The immunohistochemical distribution of the GABAA receptor α1, α2, α3, ß2/3 and γ2 subunits in the human thalamus.

The GABAA receptor is the most abundant inhibitory receptor in the human brain and is assembled from a variety of different subunit subtypes which determines their pharmacology and physiology. To determine which GABAA receptor subunit proteins are found in the human thalamus we investigated the distribution of five major GABAA receptor subunits α1, α2, α3, ß2,3 and γ2 using immunohistochemical techniques. The α1-, ß2,3- and γ2- subunits which combine to form a benzodiazepine sensitive GABAA receptor showed the most intense levels of staining and were the most common subunits found throughout the human thalamus especially in the ventral and posterior nuclear groups. The next most intense staining was for the α3-subunit followed by the α2-subunit. The intralaminar nuclear group, the mediodorsal nucleus and the thalamic reticular nucleus contained α1-, ß2,3- and γ2- subunits staining as well as the highest levels of the α2- and α3- subunits. The sensory dorsal lateral geniculate nucleus contained very high levels of α1- and ß2,3- and γ2-subunits. The highest densities of GABAA receptors found throughout the thalamus which contained the subunits α1, ß2,3, and γ2 included nuclei which are especially involved in the control or the modulation of the cortico-basal ganglia-thalamo-cortical motor circuits and are thus important in disorders such as Huntington's disease where the GABAergic projections of the basal ganglia are compromised. In addition the majority of receptors in the thalamic reticular nucleus contain α3 and γ2 subunits whilst the intralaminar nuclei contain high levels of α2 and α3 subunits.

The distribution of calbindin, calretinin and parvalbumin immunoreactivity in the human thalamus.

UNLABELLED: Calcium-binding proteins show a heterogeneous distribution in the mammalian central nervous system and are useful markers for identifying neuronal populations. The distribution of the three major calcium-binding proteins - calbindin-D28k (calbindin), calretinin and parvalbumin - has been investigated in eight neurologically normal human thalami using standard immunohistochemical techniques. Most thalamic nuclei show immunoreactive cell bodies for at least two of the three calcium-binding proteins; the only nucleus showing immunoreactivity for one calcium-binding protein is the centre médian nucleus (CM) which is parvalbumin-positive. Overall, the calcium-binding proteins show a complementary staining pattern in the human thalamus. In general terms, the highest density of parvalbumin staining is in the component nuclei of the ventral nuclear group (i.e. in the ventral anterior, ventral lateral and ventral posterior nuclear complexes) and in the medial and lateral geniculate nuclear groups. Moderate densities of parvalbumin staining are also present in regions of the mediodorsal nucleus (MD). By contrast, calbindin and calretinin immunoreactivity both show a similar distribution of dense staining in the thalamus which appears to complement the pattern of intense parvalbumin staining. That is, calbindin and calretinin staining is most dense in the rostral intralaminar nuclear group and in the patchy regions of the MD which show very low levels of parvalbumin staining. However, calbindin and calretinin also show low levels of staining in the ventral nuclear complex and in the medial and lateral geniculate bodies which overlaps with the intense parvalbumin staining in these regions. These results show that the calcium-binding proteins are heterogeneously distributed in a complementary fashion within the nuclei of the human thalamus. They provide further support for the concept recently proposed by Jones (Jones, E.G., 1998. VIEWPOINT: the core and matrix of thalamic organization. Neuroscience 85, 331-345) that the primate thalamus comprises of a matrix of calbindin immunoreactive cells and a superimposed core of parvalbumin immunoreactive cells which may have differential patterns of cortical projections.

Calcium-binding protein immunoreactivity delineates the intralaminar nuclei of the thalamus in the human brain.

Immunohistochemical studies have shown that the three calcium-binding proteins (calbindin-D28k, calretinin and parvalbumin) are heterogeneously distributed in the mammalian brain and are useful for delineating nuclear boundaries. We have investigated the distribution of the three calcium-binding proteins in the human thalamus in order to assist in the delineation of the equivocal nuclear boundaries of the intralaminar nuclei of the thalamus. The results show that each of the "functional" nuclear complexes in the human thalamus demonstrates a characteristic pattern of calcium-binding protein immunoreactivity. In particular, the intralaminar nuclei are characterized by a unique combination of calcium-binding protein staining which clearly delineates the component nuclei in this complex from the other nuclei of the human thalamus. The anterior group of intralaminar nuclei (central lateral nucleus, paracentral nucleus and central medial nucleus) showed intense staining for both calbindin-D28k and calretinin. By contrast, the posterior group of intralaminar nuclei (centre median nucleus and parafascicular nucleus) showed a complementary pattern of staining; the centre median nucleus showed immunoreactivity only for one calcium-binding protein, parvalbumin, while the parafascicular nucleus showed immunoreactivity for both calbindin-D28k and calretinin. No other nucleus in the human thalamus showed these particular combinations of calcium-binding protein staining. Since the intralaminar nuclei also have unique topographically organized connectional affiliations with both the cerebral cortex and the basal ganglia, these results suggest that the calcium-binding proteins may play an important role in the influence of the intralaminar nuclei on interactions between the cerebral cortex and the basal ganglia.

Localization of calcium-binding proteins and GABA transporter (GAT-1) messenger RNA in the human subthalamic nucleus.

The distribution of messenger RNA encoding the human GAT-1 (a high-affinity GABA transporter) was investigated in the subthalamic nucleus of 10 neurologically normal human post mortem cases. Further, the distribution of messenger RNA and protein encoding the three neuronally expressed calcium-binding proteins (calbindin D28k, parvalbumin and calretinin) was similarly investigated using in situ hybridization and immunohistochemical techniques. Cellular sites of calbindin D28k, parvalbumin, calretinin and GAT-1 messenger RNA expression were localized using human-specific oligonucleotide probes radiolabelled with [35S]dATP. Sites of protein localization were visualized using specific anti-calbindin D28k, anti-parvalbumin and anti-calretinin antisera. Examination of emulsion-coated tissue sections processed for in situ hybridization revealed an intense signal for GAT-1 messenger RNA within the human subthalamic nucleus, indeed the majority of Methylene Blue-counterstained cells were enriched in this transcript. Further, a marked heterogeneity was noted with regard to the expression of the messenger RNA's encoding the three calcium-binding proteins; this elliptical nucleus was highly enriched in parvalbumin messenger RNA-positive neurons and calretinin mRNA-positive cells but not calbindin messenger RNA-positive cells. Indeed, only an occasional calbindin messenger RNA-positive cell was detected within the mediolateral extent of the nucleus. In marked contrast, numerous parvalbumin messenger RNA-positive cells and calretinin messenger RNA-positive cells were detected and they were topographically distributed; parvalbumin messenger RNA-positive cells were highly enriched in the dorsal subthalamic nucleus extending mediolaterally; calretinin messenger RNA-positive cells were more enriched ventrally although some degree of overlap was apparent. Computer-assisted analysis of the average cross-sectional somatic area of parvalbumin, calretinin and GAT-1 messenger RNA-positive neurons revealed them all to be in the range of 300 microm2. The unique patterns of calcium-binding protein gene expression were similarly reflected at the protein level; an abundance of parvalbumin- and calretinin-immunopositive neurons were observed whereas only occasional intensely-labelled calbindin-immunopositive fibres were seen, no calbindin-immunopositive cells were detected. Single and double labelling studies show that parvalbumin-immunopositive neurons were mainly localized in the dorsal region of the nucleus, and calretinin-immunopositive neurons were mainly localized in the ventral region although there was overlap with double-labelled neurons located in the middle and dorsal regions. The significance of these findings, in particular the expression of GAT-1, a high-affinity GABA uptake protein, for basal ganglia signalling is discussed.

The effects of globus pallidus lesions on dopamine-dependent motor behaviour in rats.

Motor effects of bilateral lesions of the globus pallidus induced by quinolinic acid (30 and 60 nmol in 0.5 microl) were investigated in rats. Globus pallidus lesions with 60 nmol quinolinic acid produced a significant reduction of spontaneous motor activity measured by a reduced locomotor activity in an open field and by a reduced sniffing activity in an experimental chamber. In addition, D-amphetamine (1 mg/kg, i.p.)-induced hyperlocomotion and D-amphetamine (3 mg/kg, i.p.)-induced stereotyped sniffing were significantly reduced in animals with quinolinic acid lesions (60 nmol). Globus pallidus lesions with 60 nmol quinolinic acid potently reversed catalepsy induced by systemic administration of the dopamine D1 receptor antagonist SCH23390 (0.75 and 1 mg/kg, i.p.) or the dopamine D2 receptor antagonist raclopride (1.25 and 5 mg/kg, i.p.), while lesions with 30 nmol quinolinic acid exerted anti-cataleptic effects which were only partly significant. In line with current models of basal ganglia functions, these findings indicate that inactivation of the globus pallidus reduced spontaneous motor activity and motor hyperactivity after dopamine receptor stimulation. However, the present data also demonstrate that inactivation of the globus pallidus reversed motor hypoactivity induced by a blockade of dopamine D1 and D2 receptors. Therefore, a more complex functional model of the globus pallidus is required to account for the opposite effects on motor behaviour observed after lesions of this basal ganglia nucleus.

Motor depressant effects mediated by dopamine D2 and adenosine A2A receptors in the nucleus accumbens and the caudate-putamen.

We compared hypolocomotion and catalepsy mediated by striatal dopamine D2 and adenosine A2A receptors using microinfusions of the adenosine A2A receptor agonist 2-p-(2-carboxyethyl) phenethylamino-5'-N-ethylcarboxamidoadenosine hydrochloride (CGS21680) and the dopamine D2 receptor antagonist raclopride into the nucleus accumbens and the caudate-putamen. The effective doses (ED25/50) of CGS21680 and raclopride which produced equivalent reductions of spontaneous locomotion after microinfusion into the nucleus accumbens were found to induce similar degrees of catalepsy after microinfusion into the caudate-putamen. This comparable, little separation of the effective doses of a dopamine D2 receptor antagonist and an adenosine A2A receptor agonist to produce locomotor inhibition and catalepsy support the idea that adenosine A2A receptor agonists as potential antipsychotic agents may have a similar therapeutic profile as dopamine D2 receptor antagonists.

The adenosine receptor antagonist theophylline induces a monoamine-dependent increase of the anticataleptic effects of NMDA receptor antagonists.

Previous work revealed that adenosine antagonists as theophylline reversed neuroleptic-induced catalepsy and potentiated anticataleptic effects of dopamine agonists reflecting specific adenosine-dopamine receptor interactions in the central nervous system. We tested whether similar functional interactions exist between adenosine receptors and glutamate receptors of the N-methyl-D-asparte (NMDA) subtype. The present study demonstrates that the anticataleptic effects of the competitive NMDA receptor antagonist CGP37849 and the non-competitive NMDA receptor antagonist dizocilpine can be potentiated by coadministration of a threshold dose of the adenosine receptor antagonist theophylline (2.5 mg/kg, i.p.) in haloperidol (0.5 mg/kg, i.p.)-pretreated rats. This potentiation was elicited only with higher doses of CGP37849 (4 and 8 mg/kg, i.p.) or dizocilpine (0.16 mg/kg, i.p.) in haloperidol (0.5 mg/kg, i.p.), but not in reserpine (5 mg/kg, i.p.) plus alpha-methyl-p-tyrosine (100 mg/kg, i.p.)-pretreated animals. Therefore, these synergistic interactions seem to be brought about by indirect monoamine-dependent mechanisms rather than direct functional interrelationships between NMDA and adenosine A2a receptors.

Stimulation of adenosine A2a receptors in the rat striatum induces catalepsy that is reversed by antagonists of N-methyl-D-aspartate receptors.

Bilateral infusion of the selective adenosine A2a agonist CGS 21680C (1 microgram per side) into the anterodorsal striatum of rats produced profound catalepsy. Intraperitoneal coadministration of the non-competitive N-methyl-D-aspartate (NMDA) antagonist dizocilpine (0.16 mg/kg) or the competitive NMDA antagonist CGP 37849 (4 mg/kg) completely reversed CGS 21680C-induced catalepsy, while lower doses of both NMDA antagonists induced no or only weak anticataleptic effects. The adenosine A2a receptor localization to striatopallidal neurons suggests that a selective activation of the striatopallidal efferent pathway is involved in the expression of catalepsy induced by intrastriatal infusion of CGS 21680C. In addition, striatopallidal neurons seem to be an important neuronal substrate of the anticataleptic effects of NMDA antagonists.