Distribution and morphology of callosal commissural neurons within the motor cortex of normal and reeler mice

Retrograde transport of horseradish peroxidase was used to examine the cells of origin of the callosal commissural fibers (CC neurons) in the primary motor cortex of normal and reeler mice. Quantitative analysis of the intracortical, laminar distribution, and dendritic orientation of CC neurons was performed in conjunction with qualitative observation of their morphology. For comparison, similar quantitative data were obtained for the cells of origin of the corticospinal tract (CST) of normal and reeler mice from materials described previously by Terashima et al. ('83). In the normal mouse, CC neurons are distributed in a bilaminar pattern such that the largest number of cells are located in supragranular layers II and III and in infragranular layer V. The majority of CC neurons are normal (upright) pyramids, although a few in the upper zone of layer VI are inverted pyramidal cells. In the reeler mutant, CC neurons are found in all cortical layers, but two-thirds are situated in the lower half of the cortex. On the basis of the celL shape and orientation of the apical dendrite, CC neurons of the reeler were classified into six morphological types: (1) typical pyramidal, (2) inverted pyramidal, (3) tumbled, (4) hook-shaped, (5) polymorphic, and (6) simple. The apical dendrites of the CC neurons in all layers of the cortex of the reeler mouse are randomly oriented; no direct relationship between the intracortical position of the soma and orientation of the apical dendrite was found. In contrast, CST neurons in the reeler mutant are concentrated in the outer third of the cortex, and there is a relationship between the laminar distribution of these cells and the alignment of their dendrites with respect to the pial surface: the apical dendrites of CST neurons lie in superficial layers tend to be oriented obliquely, whereas those of CST neurons in the deeper of cortex most often are oriented vertically, i.e, toward the pial surface. Quantitative analysis revealed that the relative intracortical positions of CC and CST neurons are reversed in the reeler mutant although both populations exhibited greater laminar disposition, and as a consequence, there is more intermingling of the two cell groups in the reeler than in the normal mouse. Thus, the present study suggests that the normal cytoarchitectonics of the primary motor cortex are inverted in the reeler mutant mouse.     

The intracortical position of pyramidal tract neurons in the motor cortex of the reeler changes from postnatal day 10 to adulthood.

To determine whether or not the intracortical distribution pattern of pyramidal tract (PT) neurons in the motor cortex (hindlimb area) of normal and reeler mutant mice changes during early postnatal development of the cortex, we injected HRP into the pyramidal decussation of postnatal day (P) 8 and adult animals of the normal and reeler strains, and killed the animals 2 days later. In the normal P10 and adult mice, such an injection resulted in a band of labelled neurons confined to the layer of large pyramids (LP), suggesting that the intracortical localization of PT neurons does not change from P10 to adulthood in the normal strain. In the P10 and adult reeler mice, labelled PT neurons were scattered radially from the deepest zone to the superficial zone of the motor cortex. However, while the HRP-labelled PT neurons are located bilaminarly in both the deepest zone and the superficial zone of the motor cortex of the P10 reeler mouse, the majority of PT neurons were found in the upper third of the motor cortex of the adult reeler mouse. Thus, the intracortical distribution pattern of PT neurons of the reeler mouse changes during the postnatal period.     

Organization of Pyramidal Cell Apical Dendrites and Composition of Dendritic Clusters in the Mouse: Emphasis on Primary Motor Cortex

It has been proposed that neurons in sensory cortices are organized into modules that centre on clusters of apical dendrites belonging to layer V pyramidal neurons. In the present study, sections reacted for microtubule-associated protein (MAPP) were examined in order to determine the three-dimensional inter-relationships of pyramidal cell dendrites in mouse primary motor cortex (MsI) cortex. Results indicate that pyramidal cell dendrites in MsI cortex can be interpreted to be arranged in a modular fashion, and that these modules are organized similarly to those in the sensory areas of the cortex. Also included in the present study are experiments designed to determine if the clusters of apical dendrites, around which the modules are centred, are composed of dendrites belonging to one or to more than one type of projection cell. Callosal neurons in MsI cortex were labelled by the retrograde transport of horseradish peroxidase deposited onto severed callosal fibres in the contralateral hemisphere. Examination of tangential thin sections through layer IV of MsI cortex shows clusters of apical dendrites in which every dendrite is labelled with horseradish peroxidase. Adjacent clusters are composed of unlabelled dendrites, suggesting that the apical dendrites of callosal neurons aggregate to form clusters that are composed exclusively of dendrites belonging to this type of projection cell. These findings suggest a hitherto unsuspected degree of specificity in the cellular composition of cortical modules.     

Cells of origin and topographic organization of corticospinal neurons in the guinea pig by the retrograde HRP method

In the guinea pig, horseradish peroxidase (HRP) was injected in the cervical or lumbar enlargements of the spinal cord in order to examine the origin and topographical organization of corticospinal (CS) neurons. The cortex was divided into granular and agranular regions to attempt correlations with the location of labeled CS neurons. These, of all sizes, are found only in pyramidal layer V of both kinds of cortical regions and could be seen as single, grouped or organized in clusters of 3-5 or more cells. The soma diameters of HRP CS neurons ranged from 13 to 49 microns. The largest labeled cells were present in the medial part of both agranular and granular cortices and included the giant pyramidal neurons which were found only in the rostromedial agranular cortex. The predominating intermediate-size cells were mostly present in lateral granular areas. The smaller cells were distributed in rostrolateral agranular and caudal granular regions. Morphological evidence suggests that an aggregation in clusters of labeled neurons and the different groups of CS neurons identified in different cortical areas may have additional differences in organization with respect to their precise topographical relations and functional properties. Neurons projecting to cervical levels were more abundant compared to projecting to the lumbar spinal cord, being distributed widely more lateral and anterior on the hemisphere to cortical neurons projecting to lumbar enlargement. These were located only medially on the dorsal surface of the brain. The two groups of labeled neurons were distributed in both agranular and granular cortices and occur adjacent to each other with only a narrow strip of overlap. The findings i.e. laminar organization, the pyramidal shape of CS neurons of all sizes, the presence of clusters and the somatotopic distribution of CS neurons seen in the guinea pig are discussed in comparison with equivalent findings on the corticospinal tract (CST) system of other species. The presence of labeled neurons within the cortex on both sides following unilateral HRP injections in the spinal cord points to a bilateral origin of the CST. This finding is discussed on the basis of the present study and previous autoradiographic findings.     

Morphology of layer V pyramidal neurons in the cat somatosensory cortex: an intracellular HRP study

Pyramidal tract (PT) or corticopontine neurons of the cat somatosensory cortex (SI) were identified with antidromic activation on stimulation of the bulbar pyramid or pontine nuclei (PN) and stained intracellularly with HRP after examining the electrophysiological properties. Comparison of the conduction velocity of the stem axons and the soma-dendritic morphology revealed that in the cat SI, there exists two types of layer V pyramidal neurons, i.e. one has smooth apical dendrites with larger soma (51.6 +/- 9.5 x 22.7 +/- 2.8 micron) and the other has richly spinous apical dendrites with smaller soma (34.0 +/- 8.8 x 15.3 +/- 3.3 micron). The former group responded antidromically at latencies shorter than 1 ms by PT stimulation or 1.5 ms by PN stimulation, respectively. These values were consistent with the borderline latencies between two similar groups of layer V pyramidal neurons in the motor (fast and slow PTNs) and parietal (aspiny and spiny layer V corticopontine neurons) cortices in the cat.     

Differences in the laminar origin of projections from the medial prefrontal cortex to the nucleus accumbens shell and core regions in the rat

The medial prefrontal cortex (mPFC) projects to the nucleus accumbens shell, core and rostral pole. In this retrograde tract-tracing study of rat mPFC to nucleus accumbens projection neurons, the advantages of Neurobiotin are utilised in order to reveal the detailed morphology of labelled projection cells, and to permit an examination of the laminar projections to shell and core compartments The retrogradely transported Neurobiotin was found in somata, proximal and distal dendrites of neurons that project from the mPFC to the nucleus accumbens. The morphology of these projection neurons was revealed in great detail and confirmed that the projection arises wholly from pyramidal cells. Interestingly, it was also found that retrogradely labelled neurons were exclusively located in prelimbic and infralimbic regions in layers V and VI, after shell injections, but also in layer II following core sites. This observation may reflect possibly different roles for cortical laminae on the nucleus accumbens.     

Target-specific differences in somatodendritic morphology of layer V pyramidal neurons in rat motor cortex

Dendritic geometry has been shown to be a critical determinant of information processing and neuronal computation. However, it is not known whether cortical projection neurons that target different subcortical nuclei have distinct dendritic morphologies. In this study, fast blue retrograde tracing in combination with intracellular Lucifer yellow injection and diaminobenzidine (DAB) photoconversion in fixed slices was used to study the morphological features of corticospinal, corticostriatal, and corticothalamic neurons in layer V of rat motor cortex. Marked differences in the distribution of soma, somal size, and dendritic profiles were found among the three groups of pyramidal neurons. Corticospinal neurons were large, were located in deep layer V, and had the most expansive dendritic fields. The apical dendrites of corticospinal pyramidal neurons were thick, spiny, and branched. In contrast, nearly all corticostriatal neurons were small cells located in superficial layer V. Their apical dendritic shafts were significantly more slender, though spiny like those of corticospinal neurons. Corticothalamic neurons, which were located in superficial layer V and in layer VI, had small or medium-sized soma, slender apical dendritic shafts, and dendrites that were largely spine free. This study indicates that, in layer V of rat motor cortex, each population of projection neurons has a unique somatodendritic morphology and suggests that distinct modes of cortical information processing are operative in corticospinal, corticostriatal, and corticothalamic neurons.     

The Convergence of Axon Terminals from the Mediodorsal Thalamic Nucleus and Ventral Tegmental Area on Pyramidal Cells in Layer V of the Rat Prelimbic Cortex

We investigated the ultrastructural basis of the synaptic convergence of afferent fibres from the mediodorsal thalamic nucleus (MD) and the ventral tegmental area (VTA) on the prefrontal cortical neurons of the rat by examining the synaptic relationships between thalamocortical or tegmentocortical terminals labelled with anterograde markers [lesion-induced degeneration or transport of wheat germ agglutinin conjugated to horseradish peroxidase (WGA—HRP)] and randomly selected unlabelled apical dendrites of layer V pyramidal cells in the prelimbic cortex. WGA—HRP-labelled terminals from the VTA ranged in diameter from 0.7 to 2.8 μm and established synaptic contacts with large dendritic profiles, i.e. proximal segments of apical dendritic shafts and spines from layer V pyramidal cells. Symmetrical synapses, i.e. inhibitory synapses, were more often seen than asymmetrical ones. Degenerating terminals from the MD formed asymmetrical synapses on dendritic spines or occasionally on small dendritic shafts of apical dendrites from layer V pyramidal cells, which received tegmentocortical synapses, mostly within layer III. Thalamocortical synapses were more distally distributed over common apical dendrites than tegmentocortical synapses, although some of them overlapped. The numerical density of direct synaptic inputs from the MD and VTA was low. These results suggest that fibres from the VTA exert their inhibitory effects directly on pyramidal cells in layer V via synaptic junctions with apical dendrites of these pyramidal cells, and that the tegmentocortical fibres are in an ideal anatomical position to modulate the reverberatory circuits between the MD and the prelimbic cortex.     

Evidence for interlaminar inhibitory circuits in the striate cortex of the cat

An interlaminar, ascending, and GABAergic projection is demonstrated in the striate cortex of the cat. We have examined a basket cell, with soma and smooth dendrites in layers V and VI, that was injected intracellularly with HRP in the kitten. Three-dimensional reconstruction of its axon revealed a horizontal plexus in layer V and upper VI, extending about 1.8 mm anteroposteriorly and 0.8 mm mediolaterally; a dense termination in the vicinity of the soma in layers V and VI; and an ascending tuft terminating in layers II and III in register above the soma and about 250 microns in diameter. Many boutons of this cell contacted neuronal somata and apical dendrites of pyramidal cells and subsequent electron microscopy showed that these boutons formed type II synaptic contacts with these structures. A random sample of postsynaptic targets (n = 199) in layers III, V, and VI showed that somata (20.1%), dendritic shafts (38.2%), and dendritic spines (41.2%) were contacted. The fine structural characteristics of postsynaptic elements indicated that the majority originated from pyramidal cells. Direct identification of postsynaptic neurons was achieved by Golgi impregnation of four large pyramidal cells in layer V, which were contacted on their somata and apical dendrites by between three and 34 boutons of the HRP-filled basket cell. Layer IV neurons were not contacted. Golgi-impregnated neurons similar to the HRP-filled basket cell were also found in the deep layers. The axonal boutons of one of them were studied; it also formed type II synapses with somata and apical dendrites of pyramidal cells. Boutons of the HRP-filled neuron were shown to be GABA-immunoreactive by the immunogold method. This is direct evidence in favour of the GABAergic nature of deep layer basket cells with ascending projections. The existence of an ascending GABAergic pathway was also demonstrated by injecting [3H]GABA into layers II and III. The labelled amino acid was transported retrogradely by a subpopulation of GABA-immunoreactive cells in layers V and VI, in addition to cells around the injection site. The axonal pattern and mode of termination of deep basket cells make them a candidate for producing or enhancing directional selectivity, a characteristic of layer V cells.(ABSTRACT TRUNCATED AT 400 WORDS)     

Local circuitry of identified projection neurons in cat visual cortex brain slices

The relationship between pyramidal cell morphology and efferent target was investigated in layer 6 of cat primary visual cortex (area 17). Layer 6 has 2 projections, one to the lateral geniculate nucleus (LGN) and another to the visual claustrum. The cells of origin of each projection were identified by retrograde transport of fluorescent latex microspheres. The labeled cells were visualized in brain slices prepared from area 17, using an epifluorescence compound microscope modified for intracellular recording. Individual retrogradely labeled cells were penetrated and intracellularly stained with Lucifer yellow to visualize the patterns of axons and dendrites associated with each projection. The neurons that give rise to the 2 projections had very different patterns of dendrites and local axonal collaterals, but the patterns within each group were highly stereotyped. The differences between their axonal collaterals were particularly dramatic. Claustrum projecting cells had fine, horizontally directed collaterals that arborized exclusively in layer 6 and lower layer 5. Most LGN projecting cells had virtually no horizontal arborization in layer 6. Instead, they sent widespread collaterals vertically, which arborized extensively in layer 4. The apical dendrites of the 2 groups also differed markedly. Claustrum projecting cells had apical dendrites reaching to layer 1, with branches in layer 5 only, while LGN projecting cells never had an apical dendrite reaching higher than layer 3, with side branches in layers 5 and 4. Therefore, each efferent target must receive inputs from neurons whose synaptic connections within area 17 are significantly different from those of neurons projecting to other targets. This further suggests that distinct visual response properties should be associated with each projection. In addition to the claustrum and LGN projecting cells, about 20% of layer 6 pyramidal neurons lacked an efferent axon. Morphologically, most resembled LGN projecting neurons, but a few had characteristics of claustrum projecting cells. These neurons may represent cells that either failed to make an efferent connection or cells that lost an efferent axon during development. Their frequency suggests that such intrinsic, presumably excitatory, neurons may play a significant role in cortical processing.     

Neuroanatomical study of somatomotor cortex in microcephalic mice induced by cytosine arabinoside

Somatomotor cortex of mice with microcephaly induced by DNA polymerase inhibitor cytosine arabinoside (Ara-C), has been studied with a modified Golgi-Cox staining and a HRP retrograde tracing method. Microcephalic mice were prepared by prenatal injections of cytosine arabinoside on days 13.5 and 14.5 of pregnancy. Cytoarchitectonically, the cerebral cortices of adult microcephalic mice are characterized by atypical pyramidal cells with abnormal dendrites and irregular patterns of cellular lamination. Semiquantitative analyses of the abnormality of dendrites in Golgi-Cox preparation indicate that both the degree and direction of ramification are severely affected in Ara-C treated mice. In adult control cerebrum, original neurons of corticospinal tract labeled after HRP injection into the lumbar cord were situated in layer V. In the microcephalic brains, however, HRP labeled neurons, some of which had abnormal polarity, were scattered throughout all layers. This HRP study for corticospinal tract neurons also confirms the irregular pattern of the cortex in which only three layers are recognized.     

Immunohistochemical analysis of the development of somatostatin in the reeler neocortex

The development of somatostatin (SS) neurons and fibers has been examined in the dorsolateral cortex of the mouse mutant reeler. Immunohistochemistry was performed using antisera directed primarily against SS28 or SS28(1-12). In the normal mouse at postnatal day 5 (P5), somatostatin (SS) neurons are concentrated in the ventral half of the cortex, in the developing layers V and VI. In the reeler mutant, SS neurons are scattered throughout the radial extent of the cortex, being concentrated in the dorsal half of the cortex in the polymorphic and large pyramidal cell layers. By P20, when the adult pattern of SS neuron distribution is evident in the normal mouse cortex, the distribution of similar neurons in the reeler appears inverted: immunoreactive neurons are concentrated in the dorsal half of the cortex. Immunoreactive fiber distribution follows a developmental pattern similar to that observed for SS neurons. At P5, SS fibers are most dense in layer I and V-VI of the normal cortex, while in the reeler, fibers are predominant in the polymorphic and large pyramidal cell layers. By P10, many fewer immunoreactive fibers can be detected in either normal or reeler mice than at P5. Nevertheless, while SS fibers in the normal cortex are most dense in layers I and V-VI, the reeler cortex exhibits little laminar heterogeneity in the distribution of these fibers. Thus, the SS fiber distribution appears less organized in the reeler cortex. These results suggest that whatever the nature of the genetic alteration resulting in cortical cellular developmental malposition in the reeler, SS cells and fibers, representing a completely intrinsic neocortical cellular system, behave as do all other cortical elements.     

Layer-specific immunocytochemical localization of GABA(B)R1a and GABA(B)R1b receptors in the rat piriform cortex

A peculiar, layer-segregated immunoreactive distribution of GABABR1a and GABABR1b receptor antibodies is present in the piriform cortex of adult rats. The GABABR1a antibody selectively marked the neuropile in layer Ia, where afferent olfactory fibres and intrinsic GABAergic (gamma-aminobutyric acid) axons terminate on the distal apical dendrites of pyramidal neurons. The GABABR1b antibody was detected in the soma and the large basal dendrites of layer II and III neurons. The pattern of distribution observed supports the hypothesis that (presynaptic) GABABR1a receptors in the superficial molecular layer modulate neurotransmitter release in a feedforward synaptic circuit, whereas GABABR1b (postsynaptic) receptors mediate feedback inhibitory potentials on principal cells.     

Maturation of rat visual cortex. III. Postnatal morphogenesis and synaptogenesis of local circuit neurons

The postnatal development of 3 types of local circuit neurons in rat visual cortex was examined in Golgi and electron microscopic preparations. During the first postnatal week, smooth and sparsely spinous stellate, bitufted and bipolar neurons were identified in Golgi material by their characteristic dendritic arborizations. Morphological differentiation begins during this week, as each neuron sprouts dendrites which extend, branch and produce spines, and ends by day 21. This differentiation was traced by quantifying the somatic area and number of primary dendrites on stellate, bitufted and bipolar neurons in layer II/III or layer V. Neurons in deep cortex differentiate earlier than those in superficial laminae. On day 3, axons are evident as short, straight processes, however, by day 6, many axons have branches and varicosities. The increase in the complexity of the axonal trees continues during the second and third postnatal weeks. Since the axons of stellate and bitufted neurons form synapses with the somata of pyramidal neurons, an index of the synaptogenesis of these neurons was traced by counting the numbers of synapses on the somata of pyramidal neurons. The mean number of axosomatic synapses increases steadily from day 3 to day 30. Layer V pyramidal neurons form axosomatic synapses before pyramidal neurons in layer II/III. In conclusion, the morphology of local circuit neurons develops during the period after they migrate into cortex. The principle that cortical local circuit neurons develop after projection neurons only applies for the synaptogenesis of the axon, but not for the maturation of the cell body and dendrites.     

Somatodendritic minicolumns of output neurons in the rat visual cortex

The apical dendrites of the pyramidal neurons of the cerebral cortex form radial bundles in all species and areas. Using microtubule-associated protein (MAP)2 immunostaining and Voronoi tessellation analysis in the rat visual cortex, we obtained objective criteria to define dendritic bundles in tangential sections: in supragranular layers of the rat visual cortex we found bundles of 6-6.4 dendrites, at a density of 1929 bundles/mm(2) and a centre-to-centre distance of 27 micro m. Using lipophilic tracers to label different pyramidal cell populations, based on the same criteria as in MAP2-immunostained material, we found that in the rat visual cortex the bundles consist of neurons with specific targets. Neurons projecting to the ipsi- or contralateral cortex form bundles together and with neurons projecting to the striatum, but not with those projecting to the superior colliculus, dorsal division of the lateral geniculate nucleus or through the cerebral peduncle. The latter neurons form bundles with neurons projecting to the striatum. Thus, the cerebral cortex is organized in minicolumns of output neurons visible at the earliest ages studied (P3), which might have a higher probability of being interconnected than those outside.     

Transcallosal non-pyramidal cell projections from visual cortex in the cat

Non-pyramidal cells with transcallosal projections were identified in the area 17/18 border region of the cat by retrograde transport of horseradish peroxidase injected into border region of the opposite hemisphere. From several hundred neurons filled with a Golgi-like diaminobenzidine (DAB) reaction product, seven cells were identified by their radially oriented smooth dendrites as possible non-pyramidal cells. Following thin-sectioning and examination with the electron microscope, four of the neurons proved to be layer IV spiny stellate cells with incompletely filled dendritic spines, and two proved to be layer III pyramidal cells with an incompletely labelled apical dendrite and dendritic spines. The remaining neuron was a non-pyramidal cell whose essentially smooth dendrites were covered with synapses, and whose cell body formed both symmetric and asymmetric synapses with presynaptic terminals. To better assess how many non-pyramidal cells might be labelled, thin sections of the area 17/18 border were surveyed using material processed with tetramethylbenzidine (TMB), and another five labelled non-pyramidal cells with transcallosal projections were identified by the needle-like crystals of TMB reaction product they contained. During the study it became evident that both the DAB and TMB reaction products in the lightly labelled neurons tended to be associated with granules that are 0.5 microns or larger in diameter and that had the characteristics of lysosomes. These granules are also visible in the light microscope as dark puncta. The numbers of puncta in profiles of pyramidal and of non-pyramidal cells in layers II/III and IVa of the area 17/18 border region and in the control acallosal region of area 17 were counted and compared. These comparisons revealed that labelled transcallosally projecting non-pyramidal cells may constitute 10-32% of the non-pyramidal cell population at the area 17/18 border region. Similar values were also obtained for pyramidal cells in this region. Consequently, it is concluded that significant numbers of non-pyramidal cells have axons that project through the corpus callosum to the contralateral hemisphere.     

Morphologic characteristics of the population of neurons in the cat auditory cortex, projecting into the medial geniculate body

Neurons from the auditory cortex projecting into the medial geniculate body were studied in cats using the horseradish peroxidase. Such neurons were located in deep layers of the auditory cortex--predominantly in layer VI, and to a lesser extent in layer V. Dimensions of the pericarions of the labelled neurons were measured and types of neurons were determined. The overwhelming majority of cortico-geniculate neurons was pyramidal, and quantity of such neurons in layer VI of the first auditory cortex may reach 60% of the total number of cells in this layer. On the basis of the anterograde transport of HRP deep layer III and layer IV of the auditory cortex were determined as main targets of geniculocortical fibres.     

The postnatal development of layer VI pyramidal neurons in the cat's striate cortex, as visualized by intracellular Lucifer yellow injections in aldehyde-fixed tissue

The postnatal development of layer VI pyramidal neurons in the cat's striate cortex has been studied by means of intracellular injections of Lucifer yellow in aldehyde-fixed tissue (LYF technique). It is shown that the LYF technique gives results qualitatively and quantitatively similar to results obtained with other techniques (Golgi, marker-injections in viable tissue). Quantitative analysis demonstrated significant increases in soma diameter, number and length of basal dendrites, length of second order apical dendrites and, in particular, in number of spines/unit dendritic length, during the first postnatal month. Maturation of the basal dendritic tree and increase in number of spines continue in the second postnatal month. At later postnatal times soma diameter and number of spines decrease by about 20%. Dendritic varicosities are most frequent during the first postnatal week, and decrease in number steadily from thereon. The late maturation of layer VI pyramidal neurons suggests that these cells might be affected by early peripheral lesions and/or sensory deprivation to which the striate cortex of the cat has been shown to be most susceptible around the end of the first postnatal month.