Cholinergic and non-cholinergic projections from the rat basal forebrain revealed by combined choline acetyltransferase and Phaseolus vulgaris leucoagglutinin immunohistochemistry

A two-color fluorescence method is described for demonstrating immunohistochemically the anterogradely transported plant lectin Phaseolus vulgaris leucoagglutinin (PHAL, fluorescein isothiocyanate label) and choline acetyltransferase (ChAT, rhodamine label) on the same rat brain section. Application of this method to the study of projection neurons in the vertical and horizontal limbs of the diagonal band, the substantia innominata and nucleus basalis revealed that both cholinergic and non-cholinergic pathways followed similar trajectories to their targets. These included: projections from the vertical, and, to a lesser extent, horizontal limb of the diagonal band coursing through the dorsal fornix, alveus and fimbria to the hippocampus; fibers from the vertical and horizontal limbs of the diagonal band traveling anteriorly to the anterior olfactory nucleus, posterolaterally to the entorhinal cortex, and anterodorsally into the cingulum to the cingulate and retrosplenial, and, in some cases, the frontal and occipital cortices; projections, mostly non-cholinergic, from the substantia innominata traveling laterally to the piriform cortex and amygdala, and anteriorly to the anterior olfactory nucleus and olfactory bulb; and fibers from cells in the nucleus basalis coursing dorsally to the frontal and parietal cortices or laterally to the basolateral amygdala and piriform, insular and temporal cortices. Some axon terminations ended at right angles to the parent axon shaft in short protuberances resembling terminal boutons.     

Choline acetyltransferase-immunoreactive neurons in the rat entopeduncular nucleus.

Using a monoclonal antibody against choline acetyltransferase, neurons of the rat entopenduncular nucleus were found to express choline acetyltransferase immunoreactivity. These cholinergic cells were located mostly in the rostral portion of the entopeduncular nucleus with a marked decrease towards its caudal portion. To identify their target sites, a retrograde fiber tracing technique was combined with immunohistochemistry for choline acetyltransferase. After injection of wheatgerm agglutinin conjugated with horseradish peroxidase into the habenula, some of the entopedunculo-habenular cells were found to be immunoreactive for choline acetyltransferase. The cells in the peripallidal region (the substantia innominata, nucleus basalis magnocellularis and ansa lenticularis) with choline acetyltransferase immunoreactivity did not contain horseradish peroxidase. Following injection of fluorescent tracer into the frontal cerebral cortex, retrogradely labeled cells were observed in the rostral part of the entopedunucular nucleus. A majority of these entopedunculo-cortical cells exhibited choline acetyltransferase immunoreactivity, similar to the cells of the peripallidal region projecting to the neocortex. Employing two different fluorescent tracers, entopedunculo-cortical cells were shown to constitute a distinct cell population from the numerous entopedunculo-habenular cells. The present study demonstrated, in the rat entopeduncular nucleus, the presence of cholinergic neurons that projected to the neocortex and habenula.     

Atlas of cholinergic neurons in the forebrain and upper brainstem of the macaque based on monoclonal choline acetyltransferase immunohistochemistry and acetylcholinesterase histochemistry

Choline acetyltransferase immunohistochemistry was used to map the cholinergic cell bodies in the forebrain and upper brainstem of the macaque brain. Neurons with choline acetyltransferase-like immunoreactivity were seen in the striatal complex, in the septal area, in the diagonal band region, in the substantia innominata, in the medial habenula, in the pontomecencephalic tegmentum and in the oculomotor and trochlear nuclei. The ventral striatum contained a higher density of cholinergic cell bodies than the dorsal striatum. All of the structures that contained the choline acetyltransferase positive neurons also had acetylcholinesterase-rich neurons. Choline acetyltransferase positive neurons were not encountered in the cortex. Some perikarya in the midline, intralaminar, reticular and limbic thalamic nuclei as well as in the hypothalamus were rich in acetylcholinesterase but did not give a positive choline acetyltransferase reaction. A similar dissociation was observed in the substantia nigra, the raphe nuclei and the nucleus locus coeruleus where acetylcholinesterase-rich neurons appeared to lack perikaryal choline acetyltransferase activity.     

Immunocytochemical localization of cholinergic terminals in the region of the nucleus basalis magnocellularis of the rat: a correlated light and electron microscopic study

The cholinergic circuitry in the nucleus basalis magnocellularis of the rat was investigated in a correlated light and electron microscopic study by using monoclonal antibodies against the acetylcholine-synthesizing enzyme, choline acetyltransferase, following the unlabelled antibody peroxidase-antiperoxidase immunocytochemical procedure. After the immunocytochemical approach, large cholinergic cells and a few immunoreactive fibres exhibiting a varicose appearance, were detected by light microscopy in portions of the nucleus basalis magnocellularis located within the anatomical limits of the globus pallidus, mostly in its ventromedial part. Cholinergic neurons and fibre-like structures were also found within the substantia innominata on the edge of globus pallidus. The same material studied by light microscopy was analysed with the electron microscope. At the ultrastructural level, the immunopositive neurons showed the same cytological characteristics and pattern of synaptic input as cholinergic basal forebrain cells. Additionally, scarce immunoreactive preterminal axons and terminal boutons were detected in the region. The immunoreactive terminals were scattered or formed occasional clusters and appeared as heavily immunostained vesicle-filled boutons making exclusively axodendritic synaptic contacts principally with immunonegative distal dendrites. Both symmetric and asymmetric synaptic contacts established between these structures were detected, although the symmetric contacts were the more numerous. The surface of postsynaptic immunonegative dendrites in asymmetric synaptic contact with immunoreactive terminals was generally covered by terminals that lacked detectable immunoreactivity. In contrast, those in symmetric synaptic contact with labelled terminals showed much sparser input from immunonegative terminals, suggesting that they may belong to interneurons. Very rarely, cholinergic terminals were detected in asymmetric synaptic contact with dendrites which also contained positive immunoreaction product. Asymmetric contacts were frequently characterized by the presence of subjunctional dense bodies. The detection of cholinergic terminals in the region of the nucleus basalis magnocellularis of the rat indicates that this region not only contains cholinergic projecting neurons, but receives a cholinergic input itself. Results of this study provide evidence of the existence of a cholinergic transmission in the basal forebrain of the rat, and also that acetylcholine might play a role in the regulation of the extrinsic cortical cholinergic innervation. The possible sources of this innervation are discussed.     

Distribution of catecholaminergic afferent fibres in the rat globus pallidus and their relations with cholinergic neurons

The topographical distribution of catecholaminergic nerve fibres and their anatomical relationship to cholinergic elements in the rat globus pallidus were studied. Peroxidase–antiperoxidase and two-colour immunoperoxidase staining procedures were used to demonstrate tyrosine hydroxylase (TH), dopamine β-hydroxylase (DBH), phenylethanolamine N-methyltransferase (PNMT) and choline acetyltransferase (ChAT) immunoreactivities, combined with acetylcholinesterase (AChE) pharmacohistochemistry. TH immunoreactive nerve fibres were seen to enter the globus pallidus from the medial forebrain bundle. The greatest density of such fibres was found in the ventral region of the globus pallidus, which was also characterized by the greatest density of ChAT immunoreactive neurons. TH immunoreactive nerve fibres showed varicose arborizations and sparse boutons, which were occasionally seen in close opposition to cholinergic structures. In all regions of the globus pallidus, there were also larger, smooth TH immunoreactive nerve fibres of passage to the caudate putamen. A smaller number of DBH immunoreactive nerve fibres and terminal arborizations were found in the substantia innominata, internal capsule and in the globus pallidus bordering these structures. A few PNMT immunoreactive nerve fibres in the substantia innominata and internal capsule did not enter the globus pallidus. Electron microscopy revealed TH immunoreactive synaptic profiles in the ventromedial area of the globus pallidus corresponding to the nucleus basalis magnocellularis of Meynert (nBM). These made mainly symmetrical and only a few asymmetrical synaptic contacts with dendrites containing AChE reaction product. The results indicate that cholinergic structures in the nBM are innervated by dopaminergic fibres and terminals, with only a very small input from noradrenergic fibres.     

Comparative effects of quisqualic and ibotenic acid-induced lesions of the substantia innominata and globus pallidus on the acquisition of a conditional visual discrimination: differential effects on cholinergic mechanisms

Two experiments tested the hypothesis that the deficits in conditional discrimination learning produced by ibotenic acid-induced lesions of the ventral pallidum and substantia innominata are produced by loss of the magnocellular cholinergic cells in the nucleus basalis and adjacent regions. Experiment 1 replicated the previously reported deficit in conditional learning produced by ibotenate-induced lesions of the ventral pallidum/substantia innominata, but failed to demonstrate any restoration of learning by a subchronic regimen of the acetylcholinesterase inhibitor physostigmine sufficient to produce significant (30%), but equivalent, degrees of inhibition in the frontal cortex of ventral pallidum/substantia innominata-lesioned or sham-operated rats. Experiment 2 examined the effects of quisqualic acid-induced lesions of the ventral pallidum/substantia innominata. According to most of the measures of learning employed, the quisqualic acid-induced lesion of the ventral pallidum/substantia innominata failed to impair conditional learning, even though the quisqualate-induced lesion produced greater degrees of cholinergic neuron destruction than the ibotenate-induced lesion, as measured in terms of reductions in cortical choline acetyltransferase activity (44% vs 27%). Although consideration of individual data suggested that very high (60%) levels of choline acetyltransferase reduction in Experiment 2 might have detrimental effects of conditional learning, the overall failure of the quisqualate-induced lesions of the ventral pallidum/substantia innominata to impair learning is to be contrasted with the significant behavioural effects of ibotenate-induced lesions. Histological and immunocytochemical analysis showed that the quisqualate-induced lesion, unlike that produced by ibotenate, tended to produce less damage to the overlying dorsal globus pallidus and to parvocellular neurons of the ventral pallidum/substantia innominata, thus implicating these nonspecific effects of ibotenate-induced lesions in their behavioural effects. The present results question previous interpretations of the behavioural effects of ibotenate-induced lesions of the ventral pallidum/substantia innominata in terms of damage inflicted on the cortically-projecting cholinergic cells of the nucleus basalis, and suggest that quisqualic acid, although also nonspecific in its excitotoxic effects, is nevertheless more selective for producing damage to cholinergic neurons in the ventral pallidum/substantia innominata than ibotenic acid.     

Mechanothermal nociceptors in the scaly skin of the chicken leg

Corticotropin-releasing hormone (CRH) interacts with noradrenergic, dopaminergic and cholinergic systems of the brain, and these interactions are thought to be of relevance for the stress response, anxiety-related behavior, and cognitive function. CRH mediates its central effects through two high-affinity membrane receptors, CRH receptor subtypes 1 and 2. It is however unclear at present whether cholinergic or catecholaminergic cells express these receptors themselves or whether the effects of CRH are indirectly mediated through interaction with other neurotransmitter systems. Therefore, this study investigated whether choline acetyltransferase immunoreactive neurons of the murine basal forebrain and brainstem nuclei, and tyrosine hydroxylase immunoreactive neurons located within the locus coeruleus, ventral tegmental area and substantia nigra co-express CRH receptor 1, employing a double-immunocytochemical procedure. Using an antibody against the C-terminus of the CRH type 1 receptor (CRH-R1), CRH-R1-like immunoreactivity was found in all cholinergic basal forebrain nuclei except the nucleus basalis magnocellularis. In particular, the diagonal band of Broca (vertical and horizontal limbs) showed a high degree of co-localization of CRH-R1 immunoreactivity and choline acetyltransferase immunoreactivity (both limbs >90%). A less intense immunoreactivity but still high rate of co-localization was detected in the cholinergic neurons of the medial septum (80%), while lowest co-localization was observed in choline acetyltransferase immunoreactive neurons of the substantia innominata (58%). An intermediate degree of co-localization (75%) was seen in the brainstem pedunculopontine tegmental nucleus, while the other major brainstem cholinergic nucleus, the laterodorsal tegmental nucleus, showed an even higher degree of choline acetyltransferase immunoreactivity-positive cells also immunoreactive for CRH-R1 (92%). All catecholaminergic structures studied displayed a pattern of CRH-R1 immunoreactivity strongly overlapping the pattern of tyrosine hydroxylase immunoreactivity. The intensity of the CRH-R1 signal was relatively low within the ventral tegmental area and the substantia nigra pars compacta, while the CRH-R1 signal was very intense and detected in almost all of the neurons of the locus coeruleus.These results clearly demonstrate that the cholinergic and catecholaminergic systems provide direct anatomical substrates for CRH action through the CRH-R1. These findings are of particular relevance for understanding the action of recently developed CRH-R1 antagonistic drugs which may offer a new therapeutic approach to treat stress-related disorders such as anxiety and depression and their concomitant alterations in arousal and cognitive functions.     

The nucleus basalis magnocellularis: The origin of a cholinergic projection to the neocortex of the rat

The cells of origin of a neocortical cholinergic afferent projection have been identified by anterograde and retrograde methods in the rat. Horseradish peroxidase injected into neocortex labelled large, acetylcholinesterase-rich neurons in the ventromedial extremity of the globus pallidus. This same group of neurons underwent retrograde degeneration following cortical ablations. The region in which cell depletion occurred also showed significant decreases in the activities of choline acetyltransferase and acetylcholinesterase. Discrete electrolytic and kainic acid lesions restricted to the medial part of the globus pallidus each resulted in significant depletions of neocortical choline acetyltransferase and acetylcholinesterase. Hemitransections caudal to this cell group did not result in such depletions. Taken together these observations suggest that the acetylcholinesterase-rich neurons lying in the ventromedial extremity of the globus pallidus, as mapped in this study, constitute the origin of a major subcortical cholinergic projection to the neocortex. The utility of acetylcholinesterase histochemistry in animals pretreated with di-isopropylphosphorofluoridate in identifying cholinergic neurons is discussed in the light of this example; specifically, it is proposed that high acetylcholinesterase activity 4–8 h after this pretreatment is a necessary, but not sufficient, criterion for the identification of cholinergic perikarya.The neurons in question appear to be homologous to the nucleus basalis of the substantia innominata of primates, and are thus termed ‘nucleus basalis magnocellularis’ in the rat. No evidence was obtained to support the hypothesis that nucleus of the diagonal band projects to neocortex. However, striking similarities in size and acetylcholinesterase activity were observed among the putative cholinergic perikarya of the nucleus basalis magnocellularis, the nucleus of the diagonal band, and the medial septal nucleus.Kainic acid lesions of the neocortex produced uniform and complete destruction of neuronal perikarya. These lesions decreased neocortical glutamic acid decar?ylase activity, suggesting that there are GABAergic perikarya in the neocortex. However, the same lesions did not affect neocortical choline acetyltransferase. This observation suggests that there are no cholinergic perikarya in the neocortex, a conclusion that is consistent with the absence of intensely acetylcholinesterase-reactive neurons in the neocortex.     

Cholinergic projections to the substantia nigra from the pedunculopontine and laterodorsal tegmental nuclei

The cholinergic innervation of the compact and reticular parts of the substantia nigra in the rat was investigated by use of highly sensitive retrograde and anterograde tract-tracing methods in combination with choline acetyltransferase immunohistochemistry. The fluorescent tracers True Blue, propidium iodide, or fluorogold were infused preferentially into either nigral subnucleus. Cells positive for choline acetyltransferase and retrograde tracer were found in both the pedunculopontine and laterodorsal tegmental nuclei, although considerably more double-labeled somata were observed in the former than in the latter component of the pontomesencephalotegmental cholinergic complex. Approximately 2-3 times more cholinergic cells were labeled in the peduculopontine and laterodorsal tegmental nuclei when tracer injections were centered in the compact nigral subdivision than when infusions of about the same size were confined totally to the reticular part. Infusions of the anterogradely transported tracer Phaseolus vulgaris leucoagglutinin into the pontomesencephalotegmental cholinergic complex resulted in uptake and transport of that label to both nigral subnuclei, and some of the Phaseolus vulgaris leucoagglutinin-accumulating somata and proximal processes also demonstrated choline acetyltransferase-like immunoreactivity. The Phaseolus vulgaris agglutinin-labeled entities in the substantia nigra exhibited terminal-like profiles that were reminiscent of the pattern of nigral choline acetyltransferase-positive puncta demonstrated immunohistochemically by use of nickel ammonium sulfate enhancement of the final reaction product in the avidin-biotin procedure. These observations strongly support the contention that the pontomesencephalotegmental cholinergic complex is the major source of cholinergic projections to both the compact and reticular portions of the rat substantia nigra.     

Arcuate nucleus projections to brainstem regions which modulate nociception

Anterograde tracing studies were conducted in order to identify efferents from the arcuate nucleus, which contains the hypothalamic opiocortin neuronal pool. Phaseolus vulgaris leucoagglutinin (PHA-L) was stereotaxically iontophoresed into the arcuate nucleus and the terminal fields emanating from the labelled perikarya were identified immunocytochemically. PHA-L-immunoreactive (-ir) fibers were identified in nucleus accumbens, lateral septal nucleus, bed nucleus of the stria terminalis, medial and lateral preoptic areas, anterior hypothalamus, amygdaloid complex, lateral hypothalamus, paraventricular nucleus, zona incerta, dorsal hypothalamus, periventricular gray, medial thalamus and medial habenula. In the brainstem, arcuate terminals were identified in the periaqueductal gray (PAG), dorsal raphe nucleus (DRN), nucleus raphe magnus (NRM), nucleus raphe pallidus, locus coeruleus, parabrachial nucleus, nucleus reticularis gigantocellularis pars alpha, nucleus tractus solitarius and dorsal motor nucleus of the vagus nerve. Dual immunostaining was used to identify the neurochemical content of neurons in arcuate terminal fields in the brainstem. Arcuate fiber terminals established putative contacts with serotonergic neurons in the ventrolateral PAG, DRN and NRM and with noradrenergic neurons in periventricular gray, PAG and locus coeruleus. In the PAG, arcuate fibers terminated in areas with neurons immunoreactive to substance P, neurotensin, enkephalin and cholecystokinin (CCK) and putative contacts were identified with CCK-ir cells. This study provides neuroanatomical evidence that putative opiocortin neurons in the arcuate nucleus influence a descending system which modulates nociception.     

Regional decreases of cortical choline acetyltransferase after lesions of the septal area and in the area of nucleus basalis magnocellularis

Choline acetyltransferase and [³H]choline uptake have been measured in neocortical regions and hippocampus one week after lesions which destroyed the septum bilaterally, and after unilateral lesions in the area of nucleus basalis magnocellularis. Lesions of the septal area, which severely decreased choline acetyltransferase in hippocampus, only moderately decreased choline acetyltransferase in a posterior cortical region and had no effect in frontal and parietal regions. In contrast, lesions which included nucleus basalis magnocellularis decreased choline acetyltransferase markedly in frontal and parietal regions and had less of an effect in the posterior cortical regions. Lesion-induced decreases of [³H]choline uptake paralleled those of choline acetyltransferase. Lesions which included nucleus basalis magnocellularis had no effect on choline acetyltransferase in hippocampus, nucleus accumbens, olfactory tubercle, midbrain or pons-medulla.These results suggest the existence of topographically distinct cholinergic inputs to neocortex. In agreement with previous studies, cholinergic projections from the peripallidal region of nucleus basalis magnocellularis are predominantly to frontal and parietal neocortex. In contrast to previous suggestions, cholinergic projections to neocortex from the septal area are limited to the posterior regions of neocortex.     

The basal forebrain cholinergic system in the raccoon

The distribution of neurons displaying choline acetyltransferase (ChAT) immunoreactivity was examined in the raccoon basal forebrain using a rabbit antiscrum and a monoclonal antibody. Alternating sections were used for Nissl staining. ChAT-positive neurons were arranged in a continuous mass extending from the medial septum to the caudal pole of the pallidum. Based upon spatial relations to fibre tracts, the clustering of neuronal groups, and cytological criteria, the basal forebrain magnocellular complex can be subdivided into several distinct regions. Although clear nuclear boundaries were often absent, the ChAT-positive neurons were divided into: the nucleus tractus diagonalis (comprising pars septi medialis, pars verticalis and pars horizontalis); nucleus praeopticus magnocellularis; substantia innominata; and the nucleus basalis of Meynert. Comparison with Nissl-stained sections indicated the presence of varying proportions of non-cholinergic neurons clustered or arranged loosely within these basal forebrain subdivisions. These data provide a structural basis for studies concerned with the topographical and physiological aspects of the raccoon basal forebrain cholinergic projections and its comparison with the basal forebrains of other species.     

Neocortical cholinergic innervation: A description of extrinsic and intrinsic components in the rat

Electrothermic lesion of the peri-pallidal region of the rat caused a marked reduction in the activity of choline acetyltransferase in the ipsilateral fronto-parietal cortex without affecting the activity of glutamate decarboxylase. Only lesions that involved the ventral globus pallidus significantly reduced cortical choline acetyltransferase activity; and lesions limited to the thalamus, internal capsule, pyriform cortex or zone incerta were ineffective. Excito-toxin lesions of the ventral globus pallidus caused 45-5% reductions in all presynaptic markers for cholinergic neurons but did not significantly decrease presynaptic markers for noradrenergic, serotonergic or histaminergic neurons in the cortex. The maximal reductions in cortical choline acetyltransferase activity achieved with the pallidal lesion was 70%; and enzyme activity reached its nadir by four days after placement of the lesion. The pallidal lesion, which ablated the large isodendritic acetylcholinesterase positive neuronal perikarya, resulted in a profound loss in histochemically stained acetylcholinesterase-reactive fibers in the fronto-parietal cortex but not in the cingulate, pyriform and occipital cortex or hippocampal formation; analysis of the subregions in choline acetyltransferase activity. The kainate lesion of the parietal cortex to ablate intrinsic neurons did not reduce the activity of tyrosine hydroxylase, a marker for noradrenergic terminals, but depressed glutamate decarboxylase by 68%; in contrast choline acetyltransferase activity fell only 29%. The results indicate that approximately 70% of the cholinergic innervation in the fronto-parietal cortex is derived from acetylcholinesterase positive neurons in the peripallidal nucleus basalis, whereas the remainder appears to be localized in cortical intrinsic neurons.     

Afferents to the basal forebrain cholinergic cell area from pontomesencephalic--catecholamine, serotonin, and acetylcholine--neurons

The afferent input to the basal forebrain cholinergic neurons from the pontomesencephalic tegmentum was examined by retrograde transport of wheatgerm agglutinin-horseradish peroxidase in combination with immunohistochemistry. Multiple tyrosine hydroxylase-, dopamine-beta-hydroxylase-, serotonin- and choline acetyltransferase-immunoreactive fibres were observed in the vicinity of the choline acetyltransferase-immunoreactive cell bodies within the globus pallidus, substantia innominata and magnocellular preoptic nucleus. Micro-injections of horseradish peroxidase-conjugated wheatgerm agglutinin into this area of cholinergic perikarya led to retrograde labelling of a large population of neurons within the pontomesencephalic tegmentum, which included cells in the ventral tegmental area, substantia nigra, retrorubral field, raphe nuclei, reticular formation, pedunculopontine tegmental nucleus, laterodorsal tegmental nucleus, parabrachial nuclei and locus coeruleus nucleus. Of the total population of retrogradely labelled neurons, a significant (approximately 25%) proportion were tyrosine hydroxylase-immunoreactive and found in the ventral tegmental area (A10), the substantia nigra (A9), the retrorubral field (A8), the raphe nuclei (dorsalis, linearis and interfascicularis) and the locus coeruleus nucleus (A6), Another important contingent (approximately 10%) was represented by serotonin neurons of the dorsal raphe nucleus (B7), the central superior nucleus (B8) and ventral tegmentum (B9). A small proportion (less than 1%) was represented by cholinergic neurons of the pedunculopontine (Ch5) and laterodorsal (Ch6) tegmental nuclei. These results demonstrate that pontomesencephalic monoamine neurons project in large numbers up to the basal forebrain cholinergic neurons and may represent a major component of the ventral tegmental pathway that forms the extra-thalamic relay from the brainstem through the basal forebrain to the cerebral cortex.     

Dissociable effects on spatial maze and passive avoidance acquisition and retention following AMPA- and ibotenic acid-induced excitotoxic lesions of the basal forebrain in rats: differential dependence on cholinergic neuronal loss

Excitotoxic lesions of the basal forebrain were made by infusing either alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) or ibotenic acid. Acquisition and performance of spatial learning in the Morris water maze, over a ten day, two trials per day, training regimen were unaffected by the AMPA-induced lesions which reduced cortical choline acetyltransferase activity by 70%. However, acquisition was significantly impaired in rats with ibotenic acid-induced lesions that reduced cortical choline acetyltransferase by 50%. Additionally, ibotenic acid-lesioned rats swam further than either sham or AMPA-lesioned rats, in the "training" quadrant during a probe trial, in which the escape platform was removed, suggesting a perseverative search strategy. Lesions induced with AMPA, but not ibotenate, significantly impaired the acquisition of "step-through" passive avoidance. Both AMPA- and ibotenate-induced lesions significantly impaired the 96 h retention of passive avoidance, but the effect of AMPA was greater on latency measures. Histological analysis revealed that AMPA infusions destroyed more choline acetyltransferase-immunoreactive neurons than did ibotenate infusions but, unlike ibotenate, spared the overlying dorsal pallidum and also parvocellular, non-choline acetyltransferase-immunoreactive neurons in the ventral pallidal/substantia innominata region of the basal forebrain. The impairment in acquisition of the water maze following ibotenate-induced basal forebrain lesions therefore appears unrelated to damage to cholinergic neurons of the nucleus basalis of Meynert and to depend instead on damage to pallidal and other neurons in this area. The AMPA- and perhaps also the ibotenate-induced impairment in the retention of passive avoidance appears to be more directly related to destruction of cholinergic neurons of the nucleus basalis. These data are discussed in the context of cortical cholinergic involvement in mnemonic processes.     

Cholinergic neurons in the rat central nervous system demonstrated by in situ hybridization of choline acetyltransferase mRNA.

Digoxigenin-labeled RNA probes and in situ hybridization histochemistry were used to examine choline acetyltransferase gene expression in the rat central nervous system. Hybridization signal was present only in brain sections processed with the antisense riboprobe. The sense probe did not yield labeling, further validating the specificity of tissue reactivity. Telencephalic neurons containing the mRNA for the cholinergic synthetic enzyme were found in the caudate-putamen nucleus, nucleus accumbens, olfactory tubercule, islands of Calleja complex, medial septal nucleus, vertical and horizontal limbs of the diagonal band, substantia innominata, nucleus basalis, and nucleus of the ansa lenticularis. Some somata evincing hybridization signal were observed in the anterior amygdalar area, and an occasional such cell was seen in the basolateral and central amygdalar nuclei. Neurons in the cerebral cortex, hippocampus, and primary olfactory structures did not demonstrate hybridocytochemically detectable amounts of choline acetyltransferase mRNA. Thalamic cells were devoid of reactivity, with the exception of several neurons located primarily in the ventral two-thirds of the medial habenula. A few somata labeled with riboprobe were found in the lateral hypothalamus, caudal extension of the internal capsule, and zona incerta. Neurons in the pedunculopontine and laterodorsal tegmental nuclei were moderately reactive, whereas cells of the parabigeminal nucleus exhibited a very weak hybridization signal. No somata in the brainstem raphe nuclei, including raphe obscurus and raphe magnus, were observed to bind riboprobe. In contrast, motor neurons of the cranial nerve nuclei demonstrated relatively large amounts of choline acetyltransferase mRNA. Putative cholinergic somata in the ventral horns and intermediolateral cell columns of the spinal cord were also labeled with riboprobe, as were a few cells around the central canal. We conclude that hybridocytochemistry with digoxigenin-labeled riboprobes confirms the existence of cholinergic neurons (i.e. those that synthesize and use acetylcholine as a neurotransmitter) in most of the neural regions deduced to contain them on the basis of previous histochemical and immunocytochemical data. Notable exceptions are the cerebral cortex and hippocampus, which do not possess neurons expressing detectable levels of choline acetyltransferase mRNA.     

Projections from the amygdaloid complex to the magnocellular cholinergic basal forebrain in rat

The amygdaloid complex has a key role in the modulation of behavioral responses in life-threatening situations, including the direction of attentional responses to sensory stimuli. The pathways from the amygdala to the basal forebrain cholinergic system, which projects to the cortex, are proposed to contribute to the modulation. To further explore the topography and postsynaptic targets of these pathways, we investigated the projections from the different divisions of the lateral, basal, accessory basal, and central nuclei of the amygdala to the cholinergic basal forebrain in rat using a sensitive anterograde tracer, Phaseolus vulgaris leucoagglutinin. The most substantial projections from the amygdala to the basal forebrain are directed to the ventrolateral and dorsomedial aspects of the substantia innominata and the fundus of the striatum. The heaviest projections originate in the capsular, lateral, and intermediate divisions of the central nucleus as well as in the magnocellular and parvicellular divisions of the basal nucleus. Light microscopic analysis of double-stained preparations revealed that the distribution of amygdaloid efferents and cholinergic neurons overlaps most prominently in the ventrolateral substantia innominata. Despite the fact that the central nucleus efferents and cholinergic elements overlap in the ventrolateral substantia innominata, electron microscopic analysis revealed, first, that the postsynaptic targets of the central nucleus efferents are non-cholinergic, probably GABAergic, neurons. Second, 80% of the synaptic contacts were symmetric.The present data extend previous observations showing that the different amygdaloid nuclei provide projections to the selective basal forebrain areas. Further, the central nucleus efferents modulate cholinergic neurons in the basal forebrain indirectly via the GABAergic interneurons.     

Direct projection from the dorsal hypothalamic area to the nucleus raphe pallidus: a study using anterograde transport with Phaseolus vulgaris leucoagglutinin in the rat

Summary A hypothalamic projection to the nucleus raphe pallidus of the medulla was examined using the anterograde tracing technique based on Phaseolus vulgaris leucoagglutinin (PHA-L) in the rat. After the iontophoretic application of PHA-L to the dorsal hypothalamic area, labeled fibers that finally ended in the nucleus raphe pallidus were observed descending through the most medial part of the ventral tegmental area and the nucleus reticularis tegmenti pontis to reach the medial aspect of the pyramid. Many varicose fibers forming a loose plexus were observed in the nucleus raphe pallidus, especially ventrally. The ventral surface of the pyramid and the most ventral region of the nucleus reticularis paragigantocellularis lateralis (PGCL) contained labeled varicose fibers. At the electron microscopic level, the labeled profiles in the nucleus raphe pallidus were small-sized unmyelinated axons and axon terminals. Labeled axon terminals containing spherical synaptic vesicles formed synapses on spine-like protrusions or small-sized dendritic shafts. These results strongly indicate that neurons in the dorsal hypothalamic area have a direct connection with neurons in the nucleus raphe pallidus and the ventral part of the PGCL. The possible involvement of this pathway in cardiovascular regulation was discussed.     

Distribution of GABAergic and cholinergic neurons in the rat diagonal band

GABAergic neurons are coextensive with cholinergic neurons in the medial septum-diagonal band complex. Serial sectioning, sequential staining and double immunofluorescence techniques employing antibodies to glutamate decarboxylase and choline acetyltransferase revealed the distribution of these transmitter-specific neurons in the rat. Morphologically, the two types of neurons appear similar, in that they are predominantly large multipolar cells, but they are characterized by different, overlapping distributions in the diagonal band. Glutamate decarboxylase-positive cells are scattered throughout the nucleus of the vertical limb of the diagonal band, while choline acetyltransferase-positive neurons are more numerous medially and are distributed in groups corresponding to the dorsal and ventral aspects of the nucleus. In the rostral parts of the nucleus of the horizontal limb of the diagonal band, the choline acetyltransferase-positive cells tend to be located medially, whereas caudally they spread dorsal to the nucleus to become continuous with other large cholinergic neurons in the ventral pallidum and sublenticular substantia innominata. The large majority of glutamate decarboxylase-positive neurons remain in a more ventral and lateral position within the nucleus of the horizontal limb and are particularly numerous just lateral to the diagonal band fibers as they join the medial forebrain bundle. Cholinergic neurons were estimated to be about two times more numerous than GABAergic neurons. Approximately 1% of the choline acetyltransferase-positive neurons were also glutamate decarboxylase-positive in double immunofluorescence studies, but not in sequentially stained or serial sections.     

The effects of excitotoxic lesions of the substantia innominata, ventral and dorsal globus pallidus on the acquisition and retention of a conditional visual discrimination: implications for cholinergic hypotheses of learning and memory

The effects of ibotenic acid-induced lesions of the ventral pallidum/substantia innominata region, the dorsal pallidum or both on the acquisition and retention of a conditional visual discrimination have been studied in the rat. Lesions of the ventral pallidum and large lesions of the dorsal and ventral pallidum severely impaired both the acquisition and retention of the conditional discrimination. Dorsal pallidal lesions had similar, but less marked effects. The same lesions also impaired the retention of a passive avoidance task, but had no effect on a conditioned taste aversion. Neurobiological investigations revealed that the lesions destroyed cholinergic neurons in the magnocellular nucleus basalis and caused reductions in cortical choline acetyltransferase activity of about 30-40%. Tract-tracing experiments indicated that the lesions destroyed, in particular, cholinergic neurons projecting to the frontal dorsolateral cortex and also those projecting to more posterior cortex, but not the occipital lobes. Contingency analysis of the behavioural, neurochemical and neuroanatomical data indicated that those animals with the largest decreases in choline acetyltransferase activity, or the largest areas of neuronal loss in the ventral and dorsal globus pallidus, were most impaired in the retention of the conditional discrimination. The results do not, therefore, indicate a simple relationship between cholinergic neuronal loss and the retention of response rules essential for performance of the task ("reference memory"). The relevance of the results to cholinergic hypotheses of learning and memory is discussed.