic day [E]27-E35 following E24-E28 (3) H-thymidine) was characterized by a transient medial accumulation of synchronously generated S cells within the caudate nucleus adjoining the ganglionic eminence, potentially a waiting compartment. Band-like arrangements of synchronously generated S cells then formed beyond this medial band. During the second phase (sampled at E38-E45), the loosely banded S-cell distributions were transformed into clustered arrangements typical of developing striosomes. In the third phase (sampled from E52 into the postnatal period), these developed into the typical mature striosomal architecture. At adulthood, gentle mediolateral birthdate-gradients in S cells were still evident, but M cells, produced over mid to late prenatal ages, became broadly distributed, without apparent gradients or banding arrangements. These findings suggest that the maturational histories of the striosomal and matrix neurons are influenced by their generation times and local environments, and that future S cells have transient, nonstriosomal distributions prior to their aggregation into striosomal clusters, including a putative waiting compartment. Further, the eventual patterning of the striosomal compartment reflects outside-in, band-like gradient patterns of settling of synchronously generated S cells, patterns that could be related both to neural processing in the mature striatum and to patterns of vulnerability of striatal neurons.

The glycoprotein Ten-m3 mediates topography and patterning of thalamostriatal projections from the parafascicular nucleus in mice

The striatum is the key input nucleus of the basal ganglia, and is implicated in motor control and learning. Despite the importance of striatal circuits, the mechanisms associated with their development are not well established. Previously, Ten-m3, a member of the Ten-m/teneurin/odz family of transmembrane glycoproteins, was found to be important in the mapping of binocular visual pathways. Here, we investigated a potential role for Ten-m3 in striatal circuit formation. In situ hybridisation revealed a patchy distribution of Ten-m3 mRNA expression superimposed on a high-dorsal to low-ventral gradient in a subregion of the striatal matrix. A survey of afferent/efferent structures associated with the matrix identified the parafascicular thalamic nucleus (PF) as a potential locus of action. Ten-m3 was also found to be expressed in a high-dorsal to low-ventral gradient in the PF, corresponding topographically to its expression in the striatum. Further, a subset of thalamic terminal clusters overlapped with Ten-m3-positive domains within the striatal matrix. Studies in wild-type (WT) and Ten-m3 knockout (KO) mice revealed no differences in overall striatal or PF structure. Thalamostriatal terminals in KOs, however, while still confined to the matrix subregion, lost their clustered appearance. Topography was also altered, with terminals from the lateral PF projecting ectopically to ventral and medial striatum, rather than remaining confined dorsolaterally as in WTs. Behaviorally, Ten-m3 KOs displayed delayed motor skill acquisition. This study demonstrates that Ten-m3 plays a key role in directing the formation of thalamostriatal circuitry, the first molecular candidate reported to regulate connectivity within this pathway.

The non-human primate striatum undergoes marked prolonged remodeling during postnatal development

We examined the postnatal ontogeny of the striatum in rhesus monkeys (Macaca mulatta) to identify temporal and spatial patterns of histological and chemical maturation. Our goal was to determine whether this forebrain structure is developmentally static or dynamic in postnatal life. Brains from monkeys at 1 day, 1, 4, 6, 9, and 12 months of age (N = 12) and adult monkeys (N = 4) were analyzed. Nissl staining was used to assess striatal volume, cytoarchitecture, and apoptosis. Immunohistochemistry was used to localize and measure substance P (SP), leucine-enkephalin (LENK), tyrosine hydroxylase (TH), and calbindin D28 (CAL) immunoreactivities. Mature brain to body weight ratio was achieved at 4 months of age, and striatal volume increased from ∼1.2 to ∼1.4 cm(3) during the first postnatal year. Nissl staining identified, prominently in the caudate nucleus, developmentally persistent discrete cell islands with neuronal densities greater than the surrounding striatal parenchyma (matrix). Losses in neuronal density were observed in island and matrix regions during maturation, and differential developmental programmed cell death was observed in islands and matrix regions. Immunohistochemistry revealed striking changes occurring postnatally in striatal chemical neuroanatomy. At birth, the immature dopaminergic nigrostriatal innervation was characterized by islands enriched in TH-immunoreactive puncta (putative terminals) in the neuropil; TH-enriched islands aligned completely with areas enriched in SP immunoreactivity but low in LENK immunoreactivity. These areas enriched in SP immunoreactivity but low in LENK immunoreactivity were identified as striosome and matrix areas, respectively, because CAL immunoreactivity clearly delineated these territories. SP, LENK, and CAL immunoreactivities appeared as positive neuronal cell bodies, processes, and puncta. The matrix compartment at birth contained relatively low TH-immunoreactive processes and few SP-positive neurons but was densely populated with LENK-immunoreactive neurons. The nucleus accumbens part of the ventral striatum also showed prominent differences in SP, LENK, and CAL immunoreactivities in shell and core territories. During 12 months of postnatal maturation salient changes occurred in neurotransmitter marker localization: TH-positive afferents densely innervated the matrix to exceed levels of immunoreactivity in the striosomes; SP immunoreactivity levels increased in the matrix; and LENK-immunoreactivity levels decreased in the matrix and increased in the striosomes. At 12 months of age, striatal chemoarchitecture was similar qualitatively to adult patterns, but quantitatively different in LENK and SP in caudate, putamen, and nucleus accumbens. This study shows for the first time that the rhesus monkey striatum requires more than 12 months after birth to develop an adult-like pattern of chemical neuroanatomy and that principal neurons within striosomes and matrix have different developmental programs for neuropeptide expression. We conclude that postnatal maturation of the striatal mosaic in primates is not static but, rather, is a protracted and dynamic process that requires many synchronous and compartment-selective changes in afferent innervation and in the expression of genes that regulate neuronal phenotypes.

EphA7 expression identifies a unique neuronal compartment in the rat striatum

Prior studies have identified two anatomically and neurochemically distinct cellular compartments within the mammalian striatum, termed striosomes and matrix, which express μ-opioid receptors (μOR) and EphA4, respectively. Here we identify and characterize an additional compartment in the rat striatum composed of neurons that express EphA7. In situ hybridization and immunohistochemical data indicate that neurons expressing EphA7 mRNA and protein are arranged in a banded "matrisome-like" pattern confined to the matrix in the dorsal striatum. Within the ventral striatum, EphA7-positive (+) neurons have a less organized mosaic pattern that partially overlaps areas expressing μOR. Immunolabeling data demonstrate that EphA7+ striatofugal axons form distinct fascicles leaving the striatum. Within the globus pallidus, EphA7+ axons terminate primarily within ventromedial areas of the nucleus and along its striatal border. EphA7+ axons avoid regions containing dopamine neurons within the substantia nigra and preferentially innervate areas near the rostral and caudal margins of the nucleus. Within both nuclei, EphA7+ axons have similar but more restricted terminal fields than the entire population of EphA4+ matrix axons, indicating that EphA7+ axons comprise a subpopulation of matrix axons. Ligand binding data demonstrate that ephrin-A5 selectively binds areas of the striatum, globus pallidus, and substantia nigra containing EphA7+ neurons and axons, but not areas expressing only EphA4. Our findings demonstrate that EphA7 expression identifies a novel "matrisome" compartment within the matrix that binds ephrin-A5 and possesses unique axonal projections. Our findings also suggest that EphA7 and ephrin-A5 may participate in the formation of this matrisome subcompartment and its striatofugal projections.

Role of excitatory amino acids in the direct and indirect presynaptic regulation of dopamine release from nerve terminals of nigrostriatal dopaminergic neurons

In vivo experiments carried out in halothane-anaesthetized cats implanted with push-pull cannulae demonstrated that glutamate (GLU) released from corticostriatal fibers triggers the release of dopamine (DA), even in the absence of activity in nigral DA cells. As shown in vitro, using rat striatal slices or synaptosomes or in vivo in the cat, both NMDA and AMPA receptors subtypes are involved in the GLU-induced release of DA. Beside this direct regulation, GLU also exert several indirect facilitatory and inhibitory controls on DA release, particularly through cholinergic and GABAergic striatal neurons. Indeed, as shown by numerous authors, the GLU-evoked release of DA is markedly reduced in the presence of tetrodotoxin, bicuculline or atropine or by previous kainate- or ibotenate-induced lesion of striatum. Differences in the presynaptic regulation of DA release in striosomal and matrix compartments have also been found with NMDA and acetylcholine. The effect of acetylcholine was of shorter duration in the matrix than in the striosomal-enriched areas. Two opposite indirect regulations of DA release could be demonstrated: one is facilitatory and involves nicotinic receptors, the other is inhibitory, involves muscarinic receptors and mediated, at least in the matrix by dynorphin containing neurons. The NMDA-evoked responses are of larger amplitude and more sensitive to tetrodotoxin in the matrix than in the striosomes. In conclusion, GLU released from corticostriatal fibers, is able to control the release of DA from terminals of nigrostriatal neurons through direct facilitatory mechanisms (NMDA and AMPA receptors), but also through indirect facilitatory and inhibitory local circuits involving cholinergic and GABAergic neurons.

Deep homology of arthropod central complex and vertebrate basal ganglia

The arthropod central complex and vertebrate basal ganglia derive from embryonic basal forebrain lineages that are specified by an evolutionarily conserved genetic program leading to interconnected neuropils and nuclei that populate the midline of the forebrain-midbrain boundary region. In the substructures of both the central complex and basal ganglia, network connectivity and neuronal activity mediate control mechanisms in which inhibitory (GABAergic) and modulatory (dopaminergic) circuits facilitate the regulation and release of adaptive behaviors. Both basal ganglia and central complex dysfunction result in behavioral defects including motor abnormalities, impaired memory formation, attention deficits, affective disorders, and sleep disturbances. The observed multitude of similarities suggests deep homology of arthropod central complex and vertebrate basal ganglia circuitries underlying the selection and maintenance of behavioral actions.

Complete 3D visualization of primate striosomes by KChIP1 immunostaining

High-resolution 3D reconstruction and morphometric analysis of striosomes was carried out in macaque monkeys by using immunocytochemistry for the Kv4 potassium channel subunit potassium channel interacting protein 1 (KChIP1), a novel marker. The striosomes form a connected reticulum made up of two distinct planar sheets spanning several millimeters in the putamen, and long finger-like branches in the caudate nucleus and putamen. Although their spatial organization is variable, morphometric analysis of the striosomes, utilizing skeletonizations, reveals several quantitative invariant measures of striosome organization, including the following findings: 1) individual bifurcation-free striosome branches are 355 +/- 108.5 microm in diameter and 1,013 +/- 751 microm in length, and are both lognormally distributed; and 2) striosome branches exhibit three pronounced orientation preferences that are approximately orthogonal. The former finding suggests a fundamental anatomical and functional component of the striatum, whereas the latter indicates that striosomes are more lattice-like than their spatial variability suggests. The perceived variable spatial organization of the striosomes in primates belies many invariant features that may reflect striatal function, development, and pathophysiology.

Differential synaptology of vGluT2-containing thalamostriatal afferents between the patch and matrix compartments in rats

The striatum is divided into two compartments named the patch (or striosome) and the matrix. Although these two compartments can be differentiated by their neurochemical content or afferent and efferent projections, the synaptology of inputs to these striatal regions remains poorly characterized. By using the vesicular glutamate transporters vGluT1 and vGluT2, as markers of corticostriatal and thalamostriatal projections, respectively, we demonstrate a differential pattern of synaptic connections of these two pathways between the patch and the matrix compartments. We also demonstrate that the majority of vGluT2-immunolabeled axon terminals form axospinous synapses, suggesting that thalamic afferents, like corticostriatal inputs, terminate preferentially onto spines in the striatum. Within both compartments, more than 90% of vGluT1-containing terminals formed axospinous synapses, whereas 87% of vGluT2-positive terminals within the patch innervated dendritic spines, but only 55% did so in the matrix. To characterize further the source of thalamic inputs that could account for the increase in axodendritic synapses in the matrix, we undertook an electron microscopic analysis of the synaptology of thalamostriatal afferents to the matrix compartments from specific intralaminar, midline, relay, and associative thalamic nuclei in rats. Approximately 95% of PHA-L-labeled terminals from the central lateral, midline, mediodorsal, lateral dorsal, anteroventral, and ventral anterior/ventral lateral nuclei formed axospinous synapses, a pattern reminiscent of corticostriatal afferents but strikingly different from thalamostriatal projections arising from the parafascicular nucleus (PF), which terminated onto dendritic shafts. These findings provide the first evidence for a differential pattern of synaptic organization of thalamostriatal glutamatergic inputs to the patch and matrix compartments. Furthermore, they demonstrate that the PF is the sole source of significant axodendritic thalamic inputs to striatal projection neurons. These observations pave the way for understanding differential regulatory mechanisms of striatal outflow from the patch and matrix compartments by thalamostriatal afferents.

Functional mu opioid receptors are expressed in cholinergic interneurons of the rat dorsal striatum: territorial specificity and diurnal variation

Striatal cholinergic interneurons play a crucial role in the control of movement as well as in motivational and learning aspects of behaviour. Neuropeptides regulate striatal cholinergic transmission and particularly activation of mu opioid receptor (MOR) inhibits acetylcholine (ACh) release in the dorsal striatum. In the present study we investigated whether this cholinergic transmission could be modulated by an enkephalin/MOR direct process. We show that mRNA and protein of MORs are expressed by cholinergic interneurons in the limbic/prefrontal territory but not by those in the sensorimotor territory of the dorsal striatum. These MORs are functional because potassium-evoked release of ACh from striatal synaptosomes was dose-dependently reduced by a selective MOR agonist, this effect being suppressed by a MOR antagonist. The MOR regulation of cholinergic interneurons presented a diurnal variation. (i) The percentage of cholinergic interneurons containing MORs that was 32% at the beginning of the light period (morning) increased to 80% in the afternoon. (ii) The MOR-mediated inhibition of synaptosomal ACh release was higher in the afternoon than in the morning. (iii) While preproenkephalin mRNA levels remained stable, enkephalin tissue content was the lowest (-32%) in the afternoon when the spontaneous (+35%) and the N-methyl-d-aspartate-evoked (+140%) releases of enkephalin (from microsuperfused slices) were the highest. Therefore, by acting on MORs present on cholinergic interneurons, endogenously released enkephalin reduces ACh release. This direct enkephalin/MOR regulation of cholinergic transmission that operates only in the limbic/prefrontal territory of the dorsal striatum might contribute to information processing in fronto-cortico-basal ganglia circuits.

Functional architecture of the mammalian striatum: mouse vascular and striosome organization and their anatomic relationships

To determine whether the general architecture of striatal vessels and mu opioid receptor-rich striosomes is similar, we investigated 3D reconstructions of coronal sections in 10 FVB mice. The sections were stained for striosomes using a mu opioid receptor antibody (MOR1). We used computerized procedures to detect striosomes and vessels and to calculate volume, number and colocalization of striosomes and vessels. The results showed a lattice-like pattern of striosomes similar to, and often surrounding, blood vessels. Furthermore, co-localization calculations suggested that the striosomes are more vascular than the matrix. Vessel volume was 5.0+/-1.3% per microm3 in striosomes versus 3.6+/-0.9%microm3 in matrix (p=0.01). The findings emphasize the probable importance of a grid- or lattice-like structure as an organizing principle of striatal anatomy and function. In addition, the greater vascularity of the striosomes compared to the matrix suggests a unique function of this compartment in relation to humoral signals and neurotropic drugs.

Ockham's razor modeling of the matrisome channels of the basal ganglia thalamocortical loops

A functional model of the basal ganglia-thalamocortical (BTC) loops is described. In our modeling effort, we try to minimize the complexity of our starting hypotheses. For that reason, we call this type of modeling Ockham's razor modeling. We have the additional constraint that the starting assumptions should not contradict experimental findings about the brain. First assumption: The brain lacks direct representation of paths but represents directions (called speed fields in control theory). Then control should be concerned with speed-field tracking (SFT). Second assumption: Control signals are delivered upon differencing in competing parallel channels of the BTC loops. This is modeled by extending SFT with differencing that gives rise to the robust Static and Dynamic State (SDS) feedback-controlling scheme. Third assumption: Control signals are expressed in terms of a gelatinous medium surrounding the limbs. This is modeled by expressing parameters of motion in parameters of the external space. We show that corollaries of the model fit properties of the BTC loops. The SDS provides proper identification of motion related neuronal groups of the putamen. Local minima arise during the controlling process that works in external space. The model explains the presence of parallel channels as the means to avoiding such local minima. Stability conditions of the SDS predict that the initial phase of learning is mostly concerned with selection of sign for the inverse dynamics. The model provides a scalable controller. State description in external space instead of configurational space reduces the dimensionality problem. Falsifying experiment is suggested. Computer experiments demonstrate the feasibility of the approach. We argue that the resulting scheme has a straightforward connectionist representation exhibiting population coding and Hebbian learning properties.

Three-dimensional cartography of functional territories in the human striatopallidal complex by using calbindin immunoreactivity

This anatomic study presents an analysis of the distribution of calbindin immunohistochemistry in the human striatopallidal complex. Entire brains were sectioned perpendicularly to the mid-commissural line into 70-microm-thick sections. Every tenth section was immunostained for calbindin. Calbindin labeling exhibited a gradient on the basis of which three different regions were defined: poorly labeled, strongly labeled, and intermediate. Corresponding contours were traced in individual sections and reformatted as three-dimensional structures. The poorly labeled region corresponded to the dorsal part of the striatum and to the central part of the pallidum. The strongly labeled region included the ventral part of the striatum, the subcommissural part of the external pallidum but also the adjacent portion of its suscommissural part, and the anterior pole of the internal pallidum. The intermediate region was located between the poorly and strongly labeled regions. As axonal tracing and immunohistochemical studies in monkeys show a similar pattern, poorly, intermediate, and strongly labeled regions were considered as the sensorimotor, associative, and limbic territories of the human striatopallidal complex, respectively. However, the boundaries between these territories were not sharp but formed gradients of labeling, which suggests overlapping between adjacent territories. Similarly, the ventral boundary of the striatopallidal complex was blurred, suggesting a structural intermingling with the substantia innominata. This three-dimensional partitioning of the human striatopallidal complex could help to define functional targets for high-frequency stimulation with greater accuracy and help to identify new stimulation sites.

The cat neostriatum: relative distribution of cholinergic neurons versus serotonergic fibers

The distribution of choline acetyltransferase (ChAT)-containing neurons and serotonin (5-HT)-containing nerve fibers in the cat neostriatum was investigated by use of immunohistochemical techniques. Both ChAT- and 5-HT-staining techniques were applied to alternate brain sections, thereby allowing a precise comparison of the distribution pattern of ChAT-immunopositive cells (ChAT cells) and 5-HT-immunopositive fibers (5-HT fibers). In the neostriatum, ChAT cells were strongly stained throughout their cell bodies and proximal (first-order) dendrites. The majority of them were multipolar cells with a soma diameter of 20-50 microm (long axis)x10-30 microm (short axis). In the caudate nucleus, ChAT cells were evenly and diffusely distributed except for the dorsolateral region of its rostral half, in which latter region they were distributed in loosely formed clusters. In the rostral portion of the putamen, the density of ChAT-cell distribution was like that in the medial region of the caudate nucleus. In contrast, this distribution was more dense in the caudomedial region of the putamen, adjacent to the globus pallidus. 5-HT fibers in the neostriatum were dark-stained, of quite fine diameter (<0.6 microm), and they contained small, round varicosities (diameter, usually 0.5-1.0 microm, but some >1.0 microm). Such 5-HT fibers were distributed abundantly throughout the caudate nucleus and putamen. In the rostrocaudal portion of the caudate nucleus, their density was high in its dorsal and ventral components, and low in the middle component. Throughout the putamen, 5-HT fibers were distributed homogeneously in the mediolateral and dorsoventral directions. In the caudal portion of the putamen adjacent to the globus pallidus, the 5-HT fibers had a higher density while maintaining their homogenous distribution pattern. In the two main divisions of the striatum, the so-called 'patch' (acetylcholinesterase (AChE)-poor) and 'matrix' (AChE-rich) compartments, there was a near-even distribution of 5-HT fibers and terminals. The above results suggest that the 5-HT-dominated, raphe-striatal pathway is optimally arranged for modulating the activity of both the intrinsic and the projection neurons of the neostriatum.

Ephrin-A binding and EphA receptor expression delineate the matrix compartment of the striatum

The striatum integrates limbic and neocortical inputs to regulate sensorimotor and psychomotor behaviors. This function is dependent on the segregation of striatal projection neurons into anatomical and functional components, such as the striosome and matrix compartments. In the present study the association of ephrin-A cell surface ligands and EphA receptor tyrosine kinases (RTKs) with the organization of these compartments was determined in postnatal rats. Ephrin-A1 and ephrin-A4 selectively bind to EphA receptors on neurons restricted to the matrix compartment. Binding is absent from the striosomes, which were identified by mu-opioid receptor immunostaining. In contrast, ephrin-A2, ephrin-A3, and ephrin-A5 exhibit a different mosaic binding pattern that appears to define a subset of matrix neurons. In situ hybridization for EphA RTKs reveals that the two different ligand binding patterns strictly match the mRNA expression patterns of EphA4 and EphA7. Ligand-receptor binding assays indicate that ephrin-A1 and ephrin-A4 selectively bind EphA4 but not EphA7 in the lysates of striatal tissue. Conversely, ephrin-A2, ephrin-A3, and ephrin-A5 bind EphA7 but not EphA4. These observations implicate selective interactions between ephrin-A molecules and EphA RTKs as potential mechanisms for regulating the compartmental organization of the striatum.

Spatiotemporal expression gradients of the carbohydrate antigen (CD15) (Lewis X) during development of the human basal ganglia

The developmental expression pattern of the carbohydrate epitope CD15 (Lewis X, Le X) (alpha1-->3-fucosyl-N-acetyl-lactosamine) has been immunocytochemically evaluated in paraffin sections within the human basal ganglia from 10 weeks gestation to three years after birth. At 11 weeks of gestation, CD15 (Le X) positive radial glial cells were located in the anterior and dorsal parts of the lateral ganglionic eminence. Their processes ran from the subventricular zone radially in a highly ordered fashion to the dorsolateral margin of the caudate nucleus and further to the lateral rim of the putamen. At 12 weeks of gestation, strands of CD15 (Le X) material continued to the pial surface, forming a continuous CD15 (Le X) positive borderline separating the accumbens nucleus and olfactory tubercle from the piriform cortex. At 13 weeks of gestation the dorsal putamen was completely CD15 (Le X) immunoreactive along its perimeter and CD15 (Le X) patches, consisting of fine granular material, appeared at the dorsolateral margin of the putamen at this age; while the first CD15 (Le X) patches in the caudate nucleus were observed four weeks later. The matrix compartment of the caudate and dorsal putamen became gradually stained by granular CD15 (Le X) positive material into which CD15 (Le X) immunoreactive somata were embedded. The striking contrast in staining between patch and matrix compartments disappeared shortly after birth. The ventral striatum did not become immunoreactive until the last few weeks before birth. After the formation of CD15 (Le X) positive patches in the striatum (from 12 weeks of gestation), delicate CD15 (Le X) fibres, often accumulated in bundles and related to the striatal patches, became apparent coursing towards the external pallidal lamina and the globus pallidus. Immunoreactivity in the globus pallidus itself was transient, emerging from 16 weeks of gestation, reaching a peak at 21 weeks of gestation and disappearing by birth. Both processes, i.e. the occurrence of CD15 (Le X) striatopallidal fibres and the emerging immunoreactivity in their pallidal target, may be interrelated, so that ingrowing CD15 (Le X) positive axons from the striatum provoke CD15 (Le X) expression in the external and internal pallidum. The variable patterns and intensities of CD15 (Le X) expression are possibly related to periods of maturation of the striatum and the establishment of functional interactions within the basal ganglia. Differential staining of patch and matrix in the developing neostriatum suggests that a distinct phase of cellular adhesion or dishesion mediated by the CD15 (Le X) epitope occurs during establishment of the patch and matrix regions.

Distinct modifications by neurokinin1 (SR140333) and neurokinin2 (SR48968) tachykinin receptor antagonists of the N-methyl-D-aspartate-evoked release of acetylcholine in striosomes and matrix of the rat striatum

The effects of SR140333 and SR48968 (neurokinin1 and neurokinin2 tachykinin receptor antagonists, respectively) on the N-methyl-D-aspartate-evoked release of [3H]acetylcholine (previously formed from [3H]choline) were investigated in striosome-enriched areas and in the matrix of the rat striatum using an in vitro microsuperfusion method. In both striatal compartments, SR140333 and SR48968 did not modify the 50 microM N-methyl-D-aspartate-evoked release of [3H]acetylcholine. However, in low concentrations, both SR140333 (0.1 microM to 1 pM) and SR48968 (0.1 microM to 0.1 nM) markedly enhanced the 1 mM N-methyl-D-aspartate (+10 microM D-serine)-evoked release of [3H]acetylcholine in striosome-enriched areas. These responses were dopamine-dependent since they were not observed any more following the local blockade of D2 receptors by sulpiride or of dopamine synthesis by alpha-methyl-p-tyrosine. A dopamine-dependent disinhibitory effect (of lower amplitude) on the 1 mM N-methyl-D-aspartate (+10 microM D-serine)-evoked release of [3H]acetylcholine was also induced by SR48968 (0.1 microM to 0.1 nM) (but not by SR140333) in the matrix. In addition, in the matrix, as shown only in the presence of alpha-methyl-p-tyrosine, both SR140333 and SR48968 reduced the 1 mM N-methyl-D-aspartate (+10 microM D-serine)-evoked response and these non-dopamine-mediated inhibitory effects only occurred at the highest tested concentration (0.1 microM) of the antagonists. Indicating the specificity of these responses, the effects of SR140333 were reproduced by RP67580, another neurokinin1 receptor antagonist and, as expected from previous binding studies, corresponding SR140333 and SR48968 enantiomers were without effect. These results suggest that under potent stimulation of N-methyl-D-aspartate receptors, endogenously released substance P and neurokinin A (or related tachykinins) regulate differently the N-methyl-D-aspartate-evoked release of [3H]acetylcholine in striosomes and in the matrix. The inhibitory effects of these tachykinins on the evoked release of [3H]acetylcholine are mediated by dopamine. On the contrary, their facilitatory responses are only observed in the matrix under blockade of dopamine transmission.

Neuroanatomy of the basal ganglia

This article deals with the neuroanatomic aspects of the basal ganglia with regard to different neurotransmitter systems and to different diseases. A general scheme of these circuits with the overall distinction between limbic-associative and motor components and circuits is presented.

Static and dynamic state feedback control model of basal ganglia-thalamocortical loops

It is argued that a novel control architecture, the Static and Dynamic State (SDS) feedback scheme, which utilizes speed-field tracking, exhibits global stability, and allows on-line tuning by any adaptation mechanism without canceling stability if certain structural conditions are met, can be viewed as a model of basal ganglia-thalamocortical loops since (1) the SDS scheme predicts the neuronal groups that fit neuronal classification in the supplementary motor area, the motor cortex and the putamen, (2) the structural stability conditions require parallel channels, a feature that these loops provide, and (3) the SDS scheme predicts two major disorders that can be identified as Parkinson's and Huntington's diseases. Simulations suggests that the basal ganglia work outside the realm of the stability condition allowed by the robustness of the scheme and required for increased computation speeds.

Distinct regulations by septide and the neurokinin-1 tachykinin receptor agonist [pro9]substance P of the N-methyl-D-aspartate-evoked release of dopamine in striosome- and matrix-enriched areas of the rat striatum

The effects of septide (a short substance P C-terminal analogue) and of the neurokinin-1 receptor agonist [Pro9]substance P on the N-methyl-D-aspartate (50 microM)-evoked release of [3H]dopamine (continuously synthesized from [3H]tyrosine) were investigated in the absence or the presence of the selective neurokinin-1 receptor antagonist RP 67580 in selected striosome- and matrix-enriched areas of the rat striatum. Experiments were performed in vitro using a microsuperfusion procedure described previously. At a concentration of 0.1 microM, septide and [Pro9]substance P stimulated the spontaneous release of [3H]dopamine in striosome-enriched areas similarly. However, in this compartment, these peptides induced larger and opposite effects on the N-methyl-D-aspartate (50 microM)-evoked release of [3H]dopamine (estimated in the absence of magnesium). Indeed, septide markedly enhanced the N-methyl-D-aspartate response, while [Pro9]substance P largely reduced the N-methyl-D-aspartate-evoked release of [3H]dopamine. Septide also enhanced the N-methyl-D-aspartate response in the matrix, but [Pro9]substance P was without effect. When used alone, at 0.1 or 1 microM, RP 67580 reduced by about 33% the N-methyl-D-aspartate-evoked release of [3H]dopamine in striosome-enriched areas. In contrast, in the matrix, the N-methyl-D-aspartate response was enhanced in the presence of a low concentration of the antagonist, while the higher concentration was ineffective. In striosomes, the reducing effect of [Pro9]substance P and the enhancing action of septide on the N-methyl-D-aspartate response were respectively blocked in the presence of low and high concentrations of RP 67580, while the stimulatory effect of septide on the N-methyl-D-aspartate response in the matrix was prevented with both concentrations of the neurokinin-1 receptor antagonist. Finally, the co-application of [Pro9]substance P (0.1 microM) with septide (0.1 microM) abolished the enhancing effect of septide on the N-methyl-D-aspartate-evoked release of [3H]dopamine in both striatal compartments. Altogether, these results suggest that substance P and eventually one of its metabolites, substance P(6-11) or another endogenous tachykinin released under the action of N-methyl-D-aspartate, contribute to the regulation of [3H]dopamine release in both striatal compartments. They also extend previous observations which allowed us to demonstrate that the local circuits contributing to the presynaptic regulation of [3H]dopamine release differ in striosome- and matrix-enriched areas. Furthermore, in agreement with observations made in some peripheral tissues, the present results support the existence of "septide-sensitive" tachykinin receptors in the rat striatum or alternatively of septide sensitive sites on tachykinin neurokinin-1 receptors distinct from those sensitive to neurokinin-1 receptor agonists, coupled to distinct transducing systems, and thus leading to biological responses which differ from those evoked by neurokinin-1 receptor agonists.

A topographic re-evaluation of the nigrostriatal projections to the caudate nucleus in the cat with multiple retrograde tracers

The anatomical organization of the cat nigrostriatal projections to the caudate nucleus was studied by retrograde tracer techniques. Horseradish peroxidase conjugated with wheat germ agglutinin and fluorescent retrograde tracers such as Fast Blue and Diamidino Yellow were injected concomitantly in different regions of the caudate nucleus. The distribution of single and double retrogradely labeled neurons was analysed in the substantia nigra pars compacta, substantia nigra pars lateralis, retrorubral area and adjacent ventral tegmental area. Adjacent sections processed for acetylcholinesterase were used as histochemical markers for the densocellular zone of the substantia nigra. The main findings of this study are: (1) The rostral caudate nucleus receives projections mainly from the caudal substantia nigra while the caudal caudate nucleus receives projections from all rostrocaudal levels of the substantia nigra. (2) The substantia nigra pars lateralis projects very specifically to the caudal caudate nucleus. (3) The ventral retrorubral area close to the medial lemniscus projects to all rostrocaudal levels of the caudate nucleus. (4) The rostral caudate nucleus receives projections mainly from the medial substantia nigra while more caudal sectors of the caudate nucleus receive projections from the medial and lateral substantia nigra. (5) A dorsoventral inversion of nigrostriatal projections from the medial substantia nigra pars compacta and the adjacent ventral tegmental area to the caudate nucleus was established. In contrast, we found zones within the retrorubral area projecting both to the dorsal and ventral caudate nucleus. (6) Distant injections of two different fluorescent tracers regarding both the dorsoventral and the rostrocaudal coordinates, yielded double-labeled neurons that were mainly located in the medial and caudal portions of the substantia nigra and in the ventral retrorubral area. However, the number of double-labeled neurons was higher after separated injections in the dorsoventral axis, suggesting that the collateralization to the caudate nucleus occurs mainly in the dorsoventral plane. (7) A clustering organization of nigrostriatal cells projecting to the caudate nucleus was detected mainly in the intermediate rostrocaudal part of the substantia nigra pars compacta and in the retrorubral area. The results of this comprehensive study on the cat nigrostriatal pathway to the caudate nucleus show novel findings on the anatomical organization of the nigrostriatal projections which might help the understanding of the complex architecture of nigral neurons projecting to the caudate nucleus in carnivores.

Heterogeneous topographical distribution of the striatonigral and striatopallidal neurons in the matrix compartment of the cat caudate nucleus

The topographical organization of the striatonigral projection was investigated in the cat by comparing the localization and the intensity of labelling of retrogradely labelled cells in the caudate nucleus following one or multiple injections of horseradish peroxidase-wheat germ agglutinin into the center or along the rostrocaudal axis of the substantia nigra pars reticulata. Second, the localizations of retrogradely labelled striatopallidal neurons and of clusters of aggregated striatonigral neurons (as outlined by the transport of 14C-material) were compared in cats that received four horseradish peroxidase-wheat germ agglutinin injections into the internal segment of the globus pallidus and three nigral injections of 14C-amino acids into the substantia nigra pars reticulata. Two types of striatonigral neurons located predominantly within the matrix compartment were identified: poorly collateralized aggregated cells distributed in clusters and more numerous collateralized cells distributed outside the clusters. In addition, two cell types were distinguished within each cluster of aggregated neurons. Those innervating the center of the substantia nigra pars reticulata were observed after a single nigral injection of the tracer, whereas those projecting to distinct sites of the substantia nigra pars reticulata along a rostrocaudal axis were observed only after multiple injections. Striatal neurons innervating the internal segment of the globus pallidus were heterogeneously distributed predominantly within the matrix but outside the clusters of aggregated striatonigral neurons. Together, these results provide further evidence for the heterogeneity of the matrix and for the complexity of matrix striatonigral connections that send both diverging and converging signals to the substantia nigra pars reticulata.

Parvalbumin-containing GABAergic neurons in the basal ganglia output system of the rat

The output of the basal ganglia is directed through the entopeduncular nucleus (EPN) and the substantia nigra pars reticulata (SNR) and pars lateralis (SNL), which provide a gamma-aminobutyric acidergic (GABAergic) projection to various nuclei of the thalamus and brainstem. Although many neurons within the SNR and EPN have been described as modality specific, the morphological and neurochemical similarities preclude their precise identification. In the present study, the immunocytochemical localization of parvalbumin, a calcium-binding protein, is used in combination with axonal tracing to verify neuronal heterogeneity within the SNR, SNL, and EPN. The results reveal that the majority of neurons in all three centers contain parvalbumin. The parvalbumin-containing neurons are distributed in the caudal two-thirds of the EPN, the rostral part of the SNL, and the lateral two-thirds of the entire rostrocaudal extent of the SNR, the areas involved in sensorimotor function of the basal ganglia. Moreover, the nigrothalamic, nigrocollicular, and EPN-thalamic neurons possess parvalbumin immunoreactivity, whereas the EPN-habenular neurons are devoid of parvalbumin immunoreactivity. The results indicate a neurochemical heterogeneity within the GABAergic output neurons of the basal ganglia and suggest that the parvalbumin-containing neurons of the SNR, SNL, and EPN are the tonically active output neurons that form a major link in the disinhibitory neuronal circuit of the basal ganglia, especially that concerned with sensorimotor function.

The distribution of cholinergic perikarya with respect to enkephalin-rich patches in the caudate nucleus of the adult cat

The distribution of cholinergic interneurons with respect to enkephalin-rich patches in the caudate nucleus of the cat was examined using both computer-assisted 3-D reconstruction and immunocytochemical techniques. Examination of the 3-D distribution of perikarya staining for choline acetyltransferase (ChAT) revealed that these cells were not evenly distributed within the caudate nucleus but exhibited areas of increased and decreased density. Comparison of the 3-D distribution of cholinergic perikarya to that of the enkephalin-rich patches indicated that areas of increased ChAT+ cell density often corresponded to the positions of enkephalin-rich patches within the dorsal-lateral caudate nucleus. At more ventral regions, there was no clear correspondence between areas of increased ChAT+ cell density and enkephalin-rich patches. In agreement with these observations, a quantitative analysis of sections double-labeled for ChAT and enkephalin revealed that the density of cholinergic neurons within enkephalin-rich patches was twice that in the surrounding tissue in the dorsal region of the caudate nucleus. In contrast at more ventral levels, the difference in the density of ChAT+ cells in enkephalin-rich patches did not significantly differ from that in the surrounding striatal tissue. Both the results of the 3-D and the double-labeling analysis suggest that cholinergic neurons are not evenly distributed within the caudate nucleus of the cat but form loose clusters which are associated dorsally with the enkephalin-rich patches. These results also provide further evidence of heterogeneity within the striosomal compartment in the cat.

The basal ganglia and adaptive motor control

The basal ganglia are neural structures within the motor and cognitive control circuits in the mammalian forebrain and are interconnected with the neocortex by multiple loops. Dysfunction in these parallel loops caused by damage to the striatum results in major defects in voluntary movement, exemplified in Parkinson's disease and Huntington's disease. These parallel loops have a distributed modular architecture resembling local expert architectures of computational learning models. During sensorimotor learning, such distributed networks may be coordinated by widely spaced striatal interneurons that acquire response properties on the basis of experienced reward.

Compartmental organization of the peptide network in the human caudate nucleus

The mammalian striatum may be divided into a striosomal compartment and a surrounding matrix region. We have examined the distribution of leucine enkephalin (LENK) and substance P (SP) immunoreactivity in relation to striosomes defined by calbindin-D (CABD) staining in alternate 70 microns serial sections from the human caudate nucleus. The distribution of LENK immunoreactivity showed a transition from dorsal to ventral striatum: dorsally, LENK-rich patches were present in a lightly stained matrix; mid-ventrally, annular patches of LENK staining with a lighter core were seen. These patches corresponded to striosomal regions defined by CABD-poor zones. In contrast, in the ventral caudate and nucleus accumbens, LENK-poor zones matched CABD-defined striosomes. CABD staining in the matrix was intense in the dorsal caudate, diminishing ventrally. SP-rich zones in dorsal caudate and SP-poor areas in the mid-ventral region overlapped striosomes. In the ventromedial sector, the SP staining pattern was complex and did not consistently correlate with striosomes. Computer-assisted three-dimensional reconstruction of the striosomal system in the human, based on regions of either high LENK or low CABD immunoreactivity, revealed the existence of considerable long-range order. Patches appeared aligned over several millimeters to form long, horizontal structures in the caudate nucleus, with occasional orthogonal interconnecting crossbridges. Our results are in accord with previous work in the human and in other species. These three-dimensional networks are strikingly similar across individuals and may relate to the segregation of and interactions between striatal circuits.

Primate striatonigral projections: a comparison of the sensorimotor-related striatum and the ventral striatum

The striatum receives topographic cortical inputs with the limbic lobe terminating in the ventral striatum and sensorimotor cortical regions terminating in the dorsolateral striatum. The organization of striatonigral projections originating from these different striatal territories was examined in primate by using several anterograde tracers. The ventral striatum innervates a large area of the substantia nigra, including the medial pars reticulata and much of the pars compacta. Moreover, projections from separate areas of the ventral striatum overlap considerably in the substantia nigra. No mediolateral or rostrocaudal topographic order is apparent, and the area of the substantia nigra associated with the ventral striatum is extensive. In contrast, the sensorimotor-related striatum innervates a limited region of the ventrolateral substantia nigra. Similar to ventral striatonigral projections, projections originating from different areas of the sensorimotor-related striatum send converging inputs to the substantia nigra. Sensorimotor-related striatonigral projections avoid the region of the dopaminergic neurons in the dorsal pars compacta. Striatonigral projections from the sensorimotor-related and ventral striatum do not overlap in the substantia nigra. Examination of the outputs of discrete striatal loci indicates that the organization of striatonigral projections is more related to corticostriatal inputs than to a simple rostrocaudal, dorsoventral, or mediolateral topography of the striatum. Striatal projections that originate from different striatal territories are distinct and nonoverlapping, thus supporting the concept of segregated striatonigral circuits. However, areas of the striatum that receive common cortical inputs send converging inputs to the substantia nigra. This suggests that the substantia nigra is also an important link for integrating information between functionally related (sub)circuits.

NMDA and carbachol but not AMPA affect differently the release of [3H]GABA in striosome- and matrix-enriched areas of the rat striatum

The effects of alpha-amino-3-hydroxy-5-methylisoxazole-4-propionate (AMPA; 10(-3) M), N-methyl-D-aspartate (10(-3) M, in the absence of magnesium or presence of AMPA) and carbachol (10(-3) M) on the release of preloaded [3H]gamma-aminobutyric acid ([3H]GABA) from microdiscs of tissue punched out from sagittal brain slices in striosome- or matrix-enriched areas of the rat striatum have been compared. Although AMPA stimulated similarly the release of [3H]GABA in both striatal compartments, the release of [3H]GABA evoked by either N-methyl-D-aspartate (in the presence of AMPA) or carbachol was more pronounced in matrix- than in striosome-enriched areas. AMPA- and N-methyl-D-aspartate- (in the absence of magnesium) evoked responses were reduced but not abolished in the presence of tetrodotoxin (10(-6) M) in both compartments while the carbachol-evoked release of [3H]GABA was decreased by tetrodotoxin only in the matrix. The interruption of cholinergic transmission by the combined application of atropine (10(-5) M) and pempidine (10(-4) M) was without effect on the AMPA-evoked release of [3H]GABA, but it reduced the N-methyl-D-aspartate- (in the absence of magnesium or presence of AMPA) evoked release of [3H]GABA in both compartments, these reductions being of similar amplitude than those observed with tetrodotoxin.

Identification and characterization of striatal cell subtypes using in vivo intracellular recording and dye-labeling in rats: III. Morphological correlates and compartmental localization

In the first two reports of this series, in vivo intracellular recording techniques were used to characterize the electrophysiological properties of two types of striatal neurons that had been identified by their distinct response patterns to stimulation of corticostriatal afferents. In this paper, we examined whether cells showing Type I or Type II response patterns also differed with respect to their morphology or compartmental localization by combining intracellular recording and Lucifer yellow staining with immunocytochemical localization of calbindin 28 kd immunoreactivity. In the majority of cases, both Type I and Type II neurons exhibited similar morphological characteristics, with 80% of the Type I cells (13/16) and all of the Type II cells (n = 40) being small or medium spiny neurons. In each case where the morphological class of the cell was different than the spiny cell class, the cell exhibited a Type I response pattern. These spiny neurons had somata that averaged 17.1 +/- 1.3 microns in diameter and gave rise to between four and eight primary dendrites. The axons typically arose from cell bodies (7/13 for Type I and 25/40 for Type II cells) and emitted extensive local axonal collaterals. However, the axons of Type I cells more frequently originated from the dorsal surface of the somata (9/13; 69%), whereas Type II axons more frequently arose from the ventral surface of the somata (25/35; 71%), which may account for their different extracellular waveforms. In coronally sectioned tissue (n = 18), the axons always projected laterally when the somata were located in the medial striatum and projected medially when the somata were in the lateral striatal region. In a subset of experiments (N = 22), Lucifer yellow-stained neurons were localized with respect to their position within the patch and matrix compartments of the striatum using subsequent staining for calbindin 28 kd immunoreactivity. Of the 20 labeled medium spiny neurons examined (Type II: N = 13; Type I: N = 7), 19 were located in the calbindin-positive matrix compartment. The only neuron localized to the patch compartment was a medium spiny cell that exhibited a Type II paired impulse response pattern. In addition, of the two aspiny neurons from this group with beaded dendrites, one was localized to the border between adjacent patch and matrix compartments, whereas the other was located completely within the matrix compartment. Therefore, despite their distinct paired impulse response patterns, the majority of both Type I and Type II neurons were medium spiny cells located in the matrix compartment of the striatum.(ABSTRACT TRUNCATED AT 400 WORDS)

Islands and striosomes in the neostriatum of the rhesus monkey: non-equivalent compartments

Cytoarchitectonically defined cell-dense islands and regions of low acetylcholinesterase reactivity referred to as striosomes have been regarded as equivalent markers of the non-matrix compartment in the neostriatum. We examined islands and striosomes in adjacent sections to determine the degree of correspondence between the two neostriatal compartmental markers. Islands are aggregated centrally within the caudate, whereas striosomes are located throughout the entire nucleus, including the dorsolateral and ventromedial sectors. Moreover, even within the central sector, striosomes are more prevalent than islands. The present quantitative analysis suggests that islands may be further characterized as acetylcholinesterase-poor since the vast majority of islands co-localize with striosomes. However, due to the fact that striosomes are more numerous and more widely distributed throughout the neostriatum, less than a third of all striosomes are coincident with islands in adjacent sections. Comparison of each of these compartmental markers with the patterned terminal field of the prefrontal cortical projection revealed a near one-to-one correspondence between islands and terminal-free zones in the prefrontal projection. The percentage of striosomes which are aligned with fenestrations in the prefrontal projection is also quite high; however, because more striosomes than islands are found within the prefrontal terminal domain, some striosomes that fit within terminal-free zones do not have corresponding islands. These results indicate that islands and striosomes are not entirely equivalent compartmental markers and further suggest that contemporary, two-compartment models may not adequately represent the heterogeneity of the neostriatum.

Spatial organization of patch and matrix compartments in the rat striatum

The visualization of mu opiate receptors by [3H]naloxone binding was used to determine precisely the spatial organization of the patch compartment in the rat striatum and its reproducibility in different animals. Three-dimensional reconstruction of the patch network was made using maps of autoradiographic data obtained from successive coronal, sagittal or horizontal sections. The extreme rostral pole of the striatum (A 11) was characterized by a large patch territory exhibiting complex and tortuous fields with several extensions. In the intermediate part of the structure (A 9.0-10.0), about 20 serial parallel continuous patch channels running in a mediolateral axis, obliquely oriented and displaying in some cases connecting branches, could be observed. However, no channels could be distinguished in the rostrocaudal direction. More caudally, patches were rare and of small size. In addition, the laterocaudal region of the striatum was almost exclusively represented by a large matrix field. Finally, a fine discontinuous band of [3H]naloxone binding was seen in all sections, bordering and limiting the dorsolateral part of the striatum. The topographical and spatial distribution of the patch compartment was similar in all animals investigated. However, due to the tortuous shape and the labyrinthine organization of the patches, the precise degree of reproducibility from one animal to another could not be established. Nevertheless, the prominent patch compartment observed in the rostral pole of the striatum, the patch channels, oriented in the mediolateral axis as well as the large laterocaudal matrix field were observed in all cases. These results were compared with previous data obtained in the cat in which patch (striosome) channels oriented along a rostrocaudal axis are also observed.

Dendritic arbors of spiny neurons in the primate striatum are directionally polarized

Despite the relatively unfeatured cytoarchitecture of the striatum, this large subcortical region has been found to have a modular macroscopic substructure comprising the neurochemically distinct striosomes and matrix, and, within the matrix, patchy input and output arrangements called matrisomes. In the study reported here, we explored the possibility that the cellular architecture of the striatum is also more specialized than previously suspected. We injected medium spiny neurons in lightly fixed slices of the squirrel monkey caudate nucleus, reconstructed their dendritic arbors, and analyzed the orientations of these arbors with respect to the cardinal planes of the striatum. The data were unequivocal in suggesting that many spiny neurons, whether near striosomes or not, have dendritic arbors with preferred orientations along a diagonal axis running from rostral, dorsal, and medial to caudal, ventral, and lateral. This axis corresponds to the orientations of many striosomes and matrisomes in the squirrel monkey's caudate nucleus. We therefore suggest that the primate striatum is characterized not only by a macroscopic organization dividing it into striosomes and matrisomes, but also by a microscopic architecture observed by the dendritic arbors of many of its projection neurons. We obtained comparable supplementary observations for the ferret caudate nucleus, suggesting that such spatial alignment of spiny dendritic arbors may be a general feature of striatal organization. These polarized dendritic arrangements could provide a cellular framework for compartmental input-output processing within the striatum.

Local GABAergic regulation of the N-methyl-D-aspartate-evoked release of dopamine is more prominent in striosomes than in matrix of the rat striatum

Using an in vitro microsuperfusion device we have previously demonstrated that in the absence of magnesium, the N-methyl-D-aspartate-evoked release of [3H]dopamine (continuously synthesized from [3H]tyrosine) is more prominent in matrix- than in striosome-enriched areas of the rat striatum and that in the matrix, the response is partially tetrodotoxin-sensitive. Since the medium-sized GABAergic neurons are the main targets of the corticostriatal glutamatergic fibers, the involvement of local GABAergic regulation in the N-methyl-D-aspartate-evoked release of [3H]dopamine was investigated in both striatal compartments using the same experimental approach. Firstly, bicuculline alone (5 microM, 25-min application) was shown to enhance the release of [3H]dopamine similarly in both compartments revealing the existence of a tonic GABAergic control of the spontaneous release of [3H]dopamine. Secondly, the N-methyl-D-aspartate (50 microM, 25-min application)-evoked release of [3H]dopamine was markedly amplified in the presence of bicuculline (5 microM, continuous delivery). This effect being more important in striosome- than in matrix-enriched areas (5.5- and two-times the N-methyl-D-aspartate-evoked response observed in the absence of the GABAA antagonist, respectively). Thirdly, the tetrodotoxin (1 microM, continuous delivery)-resistant N-methyl-D-aspartate-evoked responses were also enhanced in the presence of bicuculline, but in this case, the amplification of the N-methyl-D-aspartate-evoked release of [3H]dopamine was less marked than in the absence of tetrodotoxin and identical in both compartments (about two-times the tetrodotoxin-resistant N-methyl-D-aspartate-evoked responses observed in the absence of bicuculline). Altogether, these results indicate that GABAergic neurons exert locally an important inhibitory regulation of the N-methyl-D-aspartate-evoked release of dopamine and that this effect is more prominent in the striosome-enriched area. Both tetrodotoxin-sensitive (striosome) and tetrodotoxin-resistant (striosome and matrix) processes intervene in this inhibitory GABAergic presynaptic regulation of dopamine release.

Species-specific patterns of glycoprotein expression in the developing rodent caudoputamen: association of 5'-nucleotidase activity with dopamine islands and striosomes in rat, but with extrastriosomal matrix in mouse

The glycoprotein 5'-nucleotidase is a cell surface phosphatase and represents a new marker for striosomes in the adult rat caudoputamen. We report here on its developmental expression in the rat and mouse striatum, and show an unexpected converse 5'-nucleotidase chemoarchitecture of the caudoputamen in these closely related species. In the rat, 5'-nucleotidase activity was first visible as neuropil staining in tyrosine hydroxylase-positive dopamine islands of the midstriatum on postnatal day 1, and by the end of the first postnatal week, 5'-nucleotidase-positive dopamine islands also appeared rostrally. This compartmental pattern persisted thereafter, so that in adult animals, in all but the caudal caudoputamen, zones of enhanced 5'-nucleotidase staining were restricted to calbindin-D28k-poor striosomes. Weak 5'-nucleotidase activity also emerged in the matrix. In striking contrast, in the mouse striatum, enhanced 5'-nucleotidase activity was preferentially associated with extrastriosomal tissue. Enzymatic reaction first appeared on embryonic day 18, and developed over the first postnatal week into a mosaic pattern in which the matrix was stained but the dopamine islands were unstained. The matrix staining itself was heterogeneous. After the second postnatal week, most of the caudoputamen was stained, and in adult mice only rostral striosomes expressed low 5'-nucleotidase activity. We conclude that in rats, 5'-nucleotidase represents one of the few substances that maintains a preferential dopamine island/striosome distribution during striatal development. In mice, 5'-nucleotidase activity is expressed preferentially in the matrix during development, and its compartmental pattern is gradually lost with maturation, except very rostrally. These findings do not suggest an instructive role of the enzyme in striatal compartment formation in either species, but do suggest the possibility that 5'-nucleotidase contributes to the differentiation of striatal compartments during development.

Heterogeneous distribution of D1, D2 and D5 receptor mRNAs in monkey striatum

The primate striatum has a compartmental organization reflected both in the topography of its afferent projections and in the segregation of its morphologically similar but neurochemically distinct efferent neurons. Discretely projecting mesostriatal neurons release dopamine (DA) which modulates the responses of striatal neurons to other afferent inputs. Multiple DA receptor (DAR) subtypes have been cloned and characterized and mapping their cellular expression is crucial for understanding the influence of DA on striatal function. We report the distribution of mRNAs for D1, D2 and D5 DAR subtypes (D2R, D2R and D5R) in the striatum of cynomolgus monkeys (Macaca fascicularis) studied by in situ hybridization histochemistry (ISH) using monkey-specific cRNA probes. Adjacent sections were stained for calbindin immunoreactivity to distinguish striosomal and matrix compartments for comparison with the patterns obtained with ISH. In the caudate nucleus, D1R mRNA was concentrated in calbindin-poor striosomes where dense grain clusters were seen overlying the majority of medium-sized neurons (diameter approximately 15 microns). D1R mRNA localization was relatively homogeneous in the putamen. By contrast, the distributions of D2R and D5R mRNAs showed no clear preference for the striosomal or matrix compartments of either caudate nucleus or putamen. In the ventral striatum (nucleus accumbens, olfactory tubercle and ventral portions of caudate nucleus and putamen), expression of D1R and D2R mRNA was sparse relative to dorsal striatum, while D5R mRNA expression was roughly equal in ventral and dorsal striatum. Circumscribed zones of hybridization associated with islands of tightly packed small cells occurred with all three DAR mRNA subtypes in the ventral striatum.(ABSTRACT TRUNCATED AT 250 WORDS)

Compartmental origin of the striato-entopeduncular projection in the rat

The mammalian neostriatum is divisible into neurochemically and cytoarchitectonically distinct striosome and matrix compartments. This compartmentalization is respected by many afferent and efferent projections of the striatum. The distribution of distinct types of neuroactive substances and receptors and the unique connections of the striosome and matrix suggest a functional segregation between these compartments. The present study examines the organization of efferent projections from each of the striatal compartments to the entopeduncular nucleus (EPN), a major output center of the basal ganglia. The fluorescent retrograde tracer fluorogold, or rhodamine-conjugated dextran, was injected into the lateral habenula or the ventrolateral nucleus of the thalamus of adult Wistar rats to identify the topographical organization of EPN-habenular and EPN-thalamic neurons. Fluorogold was then placed into the rostral or caudal parts of the EPN, identified from the previous experiment as areas containing predominantly EPN-habenular or EPN-thalamic neurons, respectively. Sections containing retrogradely labeled neurons in the neostriatum were simultaneously immunolabeled for calbindin-D28kDa, a calcium-binding protein found exclusively in the projection neurons of the matrix. The results indicate that the striatal projection to the EPN-habenular and EPN-thalamic parts of the EPN originates from striosome and matrix neurons, respectively. The duality of striatal outflow involving the EPN suggests a mechanism whereby the striosome is integrated into subcortical pathways that modulate the activity of the basal ganglia via the ascending serotoninergic projection from the dorsal raphe nucleus, whereas the matrix is involved in a loop that includes the thalamus and the cerebral cortex.

Distinct presynaptic control of dopamine release in striosomal- and matrix-enriched areas of the rat striatum by selective agonists of NK1, NK2, and NK3 tachykinin receptors

Using a sensitive in vitro microperfusion method, the effects of selective and potent agonists of NK1, NK2, and NK3 tachykinin receptors ([Pro9]SP, ([Lys5,MeLeu9,Nle10]NKA-(4-10), and [Pro7]NKB, respectively) on the presynaptic control of dopamine release were investigated in striosomal-enriched (area rich in [3H]naloxone binding sites) and matrix-enriched areas of the rat striatum. Marked differences could be demonstrated as follows: (i) when used at 0.1 microM, the NK1 agonist stimulated the release of [3H]dopamine continuously synthesized from [3H]tyrosine in both compartments, while the NK2 and NK3 agonists enhanced the release of [3H]dopamine only in the matrix; (ii) the stimulatory effect of the NK3 agonist was less pronounced than those of the NK1 and NK2 agonists; (iii) the NK1 agonist-evoked responses were tetrodotoxin (1 microM) sensitive, while those of the NK2 and NK3 agonists were, respectively, partially and totally tetrodotoxin resistant; (iv) specific receptors are involved in these responses since the stimulatory effects of the NK1 and NK2 agonists were, respectively, blocked by potent antagonists of NK1 (RP-67580; 1 microM) and NK2 (SR-48968; 1 microM) receptors, while these antagonists did not affect the NK3 agonist-evoked response; (v) the indirect stimulatory effect of the NK1 agonist was partially reduced under local blockade of cholinergic transmission in the matrix but not in the striosomal-enriched area. Interestingly, this study also revealed mismatches between autoradiographic data and receptor-mediated responses, since NK2 binding sites could not be observed in the striatum while NK3 but not NK1 binding sites were visualized in the striosomal-enriched area.

Functional heterogeneity of the matrix compartment in the cat caudate nucleus as demonstrated by the cholinergic presynaptic regulation of dopamine release

Previously, using a new in vitro microsuperfusion procedure, we have demonstrated marked differences in the cholinergic presynaptic regulation of the release of [3H]dopamine continuously synthesized from [3H]tyrosine in two close striosomal- and matrix-enriched areas of the cat caudate nucleus. A tetrodotoxin-resistant stimulatory effect of acetylcholine mediated by muscarinic receptors was observed in both compartments. However, in addition, two opposing types of tetrodotoxin-sensitive acetylcholine-evoked regulation of [3H]dopamine release were only seen in the matrix: one facilitatory, involving nicotinic receptors located on as yet unidentified neurons, and the other inhibitory, mediated by muscarinic receptors located on dynorphin-containing neurons. In the present study, using the same approach, a functional heterogeneity was demonstrated in the matrix. Indeed, in various conditions the effects of acetylcholine (50 microM) on the release of [3H]dopamine were different in a matrix-enriched area (matrix 2) distinct from that previously investigated (matrix 1); these areas being characterized by the presence or absence of islands of striatonigral cells, respectively. As in matrix 1, acetylcholine induced a short-lasting stimulation of [3H]dopamine release in matrix 2 but, in contrast to that observed in matrix 1, the acetylcholine-evoked response in matrix 2 was not modified in the presence of tetrodotoxin (1 microM). Experiments made in the presence of the tetrodotoxin and atropine (1 microM) indicated that both muscarinic and nicotinic receptors are located on dopaminergic nerve terminals in matrix 2 while muscarinic receptors are only present in matrix 1. In the absence of tetrodotoxin, the short-lasting stimulation of [3H]dopamine release was transformed into a long-lasting response in the presence of pempidine (50 microM), in matrix 2 but not in matrix 1 while prolonged responses were seen in both matrix areas in the presence of atropine. Finally, the acetylcholine short stimulatory effect on [3H]dopamine release was transformed into a long stimulatory response in the presence of bicuculline (50 microM) but not naloxone (1 microM) in matrix 2 while the reverse was observed in matrix 1. By providing further evidence for a functional heterogeneity of the matrix, our results suggest that depending on the matrix area investigated, dynorphin- or GABA-containing neurons are involved in the indirect cholinergic inhibitory control of dopamine release.

Heterogeneous development of calbindin-D28K expression in the striatal matrix

In the present study, we attempted to trace the development of the striatal matrix by analyzing the ontogenetic expression of calbindin-D28K (calbindin), a calcium binding protein selectivity expressed in medium-sized neurons of the matrix compartment of the mature rat's caudoputamen. The localization of calbindin was documented in a series of developing rat brains, as was the compartmental location of these cells relative to tyrosine hydroxylase (TH)-immunostained dopamine islands, sites of future striosomes. Medium-sized striatal neurons appeared in the striatum at embryonic day (E) 20, and from their first appearance, the calbindin-positive neurons had highly heterogeneous distributions. They first formed a latticework of patches and bands in a ventral region of the caudoputamen. By postnatal day (P) 7, this early calbindin-positive lattice had evolved into a mosaic in which circumscript pockets of low calbindin-like immunoreactivity appeared in more extensive calbindin-rich surrounds. With further development, the mosaic gradually encroached on all but the dorsolateral caudoputamen, a district that is calbindin-poor at adulthood. A special lateral branch of the striatal calbindin system was also identified, distinct from the rest of the calbindin-positive mosaic in several developmental characteristics. In the parts of the caudoputamen where the developing calbindin system and dopamine island system were both present, the dopamine islands invariably lay in calbindin-poor zones. Most dopamine islands, however, only filled parts of the corresponding calbindin-poor zones. Moreover, there were some calbindin-poor zones for which TH-positive dopamine islands could not be detected. Thus during development, calbindin was expressed in the extrastriosomal matrix of the striatum, but the matrix could be divided into calbindin-rich and calbindin-poor zones. In the calbindin-rich regions, there were patches of especially intense calbindin expression and zones of weaker expression. These results suggest that there is neurochemical heterogeneity in the striatal matrix during the prolonged developmental period in which the early calbindin-positive lattice expands to form the calbindin-positive matrix of the mature striatum. Surprisingly, calbindin expression in the matrix, although eventually distributed in strictly complementary fashion to striosomes, does not originate as a system complementary to dopamine islands. The prolonged disparity between the borders of dopamine islands and calbindin-poor zones, and the different spatiotemporal schedules of development of the islands and the calbindin gaps suggest instead that the final match between the borders of striosomes and surrounding matrix results from dynamic processes occurring early in postnatal development. Candidate mechanisms for the gradual adjustment of these borders are proposed.

Efferent connections of the centromedian and parafascicular thalamic nuclei in the squirrel monkey: a light and electron microscopic study of the thalamostriatal projection in relation to striatal heterogeneity

The organization of the thalamostriatal projections arising from the centromedian (CM) and parafascicular (Pf) thalamic nuclei in the squirrel monkey (Saimiri sciureus) was studied at both light and electron microscopic levels. Following selective injections of the anterograde axonal tracer Phaseolus vulgaris-leucoagglutinin (PHA-L) into the CM or Pf, patterns of terminal arborization within the striatum were compared to the biochemical heterogeneity of the striatum as revealed by immunohistochemical staining for the calcium-binding protein calbindin D-28k (CaBP), and histochemical staining for the enzymes acetylcholinesterase (AChE) and nicotinamide adenine dinucleotide phosphate-diaphorase (NADPH-diaphorase). The PHA-L-labeled axon terminals within the striatum were further analyzed at the ultrastructural level to characterize their pattern of synaptic organization. Dense and heterogeneous terminal fields occur in the "sensorimotor" territory of the striatum after CM injections, or in the "associative" striatal territory following Pf injections. In the associative territory labeled axons arborize in a diffuse manner predominantly within areas enriched with CaBP, AChE, or NADPH-diaphorase, representing the matrix compartment, and tend to avoid areas poor in these substances, corresponding to the patch/striosome compartment. In the sensorimotor territory labeled axons form bands that occupy a subregion of the NADPH-diaphorase-rich zone in the putamen. The terminal pattern of the CM-striatal projection suggests the existence of a more complex mosaic organization within the sensorimotor territory. Ultrastructural analysis of PHA-L-labeled elements within the striatum reveals that both CM and Pf projections form asymmetric synapses upon dendrites and spines of striatal cells. A total of 339 PHA-L-labeled boutons were examined after CM injections and compared to 293 boutons following Pf injections. After CM injections, 29% of PHA-L-labeled terminals form synapses on dendritic spines and 66% on dendritic shafts, whereas after Pf injections only 12% of synapses occur on dendritic spines compared to 81% on dendritic shafts. Labeled terminals forming axosomatic or axoaxonic synapses were not seen within the striatum following either CM or Pf injections. It is concluded that in the squirrel monkey: 1) Pf-striatal fibers profusely arborize within the matrix compartment of the associative territory, 2) CM-striatal fibers form bands that occupy a subregion of the NADPH-diaphorase-rich zone within the sensorimotor territory, and 3) that both Pf- and CM-striatal projections establish asymmetric synapses with dendrites and spines of medium-sized spiny cells.(ABSTRACT TRUNCATED AT 400 WORDS)