Expression of early genes in the rat brain after administration of corticoliberin into the neostriatum

In situ hybridization using oligonucleotide probes was used to study the effects of intrastriatal microinjection of corticoliberin on the expression of the early genes c-fos, jun B, c-jun, and NGFIA in the rat brain. Administration of corticoliberin (0.25 g) into the neostriatum induced the expression of mRNA encoded by the early genes c-fos, jun B, and NGFIA in both the neostriatum itself and in its efferent structures, particularly the nucleus accumbens and various parts of the cortex. Intrastriatal microinjection of corticoliberin had no effect on the expression of mRNA for the oncogene c-jun in the brain. These results suggest that neuronal activation in the neostriatum and its projection targets manifest as the expression of early genes is one of the mechanisms underlying the adaptive effects of corticoliberin in stress.     

The distribution and cellular localization of the serotonin 1C receptor mRNA in the rodent brain examined by in situ hybridization histochemistry. Comparison with receptor binding distribution

The regional distribution and cellular localization of mRNA coding for the serotonin 1C receptor were investigated in tissue sections of mouse and rat brain by in situ hybridization histochemistry. Several 32P-labelled riboprobes derived from mouse genomic clones were used. The serotonin 1C receptor binding sites were visualized autoradiographically and quantified using [3H]mesulergine as ligand, in the presence of spiperone to block serotonin 1C receptors. Strong hybridization signal was observed in the choroid plexus of all brain ventricles. High levels of hybridization were also seen in the anterior olfactory nucleus, pyriform cortex, amygdala, some thalamic nuclei, especially the lateral habenula, the CA3 area of the hippocampal formation, the cingulate cortex, some components of the basal ganglia and associated areas, particularly the nucleus subthalamicus and the substantia nigra. The midbrain and brainstem showed moderate levels of hybridization. The distribution of the serotonin 1C receptor mRNA corresponded well to that of the serotonin 1C receptors. The highest levels of serotonin 1C receptor binding were observed in the choroid plexus. In addition, significant levels of the serotonin 1C receptor binding were seen in the anterior olfactory nucleus, pyriform cortex, nucleus accumbens, ventral aspects of the striatum, paratenial and paracentral thalamic nuclei, amygdaloid body and substantia nigra pars reticulata. The cingulate and retrosplenial cortices as well as the caudal aspects of the hippocampus (CA3) were also labelled. Binding in brainstem and medulla was low and homogeneously distributed. No significant binding was seen in the habenular and subthalamic nuclei. Similar findings were obtained in rat brain. These results demonstrate that, in addition to their enrichment in the choroid plexus, the serotonin 1C receptor mRNA and binding sites are heterogeneously distributed in the rodent brain and thus could be involved in the regulation of many different brain functions. The combination of in situ hybridization histochemistry with receptor autoradiography opens the possibility of examining the regulation of the serotonin 1C receptor synthesis after pharmacological or physiological alterations.     

Differential distribution of somatostatin receptor subtypes in rat brain revealed by newly developed somatostatin analogs.

Somatostatin receptor subtypes were labeled with the somatostatin analogs [125I]CGP 23996 and [125I]MK 678 and the distribution of these receptors in rat brain was investigated using quantitative autoradiographic techniques. [125I]CGP 23996 and [125I]MK 678 specifically label different populations of somatostatin receptors in rat brain. In a number of brain regions striking differences in the distribution of the somatostatin receptor subtypes labeled by each peptide were observed. High levels of binding sites for both [125I]CGP 23996 and [125I]MK 678 were present in the cerebral cortex, CA1 region and subiculum of the hippocampus. In contrast, high levels of [125I]MK 678 binding were found in the dentate gyrus of the hippocampus while few [125I]CGP 23996 binding sites were observed in this brain region. [125I]CGP 23996 binding was detected in the central region of the interpeduncular nucleus whereas the dorsal and lateral subnuclei of this brain area expressed mainly somatostatin receptors with high affinity for MK 678. The locus coeruleus and regions of the superior colliculus and hypothalamus selectively express [125I]MK 678-sensitive somatostatin receptors. Furthermore, limbic structures such as the lateral septum, the nucleus accumbens and ventromedial striatum had much higher levels of [125I]MK 678 binding sites than [125I]CGP 23996 binding sites. Differences in the expression of the somatostatin receptor subtypes were also detected in the substantia nigra. [125I]CGP 23996 binding was present in the pars reticulata but not the pars compacta whereas the reverse distribution for [125I]MK 678 binding sites was observed. The differential distribution of [125I]CGP 23996 and [125I]MK 678 binding sites in rat brain supports the hypothesis that these peptides selectively label different somatostatin receptor subtypes in the central nervous system.     

Localization of m5 muscarinic receptor mRNA in rat brain examined by in situ hybridization histochemistry

The regional distribution of mRNA coding for the m5 muscarinic acetylcholine receptor subtype was investigated in tissue sections of rat brain by in situ hybridization histochemistry. The highest hybridization signal was observed in the hippocampus, but restricted to the ventral subiculum, pyramidal cells of the CA1 and, with lower intensity, of the CA2 subfields. Significant levels of hybridization were also seen in the substantia nigra pars compacta, ventral tegmental area, lateral habenula, ventromedial hypothalamic nucleus and mammillary bodies. An involvement of the m5 muscarinic receptors in the regulation of the dopaminergic nigrostriatal pathway is suggested.     

Temporal and spatial expression of preprotachykinin A mRNA in the developing filial mice brain after maternal administration of monosodium glutamate at a late stage of pregnancy

In the early stages of brain development, exposure of excessive monosodium glutamate (MSG) to neurons causes animal functional and behavioral disorders in adulthood. To investigate the effects of excessive MSG during pregnancy on the neurons in the developing brain, in situ hybridization was used. In mice, the expression of preprotachykinin A mRNA (PPT A mRNA) was assessed in neurons of in the brain after MSG treatment. Brain tissue sections were hybridized with specific digoxigenin-labeled RNA probes. The number of cells that expressed PPT A mRNA gradually decreased from 10-day-old (10d) to 60-day-old (60d) MSG-treated and normal animals. In the MSG-treated and normal mice, the PPT A mRNA-positive neurons almost disappeared in 90-day-old (90d) mice. The expression of PPT A mRNA significantly decreased at 10d in most of the brain regions of MSG-treated mice including the cerebral cortex (CC), hippocampal subregions of CA1, CA2 (CA1, CA2), habenula nucleus (HAB), hypothalamic periventricular nucleus (PE), hypothalamic arcuate nucleus (AR), median eminence (ME), amygdala nucleus (AMY), endopiriform nucleus (EN), and hypothalamic ventromedial nucleus (VMH) and dorsomedial nucleus (DMH). In the hippocampal CA4 subregions (CA4), paraventricular nucleus (PV) and caudate putamen (CPU), however, they were not significantly altered. Furthermore, in CC, hippocampal CA3 subregion (CA3), PE and EN regions the number of PPT A mRNA-positive neurons decreased at 20 days old (20d), but increased significantly in CA2 and CPU. At 30 days old (30d), the positive neuron number decreased in AMY, and they did not change in other regions. At 60d, the number of positive neurons significantly decreased in PV and ME, but increased in AMY. In the other observed regions, no changes were found. These results show that maternal administration of excessive MSG at a late stage of pregnancy significantly decreases PPT A mRNA expression in most of the brain regions of filial mice. This suggests that glutamate-induced excitotoxicity may affect the metabolism of precursors of substance P in developing brain neurons. The present study provides insights into the plasticity and vulnerability of neuron in different brain regions to glutamate excitotoxicity.     

Septal cholinergic afferents regulate expression of brain-derived neurotrophic factor andβ-nerve growth factor mRNA in rat hippocampus

Summary In situ hybridization was used to study the expression of members of the nerve growth factor family of trophic factors in rat hippocampus following stimulation of afferent cholinergic and glutamatergic pathways with quisqualate. A transient increase in brain-derived neurotrophic factor (BDNF) and-nerve growth factor (NGF) mRNA expression in the hippocampus was seen 4 h after a quisqualate injection into the medial septal nucleus. Both BDNF and NGF mRNA levels increased more than 4-fold in the granule layer of the dentate gyrus and for BDNF mRNA also in the pyramidal cells of CA1, while the levels of BDNF mRNA in CA3 increased 2-fold. The increase in BDNF and NGF mRNA levels were completely prevented by pretreatment with systemic injections of either scopolamine or diazepam. A quisqualate injection into the entorhinal cortex, containing glutamatergic afferents to the hippocampus, resulted in a 15-, 5- and 17-fold increase in the expression of BDNF mRNA in the ipsilateral granule cells, CA3 and CA1 pyramidal cells, respectively. Similar increases were also seen in the hippocampus contralateral to the injections. In contrast, the level of NGF mRNA did not increase significantly in any of the subfields in the hippocampus. The increase in BDNF mRNA after cortex injections was attenuated by diazepam but not by scopolamine. These findings imply that increased activity in afferent cholinergic and glutamatergic pathways to the hippocampus differentially regulate expression of the NGF family of neurotrophic factors in the hippocampus.     

Prostaglandin E2 receptor expression in the rat trigeminal-vascular system and other brain structures involved in pain

Prostaglandin E₂ (PGE₂) is considered to be a key mediator in migraine pathophysiology. PGE₂ acts via four receptors (EP₁-EP₄) but their distribution in the brain districts implicated in migraine has yet to be delineated. We quantified amount of mRNA and protein expression for the EP receptors in both peripheral and central structures involved in pain transmission and perception in migraine: dura mater, cerebral arteries, trigeminal ganglion, trigeminal nucleus caudalis, periaqueductal grey, thalamus, hypothalamus, cortex, pituitary gland, hippocampus and cerebellum. In the trigeminal-vascular system (TVS) we found highest expression of EP₁ and EP₂ protein in the trigeminal nucleus caudalis. EP₃ and EP₄ mRNA expression were highest in the trigeminal ganglion. Within intracranial structures EP₁ mRNA and protein expression were significantly higher in pituitary gland and cerebellum than in dorsal root ganglia (peripheral control), whereas the EP₂ mRNA and protein were highly abundant in the pituitary gland. EP₃ mRNA was mainly found in thalamus and hypothalamus. The most robust mRNA and protein expression for EP₄ receptor was seen in the dorsal root ganglion. In conclusion, all four receptors are located in areas implicated in migraine supporting the possible involvement of PGE₂ in this disease.     

Muscarinic cholinergic receptor subtypes in the hippocampus of aged rats.

In spite of the suggestion of impaired muscarinic function in adult-onset cognitive disorders, data on the expression of muscarinic receptors in the hippocampus as a function of age are inconsistent. One reason may be that the majority of investigations were unable to differentiate the five brain muscarinic receptors subtypes. In this study, using a protocol based on a combination of both kinetic and equilibrium binding approaches, we have assessed the expression and the density of M1-M5 muscarinic cholinergic receptors in the hippocampus of Fisher 344 rats aged 6, 15 and 22 months. An age-related decrease of the density of M1 receptor was found in pyramidal neurons of the CA1 subfield. In this area, other subtypes of muscarinic receptors were unchanged with the exception of a loss of M2 receptor in the radial layer. In the CA3 subfield, receptor changes involved M2, M3 and M5 subtypes, whereas in the dentate gyrus, the main changes affected M1 and M2 receptors of the granular layer and M2 and M3 receptors of the molecular layer. The above findings indicate that analysis of age-related changes of different muscarinic cholinergic receptors might represent a useful contribution to identifying the basis of cholinergic neurotransmission impairment in adult-onset cognitive dysfunction.     

Regional and temporal expression of sodium channel messenger RNAs in the rat brain during development

Summary The distribution of mRNA expression for three types of voltage gated neuronal sodium-channels was studied in the rat brain at different developmental stages (embryonal day E18, postnatal day P5 and adult). With the in-situ hybridization technique, using synthetic DNA-oligomer probes, pronounced regional and temporal variations in the expression levels of the different channel subtypes could be detected. In comparison with types I and III, sodium channel II mRNA was the most abundant subtype at all developmental stages. Maximal expression of sodium channel II mRNA was seen at P5 in virtually all parts of the grey matter, except for the cerebellum. In adult rat brain in contrast, sodium channel II mRNA levels were maximal in the granular layer of the cerebellum, whereas in all other regions expression had decreased to roughly 50% of postnatal levels. Na channel I expression was virtually absent at E18 and showed highest levels at P5, with maxima in the caudate nucleus and hippocampus. In the adult brain, expression of Na-channel I was nearly absent in the neocortex, but well detectable in the cerebellum and, at lower levels in the striatum and thalamus. Sodium channel III was mainly expressed at the embryonal stage and showed a decrease to very low levels with little regional preferences in the adult.Supported by Deutsche Forschungsgemeinschaft grant no.: Cr 30/16     

The distribution of beta-1- and beta-2-adrenergic receptors of normal and reeler mouse brain: an in vitro autoradiographic study

The densities of beta-1 and beta-2-adrenergic receptors in normal and reeler mice brains were compared using in vitro autoradiography. For the most part, the regional distribution of the beta-receptor subtypes in mouse brain was similar to that of rat [Rainbow et al. (1984) Proc. natn. Acad. Sci. U.S.A. 81, 1585-1589]. The most striking differences in beta-adrenergic receptor localization between the mouse and rat brain were in the superficial layer of the superior colliculus and the bed nucleus of stria terminalis. The superficial layer of the mouse superior colliculus had high densities of beta-1 receptors and much lower densities of beta-2 receptors, while the superficial layer of the rat superior colliculus had moderately high densities of beta-2 and lower densities of beta-1 receptors. The mouse bed nucleus of stria terminalis had greater densities of beta-1 receptors than beta-2 receptors. This structure in the rat possessed roughly equivalent densities of beta-1 and beta-2 receptors. In general, the distributional pattern of beta-1 and beta-2 receptors in the reeler mouse brain matched that of the normal mouse brain. In regions which neuronal malpositioning had been reported, such as the cerebral cortex, hippocampus, and cerebellum, beta-1 and beta-2 receptors were still present in proportions similar to normal mice; however, the pattern of beta-receptors within each of these regions was altered. The beta-receptor subtypes in these structures showed an abnormal radial distribution similar to the abnormal radial positioning of neurons in these structures.     

Comparison of the distribution of D-1 and D-2 dopamine receptors in the rat brain

The distribution of dopamine type 1 (D-1) and dopamine type 2 (D-2) receptors in the brain have been compared as assessed by the technique of autoradiography after labelling with highly selective ligands. D-1 receptors, as evidenced by the specific binding of [3H]R-(+)-8-chloro-2,3,4,5-tetrahydro-3-methyl-5-phenyl-IH-3-benzazepine -7- ol (SCH 23390), were found in high concentrations in the caudate-putamen, nucleus accumbens, islands of Calleja, olfactory tubercle and the zona reticulata of the substantia nigra. A similar but distinct distribution was seen for [3H]sulpiride, a ligand which is highly selective for D-2 receptors. Like [3H]SCH 23390, this ligand also labelled the caudate-putamen, nucleus accumbens, islands of Calleja and the olfactory tubercle; however, only a very low density of D-2 receptors could be found in the zona reticulata of the substantia nigra, while a greater degree of binding was present in the zona compacta. Additional brain areas which contained D-1 but not D-2 receptors included the cerebral cortex, accessory olfactory nucleus, amygdala, thalamus, suprachiasmatic nucleus, choroid plexus, claustrum, endopiriform nucleus, zona incerta, dorsal lateral geniculate nucleus and the dentate gyrus. D-2 receptors were also found in areas which appeared to contain only low amounts of D-1 receptors such as the glomerular layer of the olfactory bulb, bed nucleus of the stria terminalis, hypothalamus, habenula, stratum lacunosum moleculare of the hippocampus, intermediate lobe of the pituitary, lateral mammillary nucleus, periaqueductal gray, inferior colliculus, nodulus of the cerebellum and the dorsal horn of the spinal cord. The results show the precise localization of dopamine receptors throughout the brain and provide a means of direct comparison between the distribution of dopamine receptor subtypes. These subtypes are pharmacologically and anatomically distinct entities and their comparison indicates areas where additional biochemical and neuroanatomical studies may be performed to elucidate the roles for these receptor subtypes in the central nervous system.     

Preferential limbic expression of the cannabinoid receptor mRNA in the human fetal brain

The cannabinoid receptor one (CB1) is responsible for the effects of cannabis on motor and cognitive function in the CNS. There is to date very limited information about the CB1 gene expression in the human brain, in particular during fetal development. In the present study, in situ hybridization experiments were used to examine the microscopic and macroscopic organization of the CB1 mRNA expression in normal human fetal (approximately 20 weeks of development) and adult brains. The fetal brain showed a distinct heterogeneous pattern of the CB1 mRNA expression which was low to moderate in many brain areas. The most striking feature of the fetal brain was the intense expression in the hippocampal CA region and basal nuclear group of the amygdaloid complex. Many of the same brain areas that showed positive expression of the CB1 mRNA in the fetal brain also expressed the gene in the adult brain. However, aside from an intense expression in the hippocampus which resembled that in fetal brain, the adult brain showed very high expression throughout the cerebral cortex, caudate nucleus, putamen and cerebellar cortex. These results document a different pattern of the anatomical organization of the CB1 mRNA expression in the mid-gestation fetal and adult human brain. Overall, the high CB1 mRNA expression in the fetal hippocampus and amygdala indicates that these limbic structures might be most vulnerable to prenatal cannabis exposure.     

Stable Modifications to the Expression of Neurohormones in the Rat Hypothalamus in a Model of Post-Traumatic Stress Disorder

We report studies of the neuroendocrine mechanisms of development of an anxiety state in rats using the “stress-restress” experimental model of post-traumatic stress disorder. Immunocytochemical methods demonstrated significant increases in corticoliberin expression in both the parvo- and magnocellular parts of the paraventricular nucleus persisting to 10 days after presentation of the animals with repeated stress. Decreases in vasopressin expression were seen in the paraventricular nucleus of the animals on the first day after repeated stress. Vasopressin contents in the parvocellular part of the nucleus in animals of the experimental group were no different at 10 days from those in animals of the control group, while levels in the magnocellular part were increased. These data provide evidence for the involvement of the hypothalamic component of the vasopressinergic system (along with the corticoliberinergic system) in the pathogenetic mechanisms of the analog of post-traumatic stress disorder generated in this model.     

Striatal mechanism of action of corticoliberin on behavior in dogs in conditions of dopamine deficiency

This report describes studies of the interaction of the integrative dopaminergic and corticoliberin systems in the neostriatum during performance of situational food-related conditioned reflexes. Studies were performed in dogs with chemotrodes implanted in the substantia nigra and the head of the caudate nucleus. 6-Hydroxydopamine was injected into the substantia nigra at a dose of 50 μg, and 10 μg of corticoliberin was injected into the caudate nucleus. Blood cortisol and catecholamine levels were determined. Analysis of the result showed that an interaction takes place in the neostriatum between the corticoliberin and dopaminergic systems, and that in conditions in which dopaminergic structures are excluded, the efficacy of corticoliberin in the performance of behavioral acts decreases by 30–40%, i.e., complete expression of its regulatory role of motor situational conditioned reflexes is lost. Translated from Rossiiskii Fiziologicheskii Zhurnal imeni I. M. Sechenova, Vol. 84, No. 8, pp. 728–734, August, 1998.     

Expression of reelin, its receptors and its intracellular signaling protein, Disabled1 in the canary brain: relationships with the song control system

Songbirds produce learned vocalizations that are controlled by a specialized network of neural structures, the song control system. Several nuclei in this song control system demonstrate a marked degree of adult seasonal plasticity. Nucleus volume varies seasonally based on changes in cell size or spacing, and in the case of nucleus HVC and area X on the incorporation of new neurons. Reelin, a large glycoprotein defective in reeler mice, is assumed to determine the final location of migrating neurons in the developing brain. In mammals, reelin is also expressed in the adult brain but its functions are less well characterized. We investigated the relationships between the expression of reelin and/or its receptors and the dramatic seasonal plasticity in the canary (Serinus canaria) brain. We detected a broad distribution of the reelin protein, its mRNA and the mRNAs encoding for the reelin receptors (VLDLR and ApoER2) as well as for its intracellular signaling protein, Disabled1. These different mRNAs and proteins did not display the same neuroanatomical distribution and were not clearly associated, in an exclusive manner, with telencephalic brain areas that incorporate new neurons in adulthood. Song control nuclei were associated with a particular specialized expression of reelin and its mRNA, with the reelin signal being either denser or lighter in the song nucleus than in the surrounding tissue. The density of reelin-immunoreactive structures did not seem to be affected by 4 weeks of treatment with exogenous testosterone. These observations do not provide conclusive evidence that reelin plays a prominent role in the positioning of new neurons in the adult canary brain but call for additional work on this protein analyzing its expression comparatively during development and in adulthood with a better temporal resolution at critical points in the reproductive cycle when brain plasticity is known to occur.     

Dopaminergic regulation of striatal proenkephalin mRNA and prodynorphin mRNA: contrasting effects of D1 and D2 antagonists

In situ hybridization was used to measure the levels of proenkephalin mRNA and prodynorphin mRNA in regions of rat striatum and nucleus accumbens. Chronic administration of haloperidol (2.4 mg/kg/day for 7 days) increased the levels of proenkephalin mRNA in both striatum and nucleus accumbens. In contrast, the levels of prodynorphin mRNA were not significantly affected in any region. Chronic administration of the D1 antagonist SCH 23390 (2.4 mg/kg/day for 7 days) decreased the striatal content of proenkephalin mRNA. A similar effect was seen in the lateral nucleus accumbens. The levels of prodynorphin mRNA were unaffected by SCH 23390 treatment in all regions examined. These results suggest that there is no major tonic dopaminergic regulation of prodynorphin synthesis in the basal ganglia. However, it appears that there is a tonic suppression, via D2 receptors, and a tonic enhancement, via D1 receptors, of proenkephalin synthesis, in the striatum and nucleus accumbens.     

Differential expression of NMDA and AMPA receptor subunits in rat dorsal and ventral hippocampus

Several studies have demonstrated anatomical and functional segregation along the dorsoventral axis of the hippocampus. This study examined the possible differences in the AMPA and NMDA receptor subunit composition and receptor binding parameters between dorsal and ventral hippocampus, since several evidence suggest diversification of NMDA receptor-dependent processes between the two hippocampal poles. Three sets of rat dorsal and ventral hippocampus slices were prepared: 1) transverse slices for examining a) the expression of the AMPA (GluRA, GluRB, GluRC) and NMDA (NR1, NR2A, NR2B) subunits mRNA using in situ hybridization, b) the protein expression of NR2A and NR2B subunits using Western blotting, and c) by using quantitative autoradiography, c(1)) the specific binding of the AMPA receptor agonist [(3)H]AMPA and c(2)) the specific binding of the NMDA receptor antagonist [(3)H]MK-801, 2) longitudinal slices containing only the cornus ammonis 1 (CA1) region for performing [(3)H]MK-801 saturation experiments and 3) transverse slices for electrophysiological measures of NMDA receptor-mediated excitatory postsynaptic potentials. Ventral compared with dorsal hippocampus showed for NMDA receptors: 1) lower levels of mRNA and protein expression for NR2A and NR2B subunits in CA1 with the ratio of NR2A /NR2B differing between the two poles and 2) lower levels of [(3)H]MK-801 binding in the ventral hippocampus, with the lowest value observed in CA1, apparently resulting from a decreased receptor density since the B(max) value was lower in ventral hippocampus. For the AMPA receptors CA1 our results showed in ventral hippocampus compared with dorsal hippocampus: 1) lower levels of mRNA expression for GluRA, GluRB and GluRC subunits, which were more pronounced in CA1 and in dentate gyrus region and 2) lower levels of [(3)H]AMPA binding. Intracellular recordings obtained from pyramidal neurons in CA1 showed longer NMDA receptor-mediated excitatory postsynaptic potentials in ventral hippocampus compared with dorsal hippocampus. In conclusion, the differences in the subunit mRNA and protein expression of NMDA and AMPA receptors as well as the lower density of their binding sites observed in ventral hippocampus compared with dorsal hippocampus suggest that the glutamatergic function differs between the two hippocampal poles. Consistently, the lower value of the ratio NR2A/NR2B seen in the ventral part would imply that the ventral hippocampus NMDA receptor subtype is functionally different than the dorsal hippocampus subtype, as supported by our intracellular recordings. This could be related to the lower ability of ventral hippocampus for long-term synaptic plasticity and to the higher involvement of the NMDA receptors in the epileptiform discharges, observed in ventral hippocampus compared with dorsal hippocampus.     

Comparative analysis of glutamate transporter expression in rat brain using differential double in situ hybridization

 This study compares the mRNA expression pattern for the three glutamate transporters EAAC1, GLT1 and GLAST in rat brain, using a sensitive non-radioactive in situ hybridization technique. The results confirm the predominantly neuronal localization of EAAC1 mRNA, the astroglial and ependymal localization of GLAST mRNA and the astroglial and neuronal localization of GLT1 mRNA. Further, we demonstrate, using a novel differential double hybridization protocol, that the presence of GLT1 mRNA in neurons is more widespread than previously thought, and that it encompasses the majority of neurons in the neocortex, neurons in the external plexiform layer in the olfactory bulb, neurons in dorsal and ventral parts of the anterior olfactory nucleus, the majority of neurons in the anteromedial thalamic nuclei, the CA3 pyramidal neurons in the hippocampus and neurons in the inferior olive. In addition, we demonstrate marked variations in the expression levels of GLT1 and GLAST mRNAs in different brain areas, suggesting that their mRNA levels are regulated by different mechanisms. Finally, for EAAC1 we demonstrate also a widespread distribution and a marked heterogeneity in the expression levels. EAAC1 is strongly expressed by a heretofore unrecognized group of cells in white matter tracts such as the corpus callosum, fimbria-fornix or anterior commissure. Also, strong EAAC1 expression is present in groups of scattered cells in grey matter areas of much of the forebrain and the cerebellum. These results provide more detailed information about the precise cellular localization of these three glutamate transporters and their regulation at the mRNA level. Accepted: 29 January 1998