Thalamic and callosal connections of the rat auditory cortex

This study was designed to assess the relative distributions of two entrinsic afferent fiber systems in the rat auditory cortex as indicated by the patterns of specific lesion-induced degeneration evident in Fink-Heimer preparations. The auditory cortex consists of cytoarchitectural areas 41, 20 and 36. Lesions were made in the medial geniculate body (MGB) or the corpus callosum in some rats, while in other rats, lesions were made in both the MGB and the corpus callosum. Following the thalamic lesions, degenerating terminals occur throughout the auditory region of cortex, principally in layer IV and deep layer III, but also in layer VI and in the superficial part of layer I. With the exception of the band of degenerationin layer I, the density of the thalamic degeneration is uneven, such as that patches of increased density of degeneration are seperated by regions with few degenerating terminals. Following lesions of the corpus callosum, degenerating callosal terminals are also evident thoughout the auditory region of cortex and they occur in deep layer I through layer III, superficial layer V and layer VI. The dennsity of the degenerating callosal terminals is not uniform throughout most of area 41, to the extent that there are radially-oriented bands of increased density which appears within the continuous callosal projection. Following the double lesions, degenerating terminals throughout the auditory region are distributed homogenoously within all cortical layers with the exception of deep layer Vwhish is relatively free of degeneration. The results indicate that all regions within the rat auditory cortex are subject to both thalamic and callosal influence, although the input is not completely uniform, for the zones in layers IV and VI which have decreased thalamic input appear to have increased callosal input.     

Connections between cells of the internal capsule,thalamus, and cerebral cortex in embryonic rat

The aim of our study is to understand the development of the earliest connections in the mammalian pallium by documenting the distribution of cells and fibres labelled from the dorsal and ventral thalamus, internal capsule, perirhinal, and dorsal cortex during the period between embryonic day (E) 14 and 17 by using carbocyanine dye tracing in fixed embryonic rat brains. Dye placed in the thalamus of E14 brains backlabels cells in the thalamic reticular nucleus and within the primitive internal capsule. Both anterograde and retrograde tracing confirmed that the first corticofugal projections reach the internal capsule by E14. At E15–E16, after the first cortical plate cells have migrated into the lateral cortex, some cells of the cortical plate and subplate and marginal zone, are backlabelled from the internal capsule, but still not from the dorsal thalamus, even with very long incubation periods. Crystal placement into the perirhinal cortex at E14–E15 labels numerous cells within the internal capsule, whereas no such cells are revealed from dorsal cerebral cortex until E17, suggesting that internal capsule cells establish early connections with the perirhinal and ventral but not dorsal cortex. We propose that the growth of axons from cortex to dorsal thalamus is delayed in two regions: first from E14–E15 at the lateral entrance of the internal capsule and then, from E16, closer to the thalamus, probably within the thalamic reticular nucleus. Subplate projections reach the proximity of the diencephalon at an early stage, but they might never enter the dorsal thalamus. J. Comp. Neurol. 413:1–25, 1999. © 1999 Wiley-Liss, Inc.     

Regrowth of olfactory sensory axons into transplanted neural tissue. 1. Development of connections with the occipital cortex

Presumptive occipital cortex was transplanted from 15-day-old embryos to host mice 10 days of age close to the ethmoidal lamina cribrosa, previous to partial or total bulbectomy of the host. The transplants have been studied for a period up to 150 days on histological preparations stained with iron hematoxylin and Bodian methods. The olfactory sensory axons, which degenerated as a result of bulbectomy, subsequently regenate and invade the transplant, where they form characteristic glomerular structures. The transplant's neurons show interesting relations with the newly formed glomeruli. Significance of the results is briefly discussed.     

Development of the cholinergic fibres innervating the cerebral cortex of the rat

The ontogeny of innervation of the cholinergic fibres from the basal forebrain into the cingulate, frontal, parietal and piriform cortices of the rat has been examined using a modified histochemical method of acetylcholinesterase (AChE). The method produced crisp fibre staining with enhanced visibility and a clear back-ground, and a pattern of the distribution of these fibres was comparable to that achieved by choline acetyltransferase (ChAT) immunocytochemistry. In the rat, the AChE-stained fibres developed progressively from the deep cortical white matter towards the cortex itself. In general, a few AChE-positive fibres were seen in the subcortical white matter and the cingulum bundle, entering into the cerebral cortex by about 5 postnatal days. The number of these AChE-positive processes increased dramatically during the following two weeks. Thereafter, the general appearance of the overall pattern of distribution of the AChE fibres changed little, but the staining density became gradually more intense and by about 28 days after birth it was virtually indistinguishable from that in the adult. The onset and the development of the AChE-positive fibre network varied considerably between individual cortical regions, and indicated, in general, an anterior to posterior gradient. Within the dispersed AChE fibre network in the cerebral cortex, three bands of relatively enriched cholinergic processes, namely the deep cortical, mid-cortical and superficial layers, developed in an 'inside-out' fashion. The exact position of some of these AChE-rich bands varied from one cortical region to another and during development. A striking correlation during ontogeny was observed in the cerebral cortex between the changing patterns of AChE fibre network and the activity of ChAT, the enzyme synthesizing acetylcholine. The present findings can also provide an important anatomical baseline for future studies related to the factors controlling the expression of ChAT activity and the development of cholinergic neurotransmitter system in the rat.     

A method for the dissection of the embryonic cerebral cortex into individual layers. An application to biochemical studies of glycan metabolism

A method is presented for the isolation of defined cerebral cortical layers from the prenatal rat brain. By this procedure cells at different stages of proliferation and/or differentiation can be obtained as relatively homogeneous populations. The principle of the proposed isolation procedure consists of freezing the isolated embryonic brain in a mould which flattens the two hemispheres. The different cortical layers are then isolated by carefully cutting serial cryostat sections. On embryonic day 16 (ED 16), 3 individual layers could be sampled from the cortex: the marginal zone, the zone of DNA synthesis and the mitotic region of the ventricular germinative zone. On ED 18 and ED 20, a further 3 layers could be isolated: the cortical plate, the sub-plate region, and the intermediate zone. As an example of an application of the isolation procedure for biochemical studies, maturation-dependent changes in the protein and Concanavalin A-binding glycoprotein patterns together with the activity of beta-N-acetylglucosaminidase in the different cortical layers of the embryonic brain are demonstrated at different stages of brain development.     

Cytoarchitecture of sensorimotor areas in the cat cerebral cortex

The organization of multiple motor areas in the cerebral cortex has been investigated frequently in primates but rarely in nonprimate species. To compare sensorimotor areas in cats and primates, the cytoarchitecture of frontal and parietal areas of the cat cerebral cortex was described and mapped from coronal sections stained with cresyl violet or thionine. Multiple subdivisions of areas 4 and 6 were recognized; of these, the cytoarchitecture of area 4γ is similar to that of area 4 described in other carnivores and in primates and is characterized by giant pyramidal cells in multiple rows or clusters in lamina V. In other subdivisions of area 4 (4δ, 4sfu, and 4fu), giant pyramidal cells are few or absent in lamina V, and these subdivisions resemble area 6 of primates. Area 6 of the cat cortex is heterogeneous, and differences in laminar appearance and size of pyramidal cells in lamina V distinguish its four subdivisions (6aα, 6aβ, 6aγ, and 6iffu). The adjoining prefrontal areas are distinguishable from area 6 by the presence of a thin internal granular lamina (lamina IV) and the reduced size of pyramidal cells in lamina V. Laminae are poorly differentiated in the cingulate areas, where a rostral and caudal subdivision can be distinguished on the basis of the absence or presence of lamina IV. Area 3a is characterized by a thin lamina IV and is located between frontal agranular and parietal granular (well-defined lamina IV) fields (3b, 1, 2, 2pri, 5, and 7). Insular cortex can be subdivided into granular and agranular fields. J. Comp. Neurol. 388:354–370, 1997. © 1997 Wiley-Liss, Inc.     

Suction lesions of the frontal cerebral cortex in the rat induce asymmetrical behavioral and catecholaminergic responses

Suction lesions of the right frontal cerebral cortex in rats induce a period of spontaneous hyperactivity. This hyperactivity, as measured by an increase in running wheel activity begins about one week post-operatively and continues throughout the remainder of a 30-day observation period. The increased activity is accompanied by a bilateral decrease in norepinephrine concentrations in both the ipsilateral and contralateral cortex and locus coeruleus. Identical lesions of the left frontal cerebral cortex produce neither the hyperactivity nor a decrease in norepinephrine concentrations. These experiments have reproduced many of the behavioral and biochemical asymmetries seen after middle cerebral artery ligation; however, suction lesions are both simpler to produce and histologically less variable in their effects.     

The projection of three extrathalamic cell groups to the cerebral cortex of the turtle Pseudemys

Three extrathalamic subcortical inputs to the part of the cerebral cortex that is known to receive thalamic fibers in the turtle were examined in the present study. Direct projections from the locus coeruleus, the superior medial raphe nucleus, and a wide area of the basal telencephalon that lies ventromedial to the globus pallidus were demonstrated with the horseradish peroxidase method. Fluorescence histochemistry confirmed the presence of catecholamine-containing fibers in the rostral half of dorsal cortex and also demonstrated a dense network of serotoninergic fibers. Biochemical analysis showed the concentration of both monoamines to be relatively high; the norepinephrine concentration was 709 ng/g and the serotonin concentration was 1,750 ng/g. No evidence was found to suggest the existence of either a dopamine fiber projection to cortex comparable to that of mammalian neocortex or the presence of an epinephrine pathway to turtle cortex equivalent to the epinephrine-containing fibers in the pallium of amphibians. The coexistence of the projections from the thalamus with noradrenergic projections from the locus coeruleus, serotoninergic projections from the superior medial raphe nucleus, and presumably cholinergic projections from the basal telencephalon provide at least four distinct subcortical inputs to the reptilian dorsal cortex. Neither thalamic nor similar extrathalamic inputs have been demonstrated in the dorsal pallium of amphibia. Mammalian neocortex, in contrast, has even more elaborately differentiated inputs of both types. These results support the idea that thalamic and extrathalamic inputs to cortex appear at the same time in vertebrate evolution, and that both types of inputs are required for the normal development and function of neocortex.     

Connections of the retrosplenial dysgranular cortex in the rat.

Although the retrosplenial dysgranular cortex (Rdg) is situated both physically and connectionally between the hippocampal formation and the neocortex, few studies have focused on the connections of Rdg. The present study employs retrograde and anterograde anatomical tracing methods to delineate the connections of Rdg. Each projection to Rdg terminates in distinct layers of the cortex. The thalamic projections to Rdg originate in the anterior (primarily the anteromedial), lateral (primarily the laterodorsal), and reuniens nuclei. Those from the anteromedial nucleus terminate predominantely in layers I and IV-VI, whereas the axons arising from the laterodorsal nucleus have a dense terminal plexus in layers I and III-IV. The cortical projections to Rdg originate primarily in the infraradiata, retrosplenial, postsubicular, and areas 17 and 18b cortices. The projections arising from visual areas 18b and 17 predominantly terminate in layer I of Rdg, axons from contralateral Rdg form a dense terminal plexus in layers I-IV, with a smaller number of terminals in layers V and VI, afferents from postsubiculum terminate in layers I and III-V, and the projection from infraradiata cortex terminates in layers I and V-VI. The efferent projections from Rdg are widespread. The major cortical projections from Rdg are to infraradiata, retrosplenial granular, area 18b, and postsubicular cortices. Subcortical projections from Rdg terminate primarily in the ipsilateral caudate and lateral thalamic nuclei and bilaterally in the anterior thalamic nuclei. The efferent projections from Rdg are topographically organized. Rostral Rdg projects to the dorsal infraradiata cortex and the rostral postsubiculum, while caudal Rdg axons terminate predominantely in the ventral infraradiata and the caudal postsubicular cortices. Caudal but not rostral Rdg projects to areas 17 and 18b of the cortex. The Rdg projections to the lateral and anterior nuclei also are organized along the rostral-caudal axis. Together, these data suggest that Rdg integrates thalamic, hippocampal, and neocortical information.     

Columnar organization of callosal and associational projections from rat frontal cortex

Autoradiography and HRP histochemistry were used to study the laminar and columnar distribution of callosal and associational connections of areas 6 and 10 of Krieg in the rat frontal cortex. In coronal sections through homotopic contralateral areas and ipsilateral somatosensory cortex, terminations of projections arising in frontal cortex formed discrete vertical columns; these were 250–750 μm wide and alternated with unlabeled or poorly labeled areas of approximately equal width. In reconstructions from serial coronal sections through these areas, the terminal fields formed a series of bands. The location of retrogradely labeled neurons tended to reciprocate the distribution of terminal label, although the boundaries of terminal and cell label were not always in precise register. These findings indicate that in the rat, both association and callosal projections exhibit a terminal organization remarkably similar in width and spacing to that observed in primates. Thus, a columnar mode of termination of cortico-cortical fibers may be an organizational feature common to mammalian neocortex.     

Covariation of distributions of callosal cell bodies and callosal axon terminals in layer III of cat primary auditory cortex

Primary auditory cortex in the cat is both the source and target of callosal fibers. Injection of horseradish peroxidase (HRP) in the high frequency representation of AI in one hemisphere retrogradely labels callosal cell bodies and anterogradely labels callosal axon terminals in AI of the opposite hemisphere. In tissue sections cut through layer III parallel to the cortical surface, elongated patches composed of dense aggregates of callosal cell bodies and callosal axon terminals alternate with regions containing lower concentrations of these elements. Labeling in AI is most dense in regions corresponding to the frequency representation of the injected site. In layer III of the densely labeled region, patches of high concentrations of labeled callosal axon terminals correspond with high concentrations of labeled callosal cell bodies. On the other hand, little correspondence is apparent between the distributions of the two elements in layer III in the sorrounding area of lighter labeling. Layers V and VI contain relatively few labeled callosal axon terminals and cell bodies, and our data do not suggest whether the two distributions covary in these layers.     

The mid-ventrolateral prefrontal cortex: insights into its role in memory retrieval

Although it is widely known that the prefrontal cortex plays a role in memory, the specific contribution of particular prefrontal regions in mnemonic functions remains controversial. The present investigation examined whether the mid-ventrolateral prefrontal cortex is selectively involved in active memory retrieval in situations in which mnemonic traces are embedded in ambiguous relations and automatic recollection cannot lead to successful retrieval. Thirteen subjects participated in this event-related functional magnetic resonance imaging experiment. Throughout the scanning session, trials belonging to an experimental and a control condition were administered in a pseudorandom fashion. During the encoding phase of any particular trial, subjects were presented with a stimulus-complex that was a combination of a face and a spatial location on the screen. In the experimental active retrieval condition, a question cue following the encoding phase instructed the subjects to retrieve selectively one of the two aspects of the encoded stimulus-complex, i.e. the face or the location. In the control condition, the question cue that followed the encoding phase instructed the subjects simply to recall the initially presented stimulus-complex, so as to be able to make a decision during the test phase based on simple stimulus familiarity. The comparison of the signal obtained during the retrieval phase of these two conditions yielded an increase in activity selective to the right mid-ventrolateral prefrontal region. These results therefore establish a specific link between the mid-ventrolateral prefrontal cortex and active retrieval mechanisms.     

Effects of prenatal irradiation on the development of cerebral cortex and corpus callosum of the mouse

Defects of the cerebral cortex and corpus callosum of mice subjected prenatally to gamma irradiation were evaluated as a function of dose and of embryonic age at irradiation. Pregnant mice were exposed to a gamma source at 16, 17, and 19 days of gestation (E16, E17, and E19, respectively), with total doses of 2 Gy and 3 Gy, in order to produce brain defects on their progeny. At 60 postnatal days, the brains of the offspring were analyzed qualitatively and quantitatively and compared with those of nonirradiated animals. Mice irradiated at E16 were all acallosal. Those that were exposed to 2 Gy displayed an aberrant longitudinal bundle typical of other acallosals, but this was not the case in those irradiated with 3 Gy. The corpus callosum of animals irradiated at E17 with 3 Gy was pronouncedly hypotrophic, but milder effects were observed in the other groups. Quantitative analysis confirmed a dependence of callosal midsagittal area upon dose and age at irradiation, and, in addition, indicated an interaction between these variables. The neocortex of irradiated animals was hypotrophic: layers II-III were much more affected than layer V, and this was more affected than layer VI. Quantitative analysis indicated that this effect also depended on dose and age at irradiation and that it was due to a loss of cortical neurons. Furthermore, a positive correlation was found between the number of neurons within layers II-III, and V and the midsagittal area of the corpus callosum. Ectopic neurons were found in the white matter and in layer I of animals irradiated at E16 and E17, indicating that fetal exposure to ionizing radiation interfered with the migration of cortical neuroblasts.     

Retinal input influences the size and corticocortical connectivity of visual cortex during postnatal development in the ferret

Retinal input plays an important role in the specification of topographically organized circuits and neuronal response properties, but the mechanism and timing of this effect is not known in most species. A system that shows dramatic dependence on retinal influences is the interhemispheric connection through the corpus callosum. Using ferrets, we analyzed the extent to which development of the visual callosal pattern depends on retinal influences, and explored the period during which these influences are required for normal pattern formation. We studied the mature callosal patterns in normal ferrets and in ferrets bilaterally enucleated (BE) at postnatal day 7 (P7) or P20. Callosal patterns were revealed in tangential sections from unfolded and flattened brains following multiple injections of horseradish peroxidase in the opposite hemisphere. We also estimated the effect of enucleation on the surface areas of striate and extrastriate visual cortex by using magnetic resonance imaging (MRI) data from intact brains. In BEP7 ferrets we found that the pattern of callosal connections was highly anomalous and the sizes of both striate and extrastriate visual cortex were significantly reduced. In contrast, enucleation at P20 had no significant effect on the callosal pattern, but it still caused a reduction in the size of striate and extrastriate visual cortex. Finally, retinal deafferentation had no significant effect on the number of visual callosal neurons. These results indicate that the critical period during which the eyes influence the development of callosal patterns, but not the size of visual cortex, ends by P20 in the ferret.     

Functions of the frontal cortex of the rat: A comparative review

This review summarizes the anatomical and functional organization of the frontal cortex of the rat in comparison to primates. Lesions of the primary motor or of the prefrontal cortex of both primates and rodents produce a consistent constellation of symptoms that are strikingly similar across species as diverse as rats and humans. Thus, in spite of the tremendous difference in the relative volume of the frontal cortex of mammals, as well as the obvious diversity of behavioral repertoires across mammalian phylogeny, there appears to be a remarkable unity in frontal cortex function across the class mammalia. Hence, motor and prefrontal lesions produce analogous alterations in motor control in rodents and primates even though humans walk upright and have fine control of digit movement and rats walk on all fours and have less dextrous control of distal movements. Similarly, there are analogous changes in behaviors that can be labeled response inhibition, temporal ordering, spatial orientation, social or affective behavior, behavioral spontaneity, olfaction and habituation following prefrontal cortex lesions in both primates and rodents. Finally, it is proposed that the principal function of the prefrontal cortex of mammals is the temporal organization of behavior.     

p-Nonylphenol, 4-tert-octylphenol and bisphenol A increase the expression of progesterone receptor mRNA in the frontal cortex of adult ovariectomized rats

Alkylphenols, such as p-nonylphenol (NP) and 4-tert-octylphenol (OP) and bisphenol A (BPA) are thought to mimic oestrogens in their action, and are called endocrine disrupters. We examined whether these endocrine disrupters affected progesterone receptor (PR) mRNA expression in the adult female rat neocortex. In one experiment, at 12.00 h, ovariectomized rats were given a subcutaneous injection of 10 mg of NP, 10 mg of OP or 10 mg of BPA, or sesame oil alone as control. Twenty-four hours after injection, the left side of the frontal cortex, parietal cortex and temporal cortex was collected. In a second experiment to study the time-course of the effects of BPA on PR mRNA, the ovariectomized rats were given a subcutaneous injection of 10 mg of BPA and killed 0, 6, 12 and 24 h after injection. In addition to the frontal cortex and temporal cortex, the occipital cortex was also collected. Northern blotting revealed that, in the first experiment, injection of NP, OP or BPA significantly increased PR mRNA expression in the frontal cortex but not in the parietal cortex. In the temporal cortex, BPA significantly decreased PR mRNA, but NP and OP produced no significant changes. The second experiment revealed that, in the frontal cortex, BPA induced a significant increase in PR mRNA expression at 6 h after injection, which lasted until 24 h after injection. In the temporal cortex, PR mRNA expression was significantly decreased 6 h after injection of BPA and was still significantly low 24 h after injection. No significant change was observed in the occipital cortex. These results suggest that, even in adult rats, endocrine disrupters alter the neocortical function by affecting the PR system, although the physiological significance of PR in the affected area is unknown.     

The medial frontal cortex and gastric motility: Microstimulation results and their possible significance for the overall pattern of organization of rat frontal and parietal cortex

Bilateral intracortical microstimulation (60–90 s trains of 0.5 ms pulses at 10 Hz, currents below 50 μA) of medial frontal infralimbic and prelimbic cortical areas in ketamine-anesthetized rats produces clear and consistent decreases in ongoing gastric motility. The majority of responses consists of reductions in gastric tone, reductions in the amplitude of gastric contractions, or combined reductions in tone and amplitude. Bilateral section of the vagus nerves eliminates most of the responses, suggesting that the responses are mediated by this nerve. The effective corticdal stimulation zone (the ‘visceral motor’ cortex) largely overlaps the source of the recently described direct projection from medial frontal cortex to the nucleus of the solitary tract; this pathway may be involved in producing the effect. Connections of this cortex with the limbic system suggest it may be involved in producing physiological responses to stress. The topographical, medial to lateral sequence of cortical functional areas revealed by these and other experiments (visceral motor, frontal eye fields, somatic motor, somatic sensory, visceral sensory) is discussed, as well as the possible implications of this pattern to the question of cortical evolutionary development.     

The ipsilateral cortico-cortical connexions between the cytoarchitectonic subdivisions of the primary somatic sensory cortex in the monkey

The callosal connexions of the primary somatic sensory cortex, SI, of the monkey have been studied with axonal degeneration methods after the placement of lesions of varying size in the cortex of one hemisphere and after section of the corpus callosum. For the correlation of the distribution of the degeneration with the cytoarchitectonic subdivisions of SI and with their boundaries, planar reconstructions of the extents of the subdivisions and of area 5 were made. The extent of area 5 is surprisingly large, being about the same as SI, and area 3a can be recognized as a distinct subdivision along the entire mediolateral extent of SI. The callosal fibres end in narrow, irregular bands aligned in the medio-lateral dimension and there are accentuations at the boundaries of the cytoarchitectural subdivisions. In the representations of the trunk and face, the bands of degeneration are present across the entire antero-posterior extent of SI and with increases at the boundaries, while in the limb regions the degeneration becomes restricted to the boundaries. It is suggested that the callosal connexions of the somatic sensory cortex, like those in the visual and auditory areas, are connecting those parts of the cortex in the two hemispheres that are concurrently activated by a peripheral stimulus. The parts of SI that are devoid of callosal connexions are related to the distal limbs. The callosal connexions are homo- and heterotopical; an architectonic subdivision within the callosally connected regions projects to the same and other architectonic subdivisions at the same medio-lateral level in the opposite hemisphere; the cortex containing the representation of the caudal trunk near the post-central dimple is connected with the same region in the other hemisphere and with that of the separate representation of the caudal trunk in the posterior part of the cingulate sulcus, while the representation of the occipital region at the post-central dimple is connected both with the homotopical site in the other hemisphere and with the other representation of this part of the periphery at the level of the lower end of the intraparietal sulcus.     

The organization of the rat motor cortex: A microstimulation mapping study

In conclusion, the rat primary motor cortex appears to be organized into irregularly shaped patches of cortex devoted to particular movements. The location of major subdivisions such as the forelimb or hindlimb areas is somatotopic and is consistent from animal to animal, but the internal organization of the pattern of movements represented within major subdivisions varies significantly between animals. The motor cortex includes both agranular primary motor cortex (AgL) and, in addition, a significant amount of the bordering granular somatic sensory cortex (Gr(SI)), as well as the rostral portion of the taste sensory insular or claustrocortex (Cl). The rat frontal cortex also contains a second, rostral motor representation of the forelimb, trunk and hindlimb, which is somatotopically organized and may be the rat's supplementary motor area. Both of these motor representations give rise to direct corticospinal projections^21,42,51,57, some of which may make monosynaptic connections with cervical enlargement motorneurons¹⁶. Medial to the primary motor cortex, in cytoarchitectonic field AgM, is what appears to be part of the rat's frontal eye fields, a region which also includes the vibrissae motor representation. The somatic motor cortical output organization pattern in the rat is remarkably similar to that seen in the primate, whose primary, supplementary and frontal eye field cortical motor regions have been extensively studied.     

Demonstration of synaptic input from prefrontal cortex to the habenula in the rat

Using lesion-degeneration techniques at the EM level, it is confirmed that a pathway from prefrontal cortex projects to the lateral habenula, and further that it makes synaptic contacts predominantly with dendrites of neurons in the medial sector of the lateral nucleus.The available neuroanatomical evidence points to a role for habenula as a regulator of the activity of the meso-cortical dopamine pathway by the interaction of this cortico-habenular pathway with a wide variety of limbic inputs in the medial sector of lateral habenula.