Loss-of-function mutations in MICU1 cause a brain and muscle disorder linked to primary alterations in mitochondrial calcium signaling.

Mitochondrial Ca(2+) uptake has key roles in cell life and death. Physiological Ca(2+) signaling regulates aerobic metabolism, whereas pathological Ca(2+) overload triggers cell death. Mitochondrial Ca(2+) uptake is mediated by the Ca(2+) uniporter complex in the inner mitochondrial membrane, which comprises MCU, a Ca(2+)-selective ion channel, and its regulator, MICU1. Here we report mutations of MICU1 in individuals with a disease phenotype characterized by proximal myopathy, learning difficulties and a progressive extrapyramidal movement disorder. In fibroblasts from subjects with MICU1 mutations, agonist-induced mitochondrial Ca(2+) uptake at low cytosolic Ca(2+) concentrations was increased, and cytosolic Ca(2+) signals were reduced. Although resting mitochondrial membrane potential was unchanged in MICU1-deficient cells, the mitochondrial network was severely fragmented. Whereas the pathophysiology of muscular dystrophy and the core myopathies involves abnormal mitochondrial Ca(2+) handling, the phenotype associated with MICU1 deficiency is caused by a primary defect in mitochondrial Ca(2+) signaling, demonstrating the crucial role of mitochondrial Ca(2+) uptake in humans.

Neuroradiological and Neurophysiological Characteristics of Patients With Dyskinetic Cerebral Palsy

OBJECTIVE: To investigate neuroradiological and neurophysiological characteristics of patients with dyskinetic cerebral palsy (CP), by using magnetic resonance imaging (MRI), voxel-based morphometry (VBM), diffusion tensor tractography (DTT), and motor evoked potential (MEP). METHODS: Twenty-three patients with dyskinetic CP (13 males, 10 females; mean age 34 years, range 16-50 years) were participated in this study. Functional evaluation was assessed by the Gross Motor Functional Classification System (GMFCS) and Barry-Albright Dystonia Scale (BADS). Brain imaging was performed on 3.0 Tesla MRI, and volume change of the grey matter was assessed using VBM. The corticospinal tract (CST) and superior longitudinal fasciculus (SLF) were analyzed by DTT. MEPs were recorded in the first dorsal interossei, the biceps brachii and the deltoid muscles. RESULTS: Mean BADS was 16.4+/-5.0 in ambulatory group (GMFCS levels I, II, and III; n=11) and 21.3+/-3.9 in non-ambulatory group (GMFCS levels IV and V; n=12). Twelve patients showed normal MRI findings, and eleven patients showed abnormal MRI findings (grade I, n=5; grade II, n=2; grade III, n=4). About half of patients with dyskinetic CP showed putamen and thalamus lesions on MRI. Mean BADS was 20.3+/-5.7 in normal MRI group and 17.5+/-4.0 in abnormal MRI group. VBM showed reduced volume of the hippocampus and parahippocampal gyrus. In DTT, no abnormality was observed in CST, but not in SLF. In MEPs, most patients showed normal central motor conduction time. CONCLUSION: These results support that extrapyramidal tract, related with basal ganglia circuitry, may be responsible for the pathophysiology of dyskinetic CP rather than CST abnormality.

Neuroradiological and Neurophysiological Characteristics of Patients With Dyskinetic Cerebral Palsy

OBJECTIVE: To investigate neuroradiological and neurophysiological characteristics of patients with dyskinetic cerebral palsy (CP), by using magnetic resonance imaging (MRI), voxel-based morphometry (VBM), diffusion tensor tractography (DTT), and motor evoked potential (MEP). METHODS: Twenty-three patients with dyskinetic CP (13 males, 10 females; mean age 34 years, range 16-50 years) were participated in this study. Functional evaluation was assessed by the Gross Motor Functional Classification System (GMFCS) and Barry-Albright Dystonia Scale (BADS). Brain imaging was performed on 3.0 Tesla MRI, and volume change of the grey matter was assessed using VBM. The corticospinal tract (CST) and superior longitudinal fasciculus (SLF) were analyzed by DTT. MEPs were recorded in the first dorsal interossei, the biceps brachii and the deltoid muscles. RESULTS: Mean BADS was 16.4+/-5.0 in ambulatory group (GMFCS levels I, II, and III; n=11) and 21.3+/-3.9 in non-ambulatory group (GMFCS levels IV and V; n=12). Twelve patients showed normal MRI findings, and eleven patients showed abnormal MRI findings (grade I, n=5; grade II, n=2; grade III, n=4). About half of patients with dyskinetic CP showed putamen and thalamus lesions on MRI. Mean BADS was 20.3+/-5.7 in normal MRI group and 17.5+/-4.0 in abnormal MRI group. VBM showed reduced volume of the hippocampus and parahippocampal gyrus. In DTT, no abnormality was observed in CST, but not in SLF. In MEPs, most patients showed normal central motor conduction time. CONCLUSION: These results support that extrapyramidal tract, related with basal ganglia circuitry, may be responsible for the pathophysiology of dyskinetic CP rather than CST abnormality.

Dopamine transporter inhibitory and antiparkinsonian effect of common flowering quince extract.

Common flowering quince (FQ) is the fruit of Chaenomeles speciosa (Sweet) Nakai. FQ-containing cocktails have been applied to the treatment of neuralgia, migraine, and depression in traditional Chinese medicine. The present study assessed whether FQ is effective in dopamine transporter (DAT) regulation and antiparkinsonism by utilizing in vitro and in vivo assays, respectively. FQ at concentrations of 1-1000 microg/ml concentration-dependently inhibited dopamine uptake by Chinese hamster ovary (CHO) cells stably expressing DAT (D8 cells) and by synaptosomes. FQ had a slight inhibitory action on norepinephrine uptake by CHO cells expressing the norepinephrine transporter and no inhibitory effect on gamma-aminobutyric acid (GABA) uptake by CHO cells expressing GABA transporter-1 or serotonin uptake by the serotonin transporter. A viability assay showed that FQ mitigated 1-methyl-4-phenylpyridinium-induced toxicity in D8 cells. Furthermore, in behavioral studies, FQ alleviated rotational behavior in 6-hydroxydopamine-treated rats and improved deficits in endurance performance in 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-treated mice. Furthermore, immunohistochemistry revealed that FQ markedly reduced the loss of tyrosine hydroxylase-positive neurons in the substantia nigra in MPTP-treated mice. In summary, FQ is a selective, potent DAT inhibitor and has antiparkinsonian-like effects that are mediated possibly by DAT suppression. FQ has the potential to be further developed for Parkinson's disease treatment.

'Importance sampling' in MS: use of diffusion tensor tractography to quantify pathology related to specific impairment.

Specific neurological impairments in multiple sclerosis (MS) are dependent on the pathology in clinically eloquent areas of the central nervous system. We aimed to use diffusion tensor fiber tracking to identify the pyramidal tracts and corpus callosum in MS patients, measure the apparent diffusivity within the tracts, and evaluate whether this would correlate with relevant disability scores. Dual-echo and diffusion tensor magnetic resonance imaging (DT-MRI) brain scans were obtained from 29 patients with relapsing remitting MS, and 13 age and gender matched normal controls. Voxels from pyramidal tracts and corpus callosum were automatically identified using a tractography based algorithm. Mean apparent diffusion coefficient (ADC(av)) was measured for these tracts. Scores of Expanded Disability Status Scale (EDSS) and Paced Auditory Serial Addition Test (PASAT) were obtained. The median EDSS score was 2.5 (inter-quartile range 2-3.25). The ADC(av) in the pyramidal tracts (p=0.02) and corpus callosum (p=0.0004) in patients was significantly higher than in controls. Pyramidal tracts ADC(av) was correlated with pyramidal FSS (r=0.5, p=0.008). Corpus callosum ADC(av) was correlated with PASAT (r=-0.58, p=0.001). Global T2 lesion volume did not correlate with the EDSS, but correlated with ADC(av) of the pyramidal tracts (r=0.6, p=0.0007) and corpus callosum (r=0.8, p<0.0001). T2 lesion volume within the pyramidal tracts and corpus callosum correlated with ADC(av) in the pyramidal tracts (r=0.6, p=0.0009) and corpus callosum (r=0.65, p=0.0002) respectively, but not with pyramidal FSS or PASAT score. DT-MRI quantifies pathology in specific white matter tracts and may increase the specificity of MRI in monitoring progression of motor and cognitive deficits in MS.

[Features and mechanisms of realization the respiratory influences of the extrapyramydal system structures]. / Osobennosti i mechanizmy realizatsii respiratornykh vliianii struktur ékstrapiramidnoi sistemy.

In the publication the modern condition of the problem of suprabulbar regulation of breathing is analysed. The review on structure, neurochemistry and anatomic connections of the red nucleus and substantia nigra with the medullary respiratory center is submitted. The data on the respiratory effects of GABA and apomorphine microinjected into the red nucleus and substantia nigra as well as effects of their electrostimulation after the blockade of GABA and dopamine receptors in the respiratory center are discussed. The conceptual scheme of the mechanisms of realization the respiratory influences of the extrapyramydal system is offered.

Reduced regional [14C]2-deoxyglucose uptake in response to long-term clozapine administration in rats.

Clozapine has superior effects in treating negative symptoms of schizophrenia and causes less extrapyramidal side effects than traditional antipsychotics. In this study, we investigated the effects of acute and long-term clozapine administration on [14C]2-deoxyglucose uptake (2-DG uptake) in rats, as measured using the [14C]2-deoxy-D-glucose method. The 2-DG uptake was reduced in fewer regions after chronic clozapine (46%) than after acute clozapine (97%). After chronic clozapine treatment, the 2-DG uptake was reduced in the shell, but not the core, of the nucleus accumbens. In addition, long-term clozapine treatment remained affecting 2-DG uptake in several regions of the extrapyramidal system and the thalamus. The pattern of 2-DG uptake changes after long-term clozapine administration may provide information for the regions related to the therapeutic effect of clozapine.

Extrapyramidal involvement in amyotrophic lateral sclerosis: backward falls and retropulsion.

Three patients with sporadic amyotrophic lateral sclerosis (ALS) presented with a history of backward falls. Impaired postural reflexes and retropulsion accompanied clinical features of ALS. Hypokinesia, decreased arm swing, and a positive glabellar tap were noted in two of these three patients. Cognitive impairment, tremor, axial rigidity, sphincter dysfunction, nuchal dystonia, dysautonomia, and oculomotor dysfunction were absent. Brain MRI disclosed bilateral T2 weighted hyperintensities in the internal capsule and globus pallidus in one patient. Necropsy studies performed late in the course of ALS have shown degeneration in extrapyramidal sites-for example, the globus pallidus, thalamus, and substantia nigra. Clinically, backward falls and retropulsion may occur early in ALS. This may reflect extrapyramidal involvement.

[Changes in the ultrastructure of the motor zone of the cortex and various areas of the rats pyramidal system in acute toxic action of phenazepam]. / Izmeneniia ul'trastruktury dvigatel'noi zony kory i nekotorykh otdelov ékstrapiramidnoi sistemy krys pri ostrom toksicheskom deistvii fenazepama.

As it was revealed by electron microscopic study, phenazepam (benzodiazepine representative) administered to rats in acute toxic dose caused astrocyte processes swelling mainly in reticular portion of substantia nigra and cerebral areas with high density of GABAergic synapses in initial intoxication stage (one hour after the drug introduction). Following 24 hrs changes of axodendritic and axo-somatic synapse postsynaptic part were detected similar to those of myoneural synapses induced by anti-cholinesterase agents, which allows to suggest the contribution of the increased intracellular calcium ions content. Specific benzodiazepine receptor agonist-flumazenil, introduced 30 min later than phenazepam leaded to the decrease of ultrastructural changes in postsynaptic part of the synapses in reticular zone of substantia nigra. Thus toxic effect of drugs that cause prolonged activity of peripheral and central synapses postsynaptic part results in the alike ultrastructural changes irrespective of their mediator type and is likely to be connected with the growth of intracellular calcium concentration.

Differential distribution of N-acetylaspartylglutamate and N-acetylaspartate immunoreactivities in rat forebrain.

Contradictory immunohistochemical data have been reported on the localization of N-acetylaspartylglutamate in the rat forebrain, using different carbodiimide fixation protocols and antibody purification methods. In one case, N-acetylaspartylglutamate immunoreactivity was observed in apparent interneurons throughout all allocortical and isocortical regions, suggesting possible colocalization with GABA. In another case, strong immunoreactivity was observed in numerous pyramidal cells in neocortex and hippocampus, suggesting colocalization with glutamate or aspartate. Reconciling these disparate findings is crucial to understanding the role of N-acetylaspartylglutamate in nervous system function. Antibodies to N-acetylaspartylglutamate and a structurally related molecule, N-acetylaspartate, were purified in stages, and their cross-reactivities with protein conjugates of N-acetylaspartylglutamate and N-acetylaspartate were monitored at each stage by solid-phase immunoassay. Reduction of the cross-reactivity of the anti-N-acetylaspartylglutamate antibodies of N-acetylaspartate-protein conjugates to about 1% eliminated significant staining of most pyramidal neurons in the rat forebrain. Utilizing highly purified antibodies, the distributions of N-acetylaspartylglutamate and N-acetylaspartate were examined in several major telencephalic and diencephalic regions of the rat, and were found to be distinct. N-acetylaspartylglutamate-immunoreactivity was observed in specific neuronal populations, including many groups thought to use GABA as a neurotransmitter. Among these were the globus pallidus, ventral pallidum, entopeducular nucleus, thalamic reticular nucleus, and scattered non-pyramidal neurons in all layers of isocortex and allocortex. N-acetylaspartate-immunoreactivity was more broadly distributed than N-acetylaspartylglutamate-immunoreactivity in the rat forebrain, appearing strongest in many pyramidal neurons. Although N-acetylaspartate-immunoreactivity was found in most neurons, it exhibited a great range of intensities between different neuronal types.

[The analysis of the mechanisms of both the SM and the extrapyramidal system modulating the nociceptive responses of the neurons in NRM].

The experiment was carried out on rats. Unit discharges of NRM were recorded extracellularly with glass microelectrode. Before the lesion of the head of N. Caudatus, 15 NRM neurons were recorded. The stimulation of the hindlimb areas of SM could obviously activate NRM neurons, the spontaneous discharges were increased to 59.06 +/- 25.35% (P < 0.05) and the nociceptive responses were obviously inhibited to -46.61 +/- 14.78 (P < 0.01) at 0 minute after stopping stimulation. Both effects lasted for 15 minutes, and then gradually returned to original level. After the lesion of the head of N. Caudatus, the effects of the stimulation of the hindlimb areas of SM on 19 NRM neurons were clearly decreased. The spontaneous discharges were increased to only 6.9 +/- 21.46% and the nociceptive responses were inhibited only -4.57 +/- 11.08% at 0 minute after stopping stimulation, even the nociceptive responses were slightly increased in 5-30 minutes. The difference between the two groups was statistical significance. The value of the difference was about 42% (P < 0.05) at 0 minute and lasted 30 minutes. It suggests that the effects of stimulating SM on NRM may be transmitted by the head of N. Caudatus in the extrapyramidal system.

[History of the basal ganglia system. Slow development of a major cerebral system]. / Histoire du système des ganglions de la base. La lente émergence d'un système cérébral majeur.

Initially, basal ganglia was a descriptive term for onto- and phylogenetic or topographic classifications. A variable list of structures were included as basal ganglia. A major step was made when the thalamus was separated from the "striated bodies" (Vic d'Azyr, 1786) which was sometimes taken into account in the French description of the noyaux gris centraux. Even if the term is not perfect, it is preferable to "the system of basal ganglia". The subdivisions of the putamen, the distinction between the striatum and the pallidum were not really made until the beginning of the twentieth century. Modern tracing methods were needed to demonstrate the main connections. It was not until the end of the 1960s that the importance of the striato-pallido-nigral network within the basal ganglia and the cortico-striatal connections, the main afferent system, were recognized. With the description of the cortico-striatal connections, the sub-cortical system with multiple complex "loops" was questioned. The term "extra-pyramidal system" had an exaggerated success. Initially, it designated descending non-pyramidal afferents (some which do not exist) and their source. In 1992, Spatz based his separation of this heterogeneous group on the iron content. The terms of extra-pyramidal "system" and "syndrome" should be abandoned by clinicians. Physiological interpretations have varied. The role of automatic "habitual" motricity, derived from a concept of hierarchic, Jacksonian cerebral organization, was questioned when the pyramidal network was described. Clinico-pathological analysis (hemiballism, Parkinson's disease ...) has placed new emphasis on the motor role, for a time the only role accepted as real. More recently, debate has centred on other roles, particularly in cognition and motivation. An illustration of functions other than purely motor functions of the basal ganglia is given by the syndromes of loss of psychic auto-activation secondary to bilateral lesions.

Savoxepine fails to selectively influence glucose metabolism in the rat limbic system.

The [14C]-2-deoxyglucose method was used to map the in vivo metabolic response of glucose to savoxepine, a novel tetracyclic cyano-dibenzoxepino-azepine. Savoxepine is reported to have higher affinity for dopamine (DA) receptors in the hippocampus than in the striatum and hence should have dose-dependent, anatomically selective actions. Two doses of savoxepine (0.05 mg/kg and 0.5 mg/kg) were compared with haloperidol (1 mg/kg) to test the hypothesis that low doses of savoxepine would display a selective action on limbic brain areas. Results failed to show that low dose savoxepine selectively modifies glucose utilization in the limbic system as previous biochemical studies suggested. In fact, low doses of the drug displayed a potent activity quite similar to haloperidol in effect and localization. The low dose did not produce significantly altered glucose metabolism in the nucleus accumbens or in the lateral habenular nucleus as observed with most other neuroleptics, suggesting a lack of antipsychotic action at this dose. Our findings demonstrate the difficulty of designing a neuroleptic with a preferential blockade of limbic DA receptors and point to the need for functional assessment of regional receptor binding differences.

Role of female gonadal hormones in the CNS: clinical and experimental aspects.

The large body of evidence presented indicates that in the brain the action of sex hormones cannot be thought as restricted to the regulation of endocrine functions and mating behavior. Estrogens and progesterone seem to act in numerous regions of the CNS to regulate motor as well as limbic functions. Furthermore, the data reviewed indicate that these hormones may modulate neuronal activity through a wide variety of mechanisms. More studies should focus on such mechanisms in order to better understand the role of sex hormones in the CNS and to devise ways of limiting their effects on depression, epilepsy etc. It is known that in peripheral target organs these hormones modulate cell activities by binding to specific receptors which can recognize the DNA sequence and activate the transcription of selected genes (135, 136). There is evidence supporting the hypothesis that this mechanism of action has been conserved also in the brain. First, the brain receptors for progesterone and estrogens are functionally and biochemically indistinguishable from those in the periphery (4, 5): they may be concentrated in neuronal nuclei and bind chromatin "in vitro" (7). Second, a temporal relationship has been observed between administration of steroids and the increase of polymerase II activity (137) and protein synthesis (4, 5). Third, various hormone-induced behaviors may be blocked by inhibitors of the protein synthesis (138, 139, 140, 141). However, sex hormones must be capable to regulate neuronal functions by mechanisms other then genomic. In fact, the topical application of estrogen or progesterone on nervous tissue results in a rapid change of membrane potential (60, 71). Such a rapid effect is not likely to be the consequence of nuclear action, but rather must be related to events occurring on the cell surface. It has been hypothesized that sex steroids affect the fluidity of the cell membrane, therefore modifying the ion transport or neurotransmitter receptor activity (142). If this were the case we would expect to observe a similar effect after application of any steroid. Experimental evidence demonstrates that not all the steroids affect the nervous membrane potential. Moreover, two steroids, estradiol and progesterone, have been described to modulate membrane potential in an opposite way (66, 67, 69, 75). At the moment, there is no evidence for the presence of steroid receptors on neuronal membranes which could mediate the described phenomena.(ABSTRACT TRUNCATED AT 400 WORDS)

Angiotensin-converting enzyme in discrete areas of the rat forebrain and pituitary gland.

With the use of a sensitive radioisotopic method we have examined the activity of the angiotensin-converting enzyme (ACE, E.C. 3.4.15.1) in specific nuclei of the rat forebrain and in the anterior, intermediate and posterior lobes of the pituitary gland of the rat. We reported that ACE activity is heterogeneously distributed in the rat forebrain, with a 200-fold difference between the lowest and the highest values. Highest enzyme activities were found in the subfornical organ and in the posterior lobe of the pituitary gland. High ACE activity was also detected in the intermediate and anterior lobes of the pituitary gland, the caudate nucleus, and the medial habenular nucleus. Substantial activity also existed in the globus pallidus, the median eminence, the supraoptic and paraventricular nuclei, the lateral habenular nucleus and the organon vasculosum laminae terminalis. Our results demonstrate that one of the components of the renin-angiotensin system, the angiotensin-converting enzyme, is highly localized to a few discrete brain structures and the pituitary gland. These findings suggest that angiotensin II could be formed locally in some of these structures, supporting previous immunohistochemical data.

Cortical modulation of striatal function.

The effect of bilateral section of the corticostriatal projections or of selective bilateral ablation of the frontal cortex on behavioral and biochemical parameters related to striatal function were investigated in the rat. Either lesion almost completely prevented the cataleptogenic action of haloperidol: this effect was observed as soon as 3 days and lasted for at least 3 months after surgery, paralleling a reduction in striatal glutamate uptake. Also, such lesions enhanced the apomorphine-induced stereotyped behavior (as measured 21 days after surgery). In the striatum, dopamine, dihydroxyphenylacetic acid, acetylcholine and substance P levels as well as choline acetyltransferase and glutamic acid decarboxylase activities were unaffected 10 or 21 days after either type of lesion. In the substantia nigra, substance P levels were unchanged 10 days following suction of the frontal cortex, but glutamic acid decarboxylase was reduced at 21 days postsurgery. Cortical lesions only partially prevented the reduction in striatal acetylcholine concentrations and did not affect the increase in striatal dihydroxyphenylacetic acid caused by haloperidol. Finally, lesions of the corticostriatal pathways failed to affect the apomorphine-induced increase in striatal acetylcholine levels, reduction of the potassium (20 mM) evoked [3H]acetylcholine release in striatal slices preloaded with [3H]choline and decrease of striatal dihydroxyphenylacetic acid concentrations. These findings indicate that the frontal cortex influences extrapyramidal function by a mechanism which--in behavioral terms--is antagonistic to dopamine-mediated events. As indicated by the biochemical data, this mechanism does not involve changes in striatal dopaminergic and cholinergic neuron activity. This mechanism may utilize: (1) corticostriatal glutamatergic neurons as suggested by the reduction in striatal glutamate uptake following lesions; and (2) GABAergic pathways as suggested by the reduction of nigral glutamic acid decarboxylase activity as well as by the finding that GABA receptor agonists reinstate haloperidol-induced catalepsy.

Organization of afferent connections of the feline lateral habenular nucleus.

The organization of afferent projections to the lateral habenular nucleus (LHB) was studied in adult cats. In autoradiographic experiments, tritiated leucine injections were made into regions shown in retrograde transport studies to project to LHB (lateral hypothalamus, preoptic region, entopeduncular nucleus, and midbrain raphe) to determine the locations of axonal trajectories and terminal zones. The distribution of silver grains in the ipsilateral LHB differed according to the injection site. Entopeduncular nucleus projected to the ventrolateral part, raphe to the dorsomedial part, preoptic region to the central part, anterior hypothalamus to all but the ventrolateral part, and posterior hypothalamus to all but the dorsomedial part. The lateral hypothalamus, preoptic region, and anterior portion of the entopeduncular nucleus projected primarily through the inferior thalamic peduncle and stria medullaris, while the posterior portion of the entopeduncular nucleus projected more diffusely through thalamus to enter LHB from its ventral aspect. Raphe axons reached LHB through the fasciculus retroflexus. Entopeduncular and lateral hypothalamic axons passed through the habenular commissure to the contralateral stria medullaris to reach the contralateral LHB. Silver grains in LHB were distributed similarly bilaterally, but were much less dense contralateral to the injection. The major afferent projections to the feline LHB are therefore topographically organized. The significance of this organization in relation to the possible role of the LHB in integrating limbic and extrapyramidal motor systems is discussed.

The origin of forebrain afferents to the habenula in rat, cat and monkey.

Injections of horseradish peroxidase (HRP) involving the entire habenular complex in rat, cat and squirrel monkey (Saimiri sciureus) label (1) numerous cells in anterior lateral hypothalamic area, (2) a moderate number of cells in lateral preoptic area, substantia innominata, nucleus of diagonal band and postcommissural septum, and (3) a few cells in medial hypothalamus, ipsilaterally, in all three species. Some labeled cells also occur in corresponding regions contralaterally. The contribution of these limbic structures to the innervation of habenula is thus strikingly similar in the three groups. In contrast, significant species variations are found in respect to pallidal afferents. Whereas the entopeduncular nucleus in rat stands out as the main source of forebrain habenular afferents, the same structure in cat appears to contribute less substantially than adjoining lateral hypothalamus to the innervation of habenula. In monkey habenular afferents also arise principally from lateral hypothalamic neurons. At pallidal levels, labeled cells are nevertheless abundant in the rostral pole of primate internal pallidum. More caudally, they are found in significant number along internal and accessory medullary laminae where they intermingle with acetylcholinesterase-containing neurons which do not themselves project significantly upon habenula. This heterogeneous distribution of labeled pallidal cells indicates that the pallidohabenular projections in primate may arise, at least in part, from specific neuronal subpopulations within internal pallidum.